This is an IMSAI 8080 (S/N 1017107) that came with four S-100 cards:
- Vector Graphic Z80 CPU
- MOSTEK Z80 CPU date code 7940
- other chips on board 78 (TTL, 7805)
- "Vector Graphic, Inc. PROM/RAM Board"Â REV 3
- some TTL date code 76, others 77
- PROMÂ "Page 0" 1702A "label unreadable"
- PROMÂ "Page 1" 1702A labeled: "E" #1
- Industrial Micro Systems 8K Memory Bd (C00231)
- Seals Electronics SRAM card
- hand written "1014" on back
- chip date codes 78
The chassis components include:
- Back plane IMSAI REV-1 S/N 011307
- Power board S/N 004072
- Front panel CP-A REV-4 hand dated 5/24/79
This restoration adventure began in early January, 2016.
A little history
The first IMSAI 8080 machines shipped as kits December 16, 1975, and shortly after that as both kits and assembled computers. The price: $931 assembled, according to the February 1976 product catalog—about $3800 in 2015 dollars. That's without memory. In 1976, as IMS completed its transition from a consultancy firm into a manufacturing firm, the name of the company changed to IMSAI Manufacturing Corporation. The company filed for bankruptcy in October 1979, just three years later. Reasons for its demise include the rapid shift microcomputers were making to a mass consumer-oriented market—the TRS-80, Commodore PET, and Apple II computers had all been launched—and IMSAI founder Bill Millard had his sights on starting up the lucrative Computerland franchise. IMSAI did very well at refining the Altair 8800, but like many early microcomputer companies (including MITS and Processor Technology), it couldn't successfully evolve to the next generation of products.
The total production of the IMSAI 8080 between 1975 and 1979 was more than 17,000, but probably less than 20,000 units. The Fischer-Freitas Company, who had purchased some of the IMSAI rights and materials after bankruptcy, produced approximately another 2200 systems between late 1979 and mid-1986.Â
How many IMSAI 8080's were made? (from IMSAI.net)
According to a July 2002 e-mail from the IMSAI 8080 creator Joe Killian,
"I remember passing the 17,000 mark. Â I believe it never went past 20,000" Â [between December of 1975 and September 1979].Â
Fischer-Freitas Company produced another 2100+ machines between November 1979 and June 1986. Â These machines can be identified by the Fischer-Freitas Company back panel label.Â
Systems produced after June of 1980 were supplied with the one-piece front panel masked which consisted of a laminated Mylar mask assembly adhered to a clear acrylic backing panel.
IMSAI 8080Â S/N 1017107 possibly turned off the lights as it left the San Leandro factory.Â
This particular machine is a small technological time capsule from the late 70's. I don't know if it was used or tinkered with since the original owner, but there is something pure about the computer. I may be the first to work on it in decades.  I have the opportunity to explore the machine, documenting its state and use, with minimal interference from others in the intervening years. I enjoy taking a computer that was unused for years and restoring it to a functioning state. It feels like a direct connection back to the early hobbyist era of microcomputers.
The seller said, "it came from Georgia in a town called Clayton." Of the original owner, he adds, "I do know he bought it new in 79 I think, his wife told me about how excited he was back then..."
The problem
There are multiple problems. This machine is in the process of a deep restoration. Two mud wasp nests were removed from the panel switch escutcheon. Evidence of a third mud wasp nest was found on the inside top of the case. I think the desiccated body of a spider is trapped in the coils of the transformer. There are no signs of small mammal activity, though, and no corrosion on the case or components. All of the original plastic switch levers are present and in good shape.
Images of the machine showed that it powered up and the front panel lights turned on. However, I wanted to bring up the power supply slowly and verify its output. I removed the front panel, escutcheon, and all cards from the chassis. Loose dirt and dust was cleaned out.
With all cards removed, I brought the power supply up slowly using a variac. The power supply checks out with voltages within tolerance for an unloaded supply,
yellow (-16V): reads -17.82VÂ
orange (+16V): read +17.80V or so
I also tested the continuity of connections on the main bus board.
After verifying the output of the unregulated, linear supply (supposedly capable of generating 28A at 8V), I reinstalled all four cards, plus the front panel, to the configuration the machine was received in. Each board showed good voltages on the +5V 7805 regulators. However, the machine didn't start to a WAIT state, as expected. None of the front panel switches were responsive. After removing the Seals, PROM/RAM board and jumpering the IMS SRAM for $0000, a little more function was apparent. The lights corresponding to the highest bits on the address line were flickering. RESET seemed to work—all of the status and data lights turned on. The RUN light turned on and off with the RUN/STOP switch, but the WAIT light remained off when "stopped."
Line voltage on the front panel
One of the few Bad Things about the design of the IMSAI 8080 is the position of the power switch on the front panel board. The line voltage comes directly to the front panel, though a rocker switch, then off to the power supply. After working on it a bit with the front panel cover and switch escutcheon removed, my fingers and arm were sometimes coming precariously close to the line contacts. Before I got a good shock, I decided to disconnect the front panel from the power. IÂ jumpered the molex connector to the CPA using a 14 gauge wire. Now I can turn the machine on and off using a power strip.
