Voyager 2 computer and electronics

Hi

I saw a documentary about it, and was thinking - what kind of electronics were in there?

In 1977 at least the 4004 was around probably also the 8008. Or did they create their own?

And the digital camera?

And the story, that it as some time lost 3% of its memory and new software was uploaded. Hence, no ROM, but all store in its 900Mb (?) RAM?

I know that they usually opt for old and know technology, but what was really in the Voyager 2 (and others...)?

Sonnich

Reply to
Sonnich Jensen
Loading thread data ...

If you go to this magical site called "Google," and enter "voyager 2 comput= er," one of the first sites to pop up is:

formatting link

...and there you will find the following:

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D

There are three different computer types on the Voyager spacecraft and ther= e are two of each kind. Total number of words among the six computers is ab= out 32K.=20

Computer Command System (CCS) - 18-bit word, interrupt type processors (2) = with 4096 words each of plated wire, non-volatile memory.=20

Flight Data System (FDS) - 16-bit word machine (2) with modular memories an= d 8198 words each=20

Attitude and Articulation Control System (AACS) - 18-bit word machines (2) = with 4096 words each.=20

According to my calulations, that's a total of about 68KB, or small potatoe= s compared to today's microprocessors. We probably could perform all functi= ons with one of today's boards and still have room for solid state data sto= rage and much more fault detection software. We would still need a second u= nit for redundancy. Today's microprocessors are also much faster than the c= hips used on Voyager and a comparative system would use less electrical pow= er. On the other hand, software might be more complicated as opposed to tha= t used in an interrupt type system, but it would be much more capable and m= ore flexible.=20

Let's look closer at the CCS. The CCS has two main functions: to carry out = instructions from the ground to operate the spacecraft, and to be alert for= a problem or malfunction and respond to it. Two identical 4096- word memor= ies contain both fixed routines (about 2800 words) and a variable section (= about 1290 words) for changing science sequences. The CCS issues commands t= o the AACS for movement of the scan platform or spacecraft maneuvers; to th= e FDS for changes in instrument configurations or telemetry rates and to nu= merous other subsystems within the spacecraft for specific actions. Fault-p= rotection algorithms are also stored in the CCS, occupying roughly 10 perce= nt of the CCS memory.=20

The main functions of the FDS are to collect data from, and controls the op= erations of, the scientific instruments; and to format engineering and scie= nce data for on-board storage and/or real-time transmission. The FDS also k= eeps the spacecraft "time" and provides frequency references to the instrum= ents and other spacecraft subsystems.=20

The Voyager spacecraft computers are interrupt driven computer, similar to = processors used in general purpose computers with a few special instruction= s for increased efficiency. The programming is a form of assembly language.= =20

There is no clock chip, as such, in the spacecraft. The "clock" is really a= counter, based on one of several electronically generated frequencies. The= se frequencies, based on a reference, generated by a very stable oscillator= , are converted and fed to different locations in the spacecraft as synchro= nization signals, timers, counters, etc. The "clock" signal is part of the = information telemetered to the ground and it is with ground software that w= e convert to day of year, time of day Greenwich Mean Time.=20

Voyager was built in-house at JPL; the computers were manufactured by Gener= al Electric to JPL specifications.

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

Bob M.

Reply to
Bob Myers

Someone who sounds like he knows has answered, but I can throw in a bit of indirect information.

By the way, by 1977, there were endless single IC CPUs, though what has to be factored in is when the design was done, too much earlier and those single IC CPUs were not in the picture, other issues aside.

The shuttle started off with relatively "old" computer schemes, since the design came enough before the launch that fancier new things couldn't be factored in.

I can't remember what year, but in the late seventies someone came to talk at the local ham club about an upcoming amateur radio satellite, which must have been OSCAR 7. He pointed out that it used the RCA 1802, and for two reasons, besides it being CMOS and thus low current. It was either the only IC CPU certified for use in the radiation seen in space, or one of a handful.

The second important thing was that the 1802 CPU had a built in scheme to advance the address bus while detaching the CPU data bus from memory. So you could advance the address (to address the next place in RAM) and directly put data into the RAM. It was a simple feature, but by including it, all kinds of external circuitry (such as the Altair 8800's hardware front panel) was not needed. But one could also do without a ROM and load up the RAM from scratch, useful if some data got mangled, or at some later date they wanted to change the programming. If there had been a monitor in ROM, there was no easy way to change it's contents, not back in the seventies.

I have no idea of what the camera was like. I would point out that last year, or was it the year before?, one of the Nobel prize winners had worked in digital camera work in the seventies, the idea being there for a lot longer than we've had cheap digital cameras.

There were certainly very low definition digital "sensors" (would that be the right word?) in the early seventies. I remember something like that being displayed on tv one night, pretty large blocks that didn't provide a particularly good picture. If you've ever seen the movie "Westworld", when they show what Yul Brunner is seeing, it's that sort of image, though I don't know if they actually filmed with such a low definition sensor for those shots, or just scrambled the footage to look that way.

In Popular Electronics in 1975, there was the "Cyclops" a solid state TV canera, with very low definition. The sensor was an IC RAM with the cover removed. The chip was light sensitive, and that was the sensor, and a least some later sensors were done the same way, nothing more than specially selected RAM, though with time the density went up so the density of the picture improved.

SInce the concept existed at the hobbyist level in 1975, it Voyager probably used something on that level, albeit much improved (and likely a custom built sensor costing lots of money.

Like I said, there were often long periods between design and launch, so not only were they using what was available at the time of design, but the build of that design probably came fairly long before the launch. I don't think it was so much about "tried and true" but that other options weren't yet available. And as better things came along, there probably was not much interest in rebuilding things, the time and money having to be respent, when they already had something that did the job. If it worked initially, then it was fast enough, even if what came later was much faster.

Michael

Reply to
Michael Black

formatting link

...and there you will find the following:

Actually I'm glad people don't always just do the search themselves.. It's not something I really thought about but found both yours and Michael's replies interesting. Enough to make me do a bit of googling on the subject myself.

Reply to
The Hemulen

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.