Digital PDFs

EK-RK067-UG-001

February 1978

151 pages

Original

7.2MB

Document:	RK06/RK07 Disk Drive User's Guide
Order Number:	EK-RK067-UG
Revision:	001
Pages:	151
Original Filename:

OCR Text

EK-RK067-UG-001

RK06/RK07
Disk Drive
User's Manual

digital equipment corporation • maynard, massachusetts

First Edition, February 1978

Copyright © 1978 by Digital Equipment Corporation
The material in this manual is for informational
purposes and is subject to change without notice.
Digital Equipment Corporation assumes no responsibility for any errors which may appear in this
manual.
Printed

mU.S.A.

This document was set on DIGITAL's DECset-8000
computerized typesetting system.

The following are trademarks of Digital Equipment
Corporation, Maynard, Massachusetts:
DEC
DECCOMM
DECsystem-IO
DECSYSTEM-20

DEC tape
DECUS
DIGITAL
MASSBUS

PDP
RSTS
TYPESET-8
TYPESET-II
UNIBUS

CONTENTS
Page
CHAPTER 1
1.1
1.2
1.2.1
1.2.2
1.2.3
1.2.4
1.3

INTRODUCTION
PU]~POSE AND SCOPE ..................................................................................... 1-1

GE]~ERAL INFORMATION ............................................................................ 1-2

1.4

Media ........................................................................................................... 1-2
Diagnostic and Error Detection/Correction Capabilities .............................. 1-3
Maintenance Features .................................................................................. 1-3
Reference Documentation ............................................................................ 1-4
RK611/RK06 AND RK611/RK07 DISK SUBSYSTEM PERFORMANCE
SPE,CIFICATIONS ............................................................................................. 1-4
OPlrIONS ............................................................................................................ 1-4

CHAPTER 2

SUnSYSTEM DESCRIPTION

2.1
2.2
2.3
2.4
2.4.1
2.4.2
2.4.3
2.4.4
2.5
2.5.1
2.5.2
2.5.3
2.5.4
2.5.5
2.5.6
2.5.6.1
2.5.6.2

INl'RODUCTION .............................................................................................. 2-1
RK06 AND RK07 DISK DRIVES ...................................................................... 2-1
RK06K AND RK07K DISK CARTRIDGES ..................................................... 2-3
RK611 DISK DRIVE CONTROLLER ............................................................... 2-3
Command Control ....................................................................................... 2-3
Data Control ................................................................................................ 2-3
Controller Registers ..................................................................................... 2-3
Silo Memory ................................................................................................ 2-7
RK611/RK06 AND RK611/RK07 SUBSYSTEMS ............................................ 2-8
Controller/Drive Interface ............................................................................ 2-8
Subsystem Commands ................................................................................ 2-12
Data Formatting ........................................................................................ 2-12
Servo Surface Operations ............................................................................ 2-13
Data Error Control. .................................................................................... 2-13
Bad Sector File ........................................................................................... 2-14
Data Field Format .............................................................................. 2-15
Track Format ..................................................................................... 2-15

CHAPTER 3

SITE CONSIDERATIONS AND PREPARATIONS

3.1
3.1.1
3.1.2
3.1.3
3.1.4

ENVIRONMENTAL CONSIDERATIONS ....................................................... 3-1
General ........................................................................................................ 3-1
Temperature ................................................................................................. 3-1
Relative Humidity ........................................................................................ 3-1
Heat Dissipation .......................................................................................... 3-1

III

CQNTENTS (CONT)
Page
3.1.5
3.1.6
3.1.7
3.1.8
3.1.9
3.2
3.2.1
3.2.2
3.2.3
3.2.4
3.2.5
3.2.6

Acoustics ....................................................................................................... 3-1
Altitude ........................................................................................................ 3-1
Radiated Emissions ...................................................................................... 3-2
Vibration/Mechanical Shock/ Attitude ......................................................... 3-2
Cleanliness ................................................................................................... 3-2
SITE PREPARATION ........................................................................................ 3-5
Space ............................................................................................................ 3-5
Cabling ........................................................................................................ 3-5
Power Requirements .................................................................................... 3-5
Floor Loading .............................................................................................. 3-5
Installation Constraints ................................................................................ 3-5
Fire and Safety Precautions .......................................................................... 3-5

CHAPTER 4

INST ALLATION

4.1
4.2
4.3
4.3.1
4.3.2
4.4
4.4.1
4.4.2
4.4.3
4.4.4
4.4.4.1
4.4.4.2
4.4.4.3
4.4.5
4.4.5.1
4.4.5.2
4.4.5.3
4.5
4.6
4.6.1
4.6.2
4.6.2
4.6.2.1
4.6.2.2
4.6.2.3

UNPACKING AND INSPECTION .................................................................. .4-1
SAFETY PRECAUTIONS ................................................................................. 4-6
INSTALLATION PROCEDURE ....................................................................... 4-6
RK611 Controller ........................................................................................ 4-6
RK06 or RK07 Disk Drive ......................................................................... 4-1 0
INSTALLATION CHECKS AND ADJUSTMENTS ...................................... .4-15
Controller Power Supply Voltage Checks .................................................... 4-16
Disk Drive Power Supply Voltage Checks .................................................. .4-16
Head Load/Unload Checks ....................................................................... .4-16
Head Alignment Checks and Cartridge Care ............................................... 4-17
Head Alignment Using Program Control ........................................... .4-18
Sample Printout .................................................................................. 4-21
Head Alignment Using the Field Text Box ........................................ ,.4-23
Response to Abnormal Conditions ............................................................ .4-24
Clearing an Unsafe Condition ............................................................ .4-24
Removing a Cartridge When Drive Power is Lost ............................... 4-25
Detecting Head-to-Disk Interference .................................................. 4-25
POWER CONVERSION ................................................................................... 4-25
DUAL ACCESS DRIVE INSTALLATION AND ACCEPTANCE ................ .4-26
Dual Access Installation Procedure ............................................................ .4-26
Dual Access Acceptance Procedure ............................................................ .4-26
Dual Access Acceptance Procedure ............................................................ .4-26
Dual Access Diagnostic ...................................................................... 4-26
RK611, RK06, and RK07 Diagnostics ............................................... .4-26
RK06 and RK07 Performance Exerciser ............................................ .4-26

CONTENTS (CONT)
Page
CHAPTER 5

FIE]LD ACCEPTANCE TEST PROCEDURE

5.1
5.1.1
5.1.2
5.1.3
5.1.4
5.1.5
5.2
5.2.1
5.2.1.1
5.2.1.2
5.2.2
5.2.2.1
5.2.2.2
5.2.3
5.2.4
5.2.4.1
5.2.4.2

ERROR DEFINITIONS/RATES ....................................................................... 5-1
Control Errors .............................................................................................. 5-1
Hard Errors .................................................................................................. 5-3
Soft Errors ................................................................................................... 5-3
Cartridge Errors ........................................................................................... 5-3
Seek Errors ................................................................................................... 5-3
DIAGNOSTIC TESTING ................................................................................... 5-3
RK611 Controller ........................................................................................ 5-3
General ................................................................................................ 5-3
Procedure ............................................................................................. 5-7
RK06 or RK07 Disk Drive ........................................................................... 5-8
General ................................................................................................ 5-8
Procedure ............................................................................................. 5-8
Cartridge Interchangeability Test ................................................................. 5-9
Data Reliability Testing ................................................................................. 5-9
General ..................... ,........................................................................... 5-9
Procedure .................. ".......................................................................... 5-9

CHAPTER 6

OPE:RATOR'S GUIDE

6.1
6.2
6.2.1
6.2.2
6.2.3
6.2.4
6.2.5
6.2.6
6.2.6.1
6.2.6.2
6.2.7
6.2.7.1
6.2.7.2
6.3
6.3.1
6.3.2

INT:RODUCTION ................... ".......................................................................... 6-1
OPERATOR CONTROLS/INDICATORS ........................................................ 6-1
Power ON/OFF Circuit Breaker .................................................................. 6-1
RUN/STOP Switch with STOP Indicator .................................................... 6-1
UNIT SELECT Switch with READY Indicator ........................................... 6-3
FAULT Indicator ..............'.......................................................................... 6-3
WRITE PROT Switch with PROTECT Indicator ......................................... 6-3
ACCESS A ENABLE Switch with SELECT Indicator ............................... 6-3
Single-Access Drive (A) ........................................................................ 6-3
Dual-Access Drive (A and B) ................................................................ 6-4
ACCESS B ENABLE Switch with SELECT Indicator .......................... 6-4
Single-Access Drive (B) ........................................................................6-4
Dual-Access Drive (A and B) ................................................................ 6-4
OPERATING PROCEDURES ........................................................................... 6-4
RK06K and RK07K Cartridge Loading ....................................................... 6-4
RK06K and RK07 Cartridge Unloading ..................................................... 6-12

CONTENTS (CONT)
Page
CHAPTER 7

PROGRAMMER'S GUIDE

7.1
7.2
7.2.1
7.2.2
7.2.3
7.2.4
7.2.5
7.2.6
7.2.7
7.2.8
7.2.9
7.2.10
7.2.11
7.2.12
7.2.13
7.2.14
7.2.15
7.3
7.4
7.4.1
7.4.2
7.5
7.5.1
7.5.2
7.6
7.6.1
7.6.2
7.6.3
7.7

INTRODUCTION .............................................................................................. 7-1
RK611 DEVICE REGISTERS ............................................................................ 7-2
Control/Status Register 1 (RKCS 1) ............................................................. 7-2
Word Count Register (RKWC) .................................................................... 7-6
Bus Address Register (RKBA) ...................................................................... 7-6
Disk Address Register (RKDA) ................................................................... 7-6
Control/Status Register 2 (RKCS2) ............................................................. 7-8
Drive Status Register (RKDS) .................................................................... 7-10
Drive Error Register (RKER) ..................................................................... 7-12
Attention Summary/Offset Register (RKAS/OF) ...................................... 7-15
Desired Cylinder Register (RKDC) ............................................................ 7-16
Data Buffer Register (RKDB) .......................... :......................................... 7-17
Maintenance Register (RKMR1) ................................................................ 7-17
ECC Position Register (RKECPS) .............................................................. 7-19
ECC Pattern Register (RKECPT) ............................................................... 7-19
Maintenance Register 2 (RKMR2) ............................................................. 7-20
Maintenance Register 3 (RKMR3) ............................................................. 7-20
RK611 DISK COMMANDS ............................................................................. 7-21
DIAGNOSTIC MODE (DMD) ......................................................................... 7-29
Read Simulation ......................................................................................... 7-32
Write Simulation ........................................................................................ 7-33
ERROR DETECTION AND CORRECTION .................................................. 7-34
Correctable Error (DCK, CERR) ............................................................... 7-36
Uncorrectable Error (ECH, DCK, CERR) ................................................. 7-37
PROG RAMMING EXAMPLES ...................................................................... 7-38
RK06 Device Driver Routine ...................................................................... 7-38
ECC Correction Routine' ............................................................................ 7-41
Maintenance Register Read Routine ........................................................... 7-43
PROGRAMMING CONSIDERATIONS ........................................................ 7-43

CHAPTER 8

CUSTOMER MAINTENANCE GUIDE

8.1
8.2
8.3
8.4
8.5

INTRODUCTION .............................................................................................. 8-1
PROFESSIONAL CARTRIDGE CLEANING .................................................. 8-1
USER CARTRIDGE CLEANING ......... :........................................................... 8-1
SPINDLE ASSEMBLY CLEANING ................................................................. 8-1
CARTRIDGE CARE SUMMARY .................................................................... 8-3

CONTENTS (CONT)
Page
CHAPTER 9

RK06/RK07 DUAL-ACCESS OPTION

9.1

RK611/RK06 AND RK611/RK07 DUAL-ACCESS SYSTEM
C10NFIGURATION ........................................................................................... 9-1
DUAL-ACCESS CONTROLS ........................................................................... 9-2
Port A Mode ................................................................................................ 9-2
Port B Mode ................................................................................................. 9-2
Dual-Access Mode ....................................................................................... 9-2
DUAL-ACCESS LOGIC CIRCUITS .................................................................. 9-2
Arbitration Circuits ...................................................................................... 9-3
Timing Counters and Decoders .................................................................... 9-3
Auto-Release Timer ..................................................................................... 9-3
Multiplexers ................................................................................................. 9-3
DUAL-PORT OPERATION .............................................................................. 9-3
Seizing a Drive in the Neutral State ............................................................... 9-3
Attempting to Seize an Unavailable Drive .................................................... 9-3
Caution in Issuing a Release Command to an Unavailable Drive .................. 9-4
Auto-Release Timer ..................................................................................... 9-4
START/STOP AND WRITE PROTECT IN DUAL ACCESS ........................... 9-4
ERROR HANDLING IN DUAL ACCESS ...........................................·............. 9-5
Notifying the Controller of an Error ............................................................. 9-5
Clearing the Drive Errors and Resetting the Attention .................................. 9-5
MUltiple Drive Select and CTD Parity Errors in the Dual-Access
Configuration ................. "............................................................................. 9-5
I1\rITIALIZE IN DUAL ACCESS ....................................................................... 9-5
GETTING DRIVE STATUS IN DUAL-ACCESS MODE ................................. 9-6
Status Bits Pertinent to the Dual-Port Available at the Unibus Interface ....... 9-6

9.2
9.2.1
9.2.2
9.2.3
9.3
9.3.1
9.3.2
9.3.3
9.3.4
9.4
9.4.1
9.4.2
9.4.3
9.4.4
9.5
9.6
9.6.1
9.6.2
9.6.3
9.7
9.8
9.8.1

FIGURES
Figure No.

I-I
2-1
2-2
2-3
2-4
2-5
2-6
2-7

Title

Page

RK611 /RK06-RK611 /RK07 Dual Access System Configuration ....................... 1-2
H969 Standalone Disk Drive Cabinet ................................................................... 2-2
Read/Write Heads and Data Track Orientation ................................................... 2-4
Sector Formats ..................................................................................................... 2-5
Basic RK06 or RK07 Disk Subsystem .................................................................. 2-5
RJ.~611 Data Interface .......................................................................................... 2-7
Interface Lines ..................................................................................................... 2-8
Controller/Drive Serial, 16-Bit Message Formats ................................................. 2-9

Vll

FIGURES (CONT)

Figure No.

3-1
3-2

4-1
4-2

4-3
4-4
4-5
4-6
4-7
4-8
4-9
4-10
4-11
4-12

4-13
6-1
6-2
6-3
6-4
6-5
6-6
6-7
6-8
6-9
7-1
7-2
7-3
7-4
7-5
7-6
7-7
7-8
7-9
7-10

7-11
7-12
7-13
7-14

7-15
7-16
7-17
7-18
7-19
7-20

Page

Title

Relationship of Disk Head, Disk, and Contaminants ............................................ 3-3
RK06/RK07 Disk Drive Air Supply and Distribution Subsystem ......................... 3-4
RK06/RK07 Shipping Package Configuration .................................................... .4-3
RK06 Disk Drive Logic Module Locations .......................................................... 4-4
RK07 Disk Drive Logic Module Locations ......................................................... .4-5
RK611 Controller Backplane (from Module Side) ............................................... .4-8
RK611 I/O Connector Details ............................................................................. 4-9
Serial Tag Location and Model Identifications (Tag Visible with
Front Cover Removed) ....................................................................................... 4-11
RK06/RK07 I/O Connector Details ................................................................. .4-12
Dual-Access Disk Subsystem H969 Cabinet. ...................................................... .4-13
Dual-Access Disk Subsystem H969 Cabinet ....................................................... .4-14
RK06 Head Assembly Area ................................................................................ 4-18
RK07 Head Assembly Area ................................................................................ 4-19
RK611/RK06 or RK611/RK07 Dual Access Cable Diagram ............................ .4-27
RK611/RK06 or RK611/RK07 Cable Diagram for Testing Dual Access .......... .4-28
RK06 or RK07 Disk Drive Power and Control Switches ...................................... 6-2
Typical Disk Drive with Lid Released ................................................................... 6-5
Typical Disk Cartridge ......................................................................................... 6-6
Engaging Lock Release ........................................................................................ 6-7
Releasing Protection Cover .................................................................................. 6-8
Positioning for Load ............................................................................................ 6-9
Cartridge Locked in Shroud ............................................................................... 6-10
Installing Protection Cover ................................................................................. 6-11
Engaging Lock Release ....................................................................................... 6-12
RK611 Device Registers ....................................................................................... 7-3
Control/Status Register 1..................................................................................... 7-7
Word Count Register ........................................................................................... 7-7
B us Address Register ............................................................................................ 7-7
Disk Address Register .......................................................................................... 7-7
Control/Status Register 2 ..................................................................................... 7-8
Drive Status Register .......................................................................................... 7-10
Drive Error Register ........................................................................................... 7-13
Attention Summary/Offset Register ................................................................... 7-15
Desired Cylinder Register ................................................................................... 7-17
Data Buffer Register ........................................................................................... 7-17
Maintenance Register 1 ...................................................................................... 7-17
ECC Position Register ........................................................................................ 7-19
ECC Pattern Register ......................................................................................... 7-19
Maintenance Register 2 - Controller to Drive ..................................................... 7-20
Maintenance Register 2 - Drive to Controller ..................................................... 7-21
Maintenance Register 3 - Controller to Drive ..................................................... 7-22
Maintenance Register 3 - Drive to Controller ..................................................... 7-22
Select Drive Command ....................................................................................... 7-24
Pack Acknowledge Command .................................................................. ~ ......... 7-24

Vlll

FIGURES (CONT)
Figure No.
7-21
7-22
7-23
7-24
7-25
7-26
7-27
7-28
7-29
7-30
7-31
7-32
8-1
9-1
9-2
9-3

Page

Title

Drive Clear Command ........... "........................................................................... 7-24
Unload Command .................. "........................................................................... 7-25
Start Spindle Command ......... "........................................................................... 7-25
Re:calibrate Command ........................................................................................ 7-26
Offset Command ................................................................................................ 7-26
Offset Command Coding .................................................................................... 7-27
Sel~k Command .................................................................................................. 7-27
Data Handling Commands ................................................................................. 7-28
Maintenance Clock (MCLK) Timing .................................................................. 7-31
Error Correction Process ..............................................................................."..... 7-35
Example of 8-Bit Error Correction Process ......................................................... 7-36
Data Error Flow ................................................................................................. 7-37
Spindle Assembly ................................................................................................. 8-2
RK611/RK06 or RK611/RK07 Dual Access System Configuration .................... 9-1
RK06 or RK07 Disk Drive Power and Control Switches ...................................... 9-2
Drive Status Bits Contained in A and B Messages ................................................. 9-7

TABLES
Table No.
1-1

1-2
1-3
1-4
1-5
2-1
4-1
4-2
4-3
4-4
4-5
5-1
5-2
7-1
7-2

Title

Page

RK611 Controller Performance Specifications ..................................................... 1-5
RK06 Disk Drive Performance Specifications ..................................................... .1-6
RK07 Disk Drive Performance Specifications ...................................................... 1-8
RK611/RK06 Disk Subsystem Options .............................................................. l-ll
RK611/RK07 Disk Subsystem Options .............................................................. 1-12
RK611 Unibus-Addressable Registers .................................................................. 2-6
Spe:cial Tools and Test Equipment for RK611/RK06 or RK611/RK07
Disk Subsystem Installation ................................................................................. 4-1
RK611 Control Switch Configurations for Desired Unibus Address
and Interrupt Vector ............................................................................................ 4-7
Drive Bus Cables ................................................................................................ 4-1 0
Sample Table for Backplane Wiring ................................................................... .4-10
Unit-Select Plugs for RK06/RK07 Disk Drives ................................................. .4-15
RK611/RK06 and RK611/RK07 Disk Subsystem Diagnostic Errors .................. 5-1
Diagnostic Programs Used in RK611/RK06 or RK611/RK07 Disk Subsystem
Field Acceptance Tests ......................................................................................... 5-4
COlnbined Error (CERR) ..................................................................................... 7-5
Disk Commands ................................................................................................. 7"'23

CHAPTER 1
INTRODUCTION

1.1 PURPOSE AND SCOPE
This manual provide:s information related to the capabilities, installation, operation, programming,
and user maintenance requirements of the RK611/RK06 and RK611/RK07 Subsystems.
This manual is intended primarily for operating and programming personnel. Installation and service
should be performed only by qualified DIGITAL field engineering and maintenance personnel.
The information is arranged as follows:
• Chapter 1 provides a general description of the subsystems, including capabilities, standard
and optional features, and specifications.
• Chapter 2 provides separate drive, cartridge, and controller descriptions, concluding with a
subsystem ove:rview to familiarize the user with certain hardware and technical characteristics.
• Chapter 3 contains the details to be considered when selecting a location for installation of the
subsystems.
• Chapter 4 describes the installation process for the subsystems and the dual port option. (Installation should only be performed by qualified DIGITAL field personnel.)
• Chapter 5 describes the procedures necessary to demonstrate that the drive and controller are
suitable for user acceptance.
• Chapter 6 contains step-by-step procedures for the operation and control of the subsystems.
• Chapter 7 provides the programmer with general reference material, subsystem software requirements, and recommendations to facilitate the design of user programs.
• Chapter 8 provides the user with basic maintenance requirements and procedures.
• Chapter 9 contains a description of the subsystem dual port option, including control and
operation procedures for systems with this option.
This manual will have a much higher value to the reader if he is already familiar with PDP-II processors and peripherals.

1-]

1.2 GENERAL INFORMATION
•
The RK611 /RK06 Disk Drive subsystem and the RK611/RK07 Disk Drive Subsystem, manufactured by Digital Equipment Corporation, both consist of an RK611 Disk Controller and up to eight
RK06 and/or RK07 Disk Drives. Each RK06 or RK07 Disk Drive is a free standing, high performance, moving head device that is connected to the RK611 Controller via a daisy-chain bus arrangement. The RK611 Controller, which is located in associated cabinetry, provides an interface between
the subsystem and a PDP-II system via the Unibus. Both RK06 and RK07 drives can be used in one
subsystem.
Dual access operation is a hardware option available to the RK06 and RK07 drives enabling either
drive to be accessed through two different controllers. Figure 1-1 shows the system configuration for
dual access operation. The eight RK06s or RK07s shown on this figure are daisy chained. The two
RK611 Controllers can be attached to two different CPU s.

romoN - --,

I
I

~_ _ _I---,

RK06 OR
RK07 DISK
DRIVE

RK06 OR
RK!D7 DISK
DRIVE

RK06 OR
RK07 DISK
DRIVE

RK06/RK07 SUBSYSTEM CONFIGURATION (PDP-11)
11-4326

Figure 1-1

RK611/RK06-RK611/RK07 Dual Access System Configuration

1.2.1 Media
The storage medium for the RK06 Disk Drive is a top-loading, dual-platter, RK06K Disk Cartridge,
which can be freely interchanged between any of the available RK06 Drives. It has a maximum formatted storage capacity of 6.3 million 18-bit words or 6.9 million 16-bit words. The RK07 Disk Drive
uses a similar cartridge, the RK07K Disk Cartridge, as the storage medium. It also can be interchanged between any of the available RK07 Drives. It has a maximum formatted capacity of 12.6
million 18-bit words or 13.8 million 16-bit words. Note that an RK06K cartridge cannot be used on an
RK07 Drive, nor an RK07K cartridge on an RK06 Drive.
Each dual-platter pack provides three recording surfaces that are serviced by three read/write heads
and a servo surface that is monitored by a single read head. The recording surfaces provide three data
tracks per cylinder with a maximum of 411 cylinders per RK06 disk pack or 815 cylinders per RK07
disk pack. Each track can accommodate either 20 (l8-bit word) or 22 (l6-bit word) sectors, while each
sector can accommodate 256 data words. The servo surface provides positioning signals for the heads
and timing signals for data transfer synchronization.
1-2

The high-density re:cording capability (required to achieve maximum storage capacities) is provided by
the use of the modified frequency modulation (MFM) encoding technique. In addition, each RK06
and RK07 Disk Drive and the controller contain phase-locked oscillator (PLO) circuitry which maintains a constant bit density if a disk is subjected to minor speed variations.
For head positioning, the maximum seek time in the RK06 Disk Drive (411 cylinders) is 75 ms, while
the minimum (track-to-track for one cylinder) is 8 ms, with an average access time of 38 ms. The
maximum seek timle in the RK07 Drive (815 cylinders) is 71 ms, and the minimum seek time is 6.5 ms.
The average access time is 36.5 ms.
A data transfer rate of 4.3 J,tS per 16-bit word between the controller and the Unibus provides an

average transfer rate of 232,500 words per second.

1.2.2 Diagnostic and Error Detection/Correction Capabilities
Diagnostic capabilities are enhanced by an extensive status/error reporting network, consisting of nine
U nibus-addressable~ registers which monitor subsystem activities and are visible to the software.
For extended maintenance and serviceability, the RK611 /RK06 and RK611 /RK07 subsystems provide a software controlled diagnostic mode (DMD) of operation. When diagnostic mode is initiated,
the drive interface lines are effectively disconnected from the Controller while the Unibus interface
remains intact. With this arrangement, read or write commands can be initiated using simulated data
and control techniques, while actual transmissions between the drives and the controller are disabled.
In this manner, closed loop data transfers to and from main storage can verify controller operation in
relation to data control.
The RK611 Controller provides data error detection and correction logic, which utilizes an error
correction code (ECC) technique to detect and identify (for software correction) data error bursts up
to 11 bits in length and to detect (but not identify) error bursts longer than 11 bits. In addition and in
conjunction with ECC operation, each RK06 or RK07 Disk Drive has a head offset capability which
allows the recovery of data from a cartridge that may have been written by a drive with slightly
misaligned heads.
The data recovery fe~atures, operating in conjunction with the PLO clock system and the MFM recording technique, provide increased reading and recording reliability. However if a disk pack sector proves defective, in that hardware and software attempts to recover its data are unsuccessful, the
operating system can mark the sector bad so that future attempts to use it will be prevented.

1.2.3 Maintenance Features
The following standard features enhance the reliability, availability, and serviceability of the
RK611/RK06 and RK611/RK07 subsystems.
• The RK06 and RK07 drive interface is designed to allow a specific drive to be electrically
isolated fronl the controller for maintenance purposes. This allows an off-line tester to be
utilized without taking a drive off-line or cycling the system down. With this feature, software
can reconfigure the system for continued operation while maintenance is performed.
• Service access to an RK06 or RK07 Disk Drive is from the top, front, or rear of the drive and
not from the sides. Therefore, maintenance for a given drive in a -multi drive system can be
initiated without disabling intradrive cabling.
• The logic and power board cages can be extended, following removal of the rear panel of an
RK06 or RK.07 Disk Drive, to facilitate emergency maintenance.

1-3

• Mean-time-to-repair (MTTR) for emergency maintenance is improved by the inclusion of light
emitting diodes (LEDs) on the front edge of certain logic boards. The LEDs provide a service
engineer with an indication of the functionality of certain key logic areas.
• For the implementation of cartridge alignment procedures, a protection switch is available to
the service engineer to provide write protection for the cartridge and control of head velocity to
ensure alignment integrity.
• If an RK06 or RK07 Drive loses ac power, neither the drive nor cartridge will be harmed, no
data will be lost, and the format will remain intact.

• When dual-access RK06 or RK07 Disk Drives are installed, the logic allows dual-access functionality to be tested utilizing only one of the two available controllers. This feature provides
enhanced availability for multiprocessor systems by freeing one controller for normal dualaccess operations.
1.2.4

Reference Documentation

These documents supplement the RK06/RK07 User's Manual.
Number

Title

EK -RK067 -TD-OO 1
EK-RK611-TM-002
EK-RK067-SV -001
EK-RK06-IP-001
EK-RK07-IP-001

RK06/RK07 Disk Drive Technical Description Manual
RK611 Controller Technical Manual
RK06/RK07 Disk Subsystem Service Manual
RK06 Illustrated Parts Breakdown
RK07 Illustrated Parts Breakdown

1.3 RK611/RK06 AND RK611/RK07 DISK SUBSYSTEM PERFORMANCE SPECIFICATIONS
Tables 1-1,1-2, and 1-3 are performance specification summaries for the RK611 Controller, the RK06
Disk Drive, and the RK07 Disk Drive, respectively.
1.4 OPTIONS
Table 1-4 lists the several options currently available for the RK611 /RK06 Disk Subsystem. As indicated, each drive is contained in an H969 "dedicated" cabinet and contains all the necessary cabling,
as well as an RK06K-DC data cartridge. E models are single access and F models are dual access.
Similarly, E model RK611 Controllers are single units and F model RK611 Controllers consist of two
controllers. The RK611-C is a dual-access controller for the Disk Subsystem and a kit to convert an
RK06E to an RK06F.
Table 1-5 lists the several options currently available for the RK611 /RK07 Disk Subsystem. As indicated, each drive is contained in an H969 "dedicated" cabinet and contains all the necessary cabling,
as well as an RK07-DC data cartridge. E models are single access and F models are dual access.
Similarly, E model RK711 Controllers are single units and F model RK711 Controllers consist of two
controllers. The RK711-C is a dual-access controller for the Disk subsystem and a kit to convert an
RK07E to an RK07F.

1-4

Table 1-1

RK611 Controller Performance Specifications

Characteristics

Specifications

Required mounting space

Occupies one double system unit in an expansion
box

Power requirements

+5Vdc±5%@15A
+15 Vdc ± 5%@ 175 rnA
-15 Vdc ± 5% @400mA

Number of drives/controllers

8 (max)

No. Unibus-addressable registers

Device base address

777440 (octal) switch-selectable

Device interrupt vector

000210 (octal) switch-selectable

NOTE
The RK611 Device base address overlaps the LookAhead Register in the R C11 Disk subsystem, and the
RK611 Device Interrupt Vector overlaps the same
vector in the RC11 subsystem.
Interrupt priority

Level 5 (plug-selectable)

Small peripheral controller options

Up to three slots, one of which must be quad
height. Other two can be quad or hex height.

Data transfer rates:

22 Sector
(l6-bit
data words)

20 Sector
(l8-bit
data words)

Unibus to Controller
(average word rate)

4.3 ~s/word

4.6 ~s/word

Drive to Controller (burst rate)

3.7 ~s/word

4.2 ~s/word

Time to update driv(~ status

15 ~s (maximum)

Time to update driv(~ attention

7.4 ~s

Error detection/correction

32-bit ECC word/sector (detection)
II-bit ECC word/sector (correction)

Time for correction

Less than one revolution

Maintainability

Diagnostic mode places all controller timing and
data paths under software control.

Number of sectors/track

Software-programmable for 20 or 22 sector format.

