November 02, 2003

Spacecraft computers

Someone on the Rec.Models.Rockets newsgroup asked what kind of computers were run on board 'capsules'. The best answer posted was from Mike Gerszewski, a Graduate Assistant at the University of North Dakota Space Studies department. He concentrated on US programs, and his excellent reply follows:

Mercury: No computers on board, ground systems used IBM 709 & 7090 computers developed for the Ballistic Missile Early Warning System. An additional system was added to solve communication problems with geographically distant ground stations, a 7281 I Data Communications Channel.

Ranger, Surveyor and Early Mariner Unmanned Probes used sequencers.

Apollo Age
Gemini:
The Gemini Digital Computer. IBM received the contract for the GDC on April 19, 1962. It weighed approximately 59 pounds, performed more than 7,000 calculations per second, and required 1.35 cubic feet of space. It used a magnetic core memory, which was originally deisgned for the Semi-Automatic Ground Environment (SAGE). The GDC failed at re-entry during the flight of James McDivitt and Ed White. They had to manually de-orbit and landed 80 miles off course.

Apollo: The Apollo guidance and navigation computer was designed by MIT.

Skylab: NASA went with IBM again and the system was designed around an off-the-shelf version of the IBM 4Pi processor, a direct predecessor to the System /360. This is the first time I saw when a system that implemented microcode was flown in space.

Voyager, Galileo: Used a distributed computing system designed by JPL.

Space Transportation System (Shuttle):
Initial design for the main computer was a repackaged version of the F-15 fighter jet's IBM's AP-1, called the IBM AP-101, based on the IBM 4Pi processor. The IBM AP-101 was a collective effort between IBM and Rockwell International. The size of the AP-101's memory was settled on as 32K, but later in the software engineering process the memory requirements grew to over 700K.

The first use of Open Source in space was when Debian GNU/Linux flew on the shuttle in 1997 controlling a hydroponics experiment. This was most likely on an IBM laptop, but I didn't find any concrete references to this.

I don't know who the contractor was for the main computer aboard the ISS, but the individual astronauts use IBM laptops, running MS software, for everyday ops.

So just as home computers can be traced back to military computers, so can spacecraft computers, perhaps more directly.

Posted by: Ted at 07:58 PM | category: Space Program
Comments (5) | Add Comment
Post contains 395 words, total size 3 kb.

1 And Trailblazer (see link) will use a PDA (for the communication function of its flight computer).

Posted by: chris hall at November 03, 2003 03:45 AM (aemZH)

2 Wow--that's really cool.

Posted by: Victor at November 03, 2003 08:51 AM (L3qPK)

3 The link at my name actually goes to a page of NASA history documents. Search on computer and you'll find several good resources for further info on this topic.

Enjoy!

Posted by: chris hall at November 03, 2003 09:25 AM (zH1Gw)

4 Linux was used on the Shuttle in Feb 96 to control an accelerometer experiment. The computer used was an IBM Thinkpad with NASA specified mods.

Posted by: Pete at November 03, 2003 10:28 AM (eEfb8)

5 Look it was good but more detail was needed as to what the purpose of each of the computers.

Posted by: Tom Arto at November 30, 2003 09:40 PM (2GJW8)

Hide Comments | Add Comment

Comments are disabled. Post is locked.
25kb generated in CPU 0.0166, elapsed 0.2499 seconds.
70 queries taking 0.2411 seconds, 170 records returned.
Powered by Minx 1.1.6c-pink.