NASA Parts Scroungers Resort To eBay For Parts

beggs writes: "The New York Times is running this article about NASA using ebay and other web resources to find for sale stock piles of old hardware it needs to keep the Space Shuttle fleet up and running -- things like 8086 chips from pre-PC days!" Come to think of it, this might be a better way to take care of most NASA bidding anyhow.

Comments

Interesting...

[BrianGa]

What is NASA's feedback rating, and do they take/pay by Paypal?

NASA outbid me

[Anonymous Coward]

NASA outbid me on a set of Star Wars collectable Burger King glasses from 1983. This is your tax dollars at work! Those fuckers won the auction too! They must have used some space-based technology to outbid me at the evry last second!

It's NASA's problem now...

[Matt2000]

...but it'll increasingly be all of our problems in the future. A lot of really smart people are worried about the computer industry's quick pace and are worried that it's unsustainable. If I have a computer that I bought 15 years ago that's running a critical function in my workplace, it quickly becomes more expensive (in hardware costs) to support that piece of equipment than it does to buy a whole new machine. That's fine except that we then need to convert all the data over to new formats and operating systems, interface all our surrounding systems with the new system and generally spend a bunch of time and money replicating the functions of the old machine.

Sure, it's rarely ever that simple a scenario, but the computer industry should spend a bit more time thinking about sustainable growth rather than the next 300 Mhz of CPU performance.

Re:It's NASA's problem now...

[swb]

Aren't most mission-critical business systems already taking this into account?

We're running an HP3000 running MPE here. Most of the applications running on it are running code nearly 30 years old. HP is phasing out the 3000, but I'm guessing that the platform will remain runnable for at least another 10 years (5 years of legacy support from HP, and 3rd party beyond that).

I agree with you the breakneck pace of change is kind of nutty, but I don't think the impact is as dire as you claim outside the PC arena.

Plus, some of the changes being made are more than cosmetic Mhz changes -- the increased processing power brings real new functionality. If you don't upgrade, your competitors might, leaving you behind in terms of performance and functionality.

Transmeta/FPGA?

[larien]

Couldn't Transmeta chips (which are programmable to a degree) or FPGA's be an answer in the longer term? Obviously, a lot of the reason for requiring 8086 chips is down to form factor etc, but couldn't converters be made to help out?

Re:Transmeta/FPGA?

[morcheeba]

Check out...

G4 PowerPC processor
MIPS R3000 Mongoose
Pentium III/4
RAD6000 variant of IBM's RS/6000

Re:Transmeta/FPGA?

[ender81b]

I believe the real reason NASA is looking for 8086 chips and not changing is simple - they work. Why change?

Also, don't forget that in space the chips need to be hardened against EM raditation of all kinds. It is apparently very difficult to do on modern chips, slightly easier on old ones. In the long run (next shuttle) they might use Transmeta but... if it ain't broke don't fix it. A proven technology that works is just fine.

Is it just me or isn't it kindof sad that the damm shuttle can run on a couple (3 actually I think) 8086 put to run fscking win2000 i need about 100x the processing power...

It's a damn shame.

[austus]

Go ahead an mod me down because this is basically an emotional outburst, but I really think it is completely sad that society lacks the collective intelligence to see how important organizations such as NASA really are. The possibility of impact by an asteroid or other large space object alone justifies financing NASA adequately.

Editors distort story *sigh* again

[FearUncertaintyDoubt]

Troves of old parts that NASA uncovers and buys, officials said, are used not in the shuttles themselves but in flotillas of servicing and support gear.

NASA is not so stupid as to not contract for replacement parts for the actual shuttle from subcontractors. This is just for support gear. Probably quite a bit of this gear is custom-built by NASA engineers, like programmers who build their own toolkits. As the article says, it's easier to just scrounge up a board than pay someone to redesign some piece of equipment to use updated components.

NASA isnt the only one...

[kyfho]

I used to have the same problem in the Navy. Ever try and find a D.C. power supply for a VT100, or how about a head assembly for a RL02 disk drive? I finally started scrounging in local collage electronic surplus piles to keep my systems running. Just goes to show that systems made 20-30 years ago were built to last.

Hrm.

[Jonny 290]

Oh Lord, I can see the auction titles now.

"FIRST CLASS! Slightly used moon capsule, 8/10, L@@K!"

"Tired of having no way to get to low earth orbit? Click here! BEST SHUTTLE ON EBAY!"

"VINTAGE EMPTY SATURN ROCKET STAGE--W0W! MAKES GREAT GRAIN SILO!"

(and yes, i know they're buying, not selling.)

Shuttles until 2020 (or beyond), B-52s until 2040,

[Mordant]

it seems like we can't make any cool, upgradeable large-scale systems, anymore.

What did we do when we needed large, mobile cruise-missile & artillery platforms? Why, we loaded up then-forty-year-old Iowa-class battleships, ships so old that it was tough finding personnel who knew how to work the guns!