1-5

Table 1-2

RK06 Disk Drive Performance Specifications

Characteristics

Specifications

Storage type
Medium
Disk diameter

Dual-platter magnetic cartridge, DEC RK06K-DC
355 mm (14 in.) nominal

Magnetic heads

3 read/write; 1 servo

Recording capacity (formatted)

18-bit Word

16-bit word

411
3
1233 .
20
256
18
4608
92,160
37.88 M
113.63 M
4040
192.3

411
3
1233
22
256
16
4096
90,112
37.04 M
111.11 M
4040
192.3

Cy lin ders / cartri dge
Tracks/ cylinder
Tracks / cartridge
Sectors/track
Words/sector
Bits/word
Bits/sector
Bits/track
Bits/surface
Bits/pack
Bits/inch (inner track)
Tracks/inch
Electrical
Voltage
Input Power
Single access drives

Dual access drives

Start current
Single access drives

Dual access drives

Power factor

See model designations in Section 1.5.
500 W maximum at 60 Hz
450 W nominal at 60 Hz
550 W maximum at 50 Hz
500 W nominal at 50 Hz
550 W maximum at 60 Hz
500 W nominal at 60 Hz
600 W maximum at 50 Hz
550 W nominal at 50 Hz
10.5 A rms at 115 V /60 Hz
5.3 A rms at 230 V/60 Hz
11.0 A rms at 115 V/50 Hz
5.5 A rms at 230 V /50 Hz
11.3 Arms at 115 V/60 Hz
6.1 A rms at 230 V /60 Hz
11.8 A rms at 115 V /50 Hz
6.3 A rms at 230 V/50 Hz
0.80 minimum

1-6

Table 1-2

RK06 Disk Drive Performance Specifications (Cont)

Characteristics

Specifications

AC-Iow det<~ction

Less than 90 V (rms) for one cycle or more

Power cord
Length
Plug type

2.7 m (9 ft)
NEMA 5-15P for 120 Vac (nominal) models
NEMA 6-15P for 240 Vac (nominal) models

Operating environment
Ambient tenlperature

10 0 C to 40 0 C (50 0 F to 104 0 F)

Maximum t(~mperature rate of change

20 0 C/hour (36 0 F /hour)

Relative humidity

10% to 90% for a maximum wet bulb temperature
of 28 0 C (82 0 F)

Maximum altitude

2440 m (8000 ft)

Bit transfer rate (unbuffered nominal)

4.30 M/s

Bit cell width

232.5 ns

Latency
Rotational frequency
Average
Maximum

2400 rpm ± 2.5%
12.5 ms (1/2 rotation) ±2.5%
25.0ms

Averag1e
Maximum

38 ms
75 ms

Seek Times

Start/Stop times

60 seconds maximum, 30 seconds nominal

Model Designations
RK06-EA
RK06-EB
RK06-EC
RK06-ED

90-128 Vac@ 60 ± 0.5 Hz
180-264 Vac@ 60 ± 0.5 Hz
90~128 Vac@ 50 ± 0.5 Hz
180-264 Vac@ 50 ± 0.5 Hz
NOTE
Models RK06-F A through FD are the dual-access
models corresponding to models EA through ED respectively.

1-7

Table 1-2

RK06 Disk Drive Performance Specifications (Cont)
Specifications

Characteristics
RK06K-DC Cartridge
Temperature stabilization time

If the cartridge and drive are at approximately the
same temperature, 30 min.; if there is some difference, up to two hours, depending on the difference.

Operating temperature range

16 0 C to 49 0 C (60 0 F to 120 0 F)

Allowable relative humidity
(for operation)

8% to 90% for a maximum wet bulb reading of
28 0 C (82 0 F)

Storage temperature range
(for recorded disks)
Allowable relative humidity
(for storage)

-40 0 C to 65 0 C (-40 0 F to 150 0 F)

Table 1-3
Characteristics

8% to 80% for a maximum wet bulb reading of 28 0
C (82 0 F)

RK07 Disk Drive Performance Specifications
Specifications

Storage Type
Medium
Disk diameter

Dual-platter magnetic cartridge, DEC RK07K-DC
355 mm (14 in.) nominal

Magnetic heads

3 read/write, 1 servo

Recording capacity (formatted)
Cylinders/cartridge
Tracks/ cylinder
Tracks/cartridge
Sectors / track
Words/sector
Bits/word
Bits/sector
Bits/surface
Bits/pack
Bits/inch (inner track)
Tracks/inch

18-bit word
815
3
2445
20
256
18
4608
73.43 M
220.32 M
4040
384.6

Electrical
Voltage

See model designations in Section 1.5.

Input power
Single access drives

500 W maximum at 60 Hz
450 W nominal at 60 Hz
550 W maximum at 50 Hz
500 W nominal at 50 Hz

1-8

16-bit word
815
3

2445
22
256
16
4096
75.11 M
225.33 M
4040
384.6

Table 1-3 RK07 Disk Drive Performance Specificatons (Cont)
Characteristics
Dual access drives

Specifications
550 W maximum at 60 Hz
500 W nominal at 60 Hz
600 W maximum at 50Hz
550 W nominal at 50 Hz

Start current
Single access drives

10.5 A rms at 115 V /60 Hz
5.3 A rms at 230 V /60 Hz
11.0 A rms at 115 V / 50 Hz
5.5 A rms at 230 V /50 Hz

Dual ac:cess drives

11.3 A rms at 115 V /60 Hz
6 . 1 A rms at 230 V /60 Hz
11.8 A rms at 115 V /50 Hz
6..3 A rms at 230 V/50 Hz

Power factor

0.80 minimum

AC-Iow detection

Less than 90 V (rms) for one cycle or more

Power cord
Length
Plug type

2.7 m (9 ft)
NEMA 5-15P for 120 Vac (nominal) models
NEMA 6-15P for 240 Vac (nominal) models

Operating environment
Ambient temperature
Maximum temperature rate
of change
Relative humi.dity

10 0 C to 40 0 C (50 0 F to 104 0 F)
20 0 C/hour (36 0 F /hour)
10% to 90% for a maximum wet bulb temperature
of28° C (82 0 F)

Maximum altitude

2440 m (8000 ft)

Bit transfer rate (unbuffered nominal)

4.30 M/s

Bit cell width

232.5 ns

Latency
Rotational frequency
Average
Maximum

2400 rpm ± 2.5%
12.5 ms (1/2 rotation) ± 2.5%
25.0 ms

1-9

Table 1-3

RK07 Disk Drive Performance Specifications (Cont)
Specifications

Characteristics
Seek times
Average
Maximum

36.5 ms
71.0 ms

Start/Stop times

60 seconds'maximum, 30 seconds nominal

Model Designations
90-128 Vac@ 60 ± 0.5 Hz
180-264 Vac @ 60 ± 0.5 Hz
90-128 Vac@ 50 ± 0.5 Hz
180-264 Vac @ 50 ± 0.5 Hz

RK07-EA
RK07-EB
RK07-EC
RK07-ED

NOTE
Models RK07-FA through FD are the dual access
models corresponding to models EA through ED respectively.
RK07K-DC Cartridge
Temperature stabilization time

If the cartridge and drive are at approximately the
same temperature, 30 min.; if there is some difference, up to two hours, depending on the difference.

Operating temperature range

16 0 C to 49 0 C (60 0 C to 120 0 F)

Allowable relative humidity
(for operation)

8% to 90% for a maximum wet bulb reading of 28 0
C (82 0 F)

Storage temperature range
(for recorded disks)

-40 0 C to 65 0 C (-40 0 F to 150 0 F)

Allowable relative humidity
(for storage)

8% to 80% for a maximum wet bulb reading of 28 0
C (82 0 F)

1-10

Table 1-4

RK611/RK06 Disk Subsystem Options

Option

Description

RK06-EA
RK06-EB
RK06-EC
RK06-ED

Single-access RK06, H969 cabinet, 115 Vac/60 Hz
Single-access RK06, H969 cabinet, 230 Vac/60 Hz
Single-access RK06, H969 cabinet, 115 Vac/50 Hz
Single-access RK06, H969 cabinet, 230 Vac/50 Hz

RK06-FA
RK06-FB
RK06-FC
RK06-FD

Dual-access RK06, H969 cabinet, 115 Vac/60 Hz
Dual-access RK06, H969 cabinet, 230 Vac/60 Hz
Dual-access RK06, H969 cabinet, 115 Vac/50 Hz
Dual-access RK06, H969 cabinet, 230 Vac/50 Hz

RK611-EA
RK611-EB
RK611-EC
RK611-ED

RK06-EA Drive with RK611 Controller
RK06-EB Drive with RK611 Controller
RK06-EC Drive with RK611 Controller
RK06-ED Drive with RK611 Controller

RK611-FA
RK611-FB
RK611-FC
RK611-FD

RK06-FA Drive with two RK611 Controllers
RK06-FB Drive with two RK611 Controllers
RK06-FC Drive with two RK611 Controllers
RK06-FD Drive with two RK611 Controllers

RK06C

Dual access kit

RK611-C

Dual access kit plus one Unibus controller

RK06K··DC

Drive data cartridge

RK06K··AC

Drive alignment cartridge

RK06K··EF

Drive error free data cartridge

70-12292-08
70-12292-15
70-12292-25
70-12292-40

8 ft cable
15 ft cable
25 ft cable
40 ft cable

1-11

Table 1-5 RK611/RK07 Disk Subsystem Options
Option

Description

RK07-EA
RK07-EB
RK07-EC
RK07-ED

Single-access RK07, H969 cabinet, 115 Vac/60 Hz
Single-access RK07, H969 cabinet, 230 Vac/60 Hz
Single-access RK07, H969 cabinet, 115 Vac/50 Hz
Single-access RK07, H969 cabinet, 230 Vac/50 Hz

RK07-FA
RK07-FB
RK07-FC
RK07-FD

Dual-access RK07, H969 cabinet, 115 Vac/60 Hz
Dual-access RK07, H969 cabinet, 230 Vac/60 Hz
Dual-access RK07, H969 cabinet, 115 Vac/50 Hz
Dual-access RK07, H969 cabinet, 230 Vac/50 Hz

RK711-EA
RK711-EB
RK711-EC
RK711-ED

RK07-EA drive with RK611 controller
RK07-EB drive with RK611 controller
RK07-EC drive with RK611 controller
RK07-ED drive with RK611 controller

RK711-FA
RK711-FB
RK711-FC
RK711-FD

RK07-FA drive with two RK611 controllers
RK07-FB drive with two RK611 controllers
RK07-FC drive with two RK611 controllers
RK07-FD drive with two RK611 controllers

RK07C

Dual access kit

RK611-C

Dual access kit plus one Unibus controller

RK07K-DC
RK07K-AC
RK07K-EF

Drive data cartridge
Drive alignment cartridge
Drive error free cartridge

70-12292-08
70-12292-15
70-12292-25
70-12292-40

8 ft cable
15 ft cable
25 ft cable
40 ft cable

RK6/7-TA
RK6/7-TB

RK06, RK07 Field Test Box, 115 Vac/50 Hz or 60 Hz
RK06, RK07 Field Test Box, 230 Vac/50 Hz or 60 Hz

RK07U
RK6/7-TU

Upgrade kit to convert RK06 to RK07
Upgrade kit to convert RK06-TA, TB to RK6/7-TA, TB

1-12

CHAPTER 2
SUBSYSTEM DESCRIPTION

2.1 INTRODUCT][ON
The following subsections provide separate drive, controller, and subsystem descriptions that are intended to give an overview of the RK611/RK06 and RK611/RK07 Disk Drive Subsystems and capabilities.
2.2 RK06 AND RK07 DISK DRIVES
The RK06 and RK07 are moving-head disk drives that function as random access mass storage devices ...
(Figure 2-1). The drives perform under the direction of the RK611 Disk Controller and contain all of
the response circuitry required to perform those operations for device status reporting and the storage
of data.
The storage medium designed for the RK06 Drive is a dual-platter assembly, defined as the RK06K
Disk Cartridge, which has a maximum formatted storagecapacity of 6.3 million I8-bit words or 6.9
million 16-bit words. Similarly, the RK07 Drive uses a dual-platter assembly, the RK07K Disk Cartridge, as the storag,e medium. It has a maximum formatted storage capacity of 12.6 million I8-bit
words or\D.8)1'million 18-bit words~ As stated before, an RK06K cartridge may only be used in 'an
RK06 Dnve, and an RK07K cartridge will only operate in an RK07 Drive.
RK06 and RK07 Controls and Indicators
The following manual controls and indicators are located on the front of each RK06 and RK07 Disk
Drive:
•
•
•
•
•
•

RUN/STOP pushbutton switch with STOP indicator.
UNIT SELECT plug with READY indicator
Drive FAULT indicator
WRITE PROT pushbutton switch with PROTECT indicator
ACCESS A pushbutton switch with SELECT indicator
ACCESS B pushbutton switch with SELECT indicator

The switches are selectively used (refer to Chapter 6) to place an associated drive on- or off-line,
indicate operational status and error conditions, and provide controller access (whether the drive is a
single- or dual-access type).
For drive power, an ON/OFF circuit breaker with a POWER indicator is located at the rear of each
drive.

2-1

8376-2

Figure 2-1

H969 Standalone Disk Drive Cabinet

2-2

2.3 RK06K AND ItK07K DISK CARTRIDGES
The RK06K and RK07K Disk Cartridge assemblies contain two platters (Figure 2-2). Three of the
disk surfaces (0, 1, and 2) are available for data storage (read/write surfaces), while the fourth surface
(S) is a read-only selrvo surface that is used for head positioning and the derivation of timing.
Each RK06K disk surface is divided into 411 tracks, and each RK07K disk surface contains 815
tracks. The radial location of each track is defined as a separate cylinder position that is common to all
surfaces. Under these conditions, a cylinder can be specified (cylinder address), via a Seek command
from the controller, to cause the simultaneous positioning of all four heads to one of the 411 cylinder
positions (0-410) for the RK06, or one of 815 cylinder positions (0-814) for the RK07. A single
read/write head (0, 1, or 2) can then be defined (track address), via a Read or Write command, to
select one of the three read/write surfaces. In this manner, a desired track is accessed for data transfer
operations.
Data Track Formattitng
During disk formattiing procedures, each data track is located and recorded with header information
via a Write Header command. The result of this operation effectively divides a track into a number of
unique data storage segments that are defined as sectors (Figure 2-3). For I6-bit data words, a single
track can accommodate 22 sectors, while for I8-bit data words a track can accommodate 20 sectors. In
either case, sectors are the basic, addressable storage unit within the drives, and as such are the final
coordinate of track selection during data transfer operations.
2.4 RK611 DISK DRIVE CONTROLLER
The RK6II Disk Controller presents a complete control and data interface to the PDP-II Unibus
(which serves both the central processor and main memory) and the RK06 or RK07 Disk Drives
available to the syste~m (Figure 2-4).
2.4.1 Command COllltrol
The RK611 Control interface contains all of the logic required to receive, store, and decode the 13
separate disk commands (e.g., Start Spindle, Seek, Write, etc.) that can be executed by the controller.
In addition, the control interface contains all of the logic required to receive and store the status
information (i.e., modes, conditions, faults, and errors) delivered from a selected drive. To accommodate the exchange. of information (commands/status) between the control interface and a selected
drive, information is both sent (commands) and received (status) via a pair of bidirectional serial
message lines that are defined as Message Line A and Message Line B.
2.4.2 Data Control
The RK611 data interface contains all of the logic required to transfer data (16- or I8-bit words)
between main memory and a selected drive. Since Unibus transfers require parallel data flow, while
transfers to and from a disk require serial data flow, the data interface also contains parallel-to-serial
(memory-to-disk) and serial-to-parallel (disk-to-memory) conversion logic. In addition, a bidirectional
read/write data line is used to accommodate the exchange of serial data between the RK611 data
interface and a select.ed drive.
Finally, the data interface logic contains a 66-word data buffer (Silo), that is used to compensate for
the timing differences that exist between the average Unibus data transfer rate (4.3 ~s/word) and the
burst transfer rate of the disk (3. 7 ~s/word).
2.4.3 Controller Rel~isters
The controller contains 15 Unibus-addressable registers (Table 2-1). Nine of the 15 registers provide
temporary storage for the manipulation of status and/or error information. This information, coupled
with the inherent diagnostic capabilities of the controller, allows a programmer to create diagnostics
that can be extremely effective as a fault isolation aid.

2-3

R/W

•

READ/WRITE

l?J

~ SURFACE

rTOPDISK

R/W
R

..
R/W

READ/WRITE
SURFACE

8
0

•

SERVO

~ SURFACE

r--

BOTTOM
DISK

READ/WRITE
SURFACE

[2J

HEAD ORIENTATION
3 IDENTICAL
READ/WRITE

1 SERVO SURFACE

~'5;<-----j~+-+---....II..---- CYL 410

SERVO SURFACE (TIMING & POSITION)

READ/WRITE SURFACES

RK06K DISK CARTRIDGE

3 IDENTICAL
READ/WRITE
SURFACES
(0,1,2)

'\.....Y - - - - I - + - - l " - - t - - - - - - -

~-o:::....,.,,£-~---___,.---

1 SERVO

CYL 814

CYL 0

SERVO SURFACE (TIMING & POSITION)

READ/WRITE SURFACES

RK07K DISK CARTRIDGE
ADDRESSING
1. CYLINDER ADDRESS DEFINES 1 OF 411 TRACKS (FOR RK06K) OR
1 OF 815 TRACKS (FOR RK07K).
2. TRACK ADDRESS (R/W HEAD) DEFINES ONE R/W SURFACE.
3. SECTOR ADDRESS DEFINES 1 OF 20 OR 22 SECTORS/TRACK.
MA-0922

Figure 2-2

Read/Write Heads and Data Track Orientation
2-4

NUMBER
OF WORDS

16
HEADER

HEADER HEADER
CHECK

PREAMBLE

16
DATA

HEADER
GAP

PREAMBLE

SECTOR TOLERANCE GAP

16-BIT WORDS=4096 BITS
l8-BIT WORDS=4608 BITS

22sec.TRACK=112 BITS
20sec. TRACK'96 BITS

W~RD

CD Cylinder (000-632.B for the RK06, or
000-1456 8 for the,RK07)

® Track(headO,1,2)
@ Sector (OO-23eor 00-25e )
@) Format(I=18-bit dal,~;0'16 bit data).

® Good sector flag (bOilh bits set for good sector)

091'00

C_Y_L_N_D_E_R_CD_'________~.___0__0___0__0___00~I

______

WORD
2

6-7

256
DATA WORDS

0405

SECTOR

W~RD~

_____
E_X_C_LU_S_I_V_E__
O_R_S_U_M
__O_F__W_O_R_D_S__
'_A_N_D__
2 ____--J
11-4354

Figure 2-3

Sector Formats

PDP11
MAIN
MEMORY
PDP11
CENTRAL
PROCESSOR

RK611
DISK
CONTROLLER

INTERFACE LINES

RK06 OR RK07
DISK DRIVE(S)

Central Processor

Disk Controller

Disk Drive

• Delivers Commands (13)
To Controller:

• Interprets Commands

• Stores and Retrieves Data

• Directs Drive To:

• Generates Control Signals:

Non-Data
Handling

1. Select Drive
2. Pack Acknowledge
3. Drive Clear
4 .. Unload
5. Start Spindle
6. Recalibrate
7. Offset
8. Seek
9. Read Header
10. Write Header
11. Read Data
12. Write Data
13. Write Check

Header
Handling

Data
Handling

{

1. Position R/W Heads
2. Read or Record Data
3. Send Fault, Error, or Status
Information
• Performs Parallel-to-Serial (Write)
and Serial-to-Parallel (Read) Data
Conversions for Direct Memory Access (Data)
Transfers.

1. Polled Attention
2. Select Acknowledge
3. Multiple-Drive Select
4. Sector/Index Pulses
5. Write Clock
6. Strobe
• Generates Status Messages
(Eight 16-Bit Words)

11-4195

Figure 2-4

Basic RK06 or RK07 Disk Subsystem

2-5

Table 2-1 RK611 Unibus-Addressable Registers
Address
(octal)

Type

Basic Purpose

777440

R/W

Control/Status Register 1
(RKCSI)

Decode commands/controller
status

777442

R/W

Word Count Register (RKWC)

Number of data words for transfer

777444

R/W

Bus Address Register (RKBA)

Main memory location for data
word

777446

R/W

Disk Address Register (RKDA)

Desired track/sector address

777450

R/W

Control/Status Register
(RKCS2)

Additional control information

777452

Drive Status Register (RKDS)

Drive status reports

777454

Error Register (RKER)

Additional drive status

777456

R/W

Attention Summary/Offset
Register (RKAS/OF)

Drive status change
state/head offset value

777460

R/W

Desired Cylinder Register (RKDC)

Cylinder address

777462*

Unused

777464

R/W

Data Buffer Register (RKDB)

Data word to /from Silo

777466

R/W

Maintenance Register 1 (RKMRl)

Drive message select/control

777470

ECC Position Register (RK ECPS)

Error position information following correction sequence.

777472

ECC Pattern Register (RKECPT)

Error correction pattern following
correction sequence.

777474

Maintenance Register 2 (RKMR2)

Shift register A for serial Message
A.

777476

Maintenance Register 3 (RKMR3)

Shift register B for serial Message
B.

NOTE
R/W = read/write (selected bits) in relation to
read only in relation to Unibus.
Unibus. R

*The RK611 responds to this address with SSYN. However, if the bus cycle is a DA TI, D (00: 15) is undefined.

2-6

2.4.4 Silo Memory
During direct memory access (DMA) operations in which a device bypasses the central processor and
gains a direct access to main memory, the storage depth of the Silo (66 words) compensates for the
transfer rate differe:nces that occur between controller Idrive data exchanges and controller IU nibus
data exchanges.
When writing, the Silo receives parallel data from the Unibus and transfers the data to a Data Shift
Register for serialization. The serial data is then encoded using the modified frequency modulation
(MFM) technique. When reading, the MFM encoded data is decoded and sent to the Data Shift
Register for conversion to parallel data. The parallel data is then sent to the Silo for eventual transfer
to the Unibus (Figure 2-5).

PARALLEL DATA

---,
I
I
I
I

OUTPUT
BUFFER
(1 DATA WD.)

U
N
I
B

(64 DATA
WORDS)

I
I

SI LO IC

I
I

I
I
I

SILO
MEMORY

RKDB

l__

INPUT
BUFFER
(1 DATA WD. )

HEADER
COMPARE
LOGIC

DATA SHIFT REGISTER
(PARAL.LEL-TO-SERIAL
AND
SERIAL -TO-PARALLEL)

o
R
I

V
E

I
I
I
I

r
I

---.I
PARALLEL DATA

'1 - 4196

Figure 2-5

RK611 Data Interface

2-7

2.5 RK611/RK06 AND RK611/RK07 SUBSYSTEMS
The basic RK06 or RK07 Disk Drive subsystem consists of an RK6II Disk Controller and up to eight
RK06 or RK07 Drives that can be connected to the subsystem via a daisy-chain bus arrangement. The
controller interfaces to the system via a PDP-II Unibus. However, if the optional dual-access RK06 or
RK07 Drives are installed, an expanded subsystem can be configured. With the expanded subsystem,
dual controllers (A and B) are installed which allow each drive to be accessed by either controller via
appropriate dual-port selection (A or B). When two controllers are used in this manner, they may be
connected to the same Unibus in a single-processor system, or each controller may be connected to a
separate Unibus in a dual-processor system (Figure 1-1).
The following paragraphs describe RK06 or RK07 subsystem operations in relation to the functions
performed by the controller/drive interface lines (17) under directives imposed by the U nibus-tocontroller disk commands concluding with data formatting requirements and data error control operations.
2.5.1 Controller/Drive Interface
The following information describes the operations performed by the controller interface lines (Figure
2-6) that are common to all of the drives available to the system.

f+--.

READ/WRITE DATA
MESSAGE LI NE "A"
MESSAGE LINE "8"
STR08E

WRITE GATE
CONTROLLER-TO-DRIVE (CTD)

~NIBUS INTERFAV

CONTROL CLOCK
RK611
DISK
CONTROLLER

INITIALIZE

DISK DRIVE

POLL DRIVE 22
POLL DRIVE 2'
0

POLL DRIVE 2

CONTROLLER POWER ON

SELECT ACKNOWLEDGE (SACK)
MULTIPLE DRIVE SELECT
INDEX/SECTOR PULSES
WRITE CLOCK
POLLED ATTENTION

CONTROLLER/DRIVE INTERFACE LINES
11-4197

Figure 2-6

Interface Lines

Message A Line
This line transmits (from controller to drive) a drive selection code as the initial 3-bit portion of a 16bit serial message, the remainder of which includes additional operational information (e.g., Seek,
Start Spindle, Head Select Coding, etc.) for the desired drive (Figure 2-7a). Notice, in the Figure, (hat
the control clock timing (TO-T 15) reflects the availability of each one of the 16 bits of the message'in
relation to time.

2-8

With these timing considerations, the desired drive is selected by the initial coding (TO-T2) of the serial
format. When drive selection recognition is achieved, the remaining drives ignore the remainder of the
message (T3-T 14).
When the operational information has been assimilated by the selected drive, one of four possible 16bit status messages can be returned from the drive (when requested by the controller) in a similar
manner (Figure 2-7b).
Message B Line
In conjunction with the transmission of the Message A line (from controller to drive), a 16-bit message
is simultaneously transmitted over the Message B line. The Message B line transmits additional operational information (e.g., cylinder address, offset values, etc.), including a status message selection
code. When the operational information has been assimilated by the selected drive, the selected status
message (if requeste:d by the controller) is returned (Figure 2-7b).
Control Clock Line
As long as power is available to the controller, the 465 ns Control Clock signal is continuously generated.
Strobe Line
For message synchronization, separate strobe signals are generated in a manner which depends on the
direction of the transmission:
• Controller-to-Drive Strobe - During the transmission of Message lines A and B, a single strobe
signal is generated as an enable for drive selection code identification (Figure 2-7a).
• Drive-to-CoUitrolier Strobe - During Message line A and B status reporting, separate strobe
pulses (generated by the drive) clock each message bit into the controller (Figure 2-7b).

-j465NSICONTROL

nUI

CLO~

STRO~

I T10

TIl

Tl2

Tl3

T14

Tl5

_ _ _ _ _ _ _ _ _, - -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

I I I I I I I I I I I I I I I I ~
SA~
I I I I I I I I I I I
GTO

RESVD PARITY

CD
NOTE CD:

RESVD PARITY

These bits are used only on the RK07.

a.
Figure 2-7

Controller-to-Drive Transmissions

Controller jDrive Serial, 16-Bit Message Formats

2-9

7.44JLs

---------------toi-I

STROBE

~--------------~I

SACK

STATUS
MESS
A'lJ

,
«

(HIGH)

DRIVE SELECT CODE

w f---t---t
Al

...J DRIVE SELECT CODE

RESVD

DR
AVAIL

VOL
VAL

DRIVE
DR
READY TYPE

SERVO
HEADS BRUSH DOOR
RESVD SIGNAL
HOME HOME LTCHD
PRES

20 OFFSET
WR SPINDLE
SECT
LOCK
ON
ON
FORMAl

CART SPEED
PRES
OK

FWD

PIP

CHANGE

LDG
HDS

REV

STATU~ PARITY

UNLDG
PARITY
HDS

RTZ

(!)

«
Vl ~RIVE SELECT CODE RESVD
w

I
I

::Ii!

DRIVE SELECT CODE

MESS 10
(0) (1)

RESVD FOR
ADD'L MESS

MESS ID
(1), (1)

RESVD FOR
ADD'L MESS

::Ii!

DR
SPEED
RD/WR
OFF
PARITY
LOSS TRACK UNSAFE

LIM
SEEK
WRCN' WRITE
RESVD FOR SECTOR
SERVO
NO GATE HEAD MULT INDEX TRIBIT SERVO
PARITY
NO DETON
ADD'L MESS
ERR
FAULT
HD
SEL
ERR
ERR
SIG
ERR
~~jlEE
MOTION SEEK UNSAFE

(!)

WR
LOCK

PARITY

MESS 10
(1), (0)

«
Vl

C-D
PRTY
ERR

RESVD FOR
RESVD
ADD'L MESS

INV
ADDR

AC
LOW

FAULT

NXF

:::i
w

DRIVE SERIAL NUMBER

MESS ID
(0) (0)

CD RESVD PARITY

CYLINDER DIFFERENCE/OFFSET VALUE

NOTE CD:

SEEK
INC

IrR~~S

CYL I NDER ADDRESS

SECTOR COUNT

DECODED
HEAD ADDRESS

CD
RESVD

RESVD PARITY

These bits are used only on the RK07.

b.
Figure 2-7

11-4198

Drive-to-Controller Transmissions

Controller/Drive Serial, 16-Bit Message Formats

Controller-to-Drive (CTD) Line
When the CTD line is asserted, the transmiSSIon of Message line A and B information from the
controller to the drive is enabled. The CTD signal is normally asserted, remaining so for the entire 16
clock times (Figure 2-7a).
When the CTD signal is negated, the selected drive is allowed to transmit status information to the
controller (Figure 2-7b).
Select Acknowledge (SACK) Line
When asserted, the SACK line provides a drive-to-controller indication that the drive select code has
been recognized and the remaining bits of the message are being accepted. This occurs approximately
300 ns after the receipt of bit 5 (T4) of the controller-to-drive message. The line is negated within 100 ns
of the receipt of another controller-to-drive message strobe.

2-10

Poll Drive (22 - 2°.) l..ines
These three binary-encoded drive address lines are used to poll all drives (0-7) available to the system
to determine the att(mtion status of each drive. Regardless of whether a drive is selected or not, the
polling mechanism permits the controller to determine which drive (if any) has asserted any attention
signal, without seriailly selecting each drive. The time required to update the attention status of all eight
drives is 7.4 J,ts.

The Attention signal in a drive can be set by anyone of the following conditions:
• The completion of a Start Spindle command.
• The completion of a Seek command.
• The completion of an Offset command.
• The beginning of an Unload Heads command.
• The completion of an RTC command (return offset heads to centerline).
• The occurrenlce of any fault condition.
Polled Attention Line
An Attention signal is activated within a drive when its Drive Status-Change flip-flop is set. However,
the Attention line is not asserted to the controller until the unit number (number on the drive UNIT
SELECT plug) of the drive is polled, via the Poll Drive (22 - 20) lines. Thus, each time a drive is polled,
and its Drive Status-Change flip-flop is set, Attention will be reported.

The Attention signal is available to the controller approximately 400 ns after the detection of a polling
address for the drive, and the signal is negated within approximately 160 ns after the polling address is
removed.
Write Clock Line
This drive signal is d(~rived from the servo surface and synchronized with the rotational speed of the
disk to ensure the recording of a constant bit density on the data track.
Read/Write Data Line
This line carries encoded digital read data from the drive to the controller and encoded digital write
data from the controller to the drive.
Write Gate Line
When active for a write operation, this signal enables the generation of write current in the selected
drive. The drive recognizes the assertion of this signal under the following conditions:

• Drive selection achieved.
• Volume valid set (bit T6 of Status Message AO).
• No unsafe read/write condition exists in drive.
• No drive off-track error exists.
• No write protl~ct mode exists.
Index/Sector Pulse Line
When a drive is selectt~d, this line transmits the once-per-revolution index pulse and each of the sector
pulses derived from the servo tracks. Index/Sector pulses are enabled when the drive asserts a Select
Acknowledge (SACK) and remain enabled until the drive is deselected.

2-11

Multiple Drive Select Line
When asserted, this line indicates to the controller that more than one drive has simultaneously responded to a given selection code. To accomplish this, a selected drive monitors the Index/Sector pulse
line for the presence of extraneous pulses. If pulses other than those originating from the desired drive
are detected, the Multiple Drive Select line is asserted. However, since Index/Sector pulses cannot be
generated if the heads of the desired drive are not loaded, the drive uses Control Clock pulses, inserted
on the Index/Sector pulse line, to monitor and detect multiple selections.
Finally, the Multiple Drive Select signal is asserted within 300 ns of pulse detection from another drive
and negated within 120 ns after the receipt of another message strobe.
Initialize Line
When asserted, this signal deselects all drives, resets all error conditons, and clears the Drive StatusChange flip-flop.
Controller Power on Line
When asserted, this line indicates that power has been applied to the controller, and a physical connection (via cable) has been established between the controller and the available drives. When the line
is negated, all drives are deselected, and the Strobe and Control Clock logic is disabled in the drive to
prevent any erroneous drive selection.
2.5.2 Subsystem Commands
The 13 subsystem commands (Figure 2-4) can be divided into two basic groups. One group (nondata
handling) is concerned with the various operational requirements of the drives, while the second group
(data or header handling) is concerned with the transfer of data and header information to or from a
selected device.
2.5.3 Data Formatting
When the tracks of a disk are formatted via separate Write Header commands, the following information is recorded to identify each sector in relation to its radial and rotational position on a disk (Figure
2-3).
1.

