MINISCRIBE

PRODUCT MANUAL

MODELS 9380E, 9380S, 9380SM

P/N 1093
Revision P12
June 22, 1989


(Subject to Change Without Notice)




























MiniScribe Corporation
1861 Lefthand Circle
Longmont, Colorado 80501
(303) 651-6000
REVISIONS MANUAL NO. 1093
|---------------------------------------------------------------|
| RV | EC NO. | SECT. |DESCRIPTION | DATE |
|----|--------|-------|--------------------------------|--------|
| P1 | | All |Prototype Release | |
|----|--------|-------|--------------------------------|--------|
| P2 | | All |Reformat and rewrite | |
|----|--------|-------|--------------------------------|--------|
| P3 | | All |Reformat and rewrite |07/17/87|
|----|--------|-------|--------------------------------|--------|
| P4 | | All |Reformat and rewrite |08/18/87|
|----|--------|-------|--------------------------------|--------|
| P5 | | All |Reformat and rewrite |10/15/87|
|----|--------|-------|--------------------------------|--------|
| P6 | 72446 | |Initial release to PreProduction|12/11/87|
| | | 1.0 |Change 1140 TPI to 1100 TPI. | |
| | |Preface|Correct typographical error. | |
| | |Fig 2-1|Change 0-32 to 6-32. | |
|----|--------|-------|--------------------------------|--------|
| P7 | 72452 | 2.2 |Change Single Track access time |01/26/88|
| | | |to 3.5 msec. | |
|----|--------|-------|--------------------------------|--------|
| P8 | 73051 | All |Incorporate 9380S into manual. |09/12/88|
|----|--------|-------|--------------------------------|--------|
| P9 | 73077 | All |Clarify and correct minor errors|12/12/88|
| | |Fig 7-1|Add jumper positions. | |
| | | 9.5 |Move to 7.3. | |
| | |Fig 9-2|Add connector orientation figure| |
| | |9.5,6,7|Addition. | |
|----|--------|-------|--------------------------------|--------|
| P10| 73090 |1.0 |Add 9380S Interface statement. |02/02/89|
| | |2.1 |Change default to typical and | |
| | | |capacities. | |
| | |7.6.2 |Reformat diagnostics. | |
| | |Fig 9-1|Add pin 1 & 2 to J701 and J602, | |
| | | |and 1,2, and 3 to J601 connector| |
| | |Appen- |Add Appendix B. | |
| | |dix B | | |
|----|--------|-------|--------------------------------|--------|
| P11| 73092 |2.10 |Add CSA Qualification Statement.|03/16/89|
|----|--------|-------|--------------------------------|--------|
| P12| 73139M | 2.3 |Add 9380S, 9380SM 5V current |6/22/89 |
| | | |rating of 1.2A typical. Add | |
| | | |9380SM to title. | |
| | |Appen- |Change 9380SA to 9380SM. | |
| | |dix B | | |
|----|--------|-------|--------------------------------|--------|
| | | | | |
| | | | | |
| | | | | |
| | | | | |
| | | | | |
|----|--------|-------|--------------------------------|--------|

PREFACE


This Product Manual, intended for use by engineers, designers, and
planners, describes the design characteristics of the MiniScribe
9380 series of hard disk drives.

TABLE OF CONTENTS

Section Page

1.0 INTRODUCTION............................................ 1-1

2.0 PRODUCT SPECIFICATION................................... 2-1
2.1 Model Specifications.................................... 2-1
2.2 Performance Specifications.............................. 2-1
2.3 Power Requirements...................................... 2-2
2.4 Physical Characteristics................................ 2-2
2.5 Environmental Characteristics........................... 2-2
2.6 Reliability and Maintenance............................. 2-2
2.7 Shock and Vibration..................................... 2-3
2.8 Magnetic Field.......................................... 2-3
2.9 Acoustic Noise.......................................... 2-3
2.10 Safety Standards........................................ 2-4

3.0 FUNCTIONAL DESCRIPTION.................................. 3-1
3.1 General Theory of Operations............................ 3-1
3.2 Read/Write and Control Electronics...................... 3-1
3.3 Head Positioning System................................. 3-2
3.4 Heads and Disks......................................... 3-2
3.5 Spindle Drive Mechanism................................. 3-3
3.6 Air Filtration System................................... 3-3
3.7 Automatic Carriage Retrack and Locking.................. 3-3
3.8 Fine Track Safety System................................ 3-3

4.0 ESDI ELECTRICAL INTERFACE............................... 4-1
4.1 Control Signal Drivers and Receivers.................... 4-1
4.2 Data Line Drivers and Receivers......................... 4-2

5.0 ESDI PHYSICAL INTERFACE................................. 5-1
5.1 J1/P1 Connector......................................... 5-1
5.2 J2/P2 Connector......................................... 5-1
5.3 J3/P3 Connector......................................... 5-1
5.4 J4/P4 Frame Ground Connector............................ 5-1

6.0 9380E ESDI-SERIAL MODE INTERFACE IMPLEMENTATION......... 6-1
6.1 Control Input Lines..................................... 6-2
6.1.1 Drive Select.................................... 6-3
6.1.2 Head Select..................................... 6-3
6.1.3 Write Gate...................................... 6-4
6.1.4 Read Gate....................................... 6-6
6.1.5 Command Data.................................... 6-6
6.1.6 Transfer Request (Transfer Req).................6-15
6.1.7 Address Mark Enable.............................6-17
6.2 Control Output Lines....................................6-17
6.2.1 Drive Selected..................................6-18
6.2.2 Ready...........................................6-18
6.2.3 Configuration/Status Data (Config/Status).......6-19
6.2.4 Transfer Acknowledge (Transfer Ack).............6-20
6.2.5 Attention.......................................6-20

TABLE OF CONTENTS (CONTINUED)

Section Page

6.2.6 Index...........................................6-20
6.2.7 Sector/Address Mark Found.......................6-20
6.2.8 Command Complete................................6-22
6.3 Data Transfer Lines.....................................6-22
6.3.1 NRZ Write Data..................................6-22
6.3.2 NRZ Read Data....................,..............6-22
6.3.3 Read/Reference Clock............................6-24
6.3.4 Write Clock.....................................6-24
6.4 Read,Write and Format Parameters........................6-24
6.4.1 General Summary of Critical Read-Function
Timing Parameters...............................6-24
6.4.2 General Summary of Critical Write-
Function Parameters.............................6-25
6.4.3 Fixed Sector Implementation or Drive Hard
Sectored........................................6-26
6.4.4 Address Mark Implementation (Controller Soft
Sectored).......................................6-30
6.5 Defect List Option......................................6-32

7.0 INSTALLATION............................................ 7-1
7.1 Physical Interface...................................... 7-1
7.1.1 Power and Interface Cables and Connectors....... 7-1
7.1.2 Control Signal Connector J1..................... 7-1
7.1.3 Data Transfer Connector J2...................... 7-3
7.1.4 DC Power Connector J3........................... 7-3
7.1.5 J4 Frame and Ground Connector................... 7-3
7.2 9380E Option Jumpers and Test Point Description......... 7-3
7.3 9000E Drive Options..................................... 7-4
7.3.1 Drive Addressing and Interface Termination...... 7-4
7.3.2 Spindle Control Option.......................... 7-4
7.3.3 Sector Configurations/Options................... 7-4
7.3.4 Daisy Chaining MiniScribe 9000E Drives.......... 7-5
7.4 Mounting Orientation.................................... 7-6
7.5 Cabling................................................. 7-8
7.6 Diagnostics............................................. 7-8
7.6.1 Diagnostics Mode................................ 7-8
7.6.2 General Description............................. 7-9
7.6.3 Error Message Readout........................... 7-9
7.7 Error Code Definitions..................................7-11

8.0 SCSI ELECTRICAL INTERFACE............................... 8-1
8.1 SCSI Compatibility...................................... 8-2
8.1.1 Test Unit Ready - 00H........................... 8-2
8.1.2 Rezero Unit - 01H............................... 8-2
8.1.3 Request Sense - 03H............................. 8-3
8.1.4 Format Unit - 04H............................... 8-3
8.1.5 Re-Assign Block - 07H........................... 8-5
8.1.6 Read - 08H...................................... 8-5
8.1.7 Write - 0AH..................................... 8-5

TABLE OF CONTENTS (CONTINUED)

Section Page

8.1.8 Seek - 0BH...................................... 8-5
8.1.9 Inquiry - 12H................................... 8-5
8.1.10 Mode Select - 15H............................... 8-6
8.1.11 Reserve Unit - 16H.............................. 8-6
8.1.12 Release Unit - 17H.............................. 8-6
8.1.13 Mode Sense - 1AH................................ 8-6
8.1.14 Receive Diagnostic Results - 1CH................ 8-6
8.1.15 Send Diagnostic Results - 1DH................... 8-7
8.1.16 Start/Stop Unit - 1BH........................... 8-7
8.1.17 Read Buffer - 3CH............................... 8-7
8.1.18 Read Defect List - 37H.......................... 8-7
8.1.19 Read Capacity - 25H............................. 8-7
8.1.20 Seek (Extended) - 2BH........................... 8-7
8.1.21 Read (Extended) - 28H........................... 8-7
8.1.22 Verify - 2FH.................................... 8-8
8.1.23 Write (Extended) - 2AH.......................... 8-8
8.1.24 Write Buffer - 3BH.............................. 8-8
8.1.25 Write And Verify - 2EH.......................... 8-8
8.1.26 Read Revision Level - C1H....................... 8-8
8.1.27 Format Track - E4H.............................. 8-8
8.1.28 Read Long - E8H................................. 8-8
8.1.29 Write Long - EAH................................ 8-8

9.0 9380S JUMPER INSTRUCTIONS............................... 9-1
9.1 SCSI Address Jumpers (J601)............................. 9-1
9.2 SCSI Parity Enable (J602)............................... 9-1
9.3 SCSI Terminator Power (J701)............................ 9-2
9.4 Additional Jumper Definitions........................... 9-2
9.5 50 Pin SCSI Connection - J4............................. 9-2
9.6 DC Power Connector - J3................................. 9-2
9.7 Frame/Ground Connector - J5............................. 9-3

10.0 ERROR CODES.............................................10-1

11.0 UNPACKING AND INSPECTION................................11-1
11.1 Single Pack.............................................11-1
11.2 Multipack...............................................11-3
11.3 Repacking...............................................11-5

Appendix A Definition and Measurement of Seek Time......... A-1

Appendix B 9380SA.......................................... B-1

LIST OF TABLES

Table Page

6-1 J1 Control Signal Connector Pin Assignments............. 6-1
6-2 J2 Data Signal Connector Pin Assignments................ 6-2
6-3 J16 Drive Select Jumpers 1, 2, 3........................ 6-3
6-4 Command Data Word Structure and Definition.............. 6-7
6-5 Status Response Bits.................................... 6-9
6-6 Vendor Unique Status Words..............................6-10
6-7 Configuration Response Bits.............................6-11
6-8 Configuration Responses.................................6-12
6-9 Request Configuration Modifier Bits.....................6-13
6-10 Control Command Modifier Bits...........................6-14
6-11 Track Offset Command Modifier Bits......................6-14

7-1 Sector Configuration Jumpers............................ 7-5
7-2 Diagnostic Jumper Configurations....................... 7-9

8-1 Controller SCSI Command Set............................. 8-2

1.0 INTRODUCTION

The MiniScribe 9380 Series of disk drives are low cost, high
capacity, high performance random access storage devices
utilizing 5 1/4" thin film disks as storage media. Each disk
surface employs one moveable head to access the data tracks.
The 9380 Series features capacities of 380 megabytes (unfor-
matted storage capacity). The 9000 disk drives utilize
advanced 3380 Whitney-type head flexures and sliders which
allow closer spacing of disks, allowing a higher number of
disks in a standard 5 1/4" package.

High performance is achieved by the utilization of a rotary
voice coil actuator, microprocessor control, and a closed loop
servo positioning system. The closed loop servo system and
dedicated servo surface combine to allow state-of-the-art
recording densities (1100 TPI) in a 5-1/4" package. The
read/write heads, disk platters and actuator assembly are
housed within a sealed enclosure containing a recirculating
and breather filter to supply clean air exceeding Class 100
environment.

