home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Oakland CPM Archive
/
oakcpm.iso
/
cpm
/
gendoc
/
siemens8.iqf
/
SIEMENS8.INF
Wrap
Text File
|
1985-02-09
|
4KB
|
87 lines
Tips on getting the Siemens 100-8 SSDD 8 inch
disk drives running with the Morrow DJDMA disk controller.
Two problems arise when configuring the 100-8 for the DJDMA. The first
is the jumper settings, and the second is the CP/M BIOS drive parameter tables.
The DJDMA needs some odd (compared to other controllers) jumper settings, and
the 100-8 drive parameters (step rate and settling time) are not always what
Siemens' (now World Disk Drives, Irvine, California) documentation would lead
one to expect.
Jumper Settings
The DJDMA requires the following characteristics of its disk drives:
1. Radial (daisy chained) drive select
2. Head load when drive is selected and door is closed
3. In-Use LED on from drive select
4. Stepper power from head load
5. Soft sector
6. Terminating resistors on the last drive in the daisy chain
For the 100-8 the approriate jumper settings are:
Rad Sel 0 in drive A, 1 in drive B
Rad Step 2
12 yes
16 no
18 no
20 yes
22 yes
24 yes
34 yes
36 yes
A no
B no
C no
D yes
E yes
F no
G yes
H no
J no
K yes
L yes
M no
RI yes
RR yes
V no
SS yes
HS no
Term Res. no on drive A, yes on drive B
Act LED U no
S no
R no
H yes
SE yes
TE no
BIOS disk parameters
The Morrow BIOS comes configured for Shugart 800/801 drives. They
require 8 millisecond step and settle times. The Siemens documentation
says that the 100-8 needs 3 to 8 milliseconds "access" (by which I think
they intended "step") time and 14 milliseconds settling time. I've got the
feeling they reversed the numbers. 8 milliseconds was sufficient settling
time under all conditions. I tried 8 milliseconds as the step parameter and
the operation was erratic. More often than not the drives couldn't keep up
with the step rate and would fail the track seek. I then changed the step
parameter to 14 milliseconds and all worked fine. I tried reducing the step
parameter in 2 millisecond increments and all worked fine down to 10
milliseconds. To be safe, start with 14 milliseconds and if you're
adventurous, try reducing it until you start getting slightly erratic seeks.
You can tell you've gone too far if, instead of getting a clean head movement
directly to the requested track, it tries more than once, returning to track
0 in the process. Back off 2 milliseconds and your drive will be seeking as
fast as it can. I don't know if stepping the drive at faster than 14
milliseconds hurts the long term reliability of the drive, so if you're
worried about such things, leave it at 14. I used DU-V87.COM (available on
many RCPM systems) to help pinpoint the seek problems. It allows you to
command the head directly to any track.
Len Moskowitz
Fair Lawn, N.J.
October 1984