The Header Preamble field contains 16 words, consisting of 255 Zero bits and a single One
bit (SYNC) that defines the start of valid information.

The Header field consists of two words:

Word 1 contains the cylinder address which defines the radial location of the sector in
relation to the 411 (for the RK06) or 815 (for the RK07) addressable cylinders on the
disk.

Word 2 contains the sector address-which defines the sector in relation to the number
(20 or 22) of sectors per track-and the track address, which defines one of the three
R/W heads (and therefore the appropriate R/W surface in relation to the disks) a
Format (FMT) bit that is a zero to reflect 16-bit data, or a one to reflect 18-bit data
words. Finally, the Good Sector flag (GS FLAG) bits are set to indicate that the sector
is valid for the reading or writing of data.

The Header Check field consists of one word that is an exclusive-OR sum of header words 1
and 2. The check word is generated by software and extracted during read operations to
verify the integrity of both the Header field and the check word.

2-12

The Header Gap field contains four words, consisting of 64 Zero bits, which provide the
time required by the controller to complete a header check and accommodate mechanical
tolerances within the drive.

The Data Preamble field contains 16 words, consisting of 255 Zero bits and a single One bit
(SYNC) that defines the start of the Data field.

The Data field accommodates 256 16-bit data words (4096 bits) or I8-bit data words (4608
bits), depending on the format. The extra length required by 256 I8-bit words is compensated by the reduction of the maximum number (20) of sectors allowed per track.

The Error Correction Code (ECC) field contains a 2-word correction code that is generated
by the controller, during a write operation, by the total bit configuration of the 256-word
Data field. The ECC field is read during read operations to verify the integrity of both the
data and the ECC words.

The Data Postamble field consists of one word of Zero bits, which allows the write amplifier
to be disabled, when writing, during an unused portion of the sector.

The Sector Tolerance Gap contains six or seven words (depending on the data format) of
Zero bits, which ensures a differentiation between sectors that is required by the mechanical
tolerances.

2.5.4 Servo SurfacE~ Operations
For the drive to affect head positioning and the reading and writing of data under the direction of the
controller, radial (cylinder) and rotational (sector/index) coordinates are derived from
factory-recorded servo track signals (tribits) that are continuously read by the servo head.
When a Seek command is initiated, a desired cylinder address is delivered to the drive from the controller. Within the drive, the difference between the desired and the current cylinder position is calculated. If the difference is other than zero, the head positioning circuitry is energized, and all of the
heads (R/W and servo) are driven across the radius of the disk in the direction defined by the difference information. As the search for the desired cylinder position continues, increasing proximity to the
cylinder gradually reduces the difference to zero, at which time the heads are appropriately positioned
over the desired track. As stated, the servo tracks control this process by generating pulses that are
derived from each sc~rvo track crossing and fed back to the head positioning circuits as the disk surfaces are radially traversed. When the number of pulses generated equals the number expected (difference minus track crossing equals zero), the desired cylinder is located.
Once the heads are lloaded (regardless of whether they are stationary or in motion), the servo signals
provide the generation of Sector and Index (once-per-revolution) pulses within the drive. The Sector /Index pulses define the position of each sector for the controller, providing rotational coordinates
for read and write operations.
Finally, within the controller, the clock derived from the servo signals is used to synchronize the
operation of a phased-locked oscillator (PLO). Synchronization of the PLO circuitry ensures a constant recording density, regardless of minor variations in the rotational speed of the disks.
2.5.5 Data Error Control
The RK6II Controller contains both data error detection and correction logic and a data error recovery capability. When properly supported by appropriate software procedures, these controller features
minimize the disruptive effect of data errors from the disks.

2-13

Data Error Detection
The following defines the conditions under which data errors can be detected by the controller:

• Data bit errors indicated by the Error Correction Code (ECC) logic.
• The controller has detected a mismatch resulting from a comparison of header data.
For data error correction purposes, each sector contains a 32-bit Error Correction Code (ECC). The
code is used with a combination of hardware and software procedures to provide the following:
• Detection and correction of Data field error bursts up to 11 bits in length.
• Detection, without correction, of error bursts from 12 to 32 bits in length.
• Reliable, but not absolute, detection of error bursts greater than 32 bits.
Data Error Correction
Error correction involves the use of error position (error burst location) and error pattern (error burst
image) information, which is derived from a 32-bit Error Correction Code (ECC) configuration. Each
sector contains an ECC that is written at the end of the Data field and reflects the configuration of the
recorded data. The code is used to detect and correct (if possible) the data bits defined by an error
burst pattern read from the disk. The controller contains the logic required to generate the ECC
configuration during a write and extract the code during a read. However, the actual correction of the
data bits defined by the error burst pattern is a software process that occurs after the data is in main
memory.
Data Error Recovery
Data error recovery is a recommended sequence of attempted rereads that implements, under program
control, both the head offset and error correction capabilities of the controller to improve the probability of a successful recovery. A complete error recovery routine consists of a maximum of 28 data
recovery attempts, using error correction procedures with the following operations:
1.

Using error correction, attempt to reread the sector 16 times at track centerline.

Using error correction, attempt to reread the sector twice, from each of the following head
offset positions:
RK06

RK07

+400 microinches
-400 microinches
+ 800 microinches
- 800 microinches
+ 1200 microinches
- 1200 microinches

+ 200 microinches
- 200 microinches
+400 microinches
-400 microinches
+ 600 microinches
- 600 microinches

2.5.6 Bad Sector File
The bad sector file is a list of the bad sectors in the cartridge and is recorded on the highest number
track and cylinder on the cartridge. This track is written in 22 sector, 16-bit word format.

2-14

2.5.6.1

Data Field Format - The data field in the first 10 sectors of the last track has the following

format:

Word

o
1
2
3

~}
~}

Cartridge serial number
Zeros
Zeros
Zeros for data cartridge, at least a single one for alignment cartridge
Bad sector 1
Bad sector 2

n
} Last bad sector
n+ 1
n +2
Ones filled
Ones filled
n +3
254
255

Ones filled
Ones filled

Word 0 contains the octal cartridge serial number. Words 1 and 2 are filled with zeros and are reserved
for future use. Word 3 identifies the cartridge either as a data or alignment cartridge. If word 3
contains any ones, words 4 through 255 shall also contain ones.
The list identifying bad sectors on a data cartridge is contained in two-word groups beginning with
words 4 and 5. The first word of the two-word group contains the cylinder address of the defect, and
the second word contains the track and sector address of the defect.
The defective sectors are listed in ascending order by cylinder, track, and sector address. After the pair
of words identifying thc~ location of the last bad sector, the remainder of the data field is ones filled.

2.5.6.2 Track Format - The bad sector file format consists of the same preambles, gaps, header, ECC,
and postamble as those on any other 22 sector track. The data field of sectors 0, 2, 4, 6, and 8 will
contain the addresses of all sectors identified as defective in 22 sector, 16-bit format, and the data field
of sectors, 1, 3, 5, 7, and 9 will contain the addresses of all sectors identified as defective in 20 sector,
I8-bit format. Sectors 10 through 21 will have words 0 through 3 of the data field record as specified in
2.5.6.1. Words 4 through 255 will be ones filled.
At least two even sectors and two odd sectors out of the first 10 sectors must be error free, and two
even and two odd sectors out of the last 12 sectors must be error free.

2-15

CHAPTER 3
SITE CONSIDERATIONS AND PREPARATIONS

3.1

ENVIRONMENTAL CONSIDERATIONS

3.1.1

General
The RK611/RK06 or RK611/RK07 Disk subsystem is designed to operate in a light industry or a
business environment.

3.1.2

Temperature
The operating temperature range of either Disk Subsystem is from 10° C to 40° C (50° F to 104° F);
maximum gradient is 30° F per hour. The nonoperating temperature range is from -40° C to 66° C
(-40° F to 151 ° F). The disk cartridge operating temperature range is from 15.6° C to 48.9° C (60° F
to 120° F).

3.1.3

Relative Humidiity
Humidity control is important in a data storage system as static electricity can cause errors in any CPU
with memory. Ei;ther disk subsystem is designed to operate efficiently within a relative humidity range
of 10% to 90%, with a nnaximum wet bulb temperature of 28° C (82° F) and a minimum dewpoint of 2°
C (36° F). The nonopt~rating relative humidity range is from 10% to 95%, with a maximum wet bulb
temperature of 46° C (115° F). The relative humidity range for the disk cartridge is 8% to 90%, with a
maximum wet bulb temperature of 28° C (82° F).

3.1.4

Heat Dissipatiolll
Heat dissipation of the RK06 or RK07 is 1500 Btu/hr nominal and 1700 Btu/hr maximum. By adding
this figure to the total heat dissipation for the other system components, and then adjusting the result
to compensate for such factors as the number of personnel, heat radiation from adjoining areas, sun
exposure through windows, system efficiency', etc., the approximate cooling requirements for the system can be determined. It is advisable to allow a safety margin of at least 25 percent above maximum
estimated requirements.

3.1.5

Acoustics
Most computer sites re:quire at least some degree of acoustic treatment; however, the RK06 or RK07
Disk Subsystem will not contribute unduly to the overall acoustic problem. Acoustic materials should
neither produce nor harbor dust.

3.1.6

Altitude
Computer system operation at high altitudes can result in heat dissipation problems. The maximum
altitude specified for the RK06 or RK07 is 2440 m (8000 ft). However, maximum allowable operating
temperatures are reduced by a factor of 1.8° C/I000 m (10 F/I000 ft) for operation at high-altitude
sites. Thus, the maximum allowable operating temperature at 2440 m (8000 ft), for example, would be
reduced to 28° C (82° F).

3-1

3.1.7

Radiated Emissions

The effects of radiated emissions can be reduced by:
1.
2.
3.
4.

Grounding window screens and other large metal surfaces
Shielding interconnection cables with grounded shields
Providing additional grounding to the system cabinets
In extreme radiation environments, providing a grounded cage for the system.

3.1.8 Vibration/Mechanical Shock/Attitude
The RK06 or RK07 is designed to operate within specifications in a moderately controlled environment, such as an office or computer facility, during the following sine vibration test, applied once in
each of three orthogonal axes (sweep rate = 1 octave/min):

50-60 Hz 0.002 inch double-amplitude (DA) displacement
50-5000 Hz 0.25 G pk
500-50 Hz 0.25 G pk
50-5 Hz
0.002 inch DA
It is also designed to operate while a half sine shock pulse of 3 G pk and 10 ± 3 ms duration is applied
once in either direction of three orthogonal axes (3 pulses total).
Performance of the RK06 or RK07 will not be affected by an attitude where the maximum pitch and
maxim urn roll do not exceed ± 5 percent.

3.1.9

Cleanliness

Although cleanliness is important in all facets of a computer system, it is particularly crucial in the case
of a device such as the RK06 or RK07 Disk Subsystem. Disk cartridges are not sealed units, while
loading, and are extremely vulnerable to dirt. Even such minute obstructions as smoke particles,
fingerprint smudges, or dust specks can cause head crashes and catastrophic destruction of heads
and/ or disk surfaces, as shown in Figure 3-1.
The RK06 or RK07 is capable of operating in an ambient atmosphere containing not more than one
million particles of 0.5 micron or larger per cubic foot of air.
During site preparation, there are a number of steps that may be taken to enhance subsequent cleanliness:
1.

Seal all windows in the vicinity of the Disk Subsystem.

If partitions are to be installed, consider floor-to-ceiling walls which minimize the flow of
dust.

If painting is to be done in advance of installation, select paint for walls, ceilings, and floors
that will not tend to flake or powder excessively. (Waterbound distemper is generally unsatisfactory in this respect.)

Check that the flow of air from the air-conditioning system will tend to carry lint, etc., away
from the Disk subsystem location. Provide filtration to inhibit dust and other particulants.

Select acoustical material that will neither produce nor harbor dust.

A void glass fiber tiles that could produce abrasive particles and floor coverings that tend to
crack or crumble.

3-2

Provide closed cabinets for disk storage.

Clean and vacuum sub floor areas and air-conditioning systems just prior to installation.

Place impregnated mats at each entrance to reduce the amount of dust tracked in from other
areas.

The disk drive air circulation system (Figure 3-2) is designed to supply clean air for:
1.

Disk temperature control.

Cleaning of the shroud area by low pressure (3/4 inch of water) at 70 cfm of air.

Card cage cooling.

Power suply /power amplifier assembly cooling.

To ensure that exhausted air moves freely, floor air-conditioning outlets should not be placed at the
rear of the drive.

11-4978

Figure 3-1

Relationship of Disk Head, Disk, and Contaminants

3--3

CARTIDGE
25-CFM AIR HOSE

POWER SUPPL V/POWER
AMPLIFIER ASSV

11-4982

Figure 3-2

RK06/RK07 Disk Drive Air Supply and Distribution Subsystem

3-4

3.2 SITE PREPARATION
3.2.1

Space

Provision should be made for service clearances of916 mm (36 in.) at the front and rear of the RK06 or
RK07 disk drive. Space should also be made available in the system environment for storage of disk
cartridges, each of which has a diameter of approximately 355 mm (14 in.) and a height of approximately 101 mm (4 in.). No more than five cartridges should be placed in a single stack.

3.2.2

Cabling

Each RK06 and RK07 Disk Drive is equipped with a molded line power cord with a length of2.7 m (9
ft).

3.2.3

Power Requirements

Both 60 ± 0.5 Hz and 50 ± 0.5 Hz drives operate from 90-128 Vac or 180-264 Vac power sources at a
power factor greater than 0.80. While operating at 60 Hz, the RK06 input power is 450 W nominal and
500 W maximum. While operating at 50 Hz, the input power is 500 W nominal and 550 W maximum.
While operating at 60 Hz, the input power of the RK07 is 500 W nominal and 550 W maximum, or 550
W nominal and 600 \V maximum while operating at 50 Hz.
Receptacles that will accept the 90-128 Vac plugs are designated variously as: DEC (12-05351),
NEMA (5-15R), Hubbell (5000-M9), and General Electric (4050-ITA). Receptacles that will accept the
180-264 Vac plugs an~ designated variously as: DEC (12-11204), NEMA (6-15R), Hubbell (5662ST),
and General Electric (4092-ITA). Appropriate circuit breakers are also necessary.
Digital Equipment Corporation must be notified of available input power well in advance of shipment,
so that the proper drive may be shipped.

3.2.4

Floor Loading

The weight of the RK06 or RK07 (148 kg/325Ib) alone is not sufficient to place unusual stress on most
office building or industrial plant floors. However, the added weight should be considered in relation
to the weight of the existing computer system and possible future expansion.

3.2.5

Installation Constraints

The route the equipm':!nt will travel from the receiving area to the installation site should be studied in
advance to ensure problem-free delivery. Among the factors to be considered are the height and location of loading doors; the size, capacity, and availability of elevators; the number and size of aisles and
doors en route; and any restrictions, such as bends or obstructions, in the hallways. Any constraints
should be reported to Digital Equipment Corporation as soon as possible so that the requirements of
the individual installation site may be considered when the unit is packed for shipment.

3.2.6

Fire and Safety Precautions

The RK06 and RK07 Disk Drives present no unusual or additional fire or safety hazards to an existing
computer system. Wiring should be carefully checked, however, to ensure that the capacity is adequate
for the added load and for any contemplated expansion. The 60 Hz version of the RK06 or RK07 is
UL-listed and CSA-ct~rtified. The 50 Hz version complies with IEC 435 standards.

3-5

CHAPTER 4
INSTALLA TION

This chapter includes the procedures required for unpacking and installing the RK611 /RK06 or
RK611/RK07 Disk subsystem. Installation will only be performed by qualified DIGITAL personnel.
If the equipment is installed by personnel not so qualified, the equipment will lose its CSA rating.
The drive is designed for installation as a free-standing device.
4.1 UNPACKING AND INSPECTION
When packaged for shipment, the RK06 or RK07 in its H969 cabinet weighs approximately 148 kg
(325 lb). Although excessively heavy and bulky for single-person handling, the package does not require the use of a forklift or similar equipment for moving or lifting. Table 4-1 lists the special tools
and test equipment rc~quired for the RK611 /RK06 or RK611 /RK07 Disk subsystem installation.
CAUTION
When moving or lifting the RK06 or RK07, always
grasp the frame structure. Do NOT hold any part of
the top or side covers.
Table 4-1

Special Tools and Test Equipment for RK611/RK06
or RK611/RK07 Disk Subsystem Installation

Part No.

Part

RK6/7-TA
RK6/7-TB
RK06K-AC
RK07K-AC
93··06058
93··06240
93··06076
93·,06087
93-06122-01
93-06122-02
93-06123-01
93-06123-02
93-06171
29-19633
29-20065
29-20906
29-21487
29-21832
29-22526
93-06056

Field Test Box (115 Vac, 50 or 60 Hz)
Field Test Box (230 Vac, 50 or 60 Hz)
Alignment cartridge (RK06)
Alignment cartridge (RK07)
Wrench
Glass alignment tool
Head installation/removal tool
Head alignment tool
Shim, 0.003 in.
Shim, 0.00015 in.
Shim, 0.020 in.
Shim, 0.015 in.
Head separation tool
Mirror
Magnifier
Torque wrench
Head torque wrench
Hex ball driver set
Screwdriver, 14.5-in., Phillips
Carriage roll alignment tool
4-1

The procedure for unpacking the disk drive is as follows:
1. When delivered, the RK06 or RK07 is mounted on a shipping skid and covered by a cardboard carton (Figure 4-1). First, remove the two plastic straps that hold the carton secure to
the skid.
2.

Remove the lid from the top of the cardboard carton.

3. Remove the staples that fasten the three wooden slats to the bottom flanges of the cardboard-overlay carton. Also remove the staples from that portion of the carton that extends
to the bottom of the skid.
4. Visually inspect the cabinet for outward signs of shipping damage. Retain the original packing materials and receipts in case any claims are filed for shipping damage. All damage
claims should be promptly filed with the transportation company involved. DIGITAL
should be notified immediately of any such claims.
5. Remove the rear and front covers; visually inspect for shipping damage.
6.

Inspect for loose cable connectors, terminal points, and subassemblies.

Inspect the logic backplane in both (X and Y) axes for bent pins.

8. Pull rearward on the logic assembly and power supply assembly. Both assemblies will open
90 degrees for access to components. Ensure that all logic modules are in proper locations
(refer to Figures 4-2 and 4-3).
9. Check the cable connectors and modules in the logic assembly (card nest) to ensure tightness.
10. Check the power supply assembly for loose cables, connectors, and connection of air distribution system bellows (Figure 3-2) and proper installation of the motor release pin.
11. Inspect the absolute filter area for proper attachment of hoses. Clean the foam prefilter if
necessary.
12. Check the head cam area to ensure that the head arms are engaged properly on the head
cam.
13. If any problems are encountered during execution of the above procedures, refer to the
RK06 / RK07 Disk Subsystem Service Manual for corrective procedures.
14. Close the hinged power supply and card nest assemblies and replace all access covers.
15.

Remove both bolts securing the cabinet to the shipping skid.

16. Remove the skid blocks and position the cabinet leveling feet so that the cabinet rests on its
casters.
17. Remove the cabinet from the shipping skid, using a ramp or other suitable device. At this
point there are no other shipping restraints for the drive.
WARNING
Use care in moving the drive/cabinet.

4-2

FULL TELESCOPE CAP
(9906446)

6-PANEL FOLDER
(9906976)

CRATING SLAT
(7606868)

CUSHIONED

11-4979

Figure 4-1

RK06/RK07 Shipping Package Configuration

4-3

TRACK POSITION SENSOR

M7108

8~-----------------------------------------------------------------------------------------i

SERVO ANALOG PCB M7729

I
I

7~----'_--~--~r_----------~--~-.~r_-----------------'--~-----------r--~----------r_--~

RUNL-JMAINT VEL.POTI@ I
SAFETY
~

t-----~

SW2

SERVO CONTROL LOGIC

MAINTL-JRUN
SERVO RELEASE
SW2

M7107

6t-----~--------~~--~~--------_r--r_--~--,_~r_----~r_,_--------------------------r_----~
UNLD HDS

U BRUSHES

t~ USPEED OKI

SPEED OK POT
DUAL-PORT DRIVE CONTROL LOGIC

M7130

(SEE NOTE 2)

4t-----,-----~~__----_,~~------_r--r_----------------~r_~----------------,_--~----r_----~

SEIZED

REQUESTED

INTERFACE & TIMING LOGIC M7106

INV ~NORMI
PORT BADDR
I NV SW

DRIVEY.LECT A

INTERFACE & TIMING LOGIC (M7106)(SEE NOTE 1)

I
1..-_--1

(SEE NOTE 1)

DRIVE SELECT B

I
I

CABLE BOARD

I
~----------------------------~I~

()

oI-

()

w
Z

NOTES:
1. TWO M7706 BOARDS USED WITH DUAL ACCESS ONLY; OTHERWISE
ONL V ONE IS USED AS FOLLOWS:
IN SLOT 2 WHEN B ACCESS IS SELECTED.
IN SLOT 3 WHEN A ACCESS IS SELECTED.
2. DUAL PORT LOGIC USED ONLY WITH DUAL ACCESS OPTION.

()

I-

:::>

()

I-

:::>

;::::
:::>
Q.

~-----------------()~~

l!:

CP-2456

Figure 4-2

RK06 Disk Drive Logic Module Locations

4-4

TRACK POSI1'ION SENSOR M7908

1
I

SERVO ANALOG PCB M7906

I ~OFFSET VE~~ 1

RUNL-JMAIINT

ADJ

SAFETY
SW2

MAINT L..jRUN

ADJ

SERVO RELEASE
SW 1

SERVO CONTROL LOGIC M7907
6

00 0

0000000000
0
:z4

OFFSET

2°

UNLD HDS

OBRUSHES

~ OSPEEO OK I

RTZ

'--.,.-I

LD
HDS

FWD

SEIZED

REQUESTED

II
II
II I I

INVLJNORM
PORT B ADDR
INVSW

INTERFACE & TIMING LOGIC M7706 (SEE NOTE 1)
I

SPEED OK POT
DUAL PORT DRIVE CONTROL LOGIC M7730 (SEE NOTE 2)

REV

HOS

DRIVE CONTROL LOGIC M7705

ORIVE'dLECT A

INTERFACE & TIMING LOGIC M7706 (SEE NOTE 1)

DRIVE SELECT B

CABLE BOARD

I
I

(DC POWER CONN)

I0
J2

POR!
OUT

o I~

a::

I-

0
PO~:I 0

01
01

POR~I
OUT

PO~!

I
NOTES:
1. TWO M7706 BOARDS USED WITH DUAL ACCESS ONL Y; OTHERWISE
ONL Y ONE IS USED AS FOLLOWS:
IN SLOT 2 WHEN B ACCESS IS SELECTED.
IN SLOT 3 WHEN A ACCESS IS SELECTED.
2. DUAL PORT LOGIC USED ONL Y WITH DUAL ACCESS OPTION.

z
z

u
I-

::J
~

I-

::J

Q
I-

::J
~

MA·0923

Figure 4-3

RK07 Disk Drive Logic Module Locations

4-5

4.2 SAFETY PRECAUTIONS
When performing the installation procedures in Paragraph 4.3', observe the following precautions to
avoid injury to personnel or damage to the equipment.
1.

Keep fingers and hands out of the area between the cartridge and the carriage while drive
power is ON.

Always remove ac power cables when it is necessary to work inside the drive.
CAUTION
Always remove wristwatch before servicing the drive
(linear-motor electromagnetic fields are destructive
to watches).

Use only nonmagnetic tools near the recording area of the disk cartridge. Use care when
working in the linear-motor area with ferrous materials. Keep read/write heads away from
the el~ctromagnetic area.

Avoid touching or blowing breath on read/write heads. Skin acids can etch and ruin heads;
breath can cause condensation deposits that disfigure the gliding surface.

Never move the carriage assembly forward manually without a spinning disk in place or the
head-separator tool installed.

Never remove or change modules without shutting down all internal drive power.

4.3 INSTALLATION PROCEDURE
The procedures for installing the RK6I1 Controller and the RK06 or RK07 Disk Drive are covered in
the following paragraphs.
4.3.1 RK611 Controller
To install the RK6Il Controller in a PDP-II expander box:
1.

Remove the RK6I1 from its shipping container.

Install the RK611 backplane into the designated expander box and secure with the screws
provided in the accessories bag. Install power harnesses (2), using expander box instructions.
NOTE
For RK07 installations, the M7901 must be above
ETCH E. If not, FCO no. M7901-003 must be
added. (Step 3 is FCO no. M7901-003.) If the
M7901 is above ETCH E, skip step 3. For all RK06
installations, skip step 3.

If the subsystem contains any RK07 Disk Drives, verify that the RK611 Controller has been
modified to operate with RK07 Drives as well as with RK06 Drives. To make this verification, ensure that the following changes have been made to the M790 I module:
a.
b.
c.
d.

On Side 2, the etch at pin Ell-II has been cut.
On Side 2, the etch at pin E12-5 has been cut.
On Side 1, a wire has been added, connecting Ell-IO to Ell-II.
On Side 1, a wire has been added, connecting EI2-5 to EI7-2.

4-6

Table 4-2 RK611 Control Switch Configurations
for Desired Unibus Address and Interrupt Vector
Switch Sectiol1l

Unibus Bit

Switch Section

Unibus Bit

Sl-l
SI-2
Sl-3
Sl-4
Sl-S
Sl-6*
Sl-7*
Sl-8

A12
All
A10
A09
A08
A07
A06
AOS

S2-1
S2-2
S2-3**
S2-4
S2-S
S2-6
S2-7**
S2-8

NA
D08
D07
D06
DOS
D04
D03
D02

* For the standard address of 777440, these switches should be in the ON position; all others
should be in the OFF position. Note that this address overlaps the Look-Ahead Register
in the RC 11 Disk Subsystem.

** For the standard vector of 210, these switches should be in the OFF position; all others
should be ON. Note that this vector overlaps the interrupt vector in the RCII Disk
Subsystem.
.

Set the swit,~h configuration for the desired Unibus address and interrupt vector into the
M7900 module. Use Table 4-2 as a guide when making these settings.

If the controller is to be connected to a PDP-11/40 system, remove the M7234 from the
processor. Remove WS on the M7234 and replace the module in the backplane. This jumper
must be removed to successfully run the RK06 or RK07 diagnostics.

Insert the modules into the RK611 backplane using Figure 4-4 as a guide.

Attach the transition connector assembly (PN-70-1241S-0-0) to the cabinet containing the
RK611 using the nuts and bolts provided.

Connect the flat cable (PN-BC06R-06) from the M7904 module to the Berg connector side
of the transition connector. Refer to Figure 4-S for placement of the striped edge of the
cable.

If the RK06 or RK07 subsystem is to be dual ported, repeat this procedure for the second
controller.

NOTE
Check for a different address and vector on the second controller.

4··7

I
I

UNIBUS IN

I
I
I

I
I
I
I

I
II

I
I

UNIBUS OUT

G727

t
I

G727

M7903

M7902

I
I

M7900

I
J

M7901

M7904

I
I

I
I
I

I
I
J
I

I
I

11-4981

Figure 4-4

RK611 Controller Backplane (from Module Side)

4-8

CABLE CONNECTOR FROM
RK06 OR RK07 DRIVE

LEFT
CAB UPRIGHT (REF)

ALTERNATE TRANSITION
BRACKET MOUNTING
POSITION
(SEE NOTE 3)

STRIPED
EDGE
(REF)

TRANSITION
BRACKET

I/O CABLE ROUTE FOR HORIZONTAL
MOUNTED CONTROLLER (BAll-K)
(SEE NOTE 2)

I/O CABLE ROUTE FOR VERTICAL
MOUNTED CONTROLLER (BAll-Fl

NOTES:
1. FOR INSTALLATION IN BA11·F EXPANSION BOX:
al CUT THE WRAP AND UNFOLD POWER HARNESS.
bl ADD EXTENSION POWER CABLE TO SHORTEN THE HARNESS.
2. WHEN INSTALLED IN BAll·K EXPANSION BOX. THE I/O CABLE
SHOULD BE TURNED90° AND ROUTED UP OUT OF THE BOX
AS SHOWN.
3. THE ALTERNATE BRACKET MOUNTING POSITION MAY BE USED
WITH ANY EXPANSION BOX TO SIMPLIFY INSTALLATION.
WHEN THE ALTERNATE MOUNTING POSITION IS USED. THE
BERG CONNECTOR IN THE TRANSITION BRACKET MUST BE
INVERTED SO THAT THE CABLE FROM THE RK06 OR RK07
DRIVE WILL HANG IN A DOWNWARD POSITION AS SHOWN.

RK611 CONTROLLER

'1-4984

Figure 4-5

RK611 I/O Connector Details

4-9

4.3.2 RK06 or RK07 Disk Drive
To install the disk drive:
1.

Roll the drive cabinet to its designated location. Level the cabinet by lowering the four
leveling feet attached to the cabinet until all weight is removed from the casters.

Remove the front and rear access covers from the drive.

Verify that the model number listed on the serial tag (Figure 4-6) corresponds to the site
power requirements. Refer to the voltage ranges given in Figure 4-6, and if a discrepancy
exists, correct it before continuing with the installation.

Connect a ground wire (PN-7212827-25) from the ground stud on the H969 cabinet to the
ground stud on the cabinet containing the RK611 Controller (Figures 4-8 and 4-9).

Lower the card nest assembly to its horizontal (service) position and install a drive bus cable
(PN-70-12292-xx) between the transition connector (PN -70-12415-0-0) of the controller cabinet and I/O connector Jl (A-IN) of the RK06 or RK07 (Figure 4-7 and Table 4-3). The
standard drive cable length from controller to drive is 7.62 m (25 ft). However, optional
cable lengths can be used if the total drive bus does not exceed 30.48 m (100 ft). (See Table 43 for part numbers.)
Table 4-3

Drive Bus Cables

Part No.

Length

70-12292-8
70-12292-12
70-12292-25
70-12292-40

2.44 m (8 ft)
3.66 m (12 ft)
7.62 m (25 ft)
12. 19 m (40 ft)

On some drives, there are two rows of 12 zero ohm resistors on the M9016 module. They
represent the last three digits of the drive serial number and must be clipped to give the BCD
representation of these three digits. If they have not already been clipped, remove the top
resistor of a pair to represent zero, or remove the bottom resistor for a one, for each of the
12-bit locations. Also, there is a pair of zero ohm resistors, representing the drive type,
placed to the left of the serial number resistors. If neither has been removed, clip the top one
if the drive is an RK06 or the bottom if the drive is an RK07.
If the M9016 module does not have these resistors, the serial number must be wired to the
back plane. To do this:
a.

Set up a table similar to this:
Table 4-4

Sample Table for Backplane Wiring

800 400 200 100

A
0
6
R
2

A
1
4
S
1

B
1
4
M
2

A
1
4
T
2

B
1
4
D
1

B
1
4
A

A
1
4
V
1

A
1
4
V
2

Row 1
Row 2
A
1
4
p
1

A
1
4
R
1

A
I
4
S
2

4-10

1
4
C
1

Backplane
Pin
Assignments

For a serial number ABC, enter the BCD equivalent of digit A in Row 1, Column A;
enter the BCD equivalent of digit B in Row 1, Column B; enter the BCD equivalent of
digit C in Row 1, Column C.

Enter the complement of Row 1 in Row 2.

For every "1" in Row 1, wire the appropriate backplane pin (below Row 1 and Row 2)
to pin D6H 1 (+3 V). For every "1" in Row 2, wire the appropriate pin to A6Tl
(ground).

If the drive is an RK06, wire 01 VI to AITI (ground). If the drive is an RK07, wire
01Vl to D6Hl (+3 V).

If the installation is multidrive, proceed to step 8; otherwise, proceed to step 10.