High quality mechanical construction with a sophisticated
single printed circuit assembly allows for high MTBF (50,000
hours) and maintenance-free operation throughout the life of
the drive.

The 9380S Interface conforms to the ANSI specification for
SCSI X3.131.1986 and supports the common command set.

The 9380E Series utilizes the industry standard ESDI com-
patible interface with a 10 megabit/second data transfer rate.
It also supports hard and soft sectoring along with a range
of ESDI standard options such as track offset, spindle motor
sequencing, and initiation of drive diagnostics.

Other advantages of ESDI are data separation performed on
drive, disk configuration data stored on drive along with a
self-contained defect map, data encoded in (NRZ) rather than
MFM, insuring higher data integrity over longer cable
distances.

The 9380 Series drives employ size and shock mountings
identical to the industry standard full height 5-1/4" mini-
floppy mounting dimensions. It also uses the same DC voltages
and connectors.

KEY FEATURES OF THE M9380 SERIES

o Storage capacity of 380 Mbytes unformatted.
o Identical physical size and mounting as standard 5 1/4"
Winchester disk drives.
o Power supply requirements compatible with industry
standard 5-1/4" fixed disk drives.
o 16 msec. average access time (including settling).
o 18 watt standby power requirement.
o No adjustments necessary.
o Fail-safe actuator lock at spin down or power failure.
o Dual chassis construction for protection during adverse
operating conditions.
o Start up diagnostics.
o Dedicated shipping zone.
o Single printed circuit board utilizing custom VLSI IC's
and SMT.
o Thin film metallic media for higher bit density and
resolution plus improved durability.
o MTTR of less than 30 minutes.
o Hard or soft sectoring permits use with existing ESDI
controller.

2.0 PRODUCT SPECIFICATIONS


2.1 MODEL SPECIFICATIONS

9380S 9380E

Capacity (Unformatted): 382 MB 382 MB
Number of Heads: 15 15
Number of Cylinders 1224 1224
Bytes per Track: 20832 20832
Number of Disks: 8 8
Type of Disk: sputtered sputtered
Type of Head: composite composite
Servo Head 1 1

*Typical Formatted Capacities:
Capacity Formatted: 327 MB 330 MB
Sector size (bytes): 512 512
Sectors/Track (1 spare): 36 35
Number of Cylinders: 1218 1224

*See Section 7.3.3 for Fixed Sector.

Recording Characteristics:
BPI (Bits per Inch) 20.1K 20.1K
TPI (Tracks per Inch) 1100 1100
FCI (Flux Changes per Inch) 13.4K 13.4K
Recording Code 2,7 2,7
Rotational Speed 3600 rpm 3600 rpm
Average Latency 8.33 msec 8.33 msec


2.2 PERFORMANCE SPECIFICATIONS

Data Transfer RateUp to 4 MBytes/Sec
Burst10.0 Mbits/
Sec1.25 MBytes/Sec
Data Rate Sustained
Access Times

(incl. settling time)*
Single Track 3.5 msec.
** Average (of all possible seeks) 16 msec. typ
1/3 Stroke Seek Time 17 msec. max
Full Stroke Seek Time 35 msec. max
Rezero 250 msec. max

Start time (typical) 20 seconds from power on to -Ready
Stop time (typical) 15 seconds from power removal

*Not including command transfer overhead
**See Appendix A2.3POWER REQUIREMENTS

DC Voltage Input

+12 Volts, DC
Start Surge: 4.5 amps max. peak startup current
Steady State: + 5%, 1.2 amps typical, 2.7 amps peak
seeking). Maximum ripple allowed is 1%
with equivalent resistive load.
+5 Volts, DC

9380E
+5%, .6 amps typical
Maximum ripple allowed is 2% with equivalent resistive load.

9380S, 9380SM
+5%, 1.2 amps typical
Maximum ripple allowed is 2% with equivalent resistive load.

AC Input None required

Power Dissipation IDLE MODE 18 WATTS
SEEKING MODE 22 WATTS


2.4 PHYSICAL CHARACTERISTICS

Outline Dimensions 3.25 in. (82.6mm) H X 5.75 in. (146mm)
W X 8.0 in. (203mm) L
Mounting Dimensions See Figure 2-1
Weight 6.0 Pounds


2.5 ENVIRONMENTAL CHARACTERISTICS

Temperature

Operating (stabilized) 50F (10C) to 122F (50C)
Non-operating -40F (-40C) to 140F (60C)
Thermal Gradient 18F/hr. (10C/hr.) max.

Humidity

Oper. and Non-operating 8% to 80% (noncondensing)
Maximum Wet Bulb 78F (26C)
Altitude (rel. to sea level)
Operative -200 to 10,000 feet
Non-operative 40,000 feet


2.6 RELIABILITY AND MAINTENANCE

MTBF 50,000 POH
MTTR 30 minutes
Preventive Maint. None
Comp. Design Life 5 yearsData Reliability1 recoverable error in 1010 bits
read
1 permanent error in 1012 bits read
(not recoverable in 16 reads)
1 seek error in 106 seeks

Media defect criteria (as shipped from MiniScribe)

9380S 9380E

Maximum Defects* 275 275
2 bytes in length
Cylinder zero defect free

*A single defect is defined as being less than 2 bytes long.
A multiple defect is defined as 2 bytes or longer, or as a
track with more than 1 single defect.


2.7 SHOCK AND VIBRATION

Non-operational Shock 35 G's, 11 ms pulse duration, 1/2
sine wave

Non-operational Vibration 5 to 500 Hz, 1.0 G peak acceleration

Operational Shock 2.0 G's, 11 ms duration, 1/2 sine
wave with no error rate degradation.

Operational Vibration 5 to 500 Hz, .5 G peak acceleration


2.8 MAGNETIC FIELD

The externally induced magnetic flux density may not exceed
3 gauss as measured at the top of the drive.


2.9 ACOUSTIC NOISE

IDLE MODE 43 dBa Sound Pressure Level
SEEKING MODE 50 dBa Sound Pressure Level

2.10SAFETY STANDARDS

The MiniScribe 9380 series of disk drives will comply with
the following relevant product safety standards:

UL 478
CSA C22.2 No. 0-M1982
C22.2 No. 154-M1983
IEC/VDE DIN IEC 380
VDE 0806/8.81
FCC SUBPART J OF PART 15 FOR CLASS B COMPUTING DEVICES

This equipment generates and uses radio frequency energy, and
if not installed and used properly, that is, in strict
accordance with manufacturer's instructions, may cause
interference to radio and television reception. This product
has been type-tested in a representative system, and found to
comply with the limits for a Class B computing device in
accordance with specifications in Subpart 2 of Part 15 of FCC
rules, which are designed to provide reasonable protection
against such interference in a residential installation. This
does not imply that this product guarantees FCC compliance in
any given system. It is the responsibility of the installing
systems manufacturer to insure system assembly EMC compliance.
MiniScribe maintains product compliance regarding FCC
requirements and will provide technical assistance in securing
system product compliance where appropriate.

If the system equipment does cause interference to radio or
television reception (this can be determined by turning the
equipment off and on) the user is encouraged to try to correct
the interference by using one or more of the following
measures:

o Reorient the receiving antenna.
o Relocate the computer with respect to the receiver.
o Move the computer away from the receiver.

It is recommended that shielded interface cable be used to
ensure compliance with FCC emission limits.


THIS DIGITAL APPARATUS DOES NOT EXCEED THE CLASS
A/CLASS B (whichever is applicable) LIMITS FOR RADIO
NOISE EMISSIONS FROM DIGITAL APPARATUS AS SET OUT
IN THE RADIO INTERFERENCE REGULATIONS OF THE CAN-
ADIAN DEPARTMENT OF COMMUNICATIONS
and
LE PRSENT APPAREIL NUMRIQUE N'MET PAS DE BRUITS
RADIOLECTRIQUES DPASSANT LES LIMITES APPLICABLES
AUX APPAREILS NUMRIQUES DE CLASSE A/DE CLASSE B
(selon le cas) PRESCRITES DANS LE RGLEMENT SUR LE
BROUILLAGE RADIOLECTRIQUE DICT PAR LE MINISTRE
DES COMMUNICATIONS DU CANADA.


3.0 FUNCTIONAL CHARACTERISTICS


3.1 GENERAL THEORY OF OPERATION

The 9380 Series of disk drive consists of read/write, control
and ESDI interface electronics, read/write heads, a servo
head, a head positioning actuator, thin film media and a
patented air filtration system. These components perform the
following functions:

1. Interpret and generate control signals
2. Position and maintain position over the desired track
3. Read and write data
4. Provide a contamination-free environment while perform-
ing all of the above functions.


3.2 READ/WRITE AND CONTROL ELECTRONICS

All the drive and interface electronics are packaged on a
single printed circuit board, using Surface Mount Technology
(SMT). Integrated circuits are mounted within the sealed
enclosure of the head/disk assembly in close proximity to the
read/write heads. Their functions are to provide head
selection 0 through 14 (380 MB), read data preamplification,
write data circuitry and servo data preamplification.

The microprocessor controlled printed circuit board contains
the necessary electronic circuits to perform the following
tasks:

1. read/writing of data
2. index detection
3. head positioning
4. head selection
5. drive selection
6. fault detection
7. voice coil actuator positioning
8. track 0 detection
9. recalibrate to track 0 on power-up
10. initiate diagnostics on power-up
11. track position counter
12. speed control for spindle drive motor
13. braking for the spindle drive motor
14. drive up-to-speed indication
15. monitoring for write fault conditions
16. controlling internal timings of the drive
17. generation of seek complete signals
18. data separation


3.3 9380 HEAD POSITIONING SYSTEM

The 9380 Series Read/Write heads are mounted on an integral
rotary actuator assembly which is positioned by a voice coil
motor. The voice coil, an integral part of the rotary
actuator assembly, lies inside a magnet housing when installed
in the drive. Current from the power amplifier, controlled
by the servo system, causes a magnetic field in the voice coil
which either aids or opposes the field around the permanent
magnets. This reaction in turn causes the voice coil to move
within the magnetic field. Since the heads are mounted to the
voice coil, the voice coil movement is translated directly to
the Read/Write heads and achieves positioning over the desired
track/cylinder.

Rotary actuator movement is controlled by the servo feed back
signal from the servo head. The servo head reads a dedicated
surface where servo information is pre-written at the factory.
This servo information serves as a control signal for the
actuator to provide track-crossing signals during a seek
operation, provide track-following signals during "on
cylinder" operation and provide timing information such as
index (start of a track) and servo clock.


3.4 HEADS AND DISKS

The 9380 employs composite Read/Write heads on Whitney-style
sliders and flexures to provide aerodynamic stability,
superior head/disk compliance and a higher signal to noise
ratio. Data on each of the data surfaces is read by one
Read/Write head, each of which accesses 1224 tracks.

The 9380 employs sputtered thin film media with a 130mm
diameter. The carbon overcoating formulation combined with
the low load/low mass Whitney-style heads permits highly
reliable contact start/stop operation. Thin film media yields
high amplitude signals, and very high resolution performance
compared to conventional oxide coated media. It also provides
a high abrasion-resistant surface, decreasing the potential
for damage caused by shipping and handling before installa-
tion.

A thin film servo head reads servo data encoded at the factory
on the dedicated servo disk surface. Having the servo data
on a disk surface in the middle of the disk pack reduces
offtrack conditions during thermal, shock and vibration
stresses.3.5SPINDLE DRIVE MECHANISM

A brushless DC drive motor located inside the spindle/hub
assembly rotates the disk pack at 3600 RPM. This design
allows the disk pack to be symmetrically mounted in the
unicasting. It provides a more even distribution of heat
across the disk surfaces and within the HDA improving
performance under thermal or shock and vibration stresses.
The disk pack (including spindle, hub, disks, and motor) is
dynamically balanced prior to installation to insure low
vibration and servo stability.

Dynamic braking is employed to quickly stop the drive motor
when power is removed.


3.6 AIR FILTRATION SYSTEM

The disks and Read/Write heads are assembled in an ultra-clean
air environment (Class 100 or better) and then sealed within
the head disk assembly.

