The Commodore PET 2001 wooden prototype

Recently John Feagans and Leonard Tramiel opened the wooden PET prototype for the first time in 40 years.

Photos taken Dec 2019 and Jan 2020 at the Computer History Museum, Mountain View, CA.
Photos courtesy of John Feagans, Leonard Tramiel and the Commodore International Historical Society, used with permission.

Photos © Leonard Tramiel, All rights reserved.

This one of a kind prototype hasn’t been seen in many years. These are (AFAIK) the first digital high quality photos of the prototype and the only photos of the interior ever. These photos show details that have never been seen before.
The prototype was built in 1976 and shown for the first time at the Jan 1977 CES in Chicago.  The case and monitor shell are made of wood. Originally painted “harvest yellow” it was re-painted and re-decorated several times in 1977 before finally settling on this color scheme.
The prototype had no IEEE-488 port and reduced ROM space so the kernel is different from the production version. The BASIC was Microsoft basic, not yet modified to become Commodore BASIC.

John Feagans, Leonard Tramiel and the PET Prototype

Photos © John Feagans, All rights reserved.

The prototype was at The Computer Museum in Boston until that museum closed in 2000 and The Computer History Museum in Mountain View, CA acquired much of the Boston museum's collection. It was on display at Stanford university for many years until it was recently returned to the CHM archives. This is catalog number X1279.98 at the Computer History Museum. It is not currently on public display.

Gantt chart in pencil on graph paper from 1976 when Commodore was developing the PET.
Showing project tasks, milestones and assigned engineers. Courtesy of Andy Finkle.

Link to a journal written by John Feagans when he was hired by Commodore to work on the PET.

Video from VCF-West 2022. Dr. Leonard Tramiel talks about the search for the prototype.

Below are older photos of the prototype over the years, courtesy of the Commodore International Historical Society and Commodore.ca. These are all the same machine.

From Electronics Today, Feb 1978.

Photo from a Commodore press release, date unk.
Probably from the Jan '77 CES Show

At Gametronics, Jan 1977.

At the West Coast Computer Faire, April 1977.
From BYTE Magazine Oct '77

There are subtle differences in the keyboard.

At NCC in Dallas, June 1977.
Chuck Peddle is wearing an exhibitor badge for the '77 NCC show

On the cover of the October 1977 Popular Science

The badge and keyboard are different.
The wood grain was contact paper.

At Commodore’s 25th anniversary event, Dec 1983.

Re-painted white with a new badge and the space-bar moved.

On display in the Gates Information Science building
at Stanford University ca 2000

The Sanyo M1540A Cassette Recorder

This is the cassette tape recorder Chuck Peddle used in the original Commodore PET prototype.

PET wooden prototype.
Photo courtesy of Leonard Tramiel

Giacomo Vernoni's book "Commodore Tape Recorders"

Commodore continued using these drives in later prototypes and pre-production units.

John Feagans wrote that they bought a pallet of tape drives from a local electronics store in Palo Alto. These were then modified at the Commodore factory in Santa Clara.

The drives were modified by removing the speaker and replacing the circuit board. The corner was sawed off to prevent interference with the power supply inside the PET.

Commodore continued to use the Sanyo drives for the initial production run in October of 1977, but switched to a custom built OEM drive after October.

PET 2001-8 from Oct '77

Video

Commodore PET 2001-8 (1977)

My oldest computer.

This is an early unit from 1977 with the Sanyo type cassette drive. SN# 0010165, dated 10/24/77 by the testing sticker on the main board. This is one of the first 100 units produced in Oct ‘77. Everything is factory original except for one ram chip that was replaced during repairs.

click on photos for higher resolution

The Sanyo tape recorder is the same one used in the original wooden prototype. According to John Feagans, one of the Commodore engineers who created the original PET, the Sanyo recorders were purchased from a local retail store in Palo Alto and modified in-house. They were only used in pre-production units and in the first 100 production run in the Fall of 1977.

Most people think that the earliest units had a blue screen bezel, but the first 100 were black. These were in a way, pre-production units. Commodore was still working out the production and testing processes in Santa Clara in October. They didn’t have the OEM cassette drive yet, so they continued using the retail Sanyo drive that was used in pre-production units.
Full scale production didn’t start until December using the OEM Cassette drive and the blue monitor bezel.
The opening in the case for the cassette drive and the mounting hardware are different for the OEM cassette, so the drives are not interchangeable.

The case badge is unusual. It is made of metal and is raised off the surface on standoffs. This type of case badge was only used thru the end of '77. By Jan '78 they were using adhesive labels.

The serial number sequence started with 001 for North America, so this serial number 0010165 is  #165. It's worth noting that Commodore didn't always use serial number labels in sequence and sometimes skipped blocks of numbers. The metal chassis was made in Toronto by the Commodore factory that made filing cabinets. The logic boards and final assembly was done in Santa Clara, California.

The mounting hardware for the Sanyo drive can be seen in the photo below. There is no front bracket like the one used on the OEM drive.

This has the original version 1.0 ROMs, so it's not capable of accessing IEEE floppy disk drives. The logic board is assy 320081, Part No. 320080 Rev B. It is an early version of the board with no solder mask on the top side. The ROM chips are Synertek, RAM is MPS 6550 1Kx4. 

The hand written number on the board near the front-left (0170) is a production serial number to track boards during testing and re-work. A hand written number like this can be found in this area on almost all 2001-8 boards, especially the early ones from late ‘77 thru mid ‘78.

The sticker with the date and initials is from production testing and indicate the date it passed testing and the technician’s initials who did the testing. In this case the board passed initial testing on 20 Oct but probably failed during burn-in and was repaired and re-tested on the 24th.  A sticker like this can be found on many very early boards, though it may have been removed or fallen off most.

Video