Connect drive bus cable (PN-70-12292-xx) from J2 (A OUT) of the previous drive to Jl (A
IN) of the next drive. Repeat for each additional drive. The standard length between drives
is 2.44 m (8 ft); however, optional length cables are available. Total drive bus length must
not exceed 30.48 m (l00 ft).

Daisy chain each drive, using a ground wire (PN-7212827-8) between the ground studs in the
drive cabinets.

[Q! DODDI

UNIT-SELECT PLUG

RK06-XA OR RK07-XA
RK06-XB OR RK07-XB
RK06-XC OR RK07-XC
RK06-XD OR RK07-XD

115 VAC/60 HZ
230 VAC/60 HZ
115 VAC/50 HZ
230 VAC/50 HZ
11-4980

Figure 4-6 Serial Tag Location and Model Identifications
(Tag Visible with Front Cover Removed)

4-11

DC POWER CABLE

CNPl

CNP2

CNP3

CNP4

CNP5

CNP6

CABLE BOARD
(M7706)

J2 A OUT AND
TERMINATOR

J4 BOUT

Jl A IN AND

J3 BIN

7897-14

Figure 4-7

RK06/RK07 I/O Connector Details

4-12

CABLE CHART
FROM
CONNECTION

REMARKS

TO
WITH

CONNECTION

WITH

CONTROLLER

J1A IN

DRIVE

CONTROLLER

J3B iN

DRIVE

J2A OUT

DRIVE

NEXT DRIVE

DRIVE

J4B OUT

DRIVE

NEXT DRIVE

DRIVE

CABLE. GROUND (25 FT)~
THIS GROUND STRAP IS
NOT REQ'D IF BOTH CONTROLLERS ARE MOUNTED IN THE
SAME EXPANDER BOX
OR CAB.

OR
TERMINATED
(SEE NOTE)

CABLE. DRIVE BUS (25 FT)
THE TOTAL LENGTH OF ALL DRIVE BUS
CABLES ATTACHED TO AN RK611 CONTROLLER
MUST NOT EXCEED 100 FEET.

FOR INSTALLATION OF
TRANSITION BRACKET
AND CABLE ORIENTATION.
REFER TO FIGURE 4-4

\_u_n_/ I ~~---~J

EXPANDER BOX USED FOR MOUNTING
CONTROLLER 15 LOCATED IN SEPARATE
CAB FROM DRIVE AND CAN BE
POSITIONED AT ANY VERTICAL
HEIGHT WITHIN THAT CAB.

I
I

t==l

CABLE. GROUND (25 FT)

VIEW C-C

POWER SUPPLY
(REF)

v.,oJ<.;q U,.

v..?~ .q I~ U,..............
~_( IF THIS 15 THE LAST DRIVE IN A SERIES,
l

)'--

A TERMINATOR (DEC PART NO: 70-12293-00)
IS INSERTED IN THESE CONNECTIONS

SEE CABLE
CHART
(VIEW A-A)
RK06 OR RK07 DISK DRIVE
(WITH REAR SKIN REMOVED
AND CARD NEST OPEN)
114985

Figure 4-8 Dual-Access
Disk Subsystem H969 Cabinet

4-13

POWER CORD

~5 FTI ~

CABLE, DRIVE BUS (25 FTI

CABLE, GROUN

CAB"'"jUN
D13FTLGI

CABLE, GROUN D (3 FT LG)

~~-----~

VIEW B-B
SCALE: NONE

POWER CORD'

THIS GROU
REQ'O IF B~~H STRAP IS NOT
::E MOUNTED

~NO~TROLLERS

ME EXPAN
HE
OR CAB.
DER BOX

11-4983

Figure 4-9

Dual-Access Disk Subsystem H969 Cab'met
4-14

10.

For the last drive, connect a drive bus terminator (PN-70-12293-0-0) to I/O 12 (Figure 4-7).
NOTE
For the dual-controller cabling configuration (Figure
4-7), the installation of the drive bus cable should follow the above sequence. The drive connectors used
are J3 (8 IN) and J4 (8 OUT). A second drive bus
terminator (PN-70-12293-0-0) must be connected to
J4 (8 OUT) of the last drive. The second controller
may be mounted in the same expander box as the
first or in another cabinet altogether. If the same box
is used, a second transition connector is mounted in
the cabinet to accommodate the second drive bus
cable.

11. Ensure that circuit breaker CBl is OFF. Connect the power cable to a power outlet whose
voltage and frequency meet the requirements of the drive model (Figure 4-6).
12.

Insert the proper drive-select plug into the unit select jack of the operator control panel
(Figure 4-6). Table 4-5 lists the part numbers for each of the eight drives in the daisy chain
configuration.
Table 4-5

Unit-Select Plugs for RK06/RK07 Disk Drives
Plug No.

Part Number
12-12691-0
12-12691-1
12-12691-2
12-12691-3
12-12691-4
12-12691-5
12-12691-6
12-12691-7

o
1
2
3
4
5
6
7

4.4 INSTALLATION CHECKS AND ADJUSTMENTS
To verify performance and adjust the disk subsystem:
1.

Check controller power supply voltage.

Check drive power supply voltage.

Check head load/unload.

Check head alignment.

Check response to abnormal conditions.

The required actions are detailed in Paragraphs 4.4.1 through 4.4.5.

4-15

4.4.1 Controller Power Supply Voltage Checks
To perform the required voltage checks on the controller power supply, set controller power to ON
and make measurements at the following points:
Voltage
+5 Vdc ± 0.25 V
-15 Vdc ± 0.75 V
GND
+15Vdc±0.75

Test Point
A04A2
A04B2
A04C2
A04D2

If any voltage is out of tolerance, refer to the appropriate expander box manual for adjustment and/or
repair procedures.

4.4.2 Disk Drive Power Supply Voltage Checks
To perform the required voltage checks on the disk drive power supply:
1.

Ensure that the power cable is connected to a power outlet of the proper voltage and frequency rating.

Ensure that the RUN/STOP switch is in the STOP position.

Set circuit breaker CB 1 to the ON position. The air system should start operating and the
lid-lock solenoid will energize to enable the lid.

Check the five output voltages at the logic backplane:
Backplane
Location

Voltage

B08A2
DOIE2
C08D2
C08B2
A08R2

+5Vdc±3%
+25 Vdc ± 10%
+15Vdc±5%
-15 Vdc ± 5%
-5Vdc±3%

If any voltage is not within tolerance, refer to the RK06/ RK07 Disk Subsystem Service Manual for
adjustment and/or repair procedures.
4.4.3 Head Load/Unload Checks
The following head load/unload checks are performed during the POWER ON sequence to ensure
that operational status can be attained:

With power ON, install a scratch cartridge into the drive.

Depress the appropriate access select switch, either A or B.

Press the R UN/STOP switch. After approximately one second, the spindle will begin to
rotate and the R UN/STOP light will extinguish.
NOTE
Power must be applied to the controller (or to the
Field Test Box, if used) during performance of the
load/unload checks.

4-16

When the disk reaches nominal speed (after approximately 30 to 60 seconds), the heads will
load onto the disk surfaces and begin a slow movement toward the inner limit. After reaching this limit, they then return to cylinder zero. When head motion has stopped, the
READY light on the operator control panel should be lit. The FA UL T light should not be
lit.

Press the RUN/STOP switch again. The READY light should extinguish and the heads
move to their fully retracted position. When the drive has stopped, the RUN/STOP indicator will light. The drive motor should stop in approximately 20 seconds after the
RUN/STOP switch is released.

4.4.4 Head Alignment Checks and Cartridge Care
Several principles should be understood before attempting any alignment procedures. A familiarity
with these principles will hopefully discourage any "shortcuts" leading to possible damage or improper alignment.
1. Whenever head alignment exceeds ±200 microinches maximum, the heads should be realigned to within ±50 microinches.
2.

Power must be applied to the drive for two hours prior to performing any head alignment
procedures.

3. The cartridge should be mounted and spinning for a minimum of one-half hour prior to
beginning head alignment.
4. Head aligmnent techniques assume that the drive being aligned is a properly working device.
If the device is not working properly, do not attempt alignment.
5. To protect the alignment pack, Write Protect should always be set when the alignment pack
is in the machine.
6. The SAFETY SWITCH on RK06 module M7729 or RK07 module M7906 (near the D end)
forces Write Lock and disables the Servo Unsafe circuit. It is important to remove this
special maintenance switch from its maintenance mode to re-enable the Servo Unsafe circuit.
7. Linear forcc!s applied to the carriage during adjustments may cause the heads to unload and
cause damage if the SAFETY SWITCH is not set to the MAINT position.
8. Lateral fOf(~e on a carriage can cause the heads to tilt and possibly crash. Therefore, all
procedures requiring the application of a lateral force to the carriage must be performed
with the heads unloaded.
9.

Positioner parts and fixtures are precision parts and must be handled carefully. The use of
torque wrenches set to specified levels is mandatory to guarantee that tolerances will hold
and thus prevent damaged parts.

10. When transporting cartridges by auto, always keep the cartridges in the passenger compartment - never in the trunk.
11. Always transport cartridges in their carrying case.

4-17

4.4.4.1 Head Alignment Using Program Control- To perform head alignment checks under program
control:

1. Remove the rear cover from the RK06 or the RK07.
2. Ensure that power is removed from the RK06 or RK07 and the Field Test Box.
3.

Connect the head alignment cable from the Field Test Box to the RK06 or RK07
Read/Write board (Figure 4-10 and Figure 4-11).

4. Apply power to the RK06 or RK07 and the Field Test Box.

HEAD ALIGNMENT
CABLE CONNECTOR

HEADO
HEAD 1

MA0986

Figure 4-10

RK06 Head Assembly Area

4-18

HEAD ALIGNMENT
CABLE CONNECTOR

HEADO
HEAD 1

MA-0985

Figure 4-11

RK07 Head Assembly Area

Load MAINDEC-ll-DZR6N and start at address 224.
Per program instructions, install an RK06K-AC or RK07K-AC alignment cartridge, depending on the type of drive, and WRITE LOCK the drive.
NOTE
Prior to begining head alignment procedures, power
should be applied to the drive for two hours; the cartridge should be mounted and spinning for at least
one-half hour.
To WRITE LOCK the drive, set the SAFETY SWITCH (S2) on the RK06 M7729 board or
the RK07 M7906 board to the MAINT position. This action will also prevent the heads
from unloading if a Servo Unsafe condition occurs.

4-19

Respond to program questions 1 through 5 as illustrated in the sample printout in Paragraph 4.4.4.2.

Mount the head alignment fixture on the appropriate head (Figures 4-10 and 4-11).
CAUTION
Make sure that the carriage does not tip beyond the
point where the heads will fall out of the cam tower.

8. Torque the head alignment fixture mounting screw to 5 in/lb. Loosen the head mounting
screw and then torque it to 2 in/lb. Turn both vertical screws of the head alignment fixture
fully CCW.
9

Load the heads by typing the letter R.

10. Set the torque wrench to 5 in/lb. Tighten the forward vertical screw to move the head
forward or the rear screw to move the head toward the rear. Since these screws cause oposite
actions, the screw not in use must be turned fully CCW. Adjust until:
a.

The red LED on the meter is ON (steady state).

The meter on the Field Test box reads 0 ± 50 microinches.
NOTE
If the needle deflects to the right of zero, move the
head toward spindle; if to the left of zero, move away
from spindle.

11.

Unload the heads by typing the same head number again. Tighten the head to 5 in/lb.

12.

Reload the heads by typing the letter R.

13. Check that the conditions of steps lOa and lOb are met.
14.

Unload the heads by typing the next head number.

15.

Remove the head alignment fixture and mount it on the next head.

16.

Repeat steps 8 through 15 for each data head.

17. When all heads have been aligned, restart the program under its Exerciser mode and exercise
the drive for two minutes. Recheck the alignment of all heads.
18.

Remove the alignment cartridge and the Field Test Box; clear the WRITE LOCK condition.

19.

Replace the rear cover removed in step 1.

4-20

4.4.4.2

Sample Printout

DZR6N-D - RK611/RK06-RK07 SUBSYSTEM VERIFICATION

PART 2

*** RK06-07 HEAD ALIGNMENT AID ***
FOR HELP TYPE H, ELSE CR
K
INSTRUCTIONS FOR USING RK06-RK07 HEAD ALIGNMENT AID :
***************************************************
MOUNT AN RK06 OR RK07 ALIGNMENT CARTRIDGE ON THE DESIRED

DRIVE, AND INSURE THAT THE DRIVE IS WRITE-LOCKED.
CONNECT THE ALIGNMENT INDICATOR TO THE DESIRED DRIVE,
VIA THE HEl~D ALIGNMENT CABLE ONLY, AND CYCLE UP THE
DRIVE. AFTER MOUNTING THE PACK ON THE DRIVE,
THE OPERATOR SHOULD WAIT 30 MINUTES FOR THE DRIVE
TEMPERATURE TO STABILIZE, BEFORE PROCEEDING WITH
ALIGNMENT.
RESPOND TO ALL REQUESTS FOR PARM1ETERS, BY ENTERING
THE DESIRED PARAMETER VALUE (NO <CR') NEEDED).
THERE ARE ~rwo MODES OF OPERATION : MANUAL MODE
ALLOWS SELI~CTION OF DRIVES AND HEADS BY TTY INPUT,
AND AUTO MODE ALLOWS DRIVES AND HEADS TO BE SELECTED
BY OFF-ON OPERATION OF DRIVE PORT SELECT SWITCHES.
IN EITHER ~10DE, UP TO 5 MINUTES OF SEEK EXERCISES
MAY BE REQUESTED FOR EACH DRIVE.
ALSO IN EI,]~HER MODE, A VERIFY OPERATION ALLOWS HEAD
SELECTION ~iITHOUT THE UNLOADING AND LOADING OF
THE DRIVE BY THE PROGRAM, WHICH OTHERWISE OCCURS
TO ALLOW MJl~NIPULATION OF THE ALIGNMENT TOOL.
TO RESTART EITHER MODE, TYPE A Z •
TO RESTART ALIGNMENT AID, TYPE ~ R •
TO SELECT NEW DRIVES IN MANUAL MODE, TYPE A C •
FOR HEAD AI,l:GNMENT PROCEDURE, REFER TO FIELD
TEST BOX (FtK06-07TA, RK06-07TB) OPERATOR'S MANUAL.
MANUAL OR ~.UTO MODE (M OR A)?
M

* MANUAL SELECT MODE *

ENTER DRIVE NO.

(0-7):

DRIVE SER. NO.

ALIGN, VERIFY OR EXERCISE (A, V, OR E)?
A

4-21

* MANUAL SELECT ALIGNHENT *
ENTER HEAD NO.

(0-2)

TYPE < R':> WHEN READY:
R

HEADS POSITIONED AT CYLINDER 365 (OCT)
HEAD 0 SELECTED
ENTER HEAD NO.

(0-2)

TYPE < R... WHEN READY:
R

HEADS POSITIONED AT CYLINDER 365 (OCT)
HEAD 1 SELECTED
ENTER HEAD NO." (0-2)
2

TYPE <R') WHEN READY:
R

HEADS POSITIONED AT CYLINDER 365 (OCT)
HEAD 2 SELECTED
ENTER HEAD

NO. (0-2) :

ALIGN, VERIFY, OR EXERCISE (A, V, OR E1 ?
E
TYPE <R,) WHEN

READY

*RANDOM SEEK EXERCISES IN PROGRESS ON DRIVE 0
AZ
ALIGN, VERIFY, OR EXERCISE (A, V, OR E) ?
V

* MANUAL SELECT VERIFY *
ENTER HEAD NO.

(0-2) :

HEADS POSITIONED AT CYLINDER 365 (OCT)
HEAD 2 SELECTED
ENTER HEAD NO. (0-2) :
~Z

4-22

4.4.4.3

Head Alignnrlent Using the Field Text Box

Remove the rear cover from the RK06 or RK07.

Remove power from the RK06 or RK07 and from the Field Test Box.

Connect the head alignment cable from the Field Test Box to the RK06 or RK07 read/write
board.

4. Attach the appropriate drive interface cable to the Field Test Box.
5. Apply power to the RK06 or RK07 and to the Field Test Box.
6. Install an RK06K or RK07K alignment cartridge, depending on the drive, and WRITE
LOCK the drive.
NOTE
Prior to beginning head alignment procedures, power
should be applied to the drive for two hours; the cartridge should be mounted and spinning for at least
one-half hour.
To WRITE LOCK the drive, set the SAFETY SWITCH (S2) on the RK06 M7729 board or
the RK07 ~f7906 board to the MAINT position. This action will also prevent the heads
from unloadling if a Servo Unsafe condition occurs.
7.

Mount the head alignment fixture on the appropriate head (Figures 4-10 and 4-11).
CAUTION
Make sure that the carriage does not tip beyond the
point where the heads will fall out of the cam tower.

8. Torque the head alignment fixture mounting screw to 5 in/lb. Loosen the head mounting
screw and then tighten it to 2 in/lb. Turn both vertical screws of the head alignment fixture
fully CCW.-9.

Select the dlesired head and load to the actual cylinder address that corresponds to the
underlined (for an RK06 alignment) or dotted (for an RKQ7 alignment) cylinder address
switches on the Field Test Box. Refer to the switch settings that fo1low.
Switch

Position

EXERCISE/STATUS

EXERCISE

DRIVE SELECT

Desired Drive#

COMMAND TO RK06/7 switches

SEEK
22 SECTORS
CLEAR ERROR & ATTENTION
SET VOLUME VALID

SINGLE CYCLE/CONTINUOUS

SINGLE CYCLE

FUNCTION

SEEK ONLY

4-23

Switch

Position

SYNC

INT

CLOCK

FAST

HALTON ERROR

ADDRESSING
(CYLINDER)
(UNDERLINED SWITCHES)
(DOTTED SWITCHES)
(HEAD)
(HEAD ADDRESS SWITCHES)
(SECTOR)

SWR
245,0 (for RK06)
496,0 (for RK07)
SWR
DESIRED HEAD#
ALL

PARITY (both switches)

ODD

Depress the INIT and START switches.
To change heads, simply change the head address switch setting and depress the START
switch.
10. Set the torque wrench to 5 in/lb. Tighten the forward vertical screw to move the head
forward or the rear one to move the head to the rear. Since these screws oppose each other,
the screw not in use must be fully CCW. Adjust until:
a.
b.

The red LED on the meter is ON (steady state).
The meter on the Field Test Box reads 0 ± 50 microinches.

NOTE
If the needle deflects to the right of zero, move the
head toward spindle; if to the left of zero, move away
from spindle.

11.

Unload the heads.

12. Tighten the head to 5 in/lb.
13.

Load the heads and check that the conditions of steps lOa and b are still met.

14.

Unload the heads; remove the head alignment fixture and mount it on the next head.

15. Load the heads and repeat the procedures of steps 8 through 15 until all data heads are
aligned.
4.4.5 Response to Abnormal Conditions
The following procedures are for handling abnormal conditions encountered during a system installation.
4.4.5.1 Clearing an Unsafe Condition - Clearing of an unsafe condition can be accomplished by
releasing the RUN/STOP switch on the front panel. Press the RUN/STOP switch again. If the fault
no longer exists, the FAULT light will go off. When the READY light is on, the drive is ready for
further operations.

4-24

4.4.5.2 Removing a Cartridge When Drive Power is Lost - If power to the driver is lost and it is
necessary to remove the disk cartridge, use the following procedure:
1.

Insert a small, flat-blade screwdriver into the opening above the door latch.

While pushing on the screwdriver, lift the pack loading door.

Ensure that the brush drive knob is turned fully clockwise.

Ensure that the head loading arms are fully retracted.

Remove the disk cartridge in the usual manner.

4.4.5.3 Detecting Hlead-to-Disk Interference - The read/write heads actually fly over the disk surface
at an altitude of approximately 50 millionths of an inch. When for any reason the read/write heads
touch the disk surface (a condition referred to as "head crash"), damage to either heads or disk surface
is possible. If there is an audible ticking sound after the heads load or if the heads show an oxide
buildup on the gliding surface when inspected, a head crash has probably occurred. The following
procedure should be followed for recovery from a head crash:
1.

Clean the oxide buildup from the head, using a head cleaning kit.

Reinspect the head, using an inspection mirror. If any scratches or oxide are visible, the head
must be replaced. Refer to the RK06/ RK07 Disk Subsystem Service Manual for the appropriate procedure.

If a head shows no visible scratches or oxides, install a scratch pack and load the heads onto
the pack as in the head load/unload checks.

When the drive is ready, cycle the drive down by setting the RUN/STOP switch and
reinspect the heads in question. If any oxide buildup is present, a catastrophic condition
exists. Refer to the RK06/ RK07 Disk Subsystem Service Manual for corrective procedures.
CAUTION
Never use a cartridge involved in a head crash until it
has been inspected and found usable; i.e., there are
no visible marks on the platter surfaces. Catastrophic crashes can be propagated from drive-to-drive
under such conditions.
Never use an alignment cartridge in a drive that has
experienced a head crash. Always use a scratch cartridge for testing.

4.5 POWER CONVERSION
The RK611/RK06 or RK611/RK07 Disk Subsystem is configured during manufacture to be either a
50-Hz or 60-Hz unit. (See Paragraph 1.4.) Although it is possible to change the frequency characteristics of the drive in the field, this requires such parts substitution as the power supply transformer and
spindle motor pulley, plus a belt change.
Changing of voltage range (120 Vac to 240 Vac or vice versa) merely requires a circuit breaker change
(CBl at the rear of the drive).

4-25

4.6 DUAL ACCESS DRIVE INSTALLATION AND ACCEPTANCE
4.6.1 Dual Access Installation Procedure
Site preparation and environmental considerations for the RK06 or RK07 dual access disk drive are
the same as those for the single access drives and are discussed in Chapter 3 of this manual. The
adjustment and installation procedures for the RK06 or RK07 dual access drive are also identical with
those of the single access drive, except for the added cables for dual access. Refer to the single access
installation procedures described earlier in this chapter. Figure 4-12 is a dual access cable diagram.
4.6.2 Dual Access Acceptance Procedure
4.6.2.1 Dual Access Diagnostic - The Dual Access diagnostic (DZR6GA.X) must be run first to
establish that a port can seize and release the drive. To accomplish this, perform the following procedures:
1.

On each drive to be tested, the dual access test switch located on module M7730 must be
placed in the invert position for the duration of this test. Refer to Figure 4-2 or 4-3 for the
location of this switch. When in the invert position, this switch grounds bit 0 of the port B
unit select lines and causes each drive to appear as two separate units to the RK611. The unit
select switch will define the address of port A. Port B address will respond to the address of
A + 1. Since each drive will appear as two units to the RK611, a maximum of four drives
can be tested by this program at one time.

Use only even numbered UNIT SELECT SWITCHes on the drive operator panel. Any
other drives on the system which have an address in conflict with any of the test addresses
must be deselected (both ports switched off). The even numbered UNIT SELECT SWITCH
on any drive can be removed and inserted into the current drive under test.

For this test, the RK611 is daisy chained to both ports of the RK06 or RK07. Refer to
Figure 4-13 for the cable diagram.

Run the Dual Access diagnostic (DZR6G.X) to test dual access seizure and release.

After running the Dual Access diagnostics, return the dual port invert switch on module
M7730 to its normal position. Return any UNIT S ELECT SWITCHes borrowed for this
test to their former drives.

Recable the RK06 or RK07 Drives for normal operation, as shown in Figure 4-12.

4.6.2.2 RK611, RK06, and RK07 Diagnostics - With the drives cabled for normal operation, run all
RK611, RK06, and RK07 diagnostics through both ports.
4.6.2.3 RK06 and RK07 Performance Exerciser - To run the RK06 and RK07 Performance Exerciser
(DZR6P.X), all disks must first be written with a predetermined data pattern through one of the
systems. This is accomplished by starting the Performance Exerciser at address 200. The exerciser will
automatically write all disks on the drive bus and begin testing data transfers immediately, without
operator intervention. While the first system is undergoing testing, the second system can be entered
into the performance tests by starting the exerciser at address 240. This will skip over the disk write
portion of the test and allow the operator to manually begin the performance tests on the second
system. Both systems will then be undergoing performance testing simultaneously.

4-26

CONTROLLERS A AND B MAY BE OPERATED FROM
THE SAME UNIBUS OR FROM lWO SEPARATE UNIBUS

UNIBUS

) 1.-_ _ _ _ _---,

v
RK511
CONTROLLER B

RKS11
CONTROLLER A

DRIVE BUSA

DRIVE BUS B

TERMINATORS

r'l

rl r1

r1 rl r1 r1

r1 n r1 rl

rl n

OUT

J4
OUT

OUT

J4
OUT

DRIVE UNITO

DRIVE UNIT1

DRIVE UNIT 2

DRIVE UNIT3

J4
OUT
B

DRIVE UNIT4

DRIVE UNIT5

OUT

DRIVE UNIT6

J4
OUT
B

DRIVE UNIT7

11-5508

Figure 4-12 RK611/RK06 or RK61l/RK07
Dual Access Cable Diagram

4-27

CHAPTER 5
FIELD ACCEPTANCE
TEST PROCEDURE

Field acceptance testing is intended to demonstrate performance of the RK06 or RK07 Disk Drive
and/ or the RK611 Controller to the customer prior, to his acceptance of the equipment.
5.1 ERROR DEFINITIONS/RATES
The RK611 has five: registers that display the various error conditions that can exist in the Disk
subsystem. Table 5-1 lists the various error indications and gives an explanation of their significance.
Errors are broken down into five categories:
I.
2.
3.
4.
5.

Control
Cartridge
Seek
Hard errors
Soft errors

These classifications aid in the identification of the various errors encountered in acceptance testing
only.
5.1.1 Control Errors
A control error indicates an operational aberration in the execution of command and control functions. Any control error detected by a diagnostic program during acceptance testing indicates a hardware malfunction that must be corrected before testing can continue.
Table 5-1

RK611/RK06 and RK611/RK07 Disk Subsystem Diagnostic Errors

Error

Indicator Bit

Condition

Error Type

Programming Error
(PGE)

RKCS2 bit 10

Control

Illegal Function
(lLF)

RKER bitO

Illegal command in low-order
5 bits ofRKCSl.

Control

Format Error
(FPER)

RKER bit 4 or 5

Format or drive type error.

Control

Cylinder Overflow
(CaE)

RKER bit 9

Cylinder address exceeded.

Control

Invalid Disk
Address (lDAE)

RKER bit 10

Invalid cylinder or track
address detected.

Control

U nit Field Error
(UFE)

RKCS2 bit 8

Drive selection problem.

Control

5-1

Table 5-1

RK611/RK06 or RK611/RK07 Disk Subsystem Diagnostic Errors (Cont)

Error

Indicator

Bit Condition

Error Type

Unibus Parity Error
(UPE)

RKCS2 bit 13

Error in 16-bit data word from
main memory.

Control

Nonexistent Drive
(NED)

RKCS2 bit 12

Drive response problem.

Control

Drive Available
Interrupt (DAINT)

RKDS bitO

Drive is not available
(dual-access only).

Control

Drive Error
(DRERR)

Message BO:T
(FAULT)

Any 1 of 21 drive error
conditions.

Control

Bad Sector Error
(BSERR)

RKER bit 7

Data transfer attempt to/from
bad sector.

Cartridge

Header VRC Error
(HVRC)

RKER bit 8

Header check, indicates bad
header information.

Hard/Soft

Operation
Incomplete (OPI)

RKER bit 13

Desired header cannot be
found.

Seek

DTC Parity Error
(DTC PAR)

RKCSI bit13

Parity error in message from
drive.

Control

Write Check Error
(WCE)

RKCS2 bit 14

Write check, indicates data
from disk did not match data
from memory.

Hard/Soft

Nonexistent
Memory (NEM)

RKCS2 bit 11

No SSYN within 10 f.1S of
MSYN assertion.

Control

Data Late Error
(DLTERR)

RKCS2 bit 15

Data late to/from silo.

Control

Drive Timing Error
(DTE)

RKER bit 12

Write clock loss during write;
data loss during read.

Control

Data Check (DCK)

RKER bit 15

Data error detected by ECC.

Soft

Multiple Drive
Select (M DS)

RKCS2 bit 9

More than one drive
simultaneously selected.

Control

Error Correction
Hard (ECH)

RKER bit6

Data error uncorrectable
using ECC.

Hard

Controller
Time-Out (CTO)

RKCSI bit 11

Command possibly not
executed.

Control

Seek Incomplete
(SKI)

RXER Bit 1

Seek operation did not
complete.

Seek

5-2

5.1.2 Hard Errors
)\ny failure to read data correctly after a complete recovery sequence l with ECC enabled constitutes an
irrecoverable, i.e., "hard" error. Errors that are not ECC-correctable include: (1) bursts greater than
>LI bits in length and, (2) isolated dropped bits that are separated by more than 11 bits within a sector.
The allowable error rate for hard errors is one error per 10 12 bits read.
5.1.3 Soft Errors
A soft error is define:d as any read error that is correctable by ECC and/or a complete recovery
sequence. The allowable error rate for soft errors is one error per 109 bits read.
5.1.4 Cartridge Errors
A cartridge error caus,ed by imperfections in the recording surface is regarded as cartridge-attributable
or media-dependent. If the imperfection is less than 11 bits long, it is ECC-correctable and will appear
as a soft error. If longer than 11 bits, it will appear as a hard error. On a given cartridge, attributable
errors will always appl~ar at the same cylinder, sector, and track addresses. The definitions of hard and
soft errors (see 5.1.2 and 5.1.3) apply only to randomly distributed errors and do not take into account
errors that are cartridge-attributable. Imperfections in the cartridge surface may be found by mapping
the cartridge by means of the formatter program.
5.1.5 Seek Errors
Any positioning operation that terminates with se(~k incomplete, or with the positioner in an incorrect
location, constitutes a. seek error. The allowable error rate for seek errors is one error per 106 seek
operations.
5.2 DIAGNOSTIC TESTING
When all installation procedures have been completed, the tests detailed in the following subsections
should be performed. Table 5-2 briefly describes the diagnostic programs used in field acceptance
testing of the RK611 /RK06 or RK611 /RK07 Disk subsystem. The total time required to run these
tests (except DZR6Q2) in the absence of failures is 2.5 hours per drive.
5.2.1

RK611 Controller

5.2.1.1 General - In diagnostic testing, the RK611 is tested first. All five of the diskless controller
diagnostics (Table 5-2, DZR6A-DZR6E) and the RK611 functional controller diagnostic (DZR6K in
the Table) must run for two error-free passes before the remainder of the acceptance program can be
carried out. Any failure detected by any of the diagnostics must be corrected, and all programs must be
rerun before proceeding.

I A complete recovery sequence consists of 28 retries, 16 of which are at the nominal head position and 2 each at
offsets of ±400, ±800, and ± 1200 microinches (for the RK06) or ±200, ±400 and ±600 microinches (for the
RK07), respectively.

2The run time of this diagnostic depends on the number of drives being tested.

5-3

Table 5-2

Diagnostic Programs Used in RK611/RK06 or RK611/RK07
Disk Subsystem Field Acceptance Tests

Program

Function

MAINDEC-l1-DZR6A
RK611 Diskless Controller Diagnostic
Part 1

Reads and writes every RK611 register.
Tests the interrupt mechanism.
Tests the SILO loading logic. (Note: No drive is
required for the execution of this program.)

MAINDEC-II-DZR6B
RK611 Diskless Controller Diagnostic,
Part 2

Tests the loading of drive bus messages by executing Class A commands. Some Part 1 tests execute
commands that are partially maintenance mode
and partially at normal speed in order to "fool" the
controller and force errors (no drive is required).

MAINDEC-II-DZR6C
RK611 Diskless Controller Diagnostic
Part 3

Tests the loading of the drive bus message shift register for Class B commands.
Tests index and sector pulse detection.
Tests SILO and NPR transfers from memory in 16and I8-bit mode.
Tests nonexistent memory and Unibus parity error
detection.
Tests Read and Write MFM loopback.
Tests Class B instruction errors (no drive is required).

