- IMSAI 8k BASIC Listing (pdf)
- An Introduction to the DOS-A IMSAI Floppy Disk Operating System (pdf)
- IMSAI/Tarbell Bootstrap Loader (pdf)
- JRT Systems Pascal Reference Manual (pdf)
- MicroPro Word-Master Operators Manual (pdf)
- Mini-article from Dr. Dobb's (9/76:17) regarding "...the memory
clobbering phantom!" (pdf)
- Solid State Music 2P/2S Manual (pdf) (JC)
- Structured Systems Group, Inc. NAD Name and Address System manual (pdf)
- Structured Systems Group, Inc. QSort User's Manual (pdf)
- TDL Z-80 Relocating Macro Assembler User's Manual
(Technical Design Labs) (rev. 1.2; 12/9/76) (pdf)
- TDL Text Editor User's Manual Preliminary (Technical Design Labs) (12/76) (pdf)
- TDL Text Output Processor Manual (Technical Design Labs) (rev. 1.0; 2/14/77)
(pdf)
- TDL Zapple 12k BASIC Version 3 User's Manual (Technical Design Labs) (rev. 0; 1/78)
(pdf)
- TDL Zapple Monitor (Technical Design Labs) (pdf)
- TDL 8k BASIC Manual (Technical Design Labs) (pdf)
- Disk Basic Version 1.06 Documentation Update (Xitan) (pdf)
- Z80-CPU Technical Manual (Zilog) (1976) (pdf)
- Zilog Data Book (Zilog) (1979) (pdf)
CP/M and Microsoft Manuals
- CP/M Manual (pdf)
- CREF Cross-reference utility (pdf)
- LIB-80 Librarian (pdf)
- LINK-80 Linker (pdf)
- MACRO-80 Macro Assembler (pdf)
- BASIC-80 Microsoft BASIC 5.0 and greater (pdf)
The Frugal Floppy system I have is a bit flaky. Actually, it's not the
system per se, but the Pertec FD400 floppy drives. So, I wanted to replace this floppy system with a working one based on 5.25" disks.
Not only is it difficult finding a working controller capable of 5.25" disks
but if
you're lucky enough to find one, it's even harder to find a working boot disk.
In many circumstances, you can make an image of an 8" diskette using
ImageDisk and take that image and write it onto a 5.25" high-density (1.2mb)
diskette. The drive RPM (360) and transfer rate (500kbps) is the same as an 8"
drive, and
the 5.25" drive is capable of stepping 80 tracks, which is greater than the 77 tracks on 8"
disks.
So, I'm going to chronicle two things here -- repairing the 8" system and
converting to a 5.25" system. In advance, I owe a load of thanks to Herb
Johnson, Mike Stein, Tony Duell, and many others on the
ClassicCmp list for their assistance
with this project.
Implementing a new 5.25" Disk System
I tried this method first, and it actually worked better than I had
anticipated. With the help of another hobbyist (John Singleton) who had the
exact same system and had gone through the same process, I was able to
re-implement a 5.25" disk system from scratch using his CBIOS and method. I wrote a complete paper on
this, available as a FAQ
here (MS Word).
Essentially, this project required a few things:
- 64k of memory, less 2k to accommodate the monitor EPROM. I used a
CompuPro RAM17 (manual pdf) with one of the 6116 SRAM chips replaced with an EPROM. Other
configurations are possible, but essentially the top 4k of memory space is
used for the monitor, CP/M CBIOS, disk buffers and monitor variables.
- A good system monitor program. Calling it a monitor is understating its
capabilities. The FAQ
identifies in detail what capabilities the monitor program needs, but in
addition to "regular" monitor capabilities, it needs the ability to read HEX
files, cold-boot CP/M, write CP/M to disk, format blank disks, and also act
as the CP/M CBIOS.
- A working disk controller board, cabling and drives. I used the CompuPro
Disk 1 controller (manual pdf) and YE Data YE-380 1.2mb drives (used on old IMB PC/AT
machines). The YE drives are great because they have head load solenoids and
operate just like 8" drives.
- CP/M 2.2 sources...available from
http://www.cpm.z80.de/source.html (mirror).
This archive has both 8080- and Z80-neumonic source files -- use the
right one for your application.
- CP/M 2.2 "standard" distribution...available from
http://www.retroarchive.org/cpm/os/os.htm (stdcpm22.zip) (mirror)
- CP/M 2.2 assorted patches from cpm.z80.de (mirror)
- CP/M emulator for compiling. I used MyZ80 from the Walnut Creek CP/M
CD-ROM (mirror)
- Lots of patience.