Within this sealed HDA a 0.1 micron filter provides constant
internal air filtration. In addition, a breather filter
provides a clean, equalized pressure between internal air and
ambient to the HDA for the life of the drive.


3.7 AUTOMATIC CARRIAGE RETRACT AND LOCKING

If power is removed from the drive during a normal power down
or in the event of a power failure, the actuator assembly will
automatically retract and be locked in a non-data area located
at the innermost portion of the disk. The heads will land on
and take off from this area only.


3.8 FINE TRACK SAFETY SYSTEM

The 9380 Series drives utilize a fine track safety system
which inhibits write operations if an excessive offtrack
condition occurs. This aids in enhancing data reliability
under extreme shock and vibration conditions that may exceed
specified levels.

4.0 ESDI ELECTRICAL INTERFACE

The MiniScribe 9380 Series is pin and function compatible with
the serial mode of the Enhanced Small Device Interface (ESDI)
for 5-1/4" Winchester Disk Drives. In the serial mode,
interface signals (control, data and status) are transmitted
serially via handshaking request/acknowledge signals.

The Enhanced Small Device Interface can be divided into three
categories, each of which is physically separated:

a. Control signals
b. Data signals
c. DC power


4.1 CONTROL SIGNAL DRIVERS AND RECEIVERS

All control lines are digital in nature (open collector TLL)
and either provide signals to the drive (input) or signals to
the host (output). Refer to Figure 4-1 for the control signal
driver and receiver equivalent circuit and signal level
specifications.


4.2 DATA SIGNAL DRIVERS AND RECEIVERS

The data transfer signals are differential in nature and
provide data either to (write) or from (read) the drive.
Refer to Figure 4-2 for the data signal driver and receiver
equivalent circuit.


5.0 PHYSICAL INTERFACE

The electrical interface between the drive and the host
controller is via four connects:

a. J1 - Control Signals (multiplexed)
b. J2 - Data Signals (radial)
c. J3 - DC Power Input
d. J4 - Frame Ground


5.1 J1 CONNECTOR

J1 is a 34-pin board edge connector on the drive printed
circuit board. The signals on this connector control the
drive and transfer drive status to the host controller. The
34 pins are oriented with even pin numbers on the component
side of the PCB. A key slot is provided between pins 4 and
6. See Figure 5-1 for connector dimensions and Section 7.1.2
for orientation and recommended types.


5.2 J2 CONNECTOR

J2 is a 20-pin edge connector on the drive PCB. The signals
on this connector contain Read or Write data to be transferred
between the drive and host controller. The 20 pins are
oriented with even numbered pins 4 and 6. See Figure 5-2 for
connector dimensions and Section 7.1.3 for orientation and
recommended types.


5.3 J3 CONNECTOR

The DC power connector (J3) is a 4-pin connector. +5V and
+12V is supplied to the drive via this connector. J3 pin
assignments are shown in Figure 5-3. See Section 7.1.4 for
recommended type.


5.4 J4 FRAME GROUND CONNECTOR

A frame ground connection is available through J4 fasten type
connector. It is located on the shock mount frame and is
isolated from the printed circuit board. See Section 7.2.5
for recommended types.


6.0 9380 ESDI-SERIAL MODE INTERFACE IMPLEMENTATION

This section describes the interface lines, hardware, and
interface protocols necessary to implement the Serial mode
disk version of the ESDI. Pin assignments for connectors J1
and J2 are shown in Tables 6-1 and 6-2.


Table 6-1
J1 CONTROL SIGNAL CONNECTOR PIN ASSIGNMENTS

J1 Connector
Signal Ground Signal Name Source

2 1 -Head Select 23 Controller

4 3 -Head Select 22 Controller

6 5 -Write Gate Controller

8 7 -Configuration/Status Data Drive

10 9 -Transfer Acknowledge Drive

12 11 -Attention Drive

14 13 -Head Select 20 Controller

16 15 -Sector/AM Found Drive

18 17 -Head Select 21 Controller

20 19 -Index Drive

22 21 -Ready Drive

24 23 -Transfer Request Controller

26 25 -Drive Select 1 Controller

28 27 -Drive Select 2 Controller

30 29 -Drive Select 3 Controller

32 31 -Read Gate Controller

34 33 -Command Data Controller
Table 6-2
J2 DATA SIGNAL CONNECTOR PIN ASSIGNMENTS


J2 Connector
Signal Ground Signal Name Source

1 - -Drive Selected Drive

2 - -Sector/AM Found Drive

3 - -Command Complete Drive

4 - -Address Mark Enable Controller

5 6 (Reserved)

7 6 +Write Clock Controller

8 6 -Write Clock Controller

9 - (Reserved)

10 12 +Read/Reference Clock Drive

11 12 -Read/Reference Clock Drive

13 15 +NRZ Write Data Controller

14 16 -NRZ Write Data Controller

17 19 +NRZ Read Data Drive

18 - -NRZ Read Data Drive

20 - -Index Drive


6.1 CONTROL INPUT LINES

The control input signals are of two types: those to be multi-
plexed in a multiple drive system and those intended to do the
multiplexing. The control input signals to be multiplexed are
Write Gate, Read Gate, Head Select 20, Head Select 21, Head
Select 22, Head Select 23, Transfer Req and Command Data. The
signals to do the multiplexing are Drive Select 1, Drive
Select 2, and Drive Select 3.

Two removable 150 ohm resistor packs (RP4 and RP17) are used
for control input line termination. See Section 7.2.4 for
details.

Address Mark Enable is a control input in the radial cable.
It is not multiplexed.6.1.1DRIVE SELECT

Three Drive Select lines are to be decoded to
determine the drive select address. Drive select
jumper J16 consists of 3 jumpers which are configured
by the user to represent a binary address (1-7). See
Figure 7-1 for J16 locations and Table 6-3 for address
jumper configurations.


Table 6-3
J16 DRIVE SELECT JUMPERS 1,2,3

J16 Drive Select
3 2 1 Binary Address Address

out out out 0 0 0 No Select

out out in 0 0 1 1

out in out 0 1 0 2

out in in 0 1 1 3

in out out 1 0 0 4

in out in 1 0 1 5

in in out 1 1 0 6

in in in 1 1 1 7


6.1.2 HEAD SELECT 20, 21, 22, AND 23

These four lines allow selection of each individual
read/write head in a binary coded sequence. Head
Select 20 is the least significant line. Heads are
numbered 0 through 15 (M9380E). When all Head Select
lines are high (inactive), head 0 will be selected.

Addressing more heads than contained in the drive will
result in a write fault when attempting to perform a
write operation.

A 150 ohm resistor pack allows for line termination.

6.1.3 WRITE GATE

The active state of this signal enables write data to
be written on the disk.

The high-to-low transition of this signal creates a
write splice which is immediately followed by the PLO
Sync field. See Figures 6-1 and 6-2. When formatt-
ing, Write Gate should be deactivated for 2 bit times
minimum between the address field and the data field
to identify to the drive the beginning of the data PLO
Sync field. This line shall be protected from
terminator power loss by implementation of an
equivalent circuit shown in Figure 6-3.


6.1.4READ GATE

The active state of this signal, or low level, enables
data to be read from the disk. This signal should
become active only during a PLO Sync field and at
least the number of bytes defined by the drive prior
to the ID or Date Sync Bytes. The PLO Sync field
length can be determined by the response to the
Request PLO Sync Field Length command. Read Gate must
be false when passing over a write splice area.

A 150 ohm resistor pack allows for line termination.

6.1.5 COMMAND DATA

When presenting a command, 16 information bits of
serial data, plus parity, will be presented on this
line. This data is to be controlled by the hand-shake
protocol with signals Transfer Req and Transfer Ack.
Upon receipt of this serial data, the drive will
perform the required function as specified by the bit
configuration.

Data is transmitted MSB first. See Table 6-4 for the
meaning of the various bit combinations. See Figure
6-4 for timing.

No communications should be attempted unless the
Command Complete line is true. NOTE: This line must
be a logic 0 when not in use.

A 150 ohm resistor pack allows for line termination.
Table 6-4
COMMAND DATA WORD STRUCTURE AND DEFINITION

________________________________________________________
Most Least
Significant Significant
Bit Bit
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 P
CMD FUNCTION CMD MODIFIER ALL ZEROS P
CMD FUNCTION CMD PARAMETER P
BIT P. PARITY (000)

COMMAND DATA WORD STRUCTURE


COMMAND DATA DEFINITION

CMD FUNCTION CMD FUNCTION CMD MODIF CMD PARAM STATUS/CONF
BIT DEFINITION APPLIC APPLIC DATA RTN
15 14 13 12 (BITS 11-8) (BITS 11-0) TO CONTROL

0 0 0 0 Seek No Yes No

0 0 0 1 Recalibrate No No No

0 0 1 0 Request Status Yes No Yes

0 0 1 1 Request Conf. Yes No Yes

0 1 0 0 Select Hd Group No Yes No

0 1 0 1 Control Yes No No

0 1 1 0 Data Strobe Offs. Yes No No

0 1 1 1 Trk Offset Yes No No

1 0 0 0 Initiate Diag. No Yes No

1 0 0 1 Set Bytes per Sec. No Yes No

1 0 1 0 Reserved -- -- --

1 0 1 1 Reserved -- -- --

1 1 0 0 Reserved -- -- --

1 1 0 1 Reserved -- -- --

1 1 1 0 Set Conf. -- -- --

1 1 1 1 Reserved -- -- --

NOTES: 1. All unused or not applicable lower order bits must be
zero.
2. Any "Reserved" or command function received shall be
treated as an invalid.6.1.5.1Command Data Bits 15 Through 12 Decode
Definition

a. Seek (0000)

This command causes the drive to seek to
the cylinder indicated in bits 0 through
11. A Seek command will restore track
offset to zero.

b. Recalibrate (0001)

This command causes the actuator to return
to cylinder 0000. A Recalibrate command
will restore track offset to zero.

c. Request Status (0010)

This command causes the drive to send 16
bits (see Table 6-5) of standard or vendor
unique status information to the controller
as determined by the command modifier bits.
The parity utilized in all status responses
shall be odd.

Request Standard Status

When the command modifier bits (11-8) of
the Request Status command is 0000, the
drive will respond with 16 bits of standard
status. Bits 15-12 of this status are
defined as state bits which do cause
Attention to be asserted. Bits 11-0 of
this status are fault or change of status
bits that cause Attention to be asserted
each time one is set. See Section 6.2.3.2
for response protocol and format of the
status response from the drive.

Request Vendor Unique status

When the command modifier bits (11-8) of
the Request Status command is 0001 through
1111, the drive responds with vendor unique
status. Upon notification that vendor
unique status is available, the drive will
return one word of unique status as shown
in Table 6-6.
Table 6-5
STATUS RESPONSE BITS

Bit Att

15 Reserved 0
14 1=Removable media not present 0
0=If not removable 0
13 1=Write protected - removable media 0
0=If not removable 0
12 1=Write protected - fixed media 0
11 Reserved 0
10 Reserved 0
9 1=Spindle motor stopped by stop command 0
1=Spindle motor stopped for other (e.g. power on,
reset) 1
8 1=Power on reset conditions exist (reconf. or
restart Spindle motor command may be required) 1
7 1=Command data parity fault
6 1=Interface fault 1
5 1=Invalid or unimplemented command fault 1
4 1=Seek fault 1
3 1=Write gate with track offset fault 1
2 1=Vendor unique status available VD
1 1=Write fault* 1
0 1=Removable media changed (has been changed since
last status request) 1


*Conditions that can cause write fault are:

a. Write current in a head without Write Gate asserted or no
write current with Write Gate asserted and the drive
selected.
b. Multiple heads selected, no head selected, or improperly
selected with Write Gate asserted.
c. Simultaneous assertion of Read Gate and Write Gate.
d. DC voltages grossly out of tolerance with Write Gate
asserted.
e. No write data transitions with Write Gate active.
f. Open or shorted heads.
g. Offtrack condition with Write Gate asserted.
Table 6-6
VENDOR UNIQUE STATUS WORDS

Word Description

00 Microprocessor RAM error
01 Microprocessor ROM checksum error
02 Interface chip diagnostic failure
03 Sector counter fault
04 Index pulse not detected or lost
05 Spin speed not within 0.5% tol
06 Loss of fine track during idle mode
07 Reserved
08 Time out on +End Decel signal (during seek)
09 Time out on CYL Pulse (during seek)
0A Overshoot (after a seek)
0B Time out on fine track (after a seek)
0C Track zero signal not detected (after a seek)
0D Comparator mismatch (after a seek)
0E Comparator mismatch (during a seek)
0F Unexpected interrupt from processor
10 Time out on non-GB pattern (during a rezero)
11 Time out on GB1 pattern (during a rezero)
12 Time out on GB2 pattern (during a rezero)
13 Seek range error
14 Voltage unsafe
15 Track offset fault
16 Write fault
17 Reserved
18 Time out on +End Decel (during a rezero)
19 Time out on CYL Pulse (during a rezero)
1A Overshoot (after a rezero)
1B Time out on fine track (after a rezero)
1C Track 0 signal not detected (after a rezero)
1D Comparator mismatch (after a rezero)
1E Reserved
1F 6301 trap
30 Time out on non-GB pattern (adj)
31 Time out on GB1 pattern (adj)
32 Time out on GB2 pattern (adj)
39 Time out on CYL Pulse (adj)
3E Cannot adjust servo

d. Request Configuration (0011)

This command causes the drive to send 16
bits (Tables 6-7 and 6-8) of configuration
data to the controller. The parity
utilized in all configuration responses
shall be odd. The specific configuration
requested is specified by bits 11-8 of the
command as shown in Table 6-9.