MAINDEC-II-DZR6D
RK6II Diskless Controller Diagnostic
Part 4

Tests loading of drive bus message shift registers
for Class C commands.
Tests header generation for search operations.
Tests write data NPR transfers to SILO.
Tests header recognition.
Tests cylinder, track, and sector increment after
successful header search.
Tests detection of all header-type errors.
Tests ECC generation and writing.
Tests partial sector write (zero fill).
Tests 18-bit format ECC generation and data write
(no drive is required).
5-4

Table 5-2

Diagnostic Programs Used inRK611/RK06 or RK611 or RK611/RK07
Disk Subsystem Field Acceptance Tests (Cont)

Program

Function

MAINDEC-II-DZR6E
RK611 Diskless Controller Diagnostic
Part 5

Tests multisector data transfers.
Tests midtransfer seeks.
Tests cylinder overflow checking.
Tests NPR transfers to memory.
Tests ECC error detection and correction in I6-and
18-bit modes.
Tests write check in both 16- and I8-bit modes and
forces.

MAINDEC-II-DZR6G
Unibus RK06-RK07 Dual Port Drive
Diagnostic (available Q4, FY78)

Tests write check errors (no drive is required).
Ensures that port request, port seize, and timeout
function correctly.
Ensures that the release function operates correctly.
Ensures that the timer inhibiting logic operates
correctly.
Ensures that data transfers function correctly
through both parts.

MAINDEC-II-DZR6H
Unibus RK06-RK07 Drive Diagnostic
Part 1

Ensures that the drive can perform all static and
cycle-up tests.
Ensures that the drive can write and read headers
in both 20- and 22-sector formats.
Ensures that the disk can perform seek operations
by doing several seek patterns.
Checks error-detection logic by software error
forcing.

MAINDEC-II-DZR61
Unibus RK06-RK07 Drive Diagnostic,
Part 2

Ensures that the disk is capable of performing read
and write data operations in both 20- and 22-sector
formats.
Performs worst case patterns, spiral writing and
reading, and all offset operations.
Reports unloading and loading times.

5-5

Table 5-2

Diagnostic Programs Used in RK611/RK06 orRK611/RK07
Disk Subsystem Field Acceptance Tests (Cont)

Program

Function

MAINDEC-ll-DZR6J
Unibus RK06-RK07 Drive Diagnostic,
Part 3

Ensures that the disk is capable of performing all
operator intervention functions properly.
Checks error-detection logic by manual and by
software forcing of errors.

MAINDEC-II-DZR6K
RK611 Functional Controller Diagnostic

Tests those areas in the controller that cannot be
tested in a diskless environment.
Tests those areas of the drive that cannot be tested
until controller operation in a diagnostic or
maintenance mode has been tested.

MAINDEC-II-DZR6L
RK06K-RK07K Cartridge Formatter

Writes and verifies header and data information on
an RK06K or RK07K disk pack at all possible
disk pack addresses.
Uses the Bad Sector File to: (1) report the serial
number or the cartridge being formatted, (2) determine whether the cartridge is an alignment cartridge, and abort the program if it is, and (3)
identify the sectors that are to be marked "bad."

MAINDEC-II-DZR6M
RK611/RK06-RK611/RK07 Subsystem
Verification, Part 1

Provides a functional shakedown of the entire subsystem, including the Unibus interface and access
to main memory.
Employs worst case situations·'involving mechanical positioning, disk addressing, and data transfer.
Measures drive operational timing.
Provides numerous options that may be specified
by the user.
Reports errors as they occur.

MAINDEC-II-DZR6N
RK61 I /RK06-RK611 /RK07 Subsystem
Verification, Part 2

Contin ues functional shakedown provided by Part
1.
Employs worst case situations involving head offsetting, memory addressing and data transfer,
Unibus cycle contention, and multiple-drive operations.
Provides RK06 or RK07 head-alignment aid.
Provides numerous options that may be specified
by the user.
5-6

Tabl,e 5-2

Diagnostic Programs Used in RK611/RK06 or RK611/RK07
Disk Subsystem Field Acceptance Tests (Cont)

Program

Function

MAINDEC-I1-DZR6P
RK611/RK06-RK611/RK07 Performance
Exerciser

Exercises in a random overlapped routine from one
to eight RK06 or RK07 Drives connected to a
common Unibus controller in a dedicated standalone mode.
Generates read, write, and write/write check commands.
Reports system errors upon occurrence.
Maintains performance statistics on each drive.

MAINDEC-I1-DZR6Q
RK06-RK07 Drive Compatibility

Verifies the capability of each drive to write data
that can be successfully read by all other drives and
to completely overwrite data written by all other
drives.
Detects: (1) head misalignment, (2) positioner lateral misalignment, (3) spindle/cartridge interface
runout, (4) improper levels of write current, and (5)
incorrect addressing of read/write heads.
Prints summary of compatibility test results for
each drive.
Reports unexpected errors as they occur.

MAINDEC-I1-DZR6R
RK611/RK06-RK611/RK07 User
Defined Test

Provides a capability for entering, editing, saving,
recalling, and executing test programs designed by
the user.
Provides an interactive command set for use in entering, storing, retrieving, editing, and executing
tests.

5.2.1.2

Procedure
NOTE
In the following steps, repair or replace the failing
module; rerun the program that detected the failure;
then return to step 1.

Load diagnostic DZR6A; run two passes with no errors.

Load DZR6B; run two passes with no errors.

Load DZR6C; run two passes with no errors.

Load DZR6D; run two passes with no errors.

5-7

Load DZR6E; run two passes with no errors.

Load DZR6K; run two passes with no errors.
NOTE
Any failures occurring during program DZR6K may
not necessarily be limited to the RK611 Controller.
Thus, it should only be run on a known good drive.
CAUTION
Diagnostic program DZR6K modifies headers on
certain tracks. Failure to halt the program, either
during the "End of Pass" printout or by typing
CTRL C may leave the pack formatted incorrectly
and cause inaccurate error reporting in subsequent
testing. If the formatting is left modified, the
RK06K/RK07K Cartridge Formatter Program
(DZR6L) must be run (using Mode 1) before any
further testing can be performed.

At this point in the testing, the RK611 is functionally checked out. The only RK611 test not performed
at this time is the Data Reliability Test.
RK06 or RK07 Disk Drive

5.2.2

5.2.2.1 General - This section of the diagnostic procedure covers the testing of the RK06 or RK07
Drive. With a high degree of confidence, any failures detected in these tests can be attributed to the
Drive only. All programs listed in Paragraph 5.2.2.2 must run through the specified number of errorfree passes before performing the remainder of the program. Any failure detected by any of the diagnostics in this section must be corrected and all programs rerun before proceeding.
5.2.2.2

Procedure
NOTE
In the following steps, repair or replace any failing
component; rerun the program that detected the failure to verify the correction; then return to step 1.

Load diagnostic DZR6H; run two passes with no errors.

Load DZR6I; run two passes with no errors.

Load DZR6J; run one pass with no errors.

Load DZR6M; run one pass with no parameters out of the specified limits (printed by the
diagnostic).

Load DZR6N; run two passes with no errors.

Load DZR6G; run two passes with no errors.

5-8

5.2.3 Cartridge Intterchangeability Test
This test requires the use of at least two drives. If there is only one drive on the system being tested, it
will be necessary to use another system containing an RK06 or RK07 to complete this test.
Load and run diagnostic DZR6Q. If any drive fails this test, make any necessary adjustments and
rerun the program.
5.2.4

Data Reliability Testing

5.2.4.1 General - Data reliability testing consists of running diagnostic DZR6P to exercise the entire
RK611/RK06 or RK611/RK07 Disk subsystem for at least one hour. At the end of this period, the
statistics for each drive must be analyzed to determine whether or not the subsystem is to be accepted.
In analyzing the stat.istics for each drive, no control, seek, or hard errors are allowed during the testing.
Refer to Paragraph 5.1 for error definitions. The occurrence of any of these errors is sufficient to reject
a drive. If it is determined that an error is due to the cartridge, it is sufficient to reformat the cartridge,
making sure that the failing sector is flagged bad. The Data Reliability Test may then be rerun.

If the excessive errors are caused by the drive, required corrective action must be performed and the
acceptance restarted with the RK06 or RK07 diagnostic (Paragraph 5.2.2). If the errors are caused by
the controller, the acceptance must be restarted with the RK611 diagnostic testing (Paragraph 5.2.1)
after the corrective action has been taken.
5.2.4.2 Procedure -- Load and start diagnostic DZR6P. Write the cartridge on all drives on the subsystem. After one hour of running, obtain the statistics for all drives. If the total number for all drives
of "WORDS READ*65K" is less than 967, allow the subsystem to run for an additional 15 minutes
and again obtain th,e statistics. Repeat until the total "WORDS READ*65K" is greater than 967. If
the total number of SOFT DATA ERRORS is zero or one, accept the subsystem. Ifit is three or more,
the failing component (drive or controller) must be fixed. If there were two SOFT DATA ERRORS,
continue running the subsystem for an additional hour. At this point, obtain statistics and continue
running until the total "WORDS READ*65K" is greater than 1934. If the total SOFT DATA ERRO RS is still two, accept the subsystem. If the total is three or more, the failing component (drive or
controller) must be fixed and acceptance restarted.

5-9

CHAPTER 6
OPERATOR'S GUIDE

6.1 INTRODUCTION
The following material describes the function of each of the external controls available to the user of
an RK06 or RK07 Disk Drive and provides procedural information for subsystem operation.
6.2 OPERATOR CONTROLS/INDICATORS
The RK06 or RK07 Disk Drive contains the following power and control switches and indicators
(Figure 6-1).
Rear Panel (Circuit Breaker):
• AC power ON"jOFF switch with POWER indicator.
Front Panel (Control Panel):
• Combination RUN jSTOP switch with a STOP indicator.
• UNIT SELECT switch with a READY indicator.
• FAULT indicator.
• WRITE PROT switch with a PROTECT indicator.
• ACCESS A ENABLE switch with a SELECT A indicator.
• ACCESS B ENABLE switch with a SELECT B indicator.
6.2.1 Power ON/OFF Circuit Breaker
When the three-wire plug is inserted into an outlet, ac power is applied to the rear panel circuit breaker
on an RK06 or RK07 Disk Drive, and the red indicator lamp is illuminated. When the circuit breaker
is switched to the 01'~ position, ac power is applied to the drive and the blower motor is energized.
6.2.2 RUN/STOP Switch with STOP Indicator
The STOP indicator is illuminated during STOP conditions when the heads and brushes are home, the
spindle is completely stopped, and the spindle motor is not energized. When the alternating action
pushbutton is initially pressed (RUN) under these conditions, the spindle motor is energized, the
STOP indicator is extinguished, and the heads are loaded. When these initial operational conditions
are met (STOP indicator extinguished and heads loading), a RUN condition is entered. However, if
the heads do not load (due to a malfunction), the spindle motor is immediately de-energized and the
STOP indicator remains illuminated.
If the pushbutton is released (STOP) when the hc~ads are loaded, the heads will be unloaded and the
spindle motor de-energized. When these conditions are met (heads home and spindle stopped), the
STOP indicator is again illuminated.

6-1

I f the RUN /STO P switch is in the RUN position and the drive has been cycled down (heads unloaded
and spindle completely stopped) due to an Unload command, a Start Spindle command from the
controller can be used to restart the spindle and reload the heads. However, if the cycle down condition
occurred as the result of an error, a Start Spindle command is not required, since the heads will
immediately load when the error is cleared.
Finally, since the RUN/STOP switch has a mechanical memory, a power interrupt followed by power
restoration will cause the drive to automatically cycle up if the switch is currently located in the RUN
position.
NOTE
The drive cover (lid) is interlocked. This prevents the
lid from being raised when the STOPlight is extinguished (RUN) or drive power is off.

--,
I
LREARPANEL
CIRCUIT BREAKER
POWER ON/OFF SWITCH - - - ; . - -

AC INDICATOR

FRONT PANEL
CONTROL PANEL

fRUNl

WRITE

~~ READY

PROT

A B
ACCESS B ENABLE SW.
WITH ACCESS INDICATOR
(ON = DUAL ACCESS
B PORT SELECTED)

. RUN/STOP SW. WITH
STOP INDICATOR
(ON =STOP
OFF = RUN)

ACCESS A ENABLE SW.
WITH ACCESS INDICATOR
( ON = SINGLE·ACCESS DRIVE
SELECTED OR DUAL·ACCESS
A PORT SELECTED)

UNIT SELECT SW. WITH
READY INDICATOR
(ON = READY)
FAULT INDICATOR
(ON = FAULT)

WRITE PROTECT SW. WITH
PROTECT INDICATOR
( ON =WR ITE PROTECT)
7897·11

Figure 6-1

RK06 or RK07 Disk Drive Power and Control Switches

6-2

6.2.3 UNIT SELECT Switch with READY Indicator
The UNIT SELECT switch is a cam-operated assembly that is actuated by the insertion of a plastic
plug. The face side of each actuator plug contains a single number (0-7), while the other side consists of
raised plastic cam actuators that are separately configured to encode the plug with the number. With
this arrangement, the: insertion of a plug into any UNIT SELECT switch assembly will automatically
encode the associated! plug number when power is applied to the drive. When the heads of a drive are
settled (not in motion), the numbered indicator on the UNIT SELECT switch assembly associated
with the drive is illuminated (READY).
6.2.4 FAULT Indicator
The FAULT indicator is illuminated whenever one of the following error conditions is detected in the
drive:
• More than one drive selected.
• Positioner, when detented, has moved too far (e.g., the drive has been jarred).
• A parity error has occurred in a message transmitted from the controller to the drive.
• A read/write unsafe condition in the drive (e.g., serve track error signals, write gate signal
errors, etc.)
• A write lock error condition (i.e., the receipt of a Write Gate signal when the drive is in Write
Protect mode).
• Low ac voltage in the drive.
• An incomplet,e Seek operation.
• The receipt of Write Gate or Seek signals while Volume Valid is reset.
6.2.5 WRITE PROT Switch with PR OTECT Indicator
If a Write Gate signal from the controller is not currently asserted in a selected drive, initial depression
of the WRITE PROT pushbutton will immediately set the drive in Write Protect mode, and the
PROTECT indicator will be illuminated. When the alternating action pushbutton is again pressed,
write protection is removed and the PROTECT indicator is extinguished. However, if a Write Gate
signal is asserted in the selected drive when the switch is initially pressed, write protection will be
inhibited and the PROTECT indicator will remain extinguished until the Write Gate signal is negated.
Once the Write Gate signal is negated, the Write Protect mode is entered and the PROTECT indicator
is illuminated.
6.2.6 ACCESS A ENABLE Switch with SELECT Indicator
The ACCESS A ENABLE switch is an alternating action pushbutton that is found on both singleaccess drives (Access A or Access B) and drives containing the dual-access option (Port A and Port B).
6.2.6.1 Single-Acces!!t Drive (A) - If Port A is configured for single access (i.e., the M7706 Interface
and Timing Module is in Slot 3 of the drive card cage) and the switch is initially depressed, bidirectional communication with the controller is enabled via Port A. Under these conditions, when the
drive is selected by the controller, the SELECT A indicator is illuminated. When the switch is again
depressed (released), access by the controller is disabled and the SELECT A lamp cannot be illuminated.

6-3

6.2.7 ACCESS B ENABLE Switch with SELECT Indicator
The ACCESS B ENABLE switch is an alternating action pushbutton that is found on both singleaccess drives (Port A or Port B) and drives containing the dual-access option (Port A and B).
6.2.7.1 Single-Access Drive (B) - If the B port is configured for single access (i.e., the M7706 Interface
and Timing module is in Slot 2 of the drive card cage) and the switch is initially pressed, bidirectional
communication with the controller is enabled via Port B. Under these conditions, when the drive is
selected by the controller, the SELECT B indicator is illuminated. When the switch is again pressed
(released), access by the controller is disabled and the SELECT B lamp cannot be illuminated.
6.2.7.2 Dual-Access Drive (A and B) - If the ACCESS B ENABLE switch is independently pressed on
a dual-access drive, the operations performed are similar to those previously described. However, if
both the ACCESS B and ACCESS A switches are initially pressed, arbitration logic will determine, on
a priority basis, which port (B or A) will be accessed by its associated controller (B or A).
6.3 OPERATING PROCEDURES
This material describes RK06K and RK07K Cartridge loading and cycle-up procedures that are required to place an RK06 or an RK07 Disk Drive on-line. The procedure assumes that ac power is
available (red indicator lamp on the rear panel circuit breaker is on), the ac circuit breaker is on
(blower motor is energized), and the STOP indicator lamp is illuminated on the control panel. If these
conditions are met, initiate the following.
NOTE
An RK06K Cartridge may not be used on an RK07
Drive, and an RK07K Cartridge may not be used on
an RK06 Drive.
6.3.1

RK06K and RK07K Cartridge Loading
1.

Depress the drive lid release bar and raise the lid (Figure 6-2).

Prepare an RK06K or an RK07K Cartridge (Figure 6-3) for loading as follows:
a.

Lift the cartridge by grasping the top cover handle with the right hand (Figure 6-4).

Support the cartridge with the left hand by holding the protection cover handle (Figure
6-5).

Lower the top cover handle and push the handle slide to the left with the thumb of the
right hand. Again raise the handle to its full upright position to release the protection
cover (Figures 6-4 and 6-5).

Lift the cartridge from the protection cover and place it in the drive shroud with the top
cover handle recess facing the rear of the machine (Figure 6-6).

Rotate the top cover handle a few degrees clockwise and counterclockwise to ensure that the
shroud locating studs are properly seated within the cartridge housing detent slots.

Gently lower the top cover handle to a horizontal position to engage the drive spindle, and
place the protection cover over the top cover (Figures 6-7 and 6-8).

Close the lid.

6-4

Depress one or both of the ACCESS A and B ENABLE switches, depending on the port
capability.

if write protection is required, press the WRITE PROT switch.

Press the RUN /STOP switch (RUN). If the lid is properly closed and no drive errors exist
(F A ULT extinguished), the spindle wi}] turn and the STOP indicator will be extinguished in
approximately one second.

When the drive has completed the start spindle sequence and the heads are detented on Cylinder 0, the
READY indicator on the numbered UNIT SELECT switch will be illuminated.

SHROUD

BRUSH INDICATOR
DIAL
READ/WRITE &
SERVO HEADS
BRUSH ARM ASS'Y

PACK-ON
SWITCH

SPINDLE

LATCH

OPERATOR
SWITCHES
UPPER
BEZEL
LOWER
BEZEL
8174-3

Figure 6-2

Typical Disk Drive with Lid Released

6-5

TOP COVER

TOP COVER
HANDLE

HANDLE
SLIDE

PROTECTION
COVER

8365-3

Figure 6-3

Typical Disk Cartridge

6-6

TOP COVER
HANDLE

HANDLE
SLIDE

8089-4

Figure 6-4

Engaging Lock Release

6-7

PROTECTION
COVER
8089-3

Figure 6-5

Releasing Protection Cover

6-8

LID RELEASE
BAR

8365-1

Figure 6-6

Positioning for Load

6-9

Figure 6-7

Cartridge Locked in Shroud

6-10

8089·1

Figure 6-8

Installing Protection Cover

6-11

6.3.2

RK06K and RK07 Cartridge Unloading
1.

Release the R UN/STOP switch (STOP).

Wait for the STOP indicator to illuminate.

Press the lid release bar and raise the lid.

Remove the RK06K or RK07K Cartridge as follows:
a.

Remove the protection cover and hold the cover in the left hand.

Push the handle slide to the left before raising the top cover handle (Figure 6-9).

Raise the top cover handle to a full upright position to release the cartridge from the
drive spindle.

Lift the cartridge up and out of the shroud, and place it in the protection cover.

Lower the top cover handle to the horizontal position to lock the protection cover in
place.

Figure 6-9

Engaging Lock Release

6-12

CHAPTER 7
PROGRAMMER'S GUIDE

7.1 INTRODUCTION
Since the Unibus provides a unified addressing structure in which device control, status, and data ..
registers are directly addressed as memory locations, all operations on these registers are performed by
normal memory reference instructions. Under these conditions, the RK06 or RK07 subsystem does
not require a special class of input/output instructions to initiate the transfer of information or the
manipulation of data.
With the exception of a Read Header command, data transfers use the nonprocessor request (NPR)
facility of the Unibus for direct access to main memory, at a maximum transfer rate of 232,000 words
per second (4.3 J.ls per word), in clock sizes ranging from I to 65,536 words. To achieve a smooth flow
of data from a selected disk to main memory, the subsystem utilizes a first in/first out, 66-word data
buffer known as a Silo.
A phase-locked clock system (PLO) and modified frequency modulation (MFM) recording technique
add reliability to reading and writing operations regardless of minor changes in the rotational speed of
a disk. In addition, program-controlled head offset positioning can correct any slight mechanical
misalignment that may occur between read/write heads and a disk cartridge by allowing a head to be
positioned about the track centerline in incremental steps. Also, error detection and correction hardware (ECC) provides adequate information for correcting any error burst of up to 11 consecutive bits
within a sector
For increased reliability, parity checking is performed on the control information that is transferred
between the drives and the controller. In addition, the controller also detects and flags data errors from
the Unibus. Upon the indication of an error condition or the completion of a command, the controller
can interrupt the processor. Thus extensive error indicators exist for on-line diagnosis, while numerous
status indicators provide complete program control.
On multidrive systems, positioning operations can be overlapped (i.e., Seek operations may be performed on one or mon;~ drives before starting a transfer on another drive). Thus, while one drive is
reading or writing, othe:rs can be positioning to a new cyclinder for the next transfer.
The following paragraphs provide detailed descriptions of controller register content and usage, followed by disk command descriptions and programming considerations related to the use of an RK06
or RK07 subsystem with a PDP-II processor.

7-1

7.2 RK611 DEVICE REGISTERS
There are 15 usable, 16-bit device registers contained in the RK611 that are used to interface the
controller with the RK06 or RK07 Drives and the PDP-II Unibus (Figure 7-1). The 15 registers are
loaded and/or read under program control to initiate selected disk commands and monitor subsystem
status and error conditions. Device register bits are generally cleared by a Unibus Initialize (lNIT),
Controller Clear (CCLR), or Subsystem Clear (SCLR) operation. In the following descriptions (unless
otherwise specified), it should be understood that the clearing of a bit by anyone of these three
methods is implied.
NOTE
The RK611 does not recognize DATOB (MOVB,
BI CB, etc.) bus cycles. All registers must be written
as words.
~

7.2.1 Control/Status Register 1 (RKCSl)
The RKCS 1 register (Figure 7-2) can be read or written via program control and as such is used to
store the current disk command function code and operational status of the controller. In addition, the
register can initiate command execution and enable a Controller Clear operation.
Go (GO) - Bit 0 - When the GO bit is set, the disk command Function Code (FI-F4) is executed. With
the GO bit set, only two other device register bits can be set (DiagnoStic Mode excepted), as follows:
• Controller Clear (CCLR), bit 15 in RKCS 1, may be set via program control to initialize (general clear and preset) certain device registers within the controller. However, any status and/ or
error conditions set in the drives are not affected.
• Subsystem Clear (SCLR), bit 5 in RKCS2, may be set via program control to initialize both the
controller and all of the drives.
When command execution is completed, the GO bit is reset and the controller is ready to accept a new
command. However, the GO bit cannot be set if the Combined Error (CERR) bit is set. When CERR
is set, the execution of a command can only occur following the initiation of a CCLR.
Function Code (FI-F4) - Bits 1-4 - The configuration of the Function Code bits (FI-F4), with the
setting of the GO bit, allows the selected drive to respond to the following command control configuration.
Command
Select Drive
Pack Acknowledge
Drive Clear
Unload
Start Spindle
Recalibrate
Offset
Seek
Read Data
Write Data
Read Header
Write Header
Write Check

o
o
o
o
o
o
o
o
1

1
1
1
1

1
1

o
o
o
1

o
o
o

1
1

o
o
o
o
1

o
o
1
1

Spare Bit - Spare bit 5 can be written and read back.

7-2

Fl·

GO Octal
1 01
1
1 03
o
1 05
1
1 07
0111
1
1 13
o
1 15
1
1 17
o
1 21
1
1 23
o
1 25
1
1 27
o
1 31

CONTROL ANDl\ STATUS REGISTER

DESI RED CYLI NOER REGI STER

UNIBUS

r-~~__- ' ' -__~__- '____~__~____T-__~____r -__- r__- , , -__~__~~~~__~__--,ADORESS

RKDC

777440

I I

UNIBUS
ADDRESS

R/W
0

1777460

WORD COUNT REGISTER
UNUSED
777442

BUS ADDRESS REGISTER

RKBA

~__~__~~__~__~____~__~____~__-+____~__~____~__~__~____~__~____~I
DATA BUFFER

RiW

R/W

777444

01 SK ADDRESS (TRACK a SECTOR) REG.

RKDA

777464

MAINTENANCE REGISTER

RKMR1

R/W

777446
8 7

777466

8 7

CONTROL AND STATUS REGISTER 2

RKCS2

777462

ECC POSI TION REG I STER

RKECPS

R/W

777450

777470

DRIVE STATUS REG I STER

RKDS

READ ONLY

ECC PATTERN REGISTER

RKECPT

777452
15

8 7

ERROR REG I STER

RKER

777472
15

ATTENTION SUMMARY AND OFFSET

RKAS/OF

R/W

RO
1

1777476

77747
•

MAl NTENANCE REGI STER 3

RKMR3
777456

MAINTENANCE REGISTER 2

RKMR2
777454

00
11-4199

Figure 7-1

RK611 Device Registers

7-3

Interrupt Enable (IE) - Bit 6 - When the Interrupt Enable (IE) bit is set, the controller will be allowed
to interrupt the processor under any of the following conditions:
• When Controller Ready (ROY), bit 7 in RKCS 1, is set upon completion of a command.
• When any drive sets an associated Attention (ATNO-ATN7) bit (8-15) in RKASjOF, with the
set condition of Controller Ready (ROY).
• When any drive or the controller indicates the presence of an error by the setting of Controller
Error (CERR), bit 15 in RKCS1.
In addition, via program control, an interrupt can be forced by the simultaneous setting of the Interrupt Enable (IE) and Controller Ready (ROY) bits in RKCS 1. Finally, Interrupt Enable, bit 6, can be
reset via program control as well as by conventional initialization (INIT, CCLR, SCLR).
Controller Ready (ROY) - Bit 7 - Controller Ready (ROY) is effectively a read-only bit. However, the
bit can be externally set via conventional initialization (INIT, CCLR, SCLR) or internally set upon
completion of a command. The ROY bit is reset when GO, bit 0 in RKCS 1, is set.
Extended Bus Address (BAI6, BAI7) - Bits 8, 9 - The Extended Bus Address bits reflect Unibus upper
address bits 16 and 17 and as such are an extension of the 16-bit RKBA register which contains the
memory address required for the current data transfer.
Controller Drive Type (COT) - Bit 10 - This bit specifies the type of drive that will be selected by the
controller. To specify RK06 ~isk Orives, the bit must be reset. For an RK07 Orive, the bit must be 1.
Controller Time-Out (CTO) - Bit 11 - Controller Time-Out is a read-only error bit that is set to
indicate that GO, bit 0 in RKCS1, has been set for approximately 800 ms. Since this interval exceeds
the time required to execute the longest possible drive operation (i.e., a Seek from cylinder 410 to
cylinder 0 followed by a 65k word data transfer), the set condition of this bit indicates that the last
command has not been completed due to a malfunction.
Controller Format (CFMT) - Bit 12 - The Controller Format bit specifies the number of sectors for
each track in the selected drive. When bit 12 is set, a 20-sector format (18-bit data words) is defined for
the sector counter located in the drive, and when the bit is reset, a 22-sector format (16-bit data words)
is defined.
Drive-to-Controller Parity Error (DTC PAR) - Bit 13 - The OTC Parity Error is a read-only bit that is
internally set to indicate that a parity error has been detected in a serial message received from the
selected drive.
Drive Interrupt (01) - Bit 14 - In relation to program control, Orive Interrupt (01) is a read-only bit.
When set, the bit differentiates between a drive-initiated interrupt and a controller-initiated interrupt.
The 01 bit is set when any drive sets an associated Attention (ATNO-A TN7) bit (8-15 in RKAS-OF).
Thus, if the Interrupt Enable (IE) bit is set, the setting of the 01 bit with the set condition of Controller
Ready (ROY), bit 7 in RKCS 1, indicates a drive-initiated interrupt. The 01 bit is reset by Unibus
Initialize (INIT), Subsystem Clear (SCLR), or by the execution of Orive Clear commands to all drives
asserting Attention.

7-4

Combined Error/Controller Clear (CERR/CCLR) - Bit 15 - As a Combined Error (CERR) indicator,
bit 15 can be set by the controller, or anyone of the drives, to indicate that a subsystem error has
occurred (Table 7-1). However, when the bit is set via program control, a controller initialize (CCLR)
operation is enabled which clears the controller and results in the clearing of bit 15 itself. Thus, if the
bit is internally set (CERR) by an error that is followed by an external set (CCLR) to initialize the
controller, bit 15 will be cleared. However, since only controller errors will be initialized by CCLR, any
error originating in a drive will remain set in the drive.
CAUTION
When using a BIC instruction on the RKCSI register, ensure that a 1 is set in bit 15 of the mask. If this
is not done and CERR is set, a CCLR will occur, and
the RK611 will be cleated. For example, to clear the
interrupt Enable (IE) bit (bit 6 in RKCSl), the following instruction format is recommended:
BIC #100100, @RKCSI
Table 7-1

Combined Error (CERR)

Error

Indicator Bit

Condition

Programming Error (PGE)

RKCS2 bit 10

Illegal Function (ILF)

RKER bit 0

Illegal command in low-order 5 bits of
RKCSI.

Format Error (FTER)

RKER bit 4 or 5

Format or Drive Type error.

Cylinder Overflow (COE)

RKER bit 9

Cylinder address exceeded.

Invalid Disk Address (lDAE)

RKER bit 10

Invalid cylinder or track address detected.

U nit Field Error (UFE)

RKCS2 bit 8

Drive selection problem.

UNIBUS Parity Error (UPE)

RKCS2 bit 13

Error in 16-bit data word from main
memory.

Nonexistent Drive (NED)

RKCS2 bit 12

Drive response problem.

Drive Available Interrupt
(DAINT)

RKDS bit 0

Drive is not available (dual-access
only).

Drive Error (DRERR)

Message BO:T7
(FAULT)

Any 1 of 21 drive error conditions.

Bad Sector Error (BSERR)

RKER bit 7

Data transfer attempt to/from bad
sector.

Header VRC Error (HVRC)

RKER bit 8

Header Check indicates bad header
information.

7-5

Table 7-1

Combined Error (CERR) (Cont)

Error

Indicator Bit

Condition

Operation Incomplete (OPI)

RKER bit 13

Desired header cannot be found.

DTC Parity Error (DTC
PAR)

RKCSI bit 13

Parity error in message from drive.

Write Check Error (WCE)

RKCS2 bit 14

Write Check indicates data from disk
did not match data from memory.

Nonexistent Memory
(NEM)

RKCS2 bit 11

No SSYN within 10 JlS of MSYN assertion.

Data Late Error (DLTERR)

RKCS2 bit 15

Data late to /from Silo.

Drive Timing Error (DTE)

RKER bit 12

Write Clock loss during write, data
loss during read.

Data Check (DCK)

RKER bit 15

Data error detected by ECC.

Multiple Drive Select (M DS)

RKCS2 bit 9

More than one drive simultaneously
selected.

Error Correction Hard (ECH)

RKER bit 6

Data error un correctable using ECC.