Overall, the project took about two months, but that included acquiring
parts and doing testing. Once I got going, it only took a couple of weeks to
implement and test. Now, I have a complete and fully-working dual-drive system
that uses 5.25" floppy drives to emulate 8" drives. I mounted my drives in an
old DEC RX-180 external drive case -- convenient, but not IMSAI obviously. I guess I could spray paint it IMSAI blue
but for now it's vintage enough.
May, 2008: A few weeks ago, I came across a post on the Vintage
Computer Forums (http://www.vintage-computer.com/vcforum/index.php)
which mentioned a software-based disk drive emulator called the Altair
Peripheral Emulator ("APE") (http://frankbarberis.tech.officelive.com/default.aspx).
APE uses two pieces of software to emulate a CP/M compatible 4-drive floppy
disk system. A Windows program runs on the host machine, which connects to an
S-100 machine through a serial cable. A small boot loader is toggled into the
S-100 machine which loads a secondary boot loader that does the heavy lifting
to get CP/M booted.
I contacted Frank and he is very helpful and was able to create a custom
floppy that matched my S-100 serial port configuration (I/O port location and
status bits). It's a little slow because of the 9600-baud serial connection
but absolutely very functional. If you have a fast enough (4MHz) S-100 machine
and serial card, it is capable of running at 19,200 baud which should be more
than adequate for daily use.
I usually use a Windows-based laptop as a terminal for my classic machines.
For this setup, I use two USB-to-RS232 adapters connected to the two SIO ports
on my IMSAI and I run but Hyperterm and APE on the laptop. Works like a charm.
Thanks, Frank, for a great tool.
Fixing the Frugal Floppy System
The iCOM "Frugal Floppy" System is a nice dual-8" floppy system. The actual
system is manufactured by Synetic Designs Company but its based on the iCOM
FD360 controller-formatter and Pertec FD400 drives. These are the same drives
used in the original Altair 88-DCDD. FD400 drives have direct-drive DC
spindles and a soft-sectored geometry of 77 tracks containing 26, 128-byte
sectors, for a total of 256,256 bytes.
The controller-formatter is a two-board, totally TTL (i.e., no LSI logic)
controller which connects to the host through a 50-pin non-Shugart-compatible
interface. This cable connects to a host interface card in the IMSAI. This
card contains some glue logic, buffer RAM and an EPROM.
In this system, the first drive (the top one in the picture above) works
perfectly, and the system boots CP/M 1.4 with no problems at all. The bottom
drive worked for a while when I first got it but now suffers from CRC errors.
There is a light on the drive cabinet that indicates CRC errors, and I get
nothing but BDOS errors when accessing that drive.
Given this, there are a few procedures that I've done and still have yet to
do:
- Clean the heads. I first used a diskette-based wet cleaner and
that didn't seem to make much of a difference. Next, I plan on cleaning the
heads manually, using isopropyl alcohol and a swab to see if that helps.
-
Check the spindle speed. The
procedure in the manual calls for counting pulses at a tie point on the
controller card. One can also use an optical/non-contact tachometer to check
the speed (which should be 360 RPM). The rub here is that the spindle motor
does not have an optical tach disk ("zebra stripe"), so you have to make one
or copy one from another floppy drive (like the Tandon TM100) and glue it to
the spindle. For convenience, here's a copy of mine.
For the
tachometer, there are several options: neon light connected to 110v/60Hz;
xenon stroboscope; Model 100 Bar Code Reader (Portable 100 magazine,
2/88, p.32). The associated machine language program has to be modified to
count more pulses because we need to count at least one second's worth of
pulses. Further, the BASIC program needs to be modified slightly to have the
math work out right.
-
Result: Well, the initial test run
of the bar code stroboscope didn't turn out as I had hoped. I think I have
some more work to do on getting the math correct. I'm testing against a
known-good Tandon TM100 floppy drive (which has a 60Hz strobe disk on the
flywheel) and I'm getting nonsense readings.
Cromemco Dazzler
I recently decided to pull out one of the Cromemco Dazzler boards I have
and see if I could get it to work on my IMSAI. Well' with some minor
adjustments, it works just fine. Now I'm searching for a Cromemco 7A+D
analog/digital board so I can get the joystick interface working.
Copyright (c) 1998-2010 Richard A. Cini, Jr. (rcini at msn dot com) All Rights Reserved.
All copyrights of any third parties
referred to herein are hereby acknowledged. There is no warranty, either
express or implied, relating to any of the content contained herein. The site maintainer
shall in no event be liable to anyone for damages, including any loss of profits, lost
savings, or other incidental or consequential damages arising out of the use or misuse of
the information contained on this Web site. Batteries not included. You may use the information contained
herein for NON-COMMERCIAL purposes only and AT YOUR OWN RISK.
Last Updated
08/05/2010 20:09 -0400