Table 6-7
CONFIGURATION RESPONSE BITS


Most Least
Significant Significant
Bit Bit
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 P

Bit
Position Function Response

15 0 = Magnetic Disk Drive 0
14 Format Speed Tolerance Gap Required 0
13 Track Offset Option Available 1
12 Data Strobe Offset Option Available 0
11 Rotational Speed Tolerance is > 0.5% 0
10 Transfer Rate > 10 Mhz 0
9 Transfer Rate 5 Mhz < = 10 Mhz 1
8 Transfer Rate <=5 Mhz 0
7 Removable Cartridge Drive 0
6 Fixed Drive 1
5 Spindle Motor Control Option Implemented x jumpers
4 Head Switch Time > 15 usec* 0
3 RLL Encoded (Not MFM) 1
2 Controller Soft Sectored (Address Mark) x jumpers
1 Hard Sectored (Sector Pulses) x jumpers
0 Reserved 0

*Command Complete shall be negated with 15 usec. of a head change
if this bit is set to 1.
Table 6-8
CONFIGURATION RESPONSES

COM MODIFIER BITS CONFIGURATION RESPONSE
11 10 9 8

0 0 0 1 Number of cylinders - Fixed (1224)
0 0 1 0 Number of cylinders - Removed Media (0)
0 0 1 1 Number of Heads
Bits 15-8 Removable drive heads (0)
Bits 7-0 Fixed heads See Note #1
0 1 0 0 Min. unformatted bytes per track (20832)
0 1 0 1 Min. unformatted bytes per sector (hard
sec. only) See Note #2
0 1 1 0 Number of sectors per track (hard sector
only)
Bits 15-8 Reserved (0)
Bits 7-0 bytes per ISG (16)
1 0 0 0 Min. bytes per PLO sync field
Bits 15-8 Reserved (0)
Bits 7-0 bytes per PLO sync field
required when READ GATE is asserted (14)
1 0 0 1 Number of words of vendor unique status
available
Bits 15-8 Reserved (0)
Bits 7-0 Number of vendor unique status
words (1)
1 0 1 0
through Reserved (0)
1 1 1 0

1 1 1 1 Vendor Identification (20)


NOTES:

1. The number of fixed heads reported is dependent on drive
model and the configuration of Jumpers J10 and J11 (See
Figure 7-1 for location). J10 and J11 are factory set.
Model 9380E/S has 15 heads (0-14).

2. The number of unformatted bytes per sector as well as the
number of sectors per track is dependent on the sector
configuration Jumpers J12, J13 and J19 (refer to Section
7.2.3).
Table 6-9
REQUEST CONFIGURATION MODIFIER BITS

COM Modifier Bits Function
11 10 9 8

0 0 0 0 General Configuration of Drive and
format
0 0 0 1 Number of cylinders fixed
0 0 1 0 Number of cylinders, removable
0 0 1 1 Number of heads
0 1 0 0 Minimum unformatted bytes per track
0 1 0 1 Unformatted byte per sector (hard sector
only)
0 1 1 0 Sector per track (hard sector only)
0 1 1 1 Minimum bytes in ISG field
1 0 0 0 Minimum bytes per PLO sync field
1 0 0 1 Number of words of vendor unique status
available
1 1 1 1 Vendor identification


e. Vendor Identification (1111)

Bits 15-8 identify the vendor. The
vendor identification code for Mini-
Scribe is 14 hex. (1101)

f. Control (0101)

This command causes the control opera-
tions specified by bits 11-8 to be
performed as shown in Table 6-10.

The spindle motor control option is
available only when the drive is jumper
configured for that option. If the
drive is not jumper configured for the
spindle motor control option and a stop
or start spindle command is issued, the
drive will respond with invalid command
fault. Refer to Figure 7-1 for the
location of the spindle motor control
option Jumper J7.
Table 6-10
CONTROL COMMAND MODIFIER BITS

COM Modifier Bits Function
11 10 9 8

0 0 0 0 Reset interface attention and standard
status (Bits 0-11)
0 0 0 1 Reserved
0 0 1 0 Stop spindle motor (only when J7 is
installed)
0 0 1 1 Start spindle motor
0 1 0 0 Reserved
0 1 0 1 Reserved
0 1 1 0 Reserved
0 1 1 1 Reserved
1 X X X Reserved


g. Track Offset (0111)

This optional command causes the drive
to perform a track offset in the direc-
tion and amount specified by bits 11-
8 as shown in Table 6-11. Only one
offset in either direction is sup-
ported.

When a track offset command is issued,
Command Complete will be inactive for
approximately 2.5 msec to allow for
settling. A track offset command will
inhibit any attempted write operations
to the drive. If a write is attempted
with track offset a write gate with
track offset fault will be reported.


Table 6-11
TRACK OFFSET COMMAND MODIFIER BITS

COM Modifier Bits Function
11 10 9 8

0 0 0 0 Restore offset to 0
0 0 0 1 Restore offset to 0
0 0 1 0 Positive offset 1
0 0 1 1 Negative offset 1

NOTE: Seek and Recalibrate command restore offsets to zero.
h. Initiate Diagnostics (1000)

This optional command with a command
of zero causes the drive to perform
internal diagnostics. Command Complete
indicates the completion of the
diagnostics. Attention with Command
Complete indicates that a fault was
encountered and status should be
requested to determine the proper
course of action.

i. Set Unformatted Bytes per Sector (1001)

The set bytes per sector option is
available only when the drive is jumper
configured for hard sector operation.
If the drive is jumper configured for
soft sectored operation, an invalid
command fault will be reported if a set
bytes per sector command is issued.
The range of allowed number of
unformatted bytes per sector in the
command parameter is between 162 and
4095 bytes in one byte increments.
If the number of bytes in the command
parameter is not within the range of
162 and 4095 bytes, then the drive will
respond with an invalid command fault.

6.1.6 TRANSFER REQUEST (TRANSFER REQ)

This line functions as a handshake signal in conjunc-
tion with Transfer Ack during command configuration/status
transfers. See Figures 6-4 and 6-5 for timing.


6.1.7 ADDRESS MARK ENABLE (SOFT SECTORED MODE)

This signal, when active with Write Gate, causes an
Address Mark to be written. Address Mark Enable shall
be active for 24 bit times. See Figure 6-6 for
timing.

Address Mark Enable, when active without Write Gate
or Read Gate, causes a search for Address Marks. See
Figure 6-2.

The trailing edge of Address Mark Enable with Write
Gate true indicates the beginning of the PLO sync
field.


6.2 CONTROL OUTPUT LINES

The output control signals are driven with an open collector
output stage capable of sinking a maximum of 48 mA at low
level or true state with maximum voltage of 0.4V measured at
the driver. When the line driver is in the high level or
false state, the driver transistor is off and collector
leakage current is a maximum of 250 uA.

All J1 output lines are enabled by their respective Drive
Select decodes.6.2.1DRIVE SELECTED

A status line provided at the J2/P2 connector to
inform the host system of the selection status of the
drive. The Drive Selected line is driven by a TTL
open collector driver as shown in Figure 4-1. This
signal will go active only when the drive is selected
as defined in Section 6.2.2. The Drive Select lines
at J1/P1 are activated by the host system.

6.2.2 READY

This signal indicates that the spindle is up to speed.
This interface signal when true, together with Command
Complete indicates that the drive is ready to read,
write or seek. When the line is false, all writing
and seeking is inhibited.


6.2.3 CONFIGURATION/STATUS (CONFIG/STATUS) DATA

The drive presents serial data on this line upon
request from the controller. See Figure 6-7 for
typical operation. This configuration status serial
data will be presented to the interface and trans-
ferred using the hand-shake protocol with signals
Transfer Req and Transfer Ack. See Figure 6-5. Once
initiated, 16 bits plus parity will be transmitted MSB
first. The parity utilized shall be odd.

6.2.3.1 Configuration Response Bits

In response to the Request Configuration
command (see Section 6.1.5.1) 16 bits of
configuration information is returned to the
controller. There shall be no invalid
response made to a Request Configuration
command. General configuration response
flags must be used to verify the validity
of the responses.

6.2.3.1.1 General Configuration Response
Bits

If the command modifier bits (11-
8) were 0000, the general
configuration status information
shown in Table 6-7 is returned.

If other command modifier bits
were used, the specific configura-
tion status information shown in
Table 6-8 is returned for each
configuration command with those
modifiers.

Upon notification that vendor
unique status is available, the
drive product will return one word
of unique status as shown in Table
6-6.

6.2.3.2 Status Response Bits

In response to the Request Status command
(see Section 6.1.5.1), 16 bits of status
information is returned to the controller.
Refer to Table 6-5.

Bits 15-12 of the status are defined as
state bits which do not cause Attention to
be asserted. Bits 11-0 are fault or change
of status bits that cause Attention to be
asserted.6.2.4TRANSFER ACKNOWLEDGE (TRANSFER ACK)

This signal functions as a hand-shake signal along
with TRANSFER REQ during COMMAND and CONFIGURATION-
STATUS transfers. See Figures 6-5 and 6-6.

6.2.5 ATTENTION

This output is asserted when the drive wants the
controller to request its standard status. Generally,
this is a result of a fault condition or a change of
status. Writing is inhibited when ATTENTION is
asserted. ATTENTION is deactivated by the Reset
Interface Attention command (see Section 6.1.5.1).

6.2.6 INDEX

This pulse is provided by the drive once each
revolution to indicate the beginning of a track. This
signal is high and makes the transition to low to
indicate INDEX. Only the transition at the leading
edge of the pulse is accurately controlled. The
period (T) of this signal is the reciprocal of the
rotational speed, Figure 6-10. This signal is
available on the command cable J1/P1 (gated) and on
the data cable J2/P2 (ungated).

6.2.7 SECTOR/ADDRESS MARK FOUND

These two signals are mutually exclusive and therefore
share this line. The signal that is used is deter-
mined by the NRZ data transfer control implementation.
These signals are available on the command data J1/P1
(gated) and on the data cable J2/P2 (ungated).

6.2.7.1 Sector (Drive Hard Sectored)

This interface signal indicates the start
of a sector. The leading edge of the sector
pulses is the only edge that is accurately
controlled. The index pulse indicates
sector zero. See Figure 6-8.

6.2.7.2 Address Mark Found (Controller Soft
Sectored)

This signal indicates the detection of the
end of an address mark. See Figure 6-9 for
timing.