Controller Time-Out (CTO)

RKCSI bit II

Command possibly not executed.

7.2.2 Word Count Register (RKWC)
The RK WC register (Figure 7-3) is loaded with the two's complement of the number of data words to
be transferred to or from main memory. The register is incremented by 1 after each bus cycle, and
accommodates a maximum transfer of 65,536 words. The RKWC register can only be cleared by
writing all zeros via program control.
7.2.3 Bus Address Register (RKBA)
The RKBA register (Figure 7-4) is initially loaded with the low-order 16 bits of the Unibus address
that will reflect the main memory start location for a data transfer. With the low-order bit (0) always
forced to 0, the RKBA register content is combined with high-order bits 8 and 9 of the RKCS 1 register
(BA 16, BAI7) to form a complete even-numbered word address. Following each data transfer bus
cycle, the register is incremented by two to select the next even-numbered location.
7.2.4 Disk Address Register (RKDA)
For Read, Write, and Write Check commands, the RKDA register (Figure 7-5) is initially loaded to
define the desired sector (1 of 20 or 22) and track (l of 3 read/write heads) on the selected drive, from
or to which the first block of a data transfer will be initiated. If the word count value indicates that a
block of more than 256 data words is to be transferred, the Sector Address bits of the RKDA register
will be incremented to select the next consecutive sector and the next track, if necessary, until a word
count overflow indicates that data transfers are completed or an error occurs. In either case, completion of the command is indicated by the setting of the Controller Ready (RDY) bit in RKCS 1 and
an increment of the RKDA register to the next sector location.
Sector Address Bits 0-4 (SAO-SA4)
Sector Address bits 0-4 are configured (00-23) to select a value (00-19 19 ) for a 20-sector format (18-bit
data words) or configured (00-25) to select a value (00-2119) for a 22-sector format (l6-bit data words).
The Sector Address is incremented by one when the sector has been transferred.
7-6

eS1

(RO)

BA1.1 RDY

(RO)

(RO)
11-4200

Figure 7-2

Control/Status Register 1

we
we [ I G J
wee we
15
14 ,____
13 L -__
12

we
11
__

we
10
__

we
~

09
__

we
~

08
__

___ J_ _ _ _

11-4201

Figure 7-3

Word Count Register

I ~;

00
SA
14

Si~

013

SA
07

SA
05

SA
01

00
11-4202

Figure 7-4

I 0

Bus Address Register

TA
2

SA
3

SA
2

I 4

11-4203

Figure 7-5

Disk Address Register

Spares - Bits 5-7 - Spare bits 5-7 are O.
Track Address (T AO-TA2) - Bits 8-10 - The Track Address bits are configured (0-2) to select the
appropriate read/write head associated with the dt::sired track. The Track Address bits are incremented
by one after the last sector (23 8 or 25 8) on the track has been transferred and the Sector Address has
been reset to zero. Similarly, if transfers continue beyond the last sector of the last track (2) of a given
cylinder, the Track Address is reset to zero and the Cylinder Address (in RKDC) is incremented by
one. In this manner, subsequent sectors, tracks, and cylinders can be consecutively transferred until the
word count equals zero. However, if the word count does not equal zero after the transfer of the last
sector (19 or 20 10 ) on the last track (2) of the last cylinder (410 10 for an RK06 or 81410 for an RK07), a
Cylinder Overflow Error (CaE) will occur.
Spares - Bits 11-15 - Spares bits 11-15 are O.

7-7

7.2.5 Control/Status Register 2 (RKCS2)
The RKCS2 register (Figure 7-6) can be read or written via program control and is used to store the
current drive select code, subsystem operational status, and Silo control information. In addition, the
register can initiate a Subsystem Clear (SCLR) operation.
15

CS2
~----------------v~--------------~
READ ONLY (RO)
11-4204

Figure 7-6

Control/Status Register 2

Dril'e Select (DSO-DS2) - Bits 0-2 - The Drive Select bits are configured (0-78) with the unit number
of the drive to be currently selected.
Release (RLS) - Bit 3 - When the RLS bit is set and a command is initiated, the drive specified by
DSO-2 will be released. The bit is used in dual-access configurations to effect the release of a drive from
one controller and make it available to the other controller. This is necessary because once a drive is
accessed by a specific controller, it is not available to the other unless one second has elapsed since the
first controller accessed the drive.
Bus Address Increment Inhibit (BAI) - Bit 4 - When the BAI bit is set via program control, the RKBA
register is prevented from incrementing during data transfers. This is primarily a diagnostic aid.
Subsystem Clear (SCLR) - Bit 5 - When the SCLR bit is set via program control, the controller is
cleared and the Initialize line is asserted on the drive interface to clear all of the drives available to the
system.
Input Ready (IR) - Bit 6 - Input Ready is a read-only bit that is set to indicate that the Silo input buffer
is ready to accept a word. Conversely, the bit is reset to indicate that the Silo is full and cannot accept a
word. The IR bit is also set by conventional initialization (INIT, CCLR, SCLR) or by the setting of the
GO bit (0 in RKCS 1).
Output Ready (OR) - Bit 7 - Output Ready is a read-only bit that is set to indicate that a word is in the
Silo output buffer. The bit is cleared by conventional initialization (INIT, CCLR, SCLR) or by the
setting of the GO bit (0 in RKCS 1).
Unit Field Error (UFE) - Bit 8 - Unit Field Error is a read-only error bit that is set to indicate the
following:
• The Drive Select code bits (TO-T2) in the current status message (AO-A3) received by the
controller did not compare with the Drive Select (DSO - DS2) code transmitted on the Message
A line. The setting of this error bit is disabled by diagnostic mode operation.
• A Deselect bit (T3) was transmitted to a drive on the Message A line and the Select Acknowledge (SACK) signal was not negated.
Multiple Dril'e Select (MDS) - Bit 9 - Multiple Drive Select is a read-only error bit that is set by the
assertion of the MDS signal line, via the selected drive, when a determination is made that at least one
other drive has been simultaneously selected. This bit can only be cleared by a Unibus Initialize (INIT)
or a Subsystem Clear (SCLR).
7-8

Programming Error (PGE) - Bit 10 - Programming Error is a read-only error bit that is set if any
controller register is written (bits for CCLR and SCLR excepted) while the GO bit in RKCS 1 is set.
The error is disabled in diagnostic mode.
Nonexistent Memory (NEM) - Bit 11 - Nonexistent Memory is a read-only error bit that is set when
the controller attempts to execute a bus cycle and SSYN is not returned within 10 J.lS of the assertion of
MSYN.
Nonexistent Drive (NED) - Bit 12 - Nonexistent Drive is a read-only error bit that is set to indicate the
following:
• A Select Acknowledge (SACK) signal from a selected drive has not been asserted on the interface in response to a message (other than a Release) sent to the drive.
• A complete status message (AO-A3, BO-B3) was not received from a selected drive within 20 J.lS
of a messag(~ request from the controller.
Unibus Parity Error (UPE) - Bit 13 - Unibus Parity Error is a read-only error bit that is set to indicate
that a parity error has been detected in a 16-bit data word format from main memory, during the
performance of a Write or Write Check command. During I8-bit data word transfers, a Unibus parity
error cannot be detected.
Write Check Error (WCE) - Bit 14 - Write Check Error is a read-only error bit that is set to indicate
that a data word read from the disk during the execution of a Write Check command did not compare
with the corresponding data word contained in main memory. If a write check error is detected and the
BAI bit is not set, the RKBA register will contain the memory address of the next data word location
(mismatched word address plus two).
Data Late Error (DL T) - Bit 15 - Data Late Error is a read-only error bit that is set to indicate the
following:
• During the execution of a Write command, the Silo was empty when the disk required a data
word.
• During the e:xecution of a Read or a Write Check command, the Silo was full when the disk
provided the next data word.
This error bit can be forced set under the following conditions:
• By loading the Data Buffer register (RK DB) when Input Ready (IR) (bit 6) is reset.
• By reading the Data Buffer register (RK DB) when Output Ready (OR) (bit 7) is reset.
NOTE
When RK CS2 is loaded under. program control, the
controller Sector/Index pulse separator logic is initialized. This action causes the controller to ignore
the next sector or index pulse from the selected drive.
Therefore, unnecessary loading of RK CS2 can increase latency time by an amount equivalent to 1.5
sectors.

7-9

7.2.6 Drive Status Register (RKDS)
The RKDS register (Figure 7 -7) is a read-only register that is used to store the operational status of a
selected drive. However, information obtained from the drive is not immediately available to program
control until the information is validated (SV AL) by the setting of bit 5, which indicates that a complete status message has been received providing a valid update.
Status information bits set in the RKDS register can be cleared by conventional initialization (lNIT,
CCLR, SCLR). However, a Controller Clear (CCLR) operation does not affect status or error condition bits that are currently set in the drives. In addition, a Unibus Initialize (lNIT) or Subsystem Clear
(SCLR) operation can only reset status or error bits in a drive if the associated status or error condition no longer exists.

os
~--------------------------~y~--------------------------~

READ ONLY (RO)
11- 4205

Figure 7-7

Drive Status Register

Drive Available (ORA) - Bit 0 - Drive Available is a read-only bit that is conditioned (message AO:T5)
to dual-access operation as follows:
• When set by a selected drive, DRA indicates that the drive is available to accept commands
from the controller.

• If an attempt is made to select a dual-access drive that is currently accessed by the other
controller, the DRA bit will not be set. This is considered an error condition by the controller,
resulting in the setting of controller Combined Error (CERR) bit in RKCS I and the generation
of a program interrupt (if enabled).
Spare - Bit 1 - Spare bit 1 is O.
Offset (OFST) - Bit 2 - Offset is a read-only bit that is set (Message AO:T10) to indicate that the
selected drive is in Offset mode.
Drive AC Low (ACLO) - Bit 3 - Drive AC Low is a read-only bit that is set (Message BO:T6) to
indicate that ac power in the selected drive is low or has been interrupted. Following detection of an ac
low condition, the heads will unload when the next Sector pulse occurs. This action allows the transfer
of the current sector to be completed before the heads are unloaded.

If ac power is restored, the error condition can be reset within the drive by a Unibus Initialize (lNIT),
Subsystem Clear (SCLR), power up, or R UN switch reset.
Speed Loss (SPLS) - Bit 4 - Speed Loss is a read-only bit that is set (Message BO:T12) to indicate that
spindle speed is unacceptably low.
Drive Off Track (DROT) - Bit 5 - Drive Off Track is a read-only bit that is set (Message BO:T13) to
indicate that the Write Gate signal has been asserted while the heads are not properly positioned over
the track centerline.

7-10

Volume Valid (VV) -- Bit 6 - Volume Valid is a read-only bit that is set (Message AO:T6) to indicate that
the Volume Valid flip-flop has been set in the selected drive by a Pack Acknowledge command. The set
condition of the bit ensures the program that the cartridge and the unit number plug have not been
changed since the last command was issued to the drive, and power has not been removed. The bit is
reset when the cartridge, the unit n umber plug, or ac power is removed.
Drive Ready (DRDY) - Bit 7 - Drive Ready is a read-only bit that is set (Message AO:T7) to indicate
that the selected drive is up to speed and the heads are properly positioned over a valid cylinder. Under
these conditions, th(~ drive is prepared to receive a command.
Disk Drive Type (DDT) - Bit 8 - Disk Drive Type is a read-only bit that is internally conditioned
(Message AO:T8) to indicate the type of drive selected. For an RK06 Drive, the bit remains reset, and
for an RK07 Drive" the bit is set. However, before any commands can be executed, the bit must
compare with the condition of Controller Drive Type, bit 10 in RKCS 1.
Spare - Bits 9 and 110 - Spare bits 9 and 10 are 0
Write Lock (WRL) _. Bit 11 - Write Lock is a read-only bit that is set (Message AO:Tll) if the selected
drive is write protected.
Spare - Bit 12 - Spare bit 12 is O.
Positioning in Progress (PIP) - Bit 13 - Positioning in Progress is a read-only bit that is set (Message
AO:TI3) to indicate that the heads are in motion.
Current Drive Attention (CDA) - Bit 14 - Current Drive Attention is a read-only bit that is the logical
equivalent of the Drive Status-Change (DSC) bit in the drive defined by the Drive Select (DSO-DS2)
bits in RKCS2. The assertion of the Attention line indicates anyone of the following conditions:
1.

Seek operation is completed.

Offset operation is completed.

Heads are unloading (Message A I:TI4).

A FAULT condition (Message BO:T7) has been set in the drive which enables the setting of
CERR (bit 15 in RKCS 1) to indicate one or more of the following:
a.
b.
c.
d.

e.
f.
g.

AC low error (Message BO:T6).
Nonexecutable function (Message BO:T8).
Controller-to-drive message parity error (Message BO:T9).
Seek incomplete (Message BO:TI0).
• Invalid address (Message BO:T5).
• Seek and no motion (Message Bl:TI2).
• Limit detected on seek (Message Bl:TI3)*.
• Servo unsafe (Message Bl :TI4)*.
Write lock error (Message BO:Tl1).
Speed loss error (Message BO:TI2)*.
Drive off track (Message BO:TI3).

*In addition to setting the CSA bit and asserting Attention, these drive fault conditions will cause the heads to be
unloaded.

7··11

Read/Write unsafe (Message BO:TI4).
• Sector error (Message Bl :T4)*.
• Write current/no write gate (Message Bl:T5)*.
• Write gate/no transitions (Message Bl :T6)*.
• Head fault (Message Bl:T7)*.
• Multiple head select (Message Bl :T8)*
• Index error (Message Bl :T9)*.
• Tribit error (Message Bl:TI0)*.
• Servo signal error (M essage B 1: T 11)* .

NOTE
A Drive Status-Change (DSC) is also indicated by
the setting (Message AO:TI4) of bit 14 in RKMR2.
Status Valid (SV AL) - Bit 15 - Status Valid is a read-only bit that is set to indicate that the bits in both
the Drive Status (RKDS) and Error (RKER) registers have been updated for the selected drive via
receipt of a complete status message. The bit is cleared by conventional initialization (INIT, CCLR,
SCLR): initiating a new command (writing into RKCS 1), selecting a new drive (writing into RKCS2),
or whenever an Attention signal is asserted by the selected drive for a drive status change.
7.2.7 Drive Error Register (RKER)
The RKER register (Figure 7-8) is a read-only register that is used to store the error status of a selected
drive. However, error information obtained from the drive is not immediately available to program
control until the information is validated (SVAL) by the setting of bit 15 in the RKDS register, which
indicates that a complete status message frame has been received.
Error bits set in the RKER register can be cleared by conventional initialization (lNIT, CCLR,
SCLR). However, a Controller Clear (CCLR) operation does not affect error bits that are currently set
in the drive. In addition, a Unibus Initialize (INIT) or Subsystem Clear (SCLR) operation can only
reset error bits in a drive if the associated error condition no longer exists.
Illegal Function (ILF) - Bit 0 - Illegal Function is a read-only bit that is set to indicate that an illegal
command (33 8 , 35 8 , 37 8 ) has been loaded into RKCS 1.
Seek Incomplete (SKI) - Bit 1 - Seek Incomplete is a read-only bit that is set (Message BO:TI0) to
indicate that a seek operation has not been completed for one of the following conditions:
• An Invalid Address (Message BO:T5) bit has been set in the drive, indicating that the drive has
received an invalid head or cylinder address.
• A Seek and No Motion (Message B l:T 12) bit has been set in the drive, indicating that a Seek
command was received by the drive but no track crossing pulses were detected within 10 ms.
• A Limit Detected on Seek (Message Bl :TI3) bit has been set in the drive, indicating that an
inner limit or an outer limit was detected during a Seek operation. If this fault occurs, the heads
are unloaded.
• A Servo Unsafe (Message Bl:T14) bit has been set in the drive, indicating a servo runaway
condition. If this fault occurs, the servo actuator is disconnected from the servo drive and
connected to a battery that provides the necessary power to drive the heads to the home position.
*In addition to setting the CSA bit and asserting Attention, these drive fault conditions will cause the heads to be
unloaded.

7-12

Oe:

ER lOCK I UNS I OPI I DTE IWlE I'DAE I COE I"VRC I eSE I EC" I om IFMTE IDRPAR I NXF I SKI l'lF I
-y

READ ONLY (RO)
11-4206

Figure 7-8

Drive Error Register

In addition to the setting of the SKI bit (Message BO:TI0) in the RKER register, the occurrence of any
one of these faults aillows the CDA bit (bit 14) to be set in the RKDS register, the Attention signal to be
asserted for the RKAS/OF register, the DSC bit (Message AO:TI4) to be set in the RKMR2 register,
and the Fault bit (M[essage BO:T7) to be set for the RKMR3 register.
Nonexecutable Function (NXF) - Bit 2 - Nonexecutable Function is a read-only bit that is set (Message
BO:T8) to indicate that a Seek command or a Write Gate signal has been received by the selected drive
while Volume Valid was reset. The occurrence of this fault allows the CDA bit (bit 14) to be set in
RKDS, the Attention signal to be asserted for RKAS/OF, the DSC bit to be set in RKMR2, and the
Fault bit to be set in RKMR3.
Control-to-Drive Par:ity Error (DRPAR) - Bit 3 - Control-to-Drive Parity Error is a read-only bit that
is set (Message BO:T9) to indicate that the drive detected a parity error on a controller-to-drive message transmission. The occurrence of this fault allows the CDA bit (bit 14) to be set in RKDS, the
Attention signal to be asserted for RKAS/OF, the DSC bit to be set in RKMR2, and the Fault bit to
be set in RKMR3.
Format Error (FMTE) - Bit 4 - Format Error is a read-only bit that is set when the format status bit
(Message AO:T9) returned from the selected drive does not compare with the CFMT bit (bit 12) in
RKCSI.
Drive-Type Error (DTYE) - Bit 5 - Drive-Type Error is a read-only bit that is set when the drive-type
status bit (Message AO:T8) returned from the selected drive does not compare with the CDT bit (bit
10) in RKCS 1 (reset for RK06 or set for RK07).
Error Correction Har'd (ECH) - Bit 6 - Error Correction Hard is a read-only bit that is set to indicate
that a data error detected by the Error Correction Code (ECC) logic in the controller cannot be
corrected using ECC.
Bad Sector Error (BSE) - Bit 7 - Bad Sector Error is a read-only bit that is set to indicate that a data
transfer has been att(~mpted to or from a sector that has at least one of the two Good Sector Flag bits
(Header Word 2, bits 14 and 15) reset, indicating a bad sector.
Header Vertical Redlllndancy Check Error (HRVe) - Bit 8 - Header VRC Error is a read-only bit that
is set to indicate that a VRC error has been detected on a sector header during a data transfer. If the
Operation Incomplete (OPI) bit (bit 13) is also set, the sector in which the error occurred cannot be
determined. However, if the OPI bit is reset, the VRC error was detected in the sector currently defined
by the RKDA regist4~r (SAO-SA4).
Cylinder Overflow Enor (COE) - Bit 9 - Cylinder Overflow Error is a read-only bit that is set to
indicate that the RK~WC register is not equal to zero following a data transfer from cylinder 410 10 (for
RK06) or 81410 (for RK07), track 2 10 , and sector 19 10 or 2110 (last logical sector).

7-13

Invalid Disk Address Error (IDAE) - Bit 10 - Invalid Disk Address Error is a read-only bit that can be
set via an Invalid Address bit (Message BO:T5) from the drive, or by the controller, to indicate the
following:
I.

An Invalid Address (Message BO:T5) bit has been set in the selected drive, indicating that an
invalid cylinder address (41110 through 51110 for the RK06 or 815 10 through 1023 10 for the
RK07) and/or an invalid track address (310) has been detected with the receipt of a Seek
command.
In addition to setting the IDAE bit, either of these faults will set the SKI bit in the RKER
register, the CDA bit in the RKDS register (while asserting an Attention signal for
RKAS/OF), the DSC bit in RKMR2, and the Fault bit in RKMR3.

The controller detected an illegal desired cylinder (DCO-DC9) value (411 10 through 814 10 ) in
the RKDC register during the initiation of a command, if the drive (as defined by bit 10 of
RKCS 1) is an RK06. These cylinder values are allowed only on the RK07.

The controller detected an illegal track address (TAO-TA2) value (310 through 7 10 ) in the
RKDA register during the initiation of a command.

Write Lock Error (WLE) - Bit 11- Write Lock Error is a read-only bit that is set (Message BO:TII) to
indicate that the drive received the assertion of the Write Gate signal when it was in Write Protect
mode. The occurrence of this fault allows the CDA bit to be set in RKDS, the Attention signal to be
asserted for RKAS/OF, the DSC bit to be set in RKMR2, and the Fault bit to be set in RKMR3.
Drive Timing Error (DTE) - Bit 12 - Drive Timing Error is a read-only bit that is set to indicate the
following:
1.

The loss of Write Clock by the controller during a write.

The loss of encoded read data by the controller during a read.

Operation Incomplete (OPI) - Bit 13 - Operation Incomplete is a read-only bit that is set to indicate
that following the positioning of the heads to a desired cylinder and the reading of 32 headers, the
desired header could not be found. This error can result from anyone of the following:
1.

Head mispositioning

Incorrect head selection

Read channel failure

Improper pack formatting.

Drive Unsafe (UNS) - Bit 14 - Drive Unsafe is a read-only bit that is set (Message BO:TI4) to indicate
that anyone of the following Read/Write Unsafe conditions, each of which causes the heads to be
unloaded, has been detected.
1.

Sector Error (Message B I:T5) indicates that Write Gate was asserted in the drive in coincidence with the trailing edge of a sector pulse.

Write Current/No Write Gate (Message Bl :T5) indicates that write current has been detected in the drive without the assertion of Write Gate.

7-14

Write Gate/No Transitions (Message Bl :T6) indicates that the drive received the assertion
of Write Grate while no encoded write data has been received.

Head Fault (Message Bl :T7) indicates that an electrical imbalance exists in the write signals
that have been fed to the head, which could cause erroneous data to be recorded.

Multiple Head Select (Message BI:T8) indicates that more than one read/write head has
been selected sim ultaneously.

Index Error (Message Bl :T9) indicates that the once-per-revolution index pulse is either
mispositioned or is missing.

Tribit Error (Message BI:T 10) indicates that three successive tribits were missing from the
servo track.

Servo Signal Error (Message BI:Tll) indicates that no signals have been detected from the
servo head"

In addition to the setting of the UNS bit in the RKER register, the occurrence of anyone of these fault
conditions allows the CDA bit to be set in the RKDS register, the Attention signal to be asserted for
the RKAS/OF register, the DSC bit (Message AO:TI4) to be set in RKMR2, and the Fault bit (Message BO:T7) to be set in RKMR3.
Data Check (DCK) - Bit 15 - Data Check is a read-only bit that is set to indicate that a data error was
detected when the cunent sector was read.
7.2.8 Attention Summary/Offset Register (RKAS/OF)
The RKAS/OF register (Figure 7-9) can be read or written via program control and as such is used to
store the head offset value required by an Offset command and the current condition of the Attention
signal line that is monitored for each drive.

ASIOF

ATN
7

ATN
6

I I

~---~---~---~---~--~~~~---~--~---~--~--~---~---~

~-------------~y~-----------~
READ ONLY (RO)

11-4207

Figure 7-9

Attention Summary/Offset Register

7-15

Offset (OFO-OF7) - Bits 0-7 - The Offset field (OFO-OF7) defines both the magnitude (OFO-OF5)
and direction (OF7) of head movement in relation to the centerline of a track. As shown below. each
binary increment of the offset value (excluding 0 F6) produces a move of 25 microinches (for an RK06)
or 12.5 microinches (for an RK07) in a positive (+) or negative (-) direction. A positive offset (OF7
0) provides head motion toward the spindle. while a negative offset (OF7
1) provides head motion
away from the spindle. Therefore. an offset value of all zero would leave the heads positioned on the
centerline of a desired cylinder.

NOTE
X = Don't care.
Sign
OF7

OF6

OF5

OF4

0
1
0
1
0
1
0
1
0
1
0
1

X
X
X
X
X
X
X
X
X
X
X
X

0
0
0
0
0
0
0
0
1
1
1
1

0
0
0
0
0
0

0
0
0
0

1
1

0
0
0
0
0
0

0
0
1
1

Magnitude
OF3
OF2
0
0
0
0
0
0
0
0
0
0
0
0

1
1

OFI

OFO

0
0
0
0
0
0
0
0
0
0
0
0

0
0
1
1
0
0
0
0
0
0
0
0

RK06
Offset
(,uin)
0
0
+25
-25
+200
-200
+400
-400
+800
-800
+ 1200
-1200

RK07
Offset
(,uin)
0
0
+ 12.5
-12.5
+ 100
-100
+200
-200
+400
-400
+600
-600

Once an Offset command has been executed, the selected drive will remain in offset mode if read
commands (i.e., Read Data, Write Check, Read Header) or a Seek are subsequently initiated for the
current cylinder. However, write commands (i.e., Write Data, Write Header) to the current cylinder.
or any head motion initiating command to another cylinder, will disable the offset condition and allow
the heads to be repositioned over the centerline of the desired cylinder.

Attention (ATNO-ATN7) - Bits 8-15 - The Attention indicators are read-onlY bits that reflect the unit
number of each drive available to the 'system and the condition of an associated Drive Status Change
flip-flop. All of the A TN bits are continuously monitored and updated (polled) every 7.4 ,us. Under
these conditions and during the update time alloted to a drive, an A TN bit will be set by the assertion
of an Attention Line signal if the Drive Status-Change flip-flop is set. Thus, the clearing of the flip-flop
within the drive will clear the associated A TN bit. The condition of the Drive Status-Change flip-flop
is also noted in the controller by the conditioning (Message AO:TI4) of the DSC bit (bit 14) in the
RKMR2 register.
7.2.9 Desired Cylinder Register (RKDC)
The RKDC register (Figure 7-10) can be read or written via program control and is used to store the
address of the desired cylinder. Following an initial load, the value in the RKDC register will be
incremented by one whenever the track address (T AO-T A2) value in the RKDA register overflows
during a data transfer. When the RKDC register is incremented and the RKWC register is not equal to
zero, a single-cylinder seek is initiated by the controller.

7-16

DCGJ~]

11 -4208

Figure 7-10

Desired Cylinder Register

Desired Cylinder (DCO-DC9) - Bits 0-9 - For an RK06 Drive, valid cylinder addresses range from the
outer edge of the disk (0) to the center (410 80 ), requiring nine (DC-DC8) address bits to define the
range (000-6328). For an RK07 Drive, valid cylinder addresses range from the outer edge of the disk
(0) to the center (815 10 ), requiring ten (DCO-DC9) address bits to define the range (000-1456 8).
Spares - Bits 10-15 - Spare bits 10-15 are zero.
7.2.10 Data Buffer Register (RKDB)
The RKDB register (Figure 7-11) can be read or written via program control. Reading from the
register empties the Silo, while writing into the register fills the Silo. Both the RKDB register and the
Silo are cleared by conventional initialization (lNIT, CCLR, SCLR).

11-4209

Figure 7-11

Data Buffer Register

7.2.11 Maintenance Register (RKMRl)
The RKMR 1 register (Figure 7-12) can be read or written via program control and is primarily used to
provide an operational analysis of the subsystem (i.e., RK611 Controller and the selected drive). This
includes drive message analysis and parity testing and the initiation of a Diagnostic mode (D MD) of
operation in which the controller can be isolated and exercised.

MRI
~---------y----------~

'--y-----J

READ ONLY (RO)

RK06
OR

RK07
11-4210

Figure 7-12

Maintenance Register 1

7-17

- Message Select (MSO-MS3) - Bits 0-3 - The RK06 or RK07 Message Select (MSO and MSl) bits
define one of the four pairs of 16-bit status messages (AO-A3 and BO-B3) that can be simultaneously
delivered to the controller on the Message A and Message B lines. When both messages are defined by
the equivalent value, Message A is sent to the RKMR2 register as Message B is simultaneously sent to
the RKMR3 register. The select bits are cleared by conventional initialization or by loading a command (other than Select Drive) into RKCS 1.
Parity Test (PAT) - Bit 4 - When the Parity Test bit is set, the controller is forced to generate even
parity for the messages sent to the drive on the Message A and Message B lines. Similarly, the controller makes an even parity check on the status messages returned from the drive.
Diagnostic Mode (DMD) - Bit 5 - When the Diagnostic mode bit is set, the controller is effectively
disconnected from all of the drives and placed under the control of specific bits in the RKMR 1 register
(i.e., MSP, MIND, MCLK, MERD).
Maintenance Sector Pulse Bit 6 (MSP) - Bit 6 - When MSP is set in conjunction with the setting of the
D MD bit, the receipt of a sector pulse from a drive is simulated.
Maintenance Index (MIND) - Bit 7 - When MIND is set in conjunction with the setting of the DMD
bit, the receipt of an index pulse from a drive is simulated.
Maintenance Clock (MCLK) - Bit 8 - When DMD is set, MCLK replaces the internal controller clock
derived from the data separator. The toggling (set, reset, etc.) of MCLK provides the clock pulses
required to step through a controller command, including the simulated reading or writing of data.
Maintenance-Encoded Read Data (MERD) - Bit 9 - When DMD is set, MERD is used to simulate
encoded serial data.
Maintenance-Encoded Write Data (MEWD) - Bit 10 - The MEWD bit is a read-only bit that is normally unconditionally set by the controller. A reset condition of the bit, when read in conjunction with
the setting of DMD, indicates that the encoding logic has functioned properly during the simulated
execution of a write command.
Precompensation Advance (PCA) - Bit 11 - Precompensation Advance is a read-only bit that is set to
indicate that write precompensation logic within the controller has determined that the pulse required
to record the current data bit must be advanced in time to ensure the reliability of the bit during
subsequent reads.
Precompensation Delay (PCD) - Bit 12 - Precompensation Delay is a read-only bit that is set to
indicate that write precompensation logic within the controller has determined that the pulse required
to record the current data bit must be delayed in time to ensure the reliability of the bit during subsequent reads.
ECC Word (ECCW) - Bit 13 - ECC Word is a read-only bit that is normally set in the controller. A
reset condition of the bit indicates that the timing logic is reflecting the ECC field within the sector.
Write Gate (WRT GT) - Bit 15 - Write Gate is a read-only bit that is set to indicate that Write Gate has
been asserted by the controller.
Read Gate (RD GT) - Bit 14 - Read Gate is a read-only bit that is set to indicate that Read Gate has
been asserted by the controller.

7-18

7.2.12 ECC Position Register (RKECPS)
The Error Correction Code (ECC) Position register (Figure 7-13) is a read-only register that is used
store the error position value, which results from the successful execution of an ECC correction sequence. When an ECC correction sequence is not in progress, the RKECPS register contains one of
two possible values (004066 8 or 005066 8). If the CFMT bit (bit 12) in RKCS 1 is reset reflecting a 22sector format (l6-bit data words), the lesser value is contained in the register. However, if the CFMT
bit is set, reflecting a 20-sector format (I8-bit data. words), the greater value is contained in the register.