6.2.8 COMMAND COMPLETE

Command Complete is a status line provided at the
J2/P2 connector. This is an ungated output from the
drive which allows the host to monitor the drive's
Command Complete status, during overlapped commands,
without selecting the drive. This signal line will
go false in the following cases:

a. A recalibration sequence is initiated (by drive
logic) at power on if the Read/Write heads are
not over track zero.

b. Upon receipt of the first Command Data bit,
Command Complete will stay false during the
entire command sequence.

This signal is driven by an open collector driver
as shown in Figure 4-1.


6.3 DATA TRANSFER LINES

All lines associated with the transfer of data between the
drive and the host system are differential in nature and may
not be multiplexed. These lines are provided at the J2/P2
connectors on all drives.

Four pair of balanced signals are used for the transfer of
data and clock: NRZ Write Data, NRZ Read Data, Write Clock,
and Read/Reference Clock. Figure 4-2 illustrates the
recommended driver/receiver circuit.


6.3.1 NRZ WRITE DATA

This is a differential pair that defines the data to
be written on the track. This data will be clocked
by the Write Clock signal. See Figure 6-10 for
timing.

6.3.2 NRZ READ DATA

The data recovered by reading previously written
information is transmitted to the host system via the
differential pair of NRZ Read Data lines. This data
is clocked by the Read Clock signal. See Figure 6-10
for timing. These lines will be held at a zero level
until PLO sync has been obtained and data is valid.


6.3.3 READ/REFERENCE CLOCK

The timing diagram as shown in Figure 6-10 depicts the
necessary sequence of events (with associated timing
restrictions for proper read/write operation of the
drive). The Read/Reference Clock is a glitchless
clock with a maximum of two missing clock pulses when
switching from reference to read clock.

6.3.4 WRITE CLOCK

Write Clock is provided by the controller. This clock
frequency shall be dictated by the Read/Reference
Clock during the write operation. See Figure 6-10 for
timing.

Write Clock need not be continuously supplied to the
drive. Write Clock should be supplied before
beginning a write operation and should last for the
duration of the write operation.


6.4 READ, WRITE, AND FORMAT PARAMETERS

6.4.1 GENERAL SUMMARY OF CRITICAL READ-FUNCTION TIMING
PARAMETERS

Controller variations of the read timing are allowed
if the following drive-dependent parameters are met:

a. Read Initialization Time

A read operation must not be initiated until 8
usec. following a head change.

b. Read-Gate Timing

Read Gate must not be enabled or true during a
Write Splice area (Read Gate must be deactivated
one bit time minimum before a Write Splice and
must be enabled one bit time minimum after a
Write Splice area).

c. Read Propagation Delay

Data (read) at the interface is delayed by
approximately 6 bit times from the data recorded
on the disk media.
d. Read Clock Timing

Read Clock and Read Data are valid within the
number of PLO sync field bytes specified by the
drive configuration after Read Enable and PLO
sync field is encountered. The Read/Reference
Clock line may contain no transitions for up to
2 Reference Clock periods when switching from a
write to a read. The transition period will also
be 1/2 of a Reference Clock period minimum with
no shortened pulse widths.

6.4.2 GENERAL SUMMARY OF CRITICAL WRITE-FUNCTION PARAMETERS

Controller timing variations in the record-update
function are allowed if the following drive-dependent
write (and interrelated read) timing parameters are
met:

a. Read-to-Write Recovery Time

Assuming head selection is stabilized, the time
lapse from deactivating Read Gate to activating
Write Gate shall be 5 Reference Clock periods
minimum.

b. Write Clock-to-Write Gate Timing

Write Clocks must precede Write Gate by a minimum
of 2-1/2 Reference Clock periods.

c. Write Driver Plus Data-Encoder Turn-On Write Gate

The Write Driver Plus Data-Encoded Turn-On Time
(write splice width) is between 3 and 7 Reference
Clock periods.

d. Write-Driver Turn-Off from Write Gate

To account for data-encoding delays, Write Gate
must be held on for at least 2 byte times after
the last bit of the information to be recorded.

e. Write-to-Read Recovery Time

The time lapse before Read Gate or Address Mark
Enable can be activated after deactivating the
Write Gate is defined by the "ISG Bytes after
Index/Sector" in configuration data response.
f. Head Switching Time

Write Gate may not be activated until 8 usec.
after a head change and Command complete is true.

Write Gate must be deactivated at least 1 usec.
before a head change.

g. Reference Clocks Valid Time

The Read/Reference Clock lines will contain valid
Reference Clocks within two Reference Clock
periods after the deactivation of Read Gate.
Pulse widths will not be shortened during the
after switching from a read to write but clock
transitions may not occur for up to 2 Reference
Clock periods.

h. Read Clocks Valid Time

The Read/Reference Clock line will contain valid
Read clocks within 2 clock periods after PLO
synchronization is established. Pulse widths
will not be shortened during the Reference Clock
to Read Clock transition time, but missing clocks
may occur for up to 2 clock periods.

i. Write Propagation Delay

Write Data received at the I/O connector will be
delayed by the Write Data Encoder by up to 8 bit
times maximum prior to being recorded on the
media.

6.4.3 FIXED SECTOR IMPLEMENTATION

6.4.3.1 Format Rules

The Index and Sector pulses are available
for use by the controller to indicate the
beginning of a track and to define the
beginning of a sector. A suggested format
for fixed data records is shown in Figure
6-11.


6.4.3.2 Intersector Gap (ISG)

The minimum Intersector Gap size is
determined from the configuration data.
The Intersector Gap provides separation
between each sector. The gap size is chosen
to provide for:

a. Drive required Write-to-Read recovery
time (minimum time between deassertion
of Write Gate and assertion of Read
Gate).

b. Drive required head switching time.

c. Control decision making time between
sectors.

d. Variations in detecting Index and
Sector.

6.4.3.3 Address Area

The address area (Figure 6-11) provides a
positive indication of the track and sector
locations. The address area is normally
read by the controller and the address bytes
verified prior to a data area read or write.
The address area is normally only written
by the controller during a format function
and thereafter only read to provide a
positive indication of the sector location
and establish the boundaries of the data
area. The address area consists of the
following bytes.

a. PLO Sync Field

These bytes are required by the drive
to allow the drive's read-data phase-
locked oscillator to become phase and
frequency synchronized with the data
bits recorded on the media. The
controller must send NRZ 00's during
this time.

b. Byte Sync Pattern (1 Byte Minimum)

This byte establishes byte synchron-
ization (i.e. the ability to partition
this ensuing serial bit stream into
meaningful information groupings such
as bytes) and indicates to the con-
troller the beginning of the address
field information. It is recommended
that the Byte Sync Pattern contain more
than a single 1 bit for a greater
confidence level of detection.

c. Address Field

These bytes are user-defined and
interpreted by the user's controller.
A suggested format consists of 5 bytes,
which allows 2 bytes to define the head
address, 1 byte to define the sector
address and 1 byte to define flag
status.

d. ADR Check Bytes (Address Field Check
Codes)

An appropriate error-detection mechanism
is generated by the controller and
applied to the address for data-
integrity purposes. These codes are
written on the media at format time.
Data integrity is maintained by the
controller recalculating and verifying
the address-field check codes when the
address field is read. ADR check bytes
are user defined.e.ADR Pad (2 Bytes Minimum)
(Address Field Pad)

The address Field Pad bytes must be
written by the controller and are
required by the drive to ensure proper
recording and recovery of the last bits
of the address-field check codes. These
pad bytes should be 00's.

6.4.3.4 Data Area

The Data Area (Figure 6-10) is used to
record data fields. The contents of the
data fields within the Data Area are
specified by the host system. The remaining
parts of the Data Area are specified and
interpreted by the disk controller to
recover the data fields and ensure their
integrity. The Data Areas consist of:

a. Write Splice (1 Byte Minimum)

This byte area is required by the drive
to allow time for the write drivers to
turn on and reach recording amplitude
sufficient to ensure data recovery.
This byte should be allowed for in the
format, and the controller should send
00's during this byte time.

b. PLO Sync Bytes

These bytes are required when reading
to allow the drive's phase-locked
oscillator to become phase and frequency
synchronized with the data bits recorded
in the media. The controller must send
NRZ 00's during these byte times.

c. Byte Sync Pattern (1 Byte Minimum)

This byte establishes byte synchroni-
zation and indicates, to the controller,
the beginning of the data field. It is
recommended that this byte contain more
than a single one bit.

d. Data Field

The data field contains the host
system's data files.e.Data Check Bytes

The Data Check or Error Check Code bytes
are generated by the controller and
written on the media at the end of the
Data Field. Data integrity is main-
tained by the controller recalculating
and verifying the Data Field Check Codes
or applying error correction algorithms,
if applicable, when the Data Field is
read. The Data Check Field is user
defined, but should have a correction
span of 6 bits or greater.

f. Data Pad (2 Bytes Minimum)

The Data Field Pad bytes must be issued
by the controller and is required by the
drive to ensure proper recording and
recovery of the last bits of the data
field check codes. The controller should
send 00's during these byte times.

6.4.3.5 Fixed Sector, Write Gate, PLO Sync Format
Timing



6.4.4 ADDRESS MARK IMPLEMENTATION (CONTROLLER SOFT SECTORED)

This section is included as an example to give meaning
to the definitions given.

6.4.4.1 Format Rules

The purpose of a format is to organize a
data track into smaller sequentially
numbered blocks of data called sectors.

6.4.4.2 Soft Sectored Format

The format shown below in Figure 6-12 is
similar to the format commonly used for hard
sectored disk drives and indicates minimum
requirements.

This format is a soft sectored type of
sector which means that the beginning of
each sector is defined by an ID Address Mark
followed by a prewritten identification (ID)
field which contains the logical sector
address plus cylinder and head information.
The ID field is then followed by a user
supplied data field. The definitions of the
functional area shown in the soft sectored format
are identical to those described for the hard
sectored format. There are some additional
fields in this format and they are the Address
Mark Field, Address Mark Pad, and ISG.

6.4.4.3 Address Mark Field

The address mark field is a field 3 bytes
long and is found before the PLO sync field
in and the address area. The contents of
this 3 byte field is drive dependent and is
written by the drive when so commanded by
Write Gate and Address Mark Enable active
simultaneously.

Detection of Address Mark indicates the
location of the beginning of a sector.


6.4.4.4 Address Mark Pad

The Address Mark Pad byte follows the
Address Mark field and is to be considered
a part of the PLO sync field. Its purpose
is to allow for Read Gate activation delays
after detecting the Address Mark Found signal.

6.4.4.5 Intersector Gap (ISG)

The ISG is included in the format to allow
for all those items discussed in Section
6.4.3.2.

6.4.4.6 Soft Sectored Address Mark, Write Gate, PLO
Sync Format Timing

This timing is mainly to support the unique
encoding for PLO sync field. The beginning
of each PLO sync field must be specified by
the controller. The beginning of the header
PLO sync field will be specified by the
trailing edge of the Address Mark Enable
signal when Write Gate is true. See Fig. 6-1.


6.5 DEFECT LIST

The 9380E series drives are shipped with a defect list that
is written in the ESDI Rev. F recommended format.

The defect list resides on Sector 0 of cylinder 1223 and is
repeated on cylinder 1215. This allows for redundancy should
an error occur on the maximum cylinder. The Sector 0 or any
surface will contain the defects for that surface. The format
for the data field portion (see Figure 6-13) of this sector
is 256 bytes with 2 bytes of CRC (x16 = 12 = x5 = 1):

Defect locations are 5 bytes long and the bytes are defined
in Figure 6-13.

The start of the actual defect may be off by up to 7 bits due
to the one byte resolution.

The end of the defect list for each surface will be indicated
by 5 bytes of ones in the defect location field or the end of
the sector.

The CRC check bytes should be used if that capability exists
but may be ignored if multiple reads are a more desirable
approach. CRC seed is zero. (Initialized state)*.

Byte count is the number of Bytes from Index.

*Sync byte will be included in the CRC calculation.


7.0 INSTALLATION

INTRODUCTION

This installation section is intended for the customer to
install the MiniScribe Drive 9380 in minimal time. The
section includes explanation of all pertinent jumpers as well
as unpacking instructions. It also includes a diagnostic
user's guide. This will aid the customer in running diagnos-
tics and explaining any error codes that may occur.