ECPS~~__~__~__~__~__~.__~__~__~__~__~__~__~__~~
_____

,------------------------------y-------------------

READ ONLY (RO)
11- 4211

Figure 7-13

ECC Position Register

, Error Position (EPOO-EPI2) - Bit 0-12 - The Error Position bits are read-only bits that define the start
bit location of an error burst (containing from 1 to 11 error bits) within a 256-word Data field, following the completion of an ECC correction sequence. The position is valid if the error is ECC-correctable. If the detected error is not correctable using ECC, the register contains a constant (010040 8).
Spares - Bits 13-15 -- Spare bits 13-15 are O.
7.2.13 ECC Patterlll Register (RKECPT)
The Error Correction Code (ECC) Pattern register (Figure 7-14) is a read~only register that contains an
II-bit error correction pattern from the ECC polynominal generator. The pattern is valid if the error is
ECC-correctable.
Error Pattern (EPAO-EPAI0) - Bits 0-10 - The Error Pattern bits are read-only bits that provide an
II-bit correction pattern for an error burst that does not exceed 11 bits in length and is, therefore,
ECC-correctable.
Spares - Bits 11-15 -- Spare bits 11-15 are O.
15

ECPT~I

13
0

EPA I EPA I EPA
10

----y
READ ONLY (RO)
11-4212

Figure 7-14

ECC Pattern Register

7-19

7.2.14 Maintenance Register 2 (RKMR2)
RKMR2 is a read-only register that is used for Message A controller-to-drive transmissions (commands), as well as Message AO-A3 drive-to-controller transmissions (status).
Controller-to-Drive (Message A) - For controller-to-drive transmissions (Figure 7-15), the 16-bit Message A shift register assembles disk commands, along with an appropriate odd parity bit, from various
device registers and command logic within the controller. As the bits are assembled in the register, they
are serially transmitted to the drives in coincidence with the transmission of Message B, with each bit
position (0-15) directly related to an equivalent transmission time (TO-TIS). For additional information, refer to Paragraph 2.5.1.

MR2
y

HEAD SELECT
CODE
RESERVED
VOLUME
VALID

START
SPINDLE
COMM
RETURN TO
RECAL
COMM
CENTERLINE

DRIVE SELECT
CODE

11- 4213

Figure 7-15

Maintenance Register 2 - Controller to Drive

Drive-to-Controller (Message AO-A3) - For drive-to-controller transmissions (Figure 7-16) the 16-bit
Message A shift register assembles status information, along with an appropriate odd parity bit, from
various status and error registers within the drive. As the bits are serially transmitted from the drive in
coincidence with Message B status bits, they are assembled in the register with each bit position relating directly to an equivalent transmission time.
Since only one of four status messages can be stored in the Message A shift register at a time, a desired
message (A I-A3), other" than AO, must be previously defined by the execution of a Select Drive command.
7.2.15 Maintenance Register 3 (RKMR3)
RKMR3 is a read-only register that is used for Message B drive-to-controller transmissions (additional command parameters), as well as Message BO-B3 drive-to-controller transmissions (status).
Controller-to-Drive (Message B) - For contr"oller-to-drive transmissions (Figure 7-17), the 16-bit Message B shift register assembles additional disk command information (including the status message
request numeric of 0, 1, 2, or 3), along with an appropriate odd parity bit, from various device registers
and command logic within the controller. As the bits are assembled in the register, they are serially
transmitted to the drives in coincidence with the transmission of Message A, with each bit position
(0-15) directly related to an equivalent transmission time (TO-TIS). For additional information, refer
to Paragraph 2.5.1.
Drive-to-Controller (Message BO-B3) - For drive-to-controller transmissions (Figure 7-18), the 16-bit
Message B shift register assembles status information, along with an appropriate odd parity bit, from
various status and error registers within the drive. As the bits are serially transmitted from the drive in
coincidence with the Message A status bits, they are assembled in the register with each bit position
relating directly to an equivalent transmission time.

7-20

Since only one of four status messages can be stored in the Message B shift register at a time, a desired
message (B I-B3), other than BO, must be previously defined by the execution of a Select Drive command.
NOTE
Maintenance registers 2 (RKMR2) and 3 (RKMR3)
only contain valid status messages following the execution of a Select command.
7.3 RK611 DISK COMMANDS
Disk commands are divided into two groups. One group (nondata handling) is concerned with the
various operational requirements of the drive, while the second group (data or header handling) is
concerned with the transfer of data or header information to or from a selected device. (Refer to Table
7-2.)

The controller recognizes a command by the configuration of the 4-bit command code (F4-F 1) that is
loaded into RKCS 1. However, the command will not be decoded for execution until bit position zero
(GO) of the register is set. In addition, two other RKCS 1 bit positions are significant to command
flow: The Ready (RDY) bit (bit 7) is set when the execution of a command is completed, and the
combination Error jClear (CERRjCCLR) bit (bit 15) will be set if a device or controller error occurs
during execution. W'ith these considerations and the simultaneous serialization of Message A and
Message B (Figure 2··8a), using input derived from additional device registers, the disk commands can
initiate the operations shown in Table 7-2.

~
~

I I I
DRIVE
SPINDLE
OFFSET
DFtlVE
I VO~UME I
PARITySTATUS POSIT
ON
WRITE
ON FORMAT TYPE DRIVE VALID DRIVE
IN
PROGR
13

LOCK
12

1=20 SECT
0=22 SECT
10

READY
08

AVAIL

c;L

y
SELECTED
DRIVE
ADDRESS

I
~13
~1400 1200 : 100 I
12

10
80

1:: I

DRIVE SERIAL NUMBER

PARITY

J\.

v
SELECTED
DRIVE
ADDRESS

CYLI NDER 01 FFERENCE/OFFSET POSIT I ON

PARITY

y
SELECTED
DRIVE
ADDRESS

UN LOG
PARITY HEADS

CHANGE

Y
SELECTED
DRIVE
ADDRESS
11-4214

Figure 7-16

Maintenance Register 2 - Drive to Controller

7-21

MR3

~--------------------~v~----------------------~)~~
CYLINDER ADDRESS/OFFSETS

RESERVED
FOR ADDL
MESSAGE

PARITY

MESSAGE
REQUEST
11-4216

Figure 7 -17

Maintenance Register 3 - Controller to Drive

WRT CURR a.
NO WRT GATE
15

PARITY
15

CYLINDER ADDRESS BITS
14

MESSAGE 1.0.
01

v
SECTOR
COUNT

PARITY

DECODED
HEAD ADDRESS

MESSAGE I.D.
11-4215

Figure 7-18

Maintenance Register 3 - Drive to Controller

7-22

Table 7-2

Disk Commands

Command
F4

Function Code
F2
Fl

Nondata Handlinl~
Select Drive
Pack Acknowledge
Drive Clear
Unload
Start Spindle
Recalibrate
Offset
Seek

0
0
0
0
0
0
0
0

0
0
0
0
1
1
1
1

0
0
1
1
0
0
1
1

0
1
0
1
0
1
0
1

1
1
1
1
1
1
1
1

01
03
05
07
11
13
15
17

Data or Header I-Iandling
Read Data
Write Data
Read Header
Write Header
Write Check

1
1
1
1
1

0
0
0
0
1

0
0
1
1
0

0
1
0
1
0

1
1
1
1
1

21
23
25
27
31

Octal

Select Drive (01) (Figure 7-19) - This command is basically used to select a drive and obtain the return
of status information" This is accomplished by the setting of the Drive Select bits of Message A
(TO-T2). However the command can also function as a release control to allow one of the dual ports of
the selected drive to be: freed for access by a second controller. The latter is accomplished by the setting
of the Release bit of Message A (T3). When Messages A and B are sent, return status messages (AO and
BO) are requested. However, Select Drive is the only command with the ability to request (MS 1, MSO)
anyone of the three pairs of additional message types (AI-A3 and BI-B3).
Pack Acknowledge (03) (Figure 7-20) - This command is used to set the Volume Valid flip-flop in a
selected drive to indicate that the software has acknowledged the medium. This is accomplished by the
setting of the Volume Valid bit in Message A (T 11). When Messages A and B are sent, a return status is
requested.
Drive Clear (05) (Figure 7-21) - This command is used to clear all error flags in the selected drive,
provided the error(s) are no longer present. In addition, the command resets the Status-Change flipflop in the drive. These actions are accomplished by the setting of the Drive Clear bit in Message A
(T8). When Messages A and B are sent, a return status is requested.
Unload (07) (Figure 7-~~2) - This command is used to unload the heads in a selected drive and stop the
spindle. This is accomplished by the setting of the Set Medium Off-Line bit in Message A (TI 0). When
Messages A and B are: sent, a return status is requested.

7-23

MESSAGE
A

v
DR SEL CODE FROM
RKCS2 BITS 0-2
RELEASE BIT
FROM RKCS2
BIT 3 (X= 011)

MES~AGE MS' I MSI

'----y---J
STATUS REQ
FROM RKMRI
BITS 011

I
I
PARITY
ODD
(X= 0/1)
11-4217

Figure 7-19

MES!AGE

DS0

DSI

Select Drive Command

DS2

FORMAT
BIT

DR SEL CODE
FROM RKCS2
BITS 0-2

PARITY
ODD

SET VOLUME
VALID BIT

(X= 0/1)

(X=0/1)

ME~AGE I 0

PARITY
ODD
11-4218

Figure 7-20

Pack Acknowledge Command

MESSAGE
A

DR SEL CODE
FROM RKCS2
BITS 0- 2

MESSAGE

DRIVE
CLEAR BIT
FORMAT BIT

(X= 0/1)

PARITY
ODD
11-4219

Figure 7-21

Drive Clear Command
7-24

MES!AGEEE I
\..

DS2

I I
0

£')

I I
0

£')

I I

y-----1
DR SEL CODE
FROM R~:CS2
BITS 0-2

MES;AGEITI

£')

I I I

I I I I
I

PARITY
ODD
11-4220

Figure 7-22

Unload Command

Start Spindle (11) (Figure 7-23) - This command is used to start the spindle, execute a brush clean
cycle, and load the heads in a selected drive if the RUN/STOP switch is in a RUN state. This is
accomplished by the setting of the Start Spindle bit in Message A (T6). When Messages A and Bare
sent, a return status is requested.

MES!AGEEE' I OS2

I I I
1

I I I
0

I IxI
0

DR SEL. CODE
FROM HKCS2
BITS 112}-2
00

MES~AGEIT

START
SPINDLE
BIT
03

FORMAT
BIT

PARITY
ODD

(X=O/1)

I I I I I I I I I
0

£')

15
1

I I I I I
0

PARITY
ODD
11-4221

Figure 7-23

Start Spindle Command

Recalibrate (13) (Figure 7-24) - This command is used to relocate the heads to cylinder zero (address of
the outermost cylinder on the disk) and to clear the drive's Cylinder Address register. This is accomplished by setting the Recalibrate bit in Message A (T5). When Messages A and B are sent, a return
status is requested.

7-25

ME~AGE OS0

DSI

I OS2 J 0 I 0 I 1 I 0 I 0 I 0 I x

MES~AGE I 0

I
FORMAT

12
0

ODD

(X=O/1)

(X= 0/1)

06
0

I
PARITY

BIT

I
RECAll BRATE

v
DR SEL CODE
FROM RKCS2
BITS 0-2

I
PARITY
ODD

11-4222

Recalibrate Command

Figure 7-24

Offset (15) (Figures 7-25 and 7-26) - This command directs the selected drive to position its heads a
specified distance from the track centerline and to inhibit the recording logic. The distance (in microinches) and the direction (toward or away from spindle) that the heads are offset are determined by the
value loaded into the Offset register (RKAS/OF). This is accomplished by configuring the Cylinder
Address/Offset bits in Message B (T4-T 12). When Messages A and B are sent, a return status is
requested.
NOTE
Notice in Figures 7-25 and 7-26 that message bits are
a reversal of register bits.

MESSAGE
A

DS0

DSI

I OS2)

£)

(X= 0/1)

(X=O/1)

PARITY
ODD

FORMAT
BIT

DR SEL CODE
FROM RKCS2
BITS 0-2
00

MESSAGE
B
v
OFFSET DISTANCE VALUE
(4-9) AND DIRECTION BIT (10)
FROM RKAS/OF BITS 0-6

FORCED
SET

PARITY
ODD

(X= 0/1)
11-4223

Figure 7-25

Offset Command

7-26

OF7

OF6

OF5

OF4

OF3

OF2

OFI

OF0

CTiONJ

DIRE
B IT

~ 251'in

XNU

RKAS/OF
REG
or
50l'in
or
I00l'in or
200l'in or
400l'in or
800l'In or

12.51' in
251' in
501' In
1001' in
2001' in
4001' in
~~
RK06 or RK07

*POS= 0
NEG:: 1

NOTE:
*POS equals head motion toward spindle.

11-4224

Figure 7-26

Offset Command Coding

Seek (17) (Figure 7-27) - This command directs the selected drive to relocate the heads over a new
cylinder. The new cylinder address is derived from the Desired Cylinder register (RKDC). When
Messages A and B are: sent, a return status is requested. When the seek is completed, a drive Attention
signal is asserted for ]polling.

rr
00

MES!AGE

I DS2 I

y-~

DR SEL CODE
FROM nKCS2
BITS ~'-2

ME~AGEIT

SEEK
BIT

FORMAT
BIT

PARITY
ODD
(X=O/Il

I DC01 DCl I DC21 DC' I DC41 DC5 I DCS I DC7

DC8

I Dcol

DESIRED CYLI NDER ADDRESS
FROM RKDC BITS 0-9

PARITY
ODD

(x= 0/1)
11-4225

Figure 7-27

Seek Command

Read Data (21) - The following sequence is executed entirely by the RK61 1. A Seek to the cylinder in
RKDC is performed. When the Attention signal from the selected drive becomes true, a drive status is
requested. When the status is obtained, a Clear command is executed to clear the Attention signal.
Headers are read and f;;ompared with the desired disk address until the correct sector is found. Transfer
of data through the Silo data buffer to memory is initia ted. When the sector data transfer is complete,
the ECC logic is checked to ensure that the data read from the disk was error-free. If a data error
occurred, the ECC correction algorithm is initiated to determine whether the error is correctable; when
finished, the command is terminated to allow software to apply the correction information. Assuming
no data errors, the word count in RK WC is checked; if nonzero, the data transfer operation is repeated
into the next sector. The word count is checked at the end of each sector until it reaches zero, at which
time the command is terminated by setting the Ready bit (Figure 7-28).

7-27

MES~AGE 050 I 051 I OS2)

05
0

08
0

SEEK
BIT

~TA0 I

TAl

I TA2)

I
FORMAT

DR SEL CODE
FROM RKCS2
BITS 0-2

PARITY
ODD

TRACK ADD
FROM RKDA
BITS 8-10

BIT

(X=O/1)

MESSAGE
B
y

PARITY
ODD

DESIRED CYLINDER ADD
FROM RKDC BITS 0-9

(X=O/1)
11 - 4226

a.
00

First Message
06

MESSAGE
A
FORMAT BIT
(X = 011)

MES~AGE I 0

DRIVE
CLEAR BIT
07
09
08

I
PARITY
ODD

11-4227

b.
Figure 7-28

Second Message
Data Handling Commands

Write Data (23) - The following sequence is executed entirely by the RK611. A Seek to the cylinder in
RKDC is performed. When the Attention signal from the selected drive becomes true, a drive status is
requested. When the status is obtained, a Clear command is executed to clear tile Attention signal.
Transfer of data from memory to the Silo is begun, and headers are read and compared with the
desired disk address until the correct sector is found. Preamble, Data (256 words), and ECC bits (32)
are written on the disk. If the word count in RKWC goes to zero during the sector, the rest of the
sector is zero-filled. After the sector transfer, the word count in RKWC,is checked and, if nonzero, the
data transfer operation is continued into the next sector. The word count in RKWC is checked at the
end of each sector and, when it equals zero, the command is terminated by setting the Ready bit
(Figure 7 -28).
Read Header (25) - The following sequence is executed. A Seek to the cylinder in RKDC is performed.
When the Attention signal from the selected drive becomes true, a drive status is requested. When the
status is obtained, a Clear command is executed to clear the Attention signal. The three words of the
first header encountered are read and placed in the Silo. The Ready bit is set. The three header words
may now be retrieved by examining RKDB three times (Figure 7-28).

7-28

Write Header (27) - The following sequence is executed. A Seek to the cylinder in RKDC is performed.
When the Attention signal from the selected drive becomes true, a drive status is requested. When the
status is obtained, a Clear command is executed to clear the Attention signal. The RK611 then waits
for Index from the drive. When Index is detected, zeros are written until sector pulse is detected. Then
the header preambl(~, including sync 1 and the three header words are written. The all-zero gap, the
data preamble (including sync 1) and all-zero data, ECC, postamble, and end-of-sector gap are written. This is repeated in each successive sector until Index is encountered again and the command is
terminated. The Ready bit is set.

NOTE
All three words of the header (including the check
word) are prepared by software and treated as data
by the controller. Only one complete track can be
formatted at a time. Either 20 or 22 headers (60 or
66 words) will be written depending on the state of
the format bit (RKCSI bit 12) at the beginning of the
command. RKWC must be set to -6010 or -6610' depending on format (Figure 7-28).

Write Check (31) - The following sequence is executed entirely by the RK611. A seek to the cylinder in
RKDC command is performed. When the Attention signal from the selected drive becomes true, a
drive status is requested. When the status is obtained, a Clear command is executed to clear the
Attention signal. The: selected drive provides data as in a Read command, and data is obtained from
memory as in a Write command. The data are compared on a word for word basis until the word count
reaches zero or until a failure to compare occurs . If the data fails to compare, the command is terminated immediately (Figure 7-28).
"
7.4 DIAGNOSTIC MODE (DMD)
When Diagnostic mode (DMD) is entered under program control, the drive interface lines are effectively disconnected from the controller, while the Unibus interface remains unchanged. Under these
conditions, read/write commands (i.e., Read, Write, Read Header, Write Header, Write Check) can
be initiated (using simulated data transfer techniques), while actual transmissions to and from the
drives are disabled. In DMD, read/write commands are decoded normally, and messages are assembled in the shift registers for serial transmission. However, the output of each shift register is redirected
to provide an alternate input to the opposite register, resulting in a message swap at the conclusion of a
simultaneous transmission. Thus, message transmission to a drive is simulated by a message swap
between shift register A (RKMR2) and shift register B (RKMR3).
Diagnostic mode is entered by the setting of the DMD bit in Maintenance register 1 (RKMRl), and
subsequent operations are controlled and monitored by additional bits within the register. Data transfers are simulated by the continuous setting and resetting of the control bits (i.e., MCLK, MIND,
MSP, MERD, PAT) in a predetermined sequence, while the program monitors the read-only bits (i.e.,
MEWD, WRTGT, RDGT, ECCW, PCA, PCD) for proper controller operation. In this manner, the
data paths existing b(~tween main memory, the Silo, and the encoder/separator logic can be exe"rcised
and evaluated. In addition, since both read and write data originates in main memory and is effectively
returned (directly or by monitor), a software comparison can be made to validate the integrity of the
data paths.

7-29

Controller Modifications - In Diagnostic mode, the operation of the controller is modified by disabling
the generation of certain error and control signals, while simulating and monitoring the generation of
other signals, as follows.

Generation of these error conditions is inhibited:
1.

Controller Time-Out (CTO) (bit 11 in RKCS1)

Programming Error (PGE) (bit 10 in RKCS2)

Unit Field Error (UFE) (bit 13 in RKCS2)

Drive Timing Error (DTE) (bit 12 in RKER)

Generation of these signals is simulated:
1.

Index Pulse (MIND)

Sector Pulse (MSP)

Data/Control Clock (MCLK)

Read Data (MERD)

Even Parity (PAT) on Messages A and B (can only be done in real-time).

These signals are monitored:
1.

Precompensation Advance (PCA)

Precompensation Delay (PCD)

Timing in ECC area of sector (ECCW)

Write Data (MEWD)

Read Gate (RD)

Write Gate (WRT)

Timing Considerations - The Maintenance Clock (MCLK) bit (bit 8 in RKMR1) can be alternately set
and reset (by repeatedly loading RKMR 1) to provide a program-controlled clock. Pulses generated by
M CLK replace the data clock, used to control the transfer of data bits to and from the
Encoder /Separator logic, and to generate control clock, used to synchronize message shifting operations. Predicating the generation of controller timing on the system clock allows a diagnostic program
to monitor response conditions following the initiation of each write operation to RKMR 1.

For the proper simulation of data transfers in Diagnostic mode, the following timing considerations
must be adhered to (Figure 7-29).

7-30

* ACTUAL DATA

SIMULATED READ TIMING

~ C~':L
MCLK

+ C"~:L -+- ~
C"E':L

0PRE-COMPENSATION

IMEWD :