7.1 PHYSICAL INTERFACE

The electrical interface between the MiniScribe 9380 Series
and the host system is accomplished via four connectors: J1,
J2, J3 and J4 (ground spade connection). The connectors and
their recommended mating connectors are described below. Also
refer to Section 5.0.


7.1.1 POWER AND INTERFACE CABLES AND CONNECTORS

Figure 7-1 shows locations of the power and interface
connectors. Pin assignments for J1, J2 and J3 are
listed in Section 6.1.

The interface connection is made through connectors
J1 and J2 on the 9380 printed circuit board. The
control cable interconnects the controller and J1; the
data cable interconnects the controller and J2. See
Figure 7-3 for J1 and J2 orientation.

7.1.2 CONTROL SIGNAL CONNECTOR J1

J1 is a 34-pin board edge connector on the drive
printed circuit board. The signals on this connector
control the drive and transfer drive status to the
host controller. A key slot is provided between pins
4 and 6.

Recommended mating connector: AMP ribbon connector
88373-3, or equivalent (key slot between pins 4 and 6).

Recommended cable: 3M Scotchplex 3365/34 or equivalent.


7.1.3 DATA TRANSFER CONNECTOR J2

J2 is a 20-pin board connector on the printed circuit
board. The signals on this connector contain read or
write data. A key slot is provided between pins 4 and
6.

Recommended mating connector: AMP ribbon connector
88373-6 or equivalent.

Recommended cable: 3M Scotchflex 3365-20, or equiv-
alent (keyslot between pins 4 and 6).

7.1.4 DC POWER CONNECTOR - J3

J3 is a 4-pin keyed connector on the printed circuit
board. DC power (+5V and +12V) is supplied to the
drive via this connector.

Recommended connector: AMP 350543-1
Pins: AMP pin 350078-4

Note:1 Equivalents of the above are permissible
Note:2 Suggested wire size: 18 AWG

7.1.5 J4 FRAME GROUND CONNECTOR

J4 is a fasten-type connection. If wire is used, the
hole in J4 will accommodate wire size of 18 AWG max.

Recommended mating connector: AMP pin 61761-2, or
equivalent


7.2 9380E OPTION JUMPERS AND TEST POINT DESCRIPTION

Option Jumpers

J7 Start/Stop Spindle Motor Enable
J9 Diagnostic Jumper
J10, J11 Head Configuration

J10 J11 HEADS
S S 13
S 0 7
0 S 11
0 0 15

J12 J13 J19 SECTOR
0 0 0 35
0 S 0 36
S 0 0 34
S S S Soft(Controller will Select Sector #)

Drive Select Address Configuration

J16 1 2 3 Drive 1 2 3 Drive
0 0 0 No Select 0 0 1 4
1 0 0 1 1 0 1 5
0 1 0 2 0 1 1 6
1 1 0 3 1 1 1 7

Terminators
RP4
RP17

These 7 Jumpers must be Installed for Drive Operation

J14 J20 J21 J24 J27 J29 J30


7.3 9380E DRIVE OPTIONS


7.3.1 DRIVE ADDRESSING AND INTERFACE TERMINATION

Figure 7-1 shows the locations of the three J16 drive
address jumpers for drive address selection (drive
address 1 through 7) and interface terminators RP4 and
RP17 on the printed circuit board. Only one drive
address jumper is installed on a 9000 drive, and the
drive is addressed as Drive 1 at the factory. See
Table 6-1 Drive Addressing. See Table 6-3 for drive
address configurations and Section 6.1.1 for more
information.

Terminator packs RP4 and RP17 provide proper termina-
tion for interface lines. When daisy chaining
multiple 9000 drives, the terminators are installed
only in the last drive on the daisy chain.

7.3.2 SPINDLE CONTROL OPTION

Refer to Section 6.1.5.1F. Jumper J7 selects the
spindle control option. See Figure 7-1 for location.

- When J7 is installed, the 9000 must wait for a start
spindle command to start the spindle motor.

- When J7 is NOT installed, the drive automatically
starts the spindle motor at power on.

7.3.3 SECTOR CONFIGURATION OPTIONS

The 9380E series can operate in either hard or soft
sector modes. Figure 7-1 shows the locations of
Jumpers J12, J13, J19.JUMPER J19 SELECTS HARD OR SOFT SECTORING MODE.

- When J19, J13 and J12 are installed, the drive is
configured to operate in soft sectored mode.
ADDRESS MARK GENERATION and DETECTION are enabled,
and the SECTOR/ADDRESS MARK FOUND interface signal
is used to report ADDRESS MARK FOUND.

- When J19 is NOT installed, the drive is configured
to operate in the hard sectored mode. The SECTOR/-
ADDRESS MARK FOUND interface signal is used to
transmit sector pulses to the host controller. The
number of bytes/sector may be specified using the
SET BYTES PER SECTOR command or by selecting a
default sector configuration with option Jumpers J12
and J13. Three fixed sectored and one soft sectored
format is available as shown in Table 7-1. Also
refer to Sections 6.1.5.1I, 6.1.7, 6.2.7.

7.3.4 DAISY CHAINING THE MINISCRIBE 9380 DRIVES

Up to seven 9380 Series drives may be connected to a
single host controller/formatter. Control signals at
J1 are transmitted via standard, daisy-chain inter-
connection. Data signals at J2 are transmitted via
radially connected data-transfer lines. Figure 7-4
shows connections for a system configuration using
multiple 9000 drives.

Note: Interface terminators RP4 and RP17 are
installed only in the last physical drive in the
control chain. Maximum number of drives = 7.


Table 7-1
SECTOR CONFIGURATION JUMPERS

Sector J19 J12 J13 Unformatted Formatted Number
Configuration Bytes Bytes of
Per Sector Per Sector Sectors

Soft on on on Soft sectored mode
Fixed off on off 612 530 34
Fixed off off on 578 512 36
Fixed off off off 594 512 35

For all sector configurations, the configuration response given is
as follows:

36 Sectors 35 and 34 Sectors

Minimum Bytes per ISG Field 16 bytes 16 bytes
ISG Bytes after Index/sector 12 bytes 12 bytes
Minimum Bytes per PLO Sync Field 12 bytes 14 bytes


7.4 MOUNTING ORIENTATION

The MiniScribe 9380 Series uses industry-standard mounting for
5-1/4" Winchester Disk Drives (the same as for 5-1/4" flexible
disk drives). See Figure 2-1. The 9380 Series are designed
to be used in applications where the unit may experience shock
and vibration at greater levels than larger and heavier disk
drives. The drive may be mounted in any attitude.

The 9380 drive is mounted using 6-32 UNC screws 1/4" maximum
penetration. The customer must allow adequate ventilation to
the drive to insure reliable operation.

In as much as the drive frame acts as a heat sink to dissipate
heat from the unit, the enclosure and mounting structure
should be designed to allow natural convection of heat around
the HDA and outer frame. If the enclosure is small or if
natural convection is restricted, a fan may be required.


7.5 CABLING

Attach interface cables with connectors P1, P2, P3 and P4 to
J1, J2, J3 and J4, respectively. (See Figure 7-3). Insure
that connectors P1 and P2 have keys installed as indicated in
Figures 5-1 and 5-2. If multiple drives are to be inter-
connected, remove the terminator packs in all but the last
drive in the daisy chain. Refer to Figure 7-1 for terminator
locations and Figure 7-2 for cable inter-connections for
multiple drive systems.


7.6 DIAGNOSTICS


7.6.1 DIAGNOSTIC MODE

This section covers the Exercise mode for the
MiniScribe 9380 Series of drives. Error identifica-
tion and error message definitions are also in-
cluded.7.6.2GENERAL DESCRIPTION

The microprocessor performs a series of "wake up"
diagnostics upon application of power. These
diagnostics must pass in order for the drive to become
ready.

If an error is detected, the drive microprocessor will
transmit a vendor unique status word consisting of an
error code message through the interface. Refer to
Section 6.1.5.1C and Table 6-6.

If no errors are detected, the microprocessor checks
the configuration of the diagnostic Jumper J9. If J9
is installed before power up or if the Initiate
Diagnostic command is initiated (see Section 6.1.5.1H)
the drive enters a Diagnostic Seek Test routine. If
J9 is not installed or the Initiate Diagnostic Command
is not initiated, the drive will become ready and

remain in an Idle state waiting for control signals.

If an error is detected during wake up or initiated
diagnostics an error word will be transmitted through
the interface as a vendor unique status word (see
Section 6.1.5.1.C). Also the error coded message will
be flashed out through the drive LED. Refer to
Section 7.5.3 for LED interpretation.


Table 7-2
DIAGNOSTIC JUMPER CONFIGURATIONS

J9 Mode

Off Normal Operation

Installed before Crescendo Seek Test
power up

Installed before Random Seek Test
power up and removed
after the LED flashes
Microcode Rev level


7.6.3 ERROR MESSAGE READOUT

Flashing error codes are displayed through the drive
LED by the Microprocessor to indicate hardware
failures that occur while in the diagnostic mode (J9
installed) only. Failures during normal operation
(connected to the host) will be reported through the
ESDI interface.

Error codes are displayed in a "morse-code" type
manner. Flashes or bits are interpreted and converted
into a hexadecimal error code. "Zeroes" are indicated
by a short (1/2 second) flashing mode. "Ones" are
indicated by a short (1/2 second) continuous "on"
mode. Error codes are separated by a one-second LED
"off" time. All error codes and the revision level
are 6 bits long.

Zero = 1/2 second "flashing" mode
One = 1/2 second "continuous" on mode
Between Bits = 1/2 second off
Between repeat cycles (words) = 1.0 second off

EXAMPLE: "1A" 011010
0 1/2 sec FLASHING
1/2 sec OFF
1 1/2 sec ON
1/2 sec OFF
1 1/2 sec ON
1/2 sec OFF
0 1/2 sec FLASHING
1/2 sec OFF
1 1/2 sec ON
1/2 sec OFF
0 1/2 sec FLASHING
1.0 sec OFF

Listed below are the binary to hexadecimal conversion
values:

00 000000 10 010000 30 110000
01 000001 11 010001 31 110001
02 000010 12 010010 32 110010
03 000011 13 010011 39 111001
04 000100 14 010100 3E 111110
05 000101 15 010101
06 000110 16 010110
07 000111 17 010111
08 001000 18 011000
09 001001 19 011001
0A 001010 1A 011010
0B 001011 1B 011011
0C 001100 1C 011100
0D 001101 1D 011101
0E 001110 1E 011110
0F 001111 1F 011111

7.7 ERROR CODE DEFINITIONS

Code 00 - Microprocessor RAM error
Code 01 - Microprocessor ROM checksum error
Code 02 - Interface chip diagnostic failure
Code 03 - Sector counter fault
Code 04 - Index pulse not detected or lost
Code 05 - Spin speed not within 0.5% tol
Code 06 - Loss of fine track during idle mode
Code 07 - Reserved
Code 08 - Time out on +End Decel signal (during seek)
Code 09 - Tin out on CYL Pulse (during seek)
Code 0A - Overshoot (after a seek)
Code 0B - Time out on fine track (after a seek)
Code 0C - Track zero signal not detected (after a seek)

Code 0D - Comparator mismatch (after a seek)
Code 0E - Comparator mismatch (during a seek)
Code 0F - Unexpected interrupt from processor
Code 10 - Time out on non-GB pattern (during a rezero)
Code 11 - Time out on GB1 pattern (during a rezero)
Code 12 - Time out on GB2 pattern (during a rezero)
Code 13 - Seek range error
Code 14 - Voltage unsafe
Code 15 - Track offset fault
Code 16 - Write fault
Code 17 - Reserved
Code 18 - Time out on +End Decel sig (during a rezero)
Code 19 - Time out on CYL Pulse (during a rezero)
Code 1A - Overshoot (after a rezero)
Code 1B - Time out on fine track (after a rezero)
Code 1C - Track 0 signal not detected (after a rezero)
Code 1D - Comparator mismatch (after a rezero)
Code 30 - Time out on non-GB pattern (adj)
Code 31 - Time out on GB1 pattern (adj)
Code 32 - Time out on GB2 pattern (adj)
Code 39 - Time out on CYL Pulse (adj)
Code 3E - Cannot adjust servo

8.0 SCSI ELECTRICAL INTERFACE

The Small Computer System Interface (SCSI) on the M9-SCSI uses
single ended drivers and receivers, which allow a maximum
cable length of 20 feet (6 meters) from the first to the last
device (up to 8 devices) on the bus.