--'Y

.. -

---:--1

~~~~~~ST~~EE ~~~~DRVALS

fL-fT
I

.. -

* ACTUAL "'ATA

SIMULATED WRITE TIMING

*DATA SEOUENCE:
DATA ONE, DATA ZERO
FOLLOWING A ONE, AND
DATA Z:ERO FOLLOWING A ZERO.
'1 -4228

Figure 7-29

Maintenance Clock (MCLK) Timing

Data Clock
1.

A set followed by a reset of the MCLK bit equals one data clock cycle.

Two data clock cycles are required to provide a bit cell for a single data bit (1 or 0); therefore, MCLK must be set and reset twice (1, 0, 1,0).

A I6-bit word time is equal to 32 data clock cycles.

An I8.. bit word time is equal to 36 data clock cycles.

An NPR tinle for a I6-bit word is equal to 37 clock cycles.

An NPR tinle for an I8-bit word is equal to 40 clock cycles.

7-31

Control Clock - One control clock cycle is equal to four times the duration of a data clock cycle;
therefore, the MCLK bit must be set and reset four times (1,0, 1,0, 1,0, 1,0) to derive a control clock
cycle.
Data Bit Simulation - To simulate read data transfers (Read, Write Check, Read Header), it is necessary to set and reset the Read Data (MERD) bit in conjunction with MCLK. A data bit of one can be
simulated by the setting of the MERD bit. However, a data bit of zero can be simulated by either the
setting or resetting of the MERD bit, depending on the condition of the previous data bit, i.e., a data
zero following a one requires a reset condition, while a data zero following a zero requires a set
condition.
To simulate write data transfers (Write, Write Header), the MERD bit is set and reset in conjunction
with MCLK in the same manner described for read simulation. However for write data transfers, it is
necessary to monitor the condition of the resultant Write Data (MEWD) bit.
7.4.1 Read Simulation
The following sequence of operations simulates the reading of a complete sector in Diagnostic mode.
1. With the exception of RKCSl, load the device registers required to support a normal Read
command.

2. Set the Diagnostic mode (DMD) bit (bit 5 in RKMRl).
3. Load Command (FI-F4) bits (bits 1-4) and GO bit (bit 0) in RKCSI.
4. Toggle the Maintenance Clock (MCLK) bit (bit 8 in RKMRl) a sufficient number of times
to assemble Message A (MR2) and Message B (MR3), and simulate transmission (message
swap between MR2 and MR3).
5. Toggle (1, 0) the Maintenance Sector Pulse (MSP) bit (bit 6 in RKMRl) to simulate the
generation of a sector pulse.
6. Toggle MCLK in conjunction with the Read Data (MERD) bit (bit 9 in RKMRl) to simulate the transfer of 128 of the 255 Header Preamble Zeros.
7. Check for the setting of the Read Gate (RDGT) signal (bit 15 in RKMRl).
8. Toggle MCLK in conjunction with MERD to simulate the transfer of the remaining 127
Header Preamble Zeros, concluding with the generation of a single One bit to reflect the
SYNC bit.
9. Simulate the transfer of the three header words.
10. Simulate the transfer of the Gap bits (64 zeros).

7-32

11. If header c:ompare is correct, following the simulation of 128 of the 255 Data Preamble
Zeros, the RDGT signal bit will again be set and the remaining sector bits must be generated
as follows:
a.

Remaining eight words of Data Preamble (127 zeros, lOne bit)

256 words of 16- or 18-bit data (4096 or 4608 bits)

Two words of ECC (32 bits)

One word of Postamble (16 bits)

Seven words of Gap (112 zeros) for 16-bit data or six words of Gap (96 zeros) for 18-bit
data: both require approximately six NPR cycles (6 X 37 or 40 data clock cycles) to
reach the end of a sector.

12. If the word count in RKWC is not equal to zero (indicating that another sector must be
read), return to step 5 of the sequence and continue simulation. However, if the word count
is equal to zero, 37 additional NPR cycles are required to complete the simulated transfers
by unloading the Silo.
The 12 steps just deslcribed constitute a basic read simulation sequence. However, variations can be
used to provide diagnostic information for error condition analysis or verification of the Error Correction Code (ECC) logiic and its operation.
7.4.2 Write Simulatiion
The following sequence of operations simulates the writing of a complete sector in Diagnostic mode.
1. With the exception of RKCSI, load the device registers required to support a normal Write
command.
2.

Set the Diagnostic mode (DMD) bit (bit 5 in RKMRl).

Load Command (FI-F4) bits (bits 1-4) and set GO bit (bit 0) in RKCS 1.

Toggle the Maintenance Clock (MCLK) bit (bit 8 in RKMRI) to produce the number of
data clock <cycles required to generate an NPR for each data word reflected by the word
count in RK WC. For example, a WC value of 6410 will fill the Silo. Therefore, 64 NPR
cycles must be generated to complete the operation (Le., 37 data clock cycles times 64 for 16bit data or 40 data cycles times 64 for I8-bit data, if the word count is greater than 6410).

5. Toggle (1, 0) the Maintenance Sector Pulse (MSP) bit (bit 6 in RKMR1) to simulate the
generation of a sector pUlse.
6. Toggle MCLK in conjunction with the Read Data (MERD) bit (bit 9 in RKMR1) to simulate the writing of 128 of the 255 Header Preamble Zeros.

7-33

7. Check for the setting of the Read Gate (RDGT) signal bit (bit 15 in RKMRl).
S. Toggle MCLK in conjunction with MERD to simulate the writing of the remaining 127
Header Preamble Zeros, concluding with the generation of a single One bit to reflect the
SYNC bit.
9. Simulate the reading of the three header words.
10. Simulate the Gap bits (64 zeros).
11. If header compare is correct, the Write Gate (WRTGT) signal bit (bit 14 in RKMRl) will be
set, and the remaining sector bits may be written as follows:
a.

16 words of Data Preamble (255 zeros, lOne bit)

256 words of 16- or IS-bit data (4096 or 460S bits)

Two words of ECC (32 bits)

One word of Postamble (16 bits)

12. If the number of data words to be written exceeds the length of one sector, return to step 5 of
the sequence and continue; otherwise, write simulation is complete.
7.5 ERROR DETECTION AND CORRECTION
When a write data command (Write) is executed, a 32-bit Error Correction Code (ECC) is generated
by the data and, following the recording of the data field, is written in the ECC field. When a read data
command (Read, Write Check) is executed, the ECC field is also read to verify the integrity of the
recorded data. If a read error is detected, the read function of the current command is disabled, and the
ECC logic attempts to define both the position of the first error bit within the data field and the pattern
of the error over an II-bit range. If the data error is confined to no more than an II-bit error burst, the
error is correctable and the following indicators are set:
1.

Data Check (DCK) error bit (bit 15 in RKER).

Controller Error (CERR) bit (bit 15 in RKCSl).

Ready (RDY) bit (bit 7 in RKCS 1).

However, if the detected data error exceeds 11 bits, the error cannot be corrected and the following
indicators are set:
1.

Error Correction Hard (ECH) bit (bit 6 in RKER).

Data Check (DCK) error bit (bit 15 in RKER).

Controller Error (CERR) bit (bit 15 in RKCS 1).

Ready (RDY) bit (bit 7 in RKCS 1).

7-34

Thus, within less than one complete disk revolution, the program can determine if the error is correctable (DCK, CERR) or uncorrectable (ECH, DCK, CERR) via a check of the indicator bits. If the
error is correctable, the ECC Position (RKECPS) register can be examined to determine the location
of the first error bit within the error field, while the ECC Pattern (RKECPT) register can be examined
to determine which of the 11 error burst bits must be corrected in main memory. In this regard,
correction results from the exclusive-ORing of the 11 bits contained in the Pattern register with the 11
bits defined by the error burst (Figure 7-30).
However, if the error is detected as uncorrectable, the program can reread the error sector during
subsequent disk revolutions, utilizing both ECC and head offset operations (if necessary) to improve
the chances for an eventual correction and/or recovery. Thus, a standard data recovery routine consists of a sequence of 28 rereads, during which 16 ECC cycles are attempted at track center, and 12
ECC cycles are attenlpted in the six offset positions (two attempts per offset position).

4096
ECC POSITION REGISTER
= 2055
( FIRST BIT OF ERROR
BURST IS BIT 2055)

EIIT
DATA STREAM (16 BIT
WORD ASSUMED)

WORD
1

1024

5'12

256

128

(ll

£)

{ll

BIT
16

2048

BIT
32
WORD
2

BIT
2055

..... ..
,

BIT
2065

.... ..... .

BIT
4049

BIT
4065

ERROR BURST (ONE BIT IN ERROR
BURST CAUSES DATA FIELD BIT TO
COMPLEMENT )

CONTENTS OF ECC
PATTERN I~EGISTER

CONTENTS OF 11 BIT FIELD IN DATA
BLOCK AFTER CORRECTION
NOTE:
Bits 0,3, and 8 or ECC pattern register
are assumed in error.
11-4229

Figure 7-30

Error Correction Process

7-35

7.5.1 Correctable Error (DCK, CERR)
Figure 7-31 provides an example of the ECC correction process that is initiated by the controller when
a correctable data error is detected (DCK, CERR).
In the example, an II-bit error burst (bits 20-30) is detected in word 2 (bits 17-32) of a recorded data
field. The II-bit error burst, containing eight erroneous data bits (20-27), is delivered to main memory.
Under ECC control, the RK6II loads the ECC Position (RKECPS) register with a value that is equal
to the location of the first bit of the error burst (2010). In addition, the ECC Pattern (RKECPT) register
is loaded with an image of the II-bit burst, that includes a One bit for every error bit (regardless of
whether the error bit is an erroneous one or zero). Consequently, the program increments (through the
Data field image area in main memory) a number of times that is equal to the value in RKECPS and
locates the start location of the error burst. The error burst is then extracted from memory and exclusively ORed with the value contained in RKECPT. The result, including a complement of each of the
eight error bits, is then returned to the 11 burst locations and the correction is complete.
One disk revolution is lost as a result of the execution of an ECC correction cycle. However, the Disk
Address (RKDA) register has been incremented to the next sector. Therefore, only the GO bit (0) in
RKCS I must be reasserted to resume reading.

DISK DATA
WORD 2

WORD 1

'r~--------------A~------------~\r~--------------A:32 :31 :30 29 28 27 26 25 24 23 22 21 20 19 18 17 16

15 14 1:3 12 11

10 09

1'1'1_1_1_1_1'1_1'1_1'1_1'1_1_1_1_1_1_1+1_1_1_1_
f.--- 1'-BIT ERROR BURST ~
I

1 + - - - - BURST PATTERN ---~
!+-ALWAYS ZERO
115

8 ERROR BITS DEFINED
00

~RESULT

PATTERN
(ECPT) REGISTER
CONTENT

CORRECTED
DATA IN

OF EXCLUSIVE OR-----l

1····- CORRECTS ERROR BITS

-I

1,1,.1.1_1.1,1.1 ~1·I'I·I'·I·I·I·I·I·I·I+101·1_
MEMORY

I+I+I+H+I+I,I·I,I+I<ECPS) ~~~~E~~
'------y---J
20
11-4230

Figure 7-31

Example of 8-Bit Error Correction Process

7-36

7.5.2 Uncorrectabl.~ Error (ECH, DCK, CERR)
When an uncorrectable data error is indicated (ECH, DCK, CERR), an error burst has been detected
that exceeds 11 bits in length-negating the value of the information contained in the ECC Position
and Pattern registers. Under these conditions, the program must initiate a data recovery routine to
provide a sequence of successive rereads (Figure 7-32).

ASSUME:
1. MUL TISECTOR READ
(SECTORS #1 TO #10)
2. SECTOR #5 HAS A
READ ERROR.

READ
COMMAND

ECC PROCESS
TO SECTOR #5
YES

YES

RE.INITIAT:-l
AGO.BIT..J

DESIRED SECTOR
ADDRESS REG.
EaUAL TO #6

RELOAD DESiRED
SECTOR ADDRESS
(TO SECTOR #1)
RE·INITIATE A
GO·BIT

YES

ERROR NOT ECC
RECOVERABLE

LOSE A REVOLU1'ION*

*ONLY APPLICABLE IF ERROR
IS UNCORRECTABLE AFTER
FIRST PASS

11·4231

Figure 7-32

Data Error Flow

7-37

Anyone of the 16 rereads at the track centerline could result in the recovery of data with or without an
ECC correction cycle. Typically, this could occur if an error-producing material (e.g., dirt specks)
either disappeared from that area of the disk surface or diminished to a point that allowed the data to
become ECC-correctable.
If error data is recovered or corrected during anyone of the 12 offset (±400, ±800, ± 1200 for an
RK06 or ±200, ±400, ±600 for an RK07) attempts (two rereads per offset position), it may indicate
that the data on the entire track has been generated by a misaligned head. Under these conditions, it is
possible that the remainder of the data will be recovered in the same offset position.

It should be understood that for every required reread cycle, the Disk Address (DA) register must be
reloaded, and the GO bit (0) in RKCS 1 reasserted. Thus a disk revolution is lost when an uncorrectable error is initially detected, and an additional revolution is lost every time a reread cycle is executed.

7.6 PROGRAMMING EXAMPLES
The following material provides several examples of RK06 subsystem programming, including Device
Driver, ECC Correction, and Maintenance Register Read routines. RK07 subsystem programming is
done in a similar manner.
7.6.1 RK06 Device Driver Routine
The RK06 Device Driver routine allows a user to establish communications with a device and determine subsystem status and availability.
,TITLE RK06 DEVICE DRIVER

;+
;CALLING SEQUENCE:
JSR PC,RK06
DRIVE UNAVAILABLE RETURN
NORMAL RETURN
,
;INPUTS:
UNIT = DESIRED UNIT NUMBER IN BITS 0-2
DSKADR = TRACK/SECTOR ADDRESS
BUSADR = LOW ORDER 16 BITS OF UNIBUS ADDRESS
WCNT = TWO'S COMPLEMENT WORD COUNT
CYLADR = DESIRED CYLINDER ADDRESS
FUNCTN = DESIRED FUNCTION + IE + A16-A17
OFFSET = OFFSET VALUE FOR OFFSET FUNCTION
,
;OUTPUTS:
RK6ACT SET IF RK06 IS ACTIVE

7-38

RKCS1=
RKCS2=
RKDS=
RKER=
RKDC=
RKOF=
PACACK=
SELDRV=
SCLR=
SVAL=
DRDY=
VV=
DRA=
UNS:=
CERR=
DI=
IE=
UNIT:
DSKADR:
BUSADR:
WCNT:
CYLADR:
FUNCTN:
RK6ACT:
PROPRO:
OFFSET:

177440
RKCS1+10
12
14
20
16
3
1
40
100000
000200
000100
000001
040000
100000
040000
000100

.WORD
.WORD
.WORD
.WORD
.WORD
.WORD
.BYTE
.BYTE
.WORD

0
0
0
0
0
0
0
0
0

;RKCS1 (base Unibus address)
;RKCS2
;RKDS offset from RKCS1
;RKER offset from RKCS 1
;RKDC offset from RKCS 1
;RKOF offset from RKCS 1
;Basic Pack Acknowledge function
;Basic Select Drive function
;Subsystem Clear
;Status Valid
;Drive Ready
;Volume Valid
;Drive Available
;Drive Unsafe
;Controller Error/Controller Clear
;Drive Interrupt
;Interrupt Enable
; Desired unit # in bits 0-2
;Track/sector address
;Loworder 16 bits of Unibus address
;Two's complement word count
;Desired cylinder address
;Desired function + IE + A 16-A 17 + GO
;1- ifRK06 currently active
; 1- if doing Seek/Recal/Offset function
; Va.lue to load into Offset register

7-39

RK06:

1$:

2$:
3$:
4$:

TSTB
BNE
MOV
MOV
MOV
MOV
MOV
MOV
MOV
TSTB
BPL
MOV
COM
BIT
BNE
BIT
BNE
MOV
CMP
BEQ
CMP
BEQ
CMP
BNE
MOV
INCB
ADD
INCB
MOV
RTS

RK6ACT
;Is the RK06 currently active?
RK06
;Wait for it to become inactive
#RKCS2,R2
;Point to CS2 register
#SCLR,($2)
;Issue a Subsystem Clear
UNIT,(R2)
;Select the desired unit
DSKADR,-(R2)
;Load RKDA
BUSADR,-(R2)
;Load RKBA
WCNT,-(R2)
;Load RKWC
#PACACK,-(R2)
;Issue Pack ACK function
(R2)
;Wait 15 J.l,S for the serial message
1$
,
RKDS(R2),R 1
;Get Drive Status register
R1
;Complement bits
#SVAL!DRDY!VV!DRA,Rl ;Can drive accept further commands?
4$
;IfNE no
#UNS,RKER(R2)
;Is this a Seek function?
4$
;IfNE yes
CYLADR,RKDC(R2)
;Load cylinder address
#113,FUNCTN
;Is this a recalibrate function?
2$
;IfEQ yes
#117,FUNCTN
;Is this a Seek function?
2$
;IfEQ yes
#115,FUNCTN
;Is this an Offset function?
3$
;IfNE no
OFFSET,RKOF(R2)
;Load offset value
POSPRO
;Indicate positioning command
#2,(SP)
;Show a good return
RK6ACT
;Set RK06 active flag
FUNCTN,(R2)
;Load RKCS 1
PC
;Return to users program

;INTERRUPT SERVICE ROUTINE
INTR:

MOV
TSTB
BEQ
BIT
BNE
RTI

#RKCSI,R2
POSPRO
5$
#CERR !DI,(R2)
5$

;Point to CS 1 register
;Positioning in progress?
;IfEQ no
;Positioning complete or error?
;IfNE yes
;Wait for positioning to complete

7-40

5$:

10$:
20$:

HIC
CLRB
TST
BMI
TST
BMI
IvlOV
TSTB
BPL
TST
BPL

;Clear IE without doing a CCLR
;Reset positioning in progress
;Any errors?
;Yes, process directly
;Is current status valid?
;Yes it is
;Select drive to get fresh status
;Wait 15 /-LS for the serial message

#CER! IE,(R2)
POSPRO
(R2)
20$
RKDS(R2)
20$
#SELDRV,(R2)
(R@)
10$
(R2)
40$

,
;Any errors?
;IfPL no

;+
,
;The user now handles any error conditions

,40$:

CJLRB
RTI

RK6ACT

;Set RK06 inactive
;Return to user

.END

7.6.2 ECC Corredion Routine
The ECC Correction routine consists of three subroutines: the first determines if an ECC correction
cycle is necessary; the second provides a set-up procedure; the third provides an ECC correction
algorithm.
1.

Determint~ if ECC correction is necessary:

35$:

BIT
BNE
BIT
BNE

#DLT!WCE!UPE!PGE! MDS !UFE,RKCS2
;Controller error?
RETRY
;Branch if controller error
#SPAR!CTO,RKCSI
;Controller time-out or message parity?
RETRY
;Y~

BIT
BNE
BIT
BNE
TST
BMI

#DTE!HVRC,RKER
RETRY
#ECH,RKER
RETRY
RKER
DMECC

;Drive timing or header error?
;Yes
;Hard error?
;Yes
;Data check error?
;No, must be soft error

7-41

Error correction is accomplished in six steps:
a.

Calculate the byte offset to the start of the block transferred.

Calculate offset to the first bad byte in the block.

Convert correction pattern into two pattern words.

Apply ECC correction using first pattern word.

Apply ECC correction using second pattern word.

Finish or continue previous function.
RKWC,RO
RO
COUNT,RO
RO,-(SP)
DMOFST
#HVRC,RKER
40$

;Get negative words remaining
;Convert to bytes
;Calculate bytes actually transfered
;And save it
;Branch to do offset if no bytes transferred
;Header VRC error?
;Yes

DEC
BIC

RO
#777,RO

;Calculate offset to start of block
;Clear residue .

BIT
BNE

#DTE!ECH !HVR C,RKER
DMOFST

;ECC Hard Error?
;Y es do offset

MOV
DEC
MOV
BIC
BIC
CLC
ROR
ASR
ASR
ADD
MOV
CLR
DEC
BLT
ASL
ROL
BR
CALL
MOV
ADD
CALL

RKECPS,Rl
Rl
Rl,-(SP)
#tC<17>,Rl
Rl,(SP)

;Get starting bit number
;Convert to relative bit number
;Save starting relative bit number
;Isolate shift count
;Clear shift count in relative bit number
;Calculate offset to first bad byte in ....
;... the block

DMECC:MOV
ASL
ADD
MOV
BEQ
BIT
BNE
40$:

41$:

42$:

(SP)
(SP)
(SP)
(SP),RO
RKECPT,R3
(SP)
Rl
42$
R3
(SP)
41$
DMECOR
(SP)+,R3
#2,RO
DMECOR

;Set offset to first byte in the transfer
;Get correction pattern word
;Clear second pattern word
;Any shifts to perform?
;IfLTno
;This is in case the error burst. ..
;... crossed a 16-Bit word boundary
;Continue shifting
;Apply ECC correction
;Retrieve second pattern word
;Update offset to next bad(?) Word
;Apply ECC correction
;Correction completed
;Finish or continue function

7-42

ECC Correction Algorithm - This routine calculates the memory address of the data word
to be correc!ted and, using the supplied pattern word, performs an XOR of the pattern and
data words .

DMECOR: CMP
BHIS
MOY
ADD

50$:

RO,COUNT
50$

;Byte Offset Within Range?
;No

ADRS,Rl
RO,Rl

;Retrieve starting buffer address
;Calculate address of word to be corrected

MOY
(Rl),-(SP)
BIC
R3,(Rl)
BIC
(SP)+ ,R3
BIS
R3,(Rl)
RETURN

;Prepare to do an XOR
;.NOT.pattern.AND.data word
;.NOT.data word.AND.pattern
;Pattern.OR.data word

7.6.3 Maintenance R(~gister Read Routine
The Maintenance Register Read routine will transfer the content of the Maintenance registers
(RKMR 1, RKMR2, RKMR3) into a previously defined buffer; its address is in RO.

Maintenance Register Read:
R2 = Address of RKCSI
196$:
MOY

CLR
R3
MOY
#100000,(R2)
UNIT,RKCS2(R2)
MOY
R3,RKMR1(R2)
MOY
:#I,(R2)
(R2)
TSTB
BPL
.-2
MOY
RKMR2(R2),(RO)+
MOY
RKMR3(R2),(RO)+
INC
R3
CMP
R3,#4
BNE
196$
RETURN

;Clear controller
;Reselect unit
;Select drive serial message
;Issue select drive function
;Wait for the serial message
,
;Save content of MR2
;Save content of MR3
;Increment drive serial message count
;Done four messages yet?
;No, loop

7.7 PROGRAMMINIG CONSIDERATIONS
The following provides a selection of programming information for the reader's consideration.
Flagging a Bad Sector
Since a Write Header command cannot be used to selectively write individual sectors, an entire track is
written when the command is executed. Therefore, if a programmer desires to flag a bad sector, all
currently recorded sector and data information from the desired track must be extracted and preserved
elsewhere for rewriting along with the new information.
U nit Selection
'When the Controller Clear (CCLR) bit (bit 15 in RKCS1) is asserted (via the Unibus), the Drive Select
(DSO-DS2) bits (bits 0--2 in RKCS2) are cleared. Therefore, before the next command is initiated, unit
selection must be reconfigured.

7-43

Start Spindle Command
If a Start Spindle command is initiated for a drive in which the heads have already been loaded, a
controller interrupt can be returned on completion of the command, but an attention interrupt cannot
be returned.
Unload Command
If an Unload command is initiated for a drive in which the heads have already been unloaded, a
controller interrupt can be returned on completion of the command; but an attention interrupt cannot
be returned.
Drive Error Unload/Reload
When a drive is in the RUN state and the heads unload due to a drive error (refer to Paragraph 7.2.6,
Current Drive Attention - Bit 14), a successful error clear will result in a reload of the heads and the
generation of two interrupts: the first interrupt will occur with the drive error, and the second interrupt
will occur (up to 6 seconds later) when the heads reload.
Enabling an Interrupt
The Interrupt Enable (IE) bit (bit 6 in RKCS 1) must be set prior to the initiation of an operation or an
interrupt will not be generated.
Simultaneous Interrupts
If interrupt conditions simultaneously occur in the controller due to the overlapping of an operational
and data transfer command (e.g., Seek done and Read Data done), the data transfer interrupt will take
precedence and the operational interrupt will not be processed until another operation is completed.
Therefore, to ensure that no interrupts are lost, a program should scan all interrupt indIcators in
conjunction with the recognition of an interrupt.
Drive Ready Requirement
A Seek, or any other head motion command, will be ignored by a drive if the Drive Ready (DRDY) bit
is not asserted in the RKDS register when the command is initiated.
Dual-Access Operation
The programmer should be aware of a problem that can occur only with the Dual-Access Option. See
Paragraph 9.4.3.

7-44

CHAPTER 8
CUSTOMER MAINTENANCE GUIDE

8.1 INTRODUCTION
User maintenance procedures are limited to the care and cleaning (external) of the RK06K or RK07K
Disk Cartridge and the cleaning of the drive spindle assemblies.
8.2 PROFESSIONAL CARTRIDGE CLEANING
Professional cartridge cleaning is recommended every 6 months or whenever practicable. Complete
cartridge cleaning procedures must be performed by either qualified DIGITAL Field Service personnel (in emergency or troubleshooting situations) or by a professional cleaning service. Application
of cleaning procedun~s (to the actual recording surfaces) by unqualified personnel will void not only
the warranty on the serviced cartridge, but the warranty for any drive on which the cartridge is operated.
8.3 USER CARTRl[DGE CLEANING
The outer sides of a completely assembled cartridge may be cleaned by the user with a lint-free wiper,
dampened with a solution of9 percent water and 91 percent isopropyl alcohol. However, the cartridge
must not be saturated, and all excess solvent must be removed with a dry wiper. This is necessary to
prevent solvent from entering the seams of the assembly and contaminating the disks.
CAUTION
For cleaning purposes, use only a solution of 9 percent water with 91 percent isopropyl alcohol. Water,
trichloroethylene, or other solvents are not permitted.
8.4 SPINDLE ASSEMBLY CLEANING
Using a lint-free wip(~r, dampened with the isopropyl alcohol (91 percent) solution, clean the spindle
cone, magnet, and spindle rim (Figure 8-1), prior to the loading of a cartridge. However, do not
saturate the assembly; remove all excess solvent with a dry wiper. This is necessary to prevent solvent
from entering a loaded cartridge and contaminating the disks. In addition, ensure that the shroud is as
free of lint and dust as possible before loading a cartridge. Dry lint and dust may be blown from the
spindle area using filtered, dry air. However, do not use plant air which may contain water or oil;
canned air is an acceptable substitute.

8-1

CONE

RIM

7897 - 12

Figure 8-1

Spindle Assembly

8-2

8.5 CARTRIDGE CARE SUMMARY
The following listing summarizes care and cleaning considerations for the RK06K or RK07K Disk
Cartridge:
1.

Keep cartridges clean.

Use cartridges at computer room temperature only.

Manipulate cartridges by the top cover handle only.

When the protection cover is removed (for loading), do not touch disk surfaces, hub center
cone, or surfaces.

5. When the protection cover is removed (for loading), interior metal hub surfaces must be
clean.
6.

When the protection cover is removed (for loading), ensure that the disks are not moved or
rotated, since improper disk motion may generate plastic particles which can result in disk
damage.

7. When loading or unloading an RK06 or RK07 Drive, insert and remove cartridges gently.
In addition, do not use excessive force when manipulating the top cover handle.
8.

If, during operation, a cartridge makes rumbling or continuous tinging sounds (or errors
occur), discontinue use of the cartridge. Use of a damaged cartridge on other drives may
damage th1e drives, resulting in additional damage to all other cartridges used thereafter.

Each cartrildge should be professionally cleaned every 6 months and/or whenever a specific
cartridge is not operating properly.

10. Cartridges are factory-repairable only. Disassembly in the field is not permitted, and such
action will void the warranty on a cartridge, as well as any drive on which the cartridge may
be operated.

8-3

CHAPTER 9
RK06/RK07 DUAL-ACCESS OPTION

9.1 RK611/RK06 and RK611/RK07 DUAL-ACCESS SYSTEM CONFIGURATION
Dual-access operation is an option available to the RK06 Drive enabling it to be accessed through two
different controllers. Figure 9-1 shows the system configuration for dual-port operation. The eight
RK06s or RK07s shown in Figure 9-1 are daisy-chained. The two RK611 Controllers can be attached
to the same or to two different CPU s.

romoN - - -,

I
I
I

'---_ _---....;I~

RK06 OR
RK07 DISK
DRIVE

RK06 OR
RKq)7 DISK
DRIVE

RK06 OR
RK07 DISK
DRIVE

RK~6/RK07 SUBSYSTEM CONFI GURATION (PDP-11)
11-4326

Figure 9-1 RK611/RK06 orRK611/RK07 Dual Access System Configuration

9-1

9.2 DUAL-ACCESS CONTROLS
A drive with the dual-access option has three modes of operation with respect to its two ports. Figure
9-2 shows the Port A and Port B control switches that determine which mode the drive is in. When
these two switches are not pressed, all attempts to issue a command to the drive will generate a
"non-existent drive" error in the controller issuing the command.

FRONT PANEL
CONTROL PANEL

fRuNl

WRITE

PROT

A[]J
\

ACCESS B ENABLE SW.
WITH ACCESS INDICATOR
(ON = DUAL ACCESS
B PORT SELECTED)

RUN/STOP SW. WITH
STOP INDICATOR
(ON = STOP
OFF = RUN)

ACCESS A ENABLE SW.
WITH ACCESS INDICATOR
( ON = SINGLE·ACCESS DRIVE
SELECTED OR DUAL-ACCESS
A PORT SELECTED)

READY INDICATOR
(ON = READY)
FAULT INDICATOR
(ON = FAULT)

WRITE PROTECT SW. WITH
PROTECT INDICATOR
( ON = WR ITE PROTECT)
11-5329

Figure 9-2

RK06 or RK07 Disk Drive Power and Control Switches

9.2.1 Port A Mode
With only the Port A enabling switch depressed, the drive will respond to Port A. Any attempt by Port
B to issue a command to the drive will generate a "non-existent drive" error in the port B controller.
9.2.2 Port B Mode
With only the Port B enabling switch depressed, the drive will respond to Port B. Any attempt by Port
A to issue a command to the drive will generate a"nonexistent drive" error in the Port A controller.
9.2.3 Dual-Access Mode
With Port A and Port B enable switches depressed, the drive is in the programmable mode. In this
mode, the drive is initially available to both ports. Once the drive has been seiz"ed by one of the ports,
that port has full control over the drive and can Issue any legal command. A port seizes a drive if the
controller selects the drive through that port and the drive was available.
9.3 DUAL-ACCESS LOGIC CIRCUITS
For the dual-access option, the drive houses two additional modules: an M7730 dual-port board and
an M7706 interface and timing module. The two interface and timing modules interface with each
controller. The dual-port board contains logic for controlling the access from both ports and acts as a
multiplexer for commands and timing from each port to the remainder of the drive electronics. The
dual-port option also uses a different backplane than the single-port option.
9-2

The logic for the dual-port option is located on the M7730 module and consists of:
1.

Arbitration Circuits

Timing Counters and Decoders

Auto-Release Timer

Multiplexers.

9.3.1 Arbitration Cilrcuits
These circuits determine and control the state of the mUltiplexers for operating the drive through the
two ports.
9.3.2 Timing CountE~rs and Decoders
These circuits controll the timing of transmissions from the controller that has seized the drive.
9.3.3 Auto-Release Timer
This timing circuit ge:nerates a signal that will automatically release a port, if it has not been utilized
within a one second period after it has gained access.
9.3.4 Multiplexers
These circuits act as a single-pole, double-throw switch for each signal used by a port for operation of
the remainder of the drive.
9.4

DUAL-PORT OPERATION

9.4.1 Seizing a Driv.~ in the Neutral State
When an RK06 or RK07 Drive is in the neutral state, it is equally available to each port. Neither port
has priority over the other.
When a controller sell~cts a drive, the drive will switch to the correct port to connect the drive with that
controller.

If the two controllers simultaneously attempt to seize the drive, a high speed clock (in the dual port
logic) in the drive will arbitrate, allowing only one of the controllers to seize the drive.
9.4.2 Attempting to Seize an Unavailable Drive
If a neutral drive is s(~ized by a port and the other port attempts to seize the drive, the latter port will
find the drive unavailable. This will assert the Combined Error bit in Control and Status Register 1 (bit
15 in RKCS 1). Also, the Drive Available bit in the Drive Status Register (bit 0 in RKDS) will be
cleared as a result of a port attempting to seize an unavailable drive.
In addition to clearing bit 0 in RKDS, a port attempting to seize an unavailable drive will set the port
request flip-flop in that drive. The dual port logic checks the status of this flip-flop when the controller
currently operating 0111 the drive releases the drive. If the port request flip-flop is not set, this indicates
that the port other than the one just in use did not attempt to seize the drive. The drive then returns to
the neutral state and it may then be seized by either port. If, however, the port request flip-flop is set,
this means the other port attempted to seize the drive when it was unavailable. In this case, the other
port seizes the drive.
For example, assume: Port A controls the drive. Port B tries to seize the drive and finds the drive
unavailable. Thus, thle Port Request flip-flop is set. When Port A is finished with the drive, it releases
the drive with a release command. Since the Port Request flip-flop is set, Port B then seizes the drive.
9-3

9.4.3 Caution in Issuing a Release Command to an Unavailable Drive
The programmer should be careful, when designing software for use with the dual port option, to
avoid the following problem. Assume Port A has seized the drive, and Port B attempts to seize it and
finds it unavailable. Suppose Port B then attempts to cancel its port request. If the release command
from Port B is completed before Port A actually releases the drive, there is no problem. If, however,
Port A releases the drive while Port B is still in the process of issuing a release command, the Port B
controller will find the attention asserted. The assertion of attention should have been prevented by the
release of Port B. This race condition can be avoided by not allowing a port to release its port request
to an unavailable drive.
9.4.4 Auto-Release Timer
At the initiation of every command to the drive by the port that has access, a one second release timer
is started. (The exception to this is the release command.) This timer is restarted with every new
command issued by the controller having access to the drive. If a port accessing the drive does not
release the drive and the one second release timer times out, the other port's port-request flip-flop will
be examined. If it is reset, the drive will go back to the neutral state. If the other port's port-request
flip-flop had been set, the following events take place.
1.

The drive becomes deselected from the port that had access.

The drive is effectively seized by the other port.

The other port's attention is asserted, signifying that the drive is now available.

9.5 START/STOP AND WRITE PROTECT IN DUAL ACCESS
When ac power is first applied to the RK06 or RK07, the drive is initially in the neutral state. The
volume valid bit for each port is cleared. A pack acknowledge command is required for each port.
The drive will not seek or write through a port unless that port has its Volume Valid bit set.
If a port issues the Unload Heads command, the one second auto-deselect timer is disabled. This has

been instituted to allow a change of cartridge and not have the drive return to the programmable state.
If the drive were allowed to return to the programmable state after an unload command, the other port

could seize the drive and issue a conflicting command.
Once the drive is cycled up, the completion of the head-loading sequence asserts the attention bit of the
port having access to the drive. For one additional second, the drive will be seized by that port.
The port having access must issue a command to the drive within this time. Failing to issue a command
within this time will cause the dual-port logic to examine the other port's Port Request flip-flop. If the
flip-flop is not set, the drive will return to the neutral state. If the flip-flop is set, the drive will become
seized by the other port and the other port's attention will be asserted.
If the drive is unavailable to a port and the RUN/STOP switch is changed, the port is notified by the

assertion of its attention. The change is reported to the controller though the "spindle on" status bit.
When Write Protect is asserted, neither port will be able to do a write operation. There is no provision
to Write Protect the drive through one port and not have it Write Protect through the other.
If the drive is seized to a port and the operator changes the status of the write protect switch, the other

port is notified through the assertion of its attention. The change is reported to the controller through
the Write Lock Status bit.

9-4

9.6

ERROR HANDLING IN DUAL ACCESS

9.6.1 Notifying the Controller of an Error
Errors occurring while the RK06 or RK07 is seized by one of the controllers are reported to and
should be serviced by that controller. The other port is not notified of these errors. The exception is a
parity error in a message to the drive on the other port.
If the drive had been in the neutral state and an error occurred, both Attention bits would be asserted.
The fault bit would be read by both controllers when servicing the Attention set in the drive. However,
only one of the controllers would service the error in the drive.

9.6.2 Clearing the Drive Errors and Resetting the Attention
Either controller may issue a drive clear at any time. The Attention in the drive will be cleared by a
drive clear regardless of the drive's availability. However, to clear the flip-flops storing the error
information, the drive must be seized to the port issuing the Clear command as mentioned. The controller-to-drive parity error is handled differently from the other errors and is discussed in greater
detail as follows.
9.6.3 Multiple Drh'e Select and CTD Parity Errors in the Dual-Access Configuration
If a controller has seized a drive and multiple drive select is detected in the drive, the drive handles this
as an unsafe condition and unloads the heads.
If a controller has seized a drive and multiple drive select is detected on the other port, the drive detects
the condition but does not set fault and unsafe and does not unload the heads. The port detecting
multiple drive select reports it to the controller regardless of whether it had access to the drive.

Controller-to-drive (CTD) parity errors are detc~cted and stored separately for each port. Unlike all
other errors, the CTD parity error flip-flop is cleared with a Drive Clear regardless of whether or not
that controller has s<:~ized the drive. The Attention in the port is raised when a CTD parity error occurs.
9.7 INITIALIZE IN DUAL ACCESS
The RK06 or RK07 will honor an "initialize" issued separately from either port. The effect the "initialize" has depends upon the availability of the drive to the port issuing the command. The three
drive conditions are described.
1.

The drive is seized by a port. At this time, an initialize will:
a.

Clear that port's attention bit

Clear the err~r flip-flops in the drive

Causle the port to release the drive.

If the other port's port-request flip-flop is cleared, the drive will be returned to neutral. If the
other port's port-request flip-flop is set, the drive will become seized by the other port.
2.

The drive is in neutral. At this time, an initialize will:
a.

Clear the port's attention bit

Clear the error flip-flops in the drive.

9-5

The drive is seized by the other port. At this time, an initialize will:
a.

Clear the port's attention bit

Cancel the port request if one is pending.

9.8 GETTING DRIVE STATUS IN DUAL-ACCESS MODE
If the drive is seized by one of the ports, that port's controller may interrogate any of the eight status
multiplexers (AO, BO, AI, Bl, A2, B2, A3, B3) within the drive.
If the drive cannot be seized (busy with another port), only status multiplexers AO and BO may be read
through the unseized port. This situation causes the Combined Error bit (bit 15 in RKCS) to be
asserted in the controller of the un seized port. Refer to Figure 9-3 for the data bits contained in drive
status messages AO through B3.

9.8.1

Status Bits Pertinent to the Dual-Port Available at the Unibus Interface
1.

Drive available (DRA) - This bit is set to the port that has access to the drive and reset to the
other port. Read only.

Volume Valid (VV) - This bit is used to indicate when a disk cartridge may have been
changed. There is a Volume Valid bit for each port.

Ready (RD) - Indicates that the selected drive is up to speed and its heads are settled over the
specified cylinder. This bit will always be reset to a port not having access to the drive.

9-6

1~·~-----------------------------------7.44~S------------------------------------~.I
CONTROL CLOCK

JlJ1IU
I I I I I I I I I I I I I
T0

T,O

TIl

T12

I T'3 I T14 I T15 I

~
4~

1 __
1 __________________

SACI( (HIGH)

STATUS
MESS
I
A0

,
,

MES1S ID

(0) (0)
I

MESS ID

::i

(1)1 (0)

w
(!)

«
CJ)

MESS ID
(0), (1)

CJ)

VOL
VAL

w
::E

DRIVE
DR
READY TYPE

20 OFFSET WR SPINDLE
SECT
LOCK
ON
ON
FORMAT
-FWD

REV

LDG
HDS

PIP

-- -

DR
STATUS PARITY
CHANGE - - - - -

RTZ

UNLDG
PARITY
HDS

RESVD PARITY

CYLINDER DIFFERENCE/OFFSET VALUE

,
I

-- - - -

DRIVE SELECT CODE

DR
AVAIL

SERVO
HEADS BRUSH DOOR CART SPEED
RESVD SIGNAL
HOME HOME LTCHD PRES
OK
PRES

« DRIVE SELECT CODE RESVD

CJ)
CJ)

w
::E

(!)

RESVD

::i DRIVE SELECT CODE
w

DRIVE SELECT CODE

PARITY

DHIVE SERIAL NUMBER

FrESVD FOR
RESVD
ADD'L MESS

INV
ADDR

AC
LOW

FAULT

NXF

C-D
PRTY
ERR

SEEK
INC

WR
DR
RD/WR
SPEED
OFF
PARITY
LOCK
LOSS TRACK UNSAFE
f-----

-~-----

WRCNT WRITE
LIM
SEEK
R:ESVD FOR SECTOR a NO GATE HEAD MULT INDEX TRIBIT SERVO
SERVO PARITY
a NO DETON UNSAFE
ADD'L MESS
SIG
ERR
ERR
FAULT
HD
SEL
ERR
ERR
~~ITE
SEEK
MOTION
ATE
RESVD FOR
RESVD PARITY
CYLINDER ADDRESS
ADD'L MESS

I.rR~~S

MESS ID
(1)1 (1)

RESVD FOR
AIDD'L MESS

S.)

SECTOR COUNT

DECODED
HEAD ADDRESS

RESVD

RESVD PARITY

DRIVE-TO-CONTROLLER TRANSMISSIONS

NOTE:
G)-THESE BITS ARE USED ONLY ON THE RK07.
11-5328

Figure 9-3

Drive Status Bits Contained in A and B Messages

9-7

RK06/RK07 DISK DRIVE
USER'S MANUAL

Reader's Comments

Your comments and sug:gestions will help us in our continuous effort to improve the quality and usefulness of our
pUblications.

What is your general n~action to this manual? In your judgment is it complete, accurate, well organized, well
wrh~n,e~.?Isiteasytou~? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

What faults or errors have you found in the manual? _~~~~~~~~~~_ _ _~_ _ _ __

Does this manual satisfy the need you think it was intended to satisfy? _ _~_ _ _ _ _ _~~~__
Does it satisfy your needs? _ _ _~~_ _ _~_~~~_ Why? __________________________

Please send me the current copy of the Technical Documentation Catalog, which contains information on
the remainder of DIGITAL's technical documentation.

Name _~~~~~__~~_~~~_~_ Street _ _~_~~_~_~~~~~~~_ _
City _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Title _ _~~.....:..._~___~~~~_~_ __
Company _ _~_~ _ _~~~_~~_ _ State/Country _______________________
Department _ _ _ _ _~_~~~~~~_ Zip

Additional copies of this document are available from:
Digital Equipment Corporation
444 Whitney Street
Northboro, Ma 01532
Attention: Comnlunications Services (NR2/M 15)
Custo:mer Services Section
Order No. _E_K_..;R_K_06
__7,-_U_G_-O_O_l_ _ _ _ __

,- -

FoldHere -

Do Not Tear-Fold Here and Staple

FIRST CLASS
PERMIT NO. 33
MA YNARD, MASS.
BUSINESS REPLY MAIL
NO POSTAGE STAMP NECESSARY IF MAILED IN THE UNITED STATES
Postage will be paid by:

Digital Equipment Corporation
Technical Documentation Department
Maynard, Massachusetts 01754

digital equipment corporation

PRINTED IN U.S.A.