As shown in Figure 8-1 all assigned signals are terminated
with 220 ohms to +5 volts and 330 ohms to ground at each end
of the cable. Both M9 terminator networks on the M9-SCSI are
removable.

All signals driven by the M9-SCSI are Open Collector and have
the following output characteristics:

Signal Assertion 0.0 VDC to 0.4 VDC
Max. Driver Output 48 mA (sinking) @ 0.5 VDC
Signal Negation 2.5 to 5.25 VDC

All signals received by the M9-SCSI shall have the following
input characteristics:

Signal True 0.0 VDC to 0.8 VDC
Max. Total Input Load -0.4 mA @ 0.4 VDC
Signal False 2.0 to 5.25 VDC
Min. Input Hysteresis 0.2 VDC


8.1 SCSI COMPATIBILITY

The controller contains an on-board SCSI protocol controller
that controls SCSI protocol and the SCSI bus. The controller
supports SCSI arbitration and reselection capabilities.

The hexadecimal codes for the SCSI commands supported by the
controller are shown in Table 8-1.


Table 8-1
CONTROLLER SCSI COMMAND SET

Group 0 Command Hex Code Group 0 Command Hex Code

TEST DRIVE READY 00 REZERO UNIT 01
REQUEST SENSE 03 FORMAT UNIT 04
REASSIGN BLOCK 07 READ 08
WRITE 0A SEEK 0B
INQUIRY 12 MODE SELECT 15
RESERVE UNIT 16 RELEASE UNIT 17
MODE SENSE 1A RECEIVE DIAGNOSTIC 1C
SEND DIAGNOSTIC 1D START/STOP UNIT 1B

Group 1 Command Hex Code Group 1 Command Hex Code

READ BUFFER 3C READ DEFECT LIST 37
READ CAPACITY 25 SEEK (EXTENDED) 2B
READ (EXTENDED) 28 VERIFY 2F
WRITE (EXTENDED) 2A WRITE BUFFER 3B
WRITE and VERIFY 2E

Group 6 Command Hex Code

READ REVISION LEVEL C1

Group 7 Command Hex Code Group 7 Command Hex Code

FORMAT TRACK E4 READ LONG E8
WRITE LONG EA


8.1.1 TEST UNIT READY - 00H

The TEST UNIT READY command provides a means for the
Initiator to check if the logical unit is ready.

8.1.2 REZERO UNIT - 01H

The REZERO UNIT command requests that the controller
set the logical unit to logical block address zero.
8.1.3 REQUEST SENSE - 03H

The REQUEST SENSE command provides a means for the
Initiator to obtain more detailed information after
execution of a command. Typically, a REQUEST SENSE
command is issued after the previous command has
completed and a CHECK CONDITION status returned to the
Initiator.

An Initiator should issue a REQUEST SENSE command as
soon as it receives a CHECK CONDITION status code to
obtain the Sense data saved by the controller. The
Initiator can issue several REQUEST SENSE commands at
this time to obtain the Extended Sense data as well
as the Nonextended Sense data. However, when the con-
troller receives a command, other than a REQUEST
SENSE, from the same Initiator for the same LUN, it
clears the Sense data for the previous command.

Although the Nonextended Sense format is supported,
it is not recommended that this format be used in any
future products. All new development should use the
Extended Sense format and the Sense Key to process any
errors.

To determine the maximum length of the extended sense
information returned, refer to the value for the REQUEST SENSE
Length field in the INQUIRY command data format.

8.1.4 FORMAT UNIT - 04H

The FORMAT UNIT command ensures that the media is
formatted so that all data blocks can be accessed.
The controller maintains a defective sector and track
file on the disk on a cylinder that is inaccessible
to the Initiator. During the formatting process, the
Initiator may specify a set of defective blocks or
tracks to be reassigned using spare blocks or
alternate tracks as appropriate.

The FORMAT UNIT command uses four different sets of
defect information:

Manufacturers Defect List - The Manufacturers Defect
List is supplied by the manufacturer and is resident
on the disk drive.


NOTE

Usually, the user cannot directly access
the Manufacturer's Defect List. If you
need to do so, use the READ DEFECT LIST
command.

Certification Defect List - The Certification Defect
List is built by the controller while it is certify-
ing the drive after the format operation. The
controller does not perform a certification step
during the format operation so this defect list will
always be empty or null.

Initiator Defect List - The Initiator Defect List is
supplied by the Initiator during the Data Out Phase
of the FORMAT UNIT command. When the format operation
is completed, this list becomes part of the Grown
Defect List.

Grown Defect List - The Grown Defect List contains
any defects which were re-assigned using the RE-ASSIGN
BLOCK command.

The FORMAT UNIT command uses the drive geometry and
format information read from a reserved area on the
disk during power up to format the disk drive. These
parameters may be changed using the MODE SELECT
command just prior to issuing the FORMAT UNIT command.
If the information contained in the reserved area is
invalid or cannot be read, the FORMAT UNIT command
will use the following drive geometry/format informa-
tion (this is the default information returned by the
MODE SENSE command):

Number of Heads - The number of heads will be obtained
from the drive or controller switches.

Number of Logical Cylinders - The number of logical
cylinders is equal to the number of physical cylinders
minus three, minus the number of alternate cylinders.
The number of physical cylinders will be obtained from
the drive or controller switches.

Number of Logical Sectors per Track - The number of
Logical Sectors per Track is equal to the number of
physical sectors per track, minus the number of
Alternate Sectors. The number of physical sectors
per track will be obtained from the drive or con-
troller switches.

Number of Alternate Cylinders - The number of
Alternate Cylinders defaults to 3.

Number of Alternate Sectors per Track - The number of
Alternate Sectors per Track defaults to 1.
Track-to-Track Sector Skew - The controller uses the
default head skew value of 0 if the disk drive
indicates a head switch time that is less than 15
microseconds. When the disk drive indicates a head
switch time that is greater than 15 microseconds, the
head skew default value is 0Ah.

Cylinder-to-Cylinder Sector Skew - The Cylinder-to-
Cylinder sector skew defaults to 9.

The FORMAT UNIT command writes all of the MODE SELECT
parameters, including those mentioned above, to a
reserved area on the disk inaccessible to the Initiator.

8.1.5 RE-ASSIGN BLOCK - 07H

The RE-ASSIGN BLOCK command requests the controller
to re-assign the defective logical block(s) to an area
on the logical unit reserved for this purpose.

During the Data Out Phase, the Initiator transfers a
defect list that contains the logical block(s) to be
re-assigned. The controller will re-assign the
physical medium used for each logical block specified
by the Initiator. The data contained in those blocks
specified by the Initiator may be altered, but the
data in all other blocks will be preserved.

8.1.6 READ - 08H

The READ command requests that the controller transfer
data from the logical unit to the Initiator.

8.1.7 WRITE - 0AH

The WRITE command requests that the controller write
the data transferred by the Initiator to the logical
unit.

8.1.8 SEEK - 0BH

The SEEK command causes the selected LUN to seek to
the specified logical block location. If the logical
block number specifies a block on a defective track,
the seek to the alternate track is not performed until
the controller receives and processes a command which
accesses the media.

8.1.9 INQUIRY - 12H

The INQUIRY command provides a means by which the
Initiator may request information regarding the
controller and its attached peripheral device(s).
If an INQUIRY command is received from an Initiator
with a pending Unit Attention Condition, the con-
troller will execute the INQUIRY command, return a
GOOD status and will not clear the Unit Attention
Condition.

8.1.10 MODE SELECT - 15H

The MODE SELECT command provides a means by which the
Initiator may specify medium, logical unit and/or
peripheral device parameters to the controller. Any
changes in the MODE SELECT parameters take effect
immediately after the MODE SELECT command has
terminated. The MODE SELECT is a complementary
command to the MODE SENSE command, which allows the
Initiator to request that the controller send it the
current values for the parameters.

8.1.11 RESERVE UNIT - 16H

The RESERVE UNIT command is used to reserve the
specified LUN for exclusive use by the Initiator.

8.1.12 RELEASE UNIT - 17H

The RELEASE UNIT command causes the LUN (connected to
the controller and previously reserved by the RESERVE
UNIT command) to be released. Once the RELEASE UNIT
command is issued, other Initiators can access the LUN.

8.1.13 MODE SENSE - 1AH

The MODE SENSE command provides a means by which the
Initiator may receive the medium, logical unit and
peripheral device parameters from the controller.
MODE SENSE is a complementary command to the MODE
SELECT command.

The controller will send blocks of parameters that
are separated into categories (called pages). These
parameters specify various options and features which
the Initiator may change. Each page is preceded by
a Page Code and the length of the page. The Page
Length value does not include Bytes 0 or 1 of the page.

8.1.14 RECEIVE DIAGNOSTIC RESULTS - 1CH

The RECEIVE DIAGNOSTIC RESULTS command requests
analysis data be sent to the Initiator after comple-
tion of a SEND DIAGNOSTIC command.8.1.15SEND DIAGNOSTIC - 1DH

The SEND DIAGNOSTIC command requests the controller
to perform diagnostic tests on itself, on the attached
peripheral device(s), or on both. This command is
usually followed by the RECEIVE DIAGNOSTIC RESULTS
command except when the Self Test bit (SlfTst) is set to 1.

8.1.16 START/STOP UNIT - 1BH

The START/STOP UNIT command requests that the
controller enable or disable the logical unit for
further operations.

8.1.17 READ BUFFER - 3CH

The READ BUFFER command is used in conjunction with
the WRITE BUFFER command as a diagnostic function for
testing the controller's data buffer memory and the
SCSI bus integrity. There is no media access with
this command.

8.1.18 READ DEFECT LIST - 37H

The READ DEFECT LIST command requests that the
controller transfer the defect list maintained by the
controller to the Initiator.

8.1.19 READ CAPACITY - 25H

The READ CAPACITY command is used to determine the
maximum logical block number on the specified LUN
which can be accessed by the Initiator. This command
also returns the size of logical block. The informa-
tion is returned to the Initiator during the Data In
phase.

8.1.20 SEEK (EXTENDED) - 2BH


The SEEK (EXTENDED) command causes the selected LUN
to begin a seek operation to the specified logical
block location. If the logical block number specifies
a block on a defective track, the seek to the alter-
nate track is not performed until the controller
receives and processes an I/O command.

8.1.21 READ (EXTENDED) - 28H

The READ (EXTENDED) command requests that the
controller transfer data from the logical unit to the
Initiator.8.1.22VERIFY - 2FH

The VERIFY command requests that the controller verify
the data written on the logical unit.

8.1.23 WRITE (EXTENDED) -2AH

The WRITE (EXTENDED) command requests that the con-
troller write the data transferred by the Initiator
to the specified logical unit.

8.1.24 WRITE BUFFER - 3BH

The WRITE BUFFER command is used in conjunction with
the READ BUFFER command as a diagnostic function for
testing the controller's data buffer memory and the
SCSI bus integrity. There is no media access with
this command.

8.1.25 WRITE AND VERIFY - 2EH

The WRITE AND VERIFY command requests that the con-
troller write the data transferred by the Initiator
to the logical unit and verify the data written on the
logical unit.

8.1.26 READ REVISION LEVEL - C1H

The READ REVISION LEVEL command returns the current
revision level of the PROM residing on the controller
to the Initiator during the Data In phase.

8.1.27 FORMAT TRACK - E4H

The FORMAT TRACK command formats a single physical
track according to the current parameters established
with the MODE SELECT command.

8.1.28 READ LONG - E8H

The READ LONG command requests the controller to
perform a read operation of one data block and the
six ECC bytes associated with that block. The data
from the block and the ECC bytes are transferred to
the Initiator during the Data In phase.

8.1.29 WRITE LONG - EAH

The WRITE LONG command requests the controller to
perform a write operation of one data block and the
six bytes of ECC information. The data and the six
ECC bytes for the specified logical block are supplied
by the Initiator during the Data Out phase.

9.0 M9380S JUMPER INSTRUCTIONS

The following describes the M9380S SCSI configuration jumpers
available to a customer.


DEFINITIONS:

JUMPER LOCATION - each jumper location represents a pair of
pins. When applying a jumper to a location, the jumper is
placed across the pair of pins represented by the designator.
The format of the designator is JXXX-Y. This indicates that
there is a group of pairs of pins designated by XXX and that
each pair in that group is designated by Y.

example: J601-1

OFF - a jumper location is OFF if the jumper IS NOT
applied.

ON - a jumper location is ON if the jumper IS applied.


9.1 SCSI ADDRESS JUMPERS (J601)

J601 is a group of three jumper locations which define the
M9380S SCSI bus address.

J601-1 J601-2 J601-3

SCSI ADDRESS 0 OFF OFF OFF
SCSI ADDRESS 1 OFF OFF ON
SCSI ADDRESS 2 OFF ON OFF
SCSI ADDRESS 3 OFF ON ON
SCSI ADDRESS 4 ON OFF OFF
SCSI ADDRESS 5 ON OFF ON
SCSI ADDRESS 6 ON ON OFF
SCSI ADDRESS 7 ON ON ON


9.2 SCSI PARITY ENABLE (J602)

J602 is a group of two pairs of jumper pins. The first pair
is presently undefined. The second pair defines SCSI parity
enable, as follows:

J602-2

SCSI PARITY ENABLED OFF
SCSI PARITY DISABLED ON


9.3 SCSI TERMINATOR POWER (J701)

J701 is a group of two pairs of jumper pins. The first pair
controls terminator power supplied by the target while the
second pair controls power supplied from elsewhere on the bus.

J701-1 J701-2

LOCAL TERMINATOR POWER ON OFF
REMOTE TERMINATOR POWER OFF ON


9.4 ADDITIONAL JUMPER DEFINITIONS

J7 - START/STOP SPINDLE MOTOR ENABLE
J9 - DIAGNOSTIC JUMPER
J10/J11 - HEAD CONFIGURATION
J12 - SECTOR SETTING
J13 - SECTOR SETTING

These seven (7) jumpers must be installed for drive operation:

J14, J20, J21, J24, J27, J29, J30

RP701/RP702 - TERMINATOR RESISTORS


9.5 50 PIN SCSI CONNECTOR - J4

J4 is a 50 pin straight solder pin connector on the printed
circuit board. This cable connector is used to send and
receive information from the SCSI Host Adapter with keyed
slots between pins 3 and 5, 23, 25 and 27, and 47 and 49.

Recommended Mating Connector: Burndy FRS50BF-8P or equivalent.

Recommended cable is a Belden 50 pin flat cable, part number
9L28050 or equivalent.


9.6 DC POWER CONNECTOR - J3

J3 is a 4 pin keyed connector on the printed circuit board.
DC power (+5V and +12V) is supplied to the drive via this
connector.

Recommended connector: Amp 350543-1
Pins: Amp 350078-4

Note 1: Equivalents of the above are permissible.
Note 2: Suggested wire size 18AWG.


9.7 FRAME GROUND CONNECTOR - J5

J5 is a Fasten-type connector. If wire is used, the hole in
J5 will accommodate wire size of 18AWG maximum.

Recommended mating connector: Amp pin 61761-2 or equivalent.


10.0 ERROR CODES

Code 0 Self-test RESET code
Code 1 Buffer controller reset test
Code 2 Disk formatter reset test
Code 4 SCSI reset latch test

Code 5 8031 microcontroller test
Code 6 ROM checksum test
Code 7 buffer controller register test
Code 8 external data RAM test
Code 9 external parity RAM test
Code 0Ah BC parity detection test
Code 0Bh BC parity interrupt test
Code 0Ch Mac chip register test

Code 0Dh disk formatter register test
Code 0Eh disk formatter interrupt test
Code 0Fh tape formatter register test
Code 10h SCSI ESP self-diagnostic test
Code 11h SCSI ESP interrupt test
Code 12h SCSI ESP register test


CAUTION/WARNING

THE MINISCRIBE DRIVE IS A PRECISION PRODUCT.
DURING HANDLING THE PRODUCT MUST NOT BE DROP-
PED, JARRED, OR BUMPED, OTHERWISE DAMAGE TO THE
HEADS AND DISKS MAY OCCUR. THE STAND ALONE
DRIVE IS SENSITIVE TO ELECTRO-STATIC DISCHARGE
(ESD). PROPER ESD PRECAUTIONS (PERSONNEL AND
EQUIPMENT GROUNDING) ARE REQUIRED PRIOR TO
UNPACKING OR HANDLING THE DRIVE. FAILURE TO
FOLLOW THESE PRECAUTIONS COULD LEAD TO AN
IMMEDIATE CATASTROPHIC DRIVE FAILURE OR A
PREMATURE RELIABILITY FAILURE. WHEN THE DRIVE
IS REMOVED FROM THE MINISCRIBE SHIPPING CON-
TAINER AND NOT IMMEDIATELY SECURED WITHIN A
CHASSIS THROUGH ITS SHOCK MOUNTS, IT MUST BE
STORED ON A PADDED, GROUNDED, ANTISTATIC
SURFACE.

FAILURE TO COMPLY WITH THE ABOVE PROCEDURE WILL
RENDER NULL AND VOID ALL WARRANTIES.


11.0 UNPACKING AND INSPECTION


11.1 SINGLE PACK

Retain the packing materials for reuse. Refer to Figure 11-
1 for the following steps.

STEP 1: Inspect the shipping container for evidence of damage
in transit. If damage is evident, notify the carrier
immediately.

STEP 2: Open the outer carton by carefully cutting the tape
on the top of the carton.

STEP 3: Lift the inner carton out of the outer carton and
remove the outer end foam cushions.

STEP 4: Open the inner carton by carefully cutting the tape
on the top of the carton.

STEP 5: Lift the drive from the inner carton and remove the
inner end foam cushions and cardboard wrap.

STEP 6: Place the pair of end cushions, the cardboard wrap,
and the inner carton within the outer carton and store
for subsequent reuse.

STEP 7: Inspect the drive for shipping damage, loose screws
or components and correct, if possible. If damage is
evident without noticeable damage to the shipping
cartons, notify MiniScribe immediately for drive
disposition.


11.2 MULTIPACK

Retain the packing materials for reuse. Refer to Figure 11-
2 for the following steps.

STEP 1: Inspect the shipping container for evidence of dam-
age in transit. If damage is evident, notify the
carrier immediately.

STEP 2: Remove banding from container and shrink wrap if used.

STEP 3: Lift off upper cap and outer sleeve surrounding the
multipacks.

STEP 4: Open each multipack carton and remove foam cap to
access drives.

STEP 5: Remove each drive from the carton by the tabs on each
cardboard sleeve containing them.

STEP 6: Take each drive from cardboard sleeve and remove from
anti-static bag.

STEP 7: Place the drive on a protective foam pad and inspect
the drive for shipping damage, loose screws or com-
ponents and correct, if possible. If damage is
evident without noticeable damage to the shipping
carton, notify MiniScribe immediately for drive
disposition.

STEP 8: Once all the drives have been removed from the
shipping carton and the cardboard sleeves have been
returned to the slots in the foam, replace foam cap
and store for reuse.

STEP 9: When using a multipack carton for reshipping, it is
not to be shipped unless secured to a standard size
wood pallet of 40" X 48".


11.3 REPACKING

Should the MiniScribe drive require shipment, repack the drive
using the MiniScribe packing materials and follow the steps
in Section 11.1 or 11.2 in reverse order.


NOTICE

THE MINISCRIBE DRIVE WARRANTY IS VOID IF
THE DRIVE IS RETURNED TO MINISCRIBE IN OTHER
THAN THE STANDARD MINISCRIBE SHIPPING CARTON
PACKED IN ACCORDANCE WITH THE ENCLOSED
PROCEDURE.


APPENDIX A
DEFINITION AND MEASUREMENT OF SEEK TIME


The measurement of seek time is important for understanding a disk
drive's performance. Seek time is measured at the -SEEK COMPLETE
interface signal. It is defined as the duration of time in which
Seek Complete is false.

The seek time performance of a disk drive is often indicated by
its average access or seek time. This is an average value of seek
time during random activity and can be calculated. Using the seek
time for each possible move length, the number of possibilities for
each move length, and the total number of possible seeks, the
following method can be used:

n = number of cylinders
m = seek length
T(m) = seek time for m length
Number of possible seeks of m length = 2(n-m)
Number of possible seeks (total) = n(n-1)

n-1
Average Seek Time = 2(n-m) T(m)
m=1
---------------------
n(n-1)
or

Sum of seek times for all possible seeks
Average Seek Time = ----------------------------------------
Number of possible seeks

This method requires numerous measurements and is often replaced
by the measurement of the one-third stroke seek time. The average
distance traveled during random activity is one-third of the total
number of cylinders. The seek time for this move is the same as
the average seek time for constant velocity positioners. Due to
the nonlinear nature of the velocity profile for high performance,
closed-loop positioners, the one-third stroke seek time is usually
longer than the average seek time. Although the average seek time
method contains more information about system throughput, the
shorter method is most often used on automatic test equipment
throughout the industry.

Effect of Step Rate

Since seek time is measured from the receipt of the first step
pulse, the rate of the incoming pulses will have an effect on seek
time measurements. This effect is minimized in the 9000 series
because motion begins immediately after the first step pulse is
received. The velocity of this initial motion is still physically
limited by the step rate, so a faster step rate will result in a
faster seek time.
APPENDIX B

9380SM

The 9380SM is an Apple compatible 380MB SCSI interface disk drive.
*The typical formatted capacity for this disk drive is 325MB. The
9380SM utilizes PAGE 00H, PAGE 20H AND PAGE 30H.

PAGE 00H - UNIT ATTENTION
PAGE 20H - Serial Number
PAGE 3CH - Software Selectable ID

* Certain operating systems and application indicate a smaller
formatted capacity.


SOFTWARE SELECTABLE ID

The intention of the SSID function is to be able to alter the SCSI
TARGET ADDRESS via mode page 3C. The page is made up of the page
number, page length, and one data byte. The data byte contains two
fields: a three bit binary address and one bit to indicate that the
binary address saved on the disk is valid. If valid, the binary
address stored on the disk is used by the SCSI controller as the
legitimate target ID. If not valid, then the jumpered address is
used. If any of the four unused bits are saved on the disk as
active, then the address byte is considered invalid regardless of
the valid bit and the jumpered address is used.

Target ID evaluation will take place during power-on reset, SCSI
bus reset, and Bus Device Reset message.


MODE SELECT PAGE 3C


BIT 7 6 5 4 3 2 1 0
BYTE 0 | 0 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | PAGE NUMBER
BYTE 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | PAGE LENGTH
BYTE 2 | VLD | 0 | 0 | 0 | 0 | BA4 | BA2 | BA1 | DATE BYTE


NOTE: User should note that page 3C is saveable. When
executing a mode select to page 3C, the user should set
the SP (save page) bit at CDB 1 bit 0. Without this bit,
this data is not saved to the disk and is not available
at reset time.


MODE SENSE PAGE 3C


BIT 7 6 5 4 3 2 1 0
BYTE 0 | 1 | 0 | 1 | 1 | 1 | 1 | 0 | 0 | PAGE NUMBER
BYTE 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | PAGE LENGTH
BYTE 2 | VLD | 0 | 0 | 0 | 0 | BA4 | BA2 | BA1 | DATE BYTE


NOTE: User should note that byte 0 bit y is active to indicate
that page 3C is saveable, producing a value of BC, not 3C.

Regardless of the type of MODE SENSE requested, bytes 0 and 1 will
appear as listed above. Byte 2 will vary as follows:

CURRENT: Represents the current Target ID. (This is unusable
since the user must know the ID to issue the command.
It was coded to be consistent with other pages.)

CHANGEABLE: Indicates which bits in page 3C may be modified via
MODE SELECT.

DEFAULT: Indicates the jumpered device address.

SAVED: Indicates the target ID byte saved on the disk. If
valid, this ID will be used as the result of any
reset.