The shuttle uses early-70s technology. The B-52, the first prototype of which flew in 1949 (!), is still our #1 conventional heavy bomber, and is a testament to forward-thinking in terms of modularity. But it seems that the trend is towards more monolithic, use-it-and-then-throw-it-away-and-buy-a-new-one systems.

Which is great for the suppliers, but not so great for the consumers (and in the case of NASA and DoD, the taxpayers).

I can generally get about 2 years of useful life out of a desktop PC, perhaps upgrading the RAM, video adaptor and CD/DVD/latest-useful-removeable-media drive along the way. I can get about 18 months of use out of a laptop, upgrading the RAM at some point. I can get 3-5 years out of a car, a (potentially) lifetime of use out of a good watch or a gun.

But the design principles I see in operation today are very much oriented towards disposability. Which is a bit of a problem when we're talking about multibillion-dollar systems.

What's the answer? For space, let private enterprise develop their own, market-driven Pull out of the Outer Space Treaty (http://www.state.gov/www/global/arms/treaties/spa ce1.htmlmodel. ), and let's start mining those asteroids! NASA can do science, while the Solar System is pioneered by those imbued with that most useful of human motivations - pure, unadulterated greed.

For defense, I'm not so sure. The bureaucracy is so bloated and elephantine, and so many different factions are constantly trying to keep their rice-bowls from being broken, I'm unsure -what- it would take to reform their procurement methodologies. If September 11th isn't enough of a wakeup call that we need to move both quicker and smarter, I don't know what would serve.

Check with IBM...

[warpSpeed]

They could get an IBM Mainframe running Linux , then partition it out to 40,000 virtual instances, and have each one running Bochs to emulate all the 8086 chips they need.

You know, it would probably still be cheaper then maintining what they have now in the way of hardware.

Problem is that they would need a third booster to get it and the power plant off the ground....

I find this hard to believe...

[bjtuna]

I remember reading somewhere, that in order to supply parts to the military or NASA, you have to contractually agree to continue producing (or be able to produce) the purchased parts for something like 30 years, because equipment like jets and space shuttles are built to have a 30 year life span. Intel, when contracted, presumably agreed that they have to be able to make an 8086 until the space shuttle is no longer used.

So either the contract has expired and the shuttles have exceeded their lifespan, or Intel has broken its contract.

Re:I find this hard to believe...

[aallan]

So either the contract has expired and the shuttles have exceeded their lifespan, or Intel has broken its contract.

The the "design lifetime" of the shuttle was around 100 flights. Based on this the most of the shuttles have only burned a quarter to a third of their design lifetimes.

On the other hand, the shuttles have been flying for over 20 years, the first flight was in 1981. NASA was, initally at least, anticipating a much higher number of flights per year, in theory this means that they were really expecting to take them out of service during the early to mid-nineties. I remember hearing 15 years as being the expected design lifetime back in the '80's.

I guess you take their pick, depending on how you want to look at it, they're only a quarter of the way through their design lifetime, or they're outlived their design lifetime by five years (possibly more).

Al.

Is that an old computer in your basement?

[forged]

Quotting the original article:

• That old computer in your basement? NASA is not interested.

Trolls will be disappointed !

NASA is trolling??

[kyletinsley]

So to keep the shuttles flying, the space agency has begun trolling the Internet -- including Yahoo and eBay -- to find replacement parts for electronic gear that would strike a home computer user as primitive.

Considering NASA's lack of enough public support to prevent funding cuts in its budget, I find it odd that they have resorted to "trolling the internet"... Seems like the would be better off without all those negative mod points.

Doh!

Re:Shuttles until 2020 (or beyond), B-52s until 20

[GigsVT]

I can generally get about 2 years of useful life out of a desktop PC, perhaps upgrading the RAM, video adaptor and CD/DVD/latest-useful-removeable-media drive along the way. I can get about 18 months of use out of a laptop, upgrading the RAM at some point. I can get 3-5 years out of a car, a (potentially) lifetime of use out of a good watch or a gun.

Tsk Tsk, you propose a radical answer, when your "question" is fundamentally flawed.

I buy a gun, lets say a .38 caliber relvolver. A fine piece of craftsmanship. It will be working long after I am rotting in the ground somewhere if I take care of it. It does one thing, it shoots little chunks of hot lead at subsonic speeds.

If I buy a computer, and I keep using the same software that came out designed for said computer, and I take care of it, I don't keep it in humid environments, I don't let it overheat, etc, then I'm sure it will last long after I am dead too.

It's only because you want to run new, bloated software, designed for new, bloated computers, that you have to upgrade so often.

It's like trying to shoot 357 magnums out of your 38. Sure, they fit in the chamber, but that isn't the gun that bullet is designed for. Don't be surprised if it doesn't work.

Re:Shuttles until 2020 (or beyond), B-52s until 20

[Leven Valera]

The B-52, the first prototype of which flew in 1949 (!), is still our #1 conventional heavy bomber, and is a testament to forward-thinking in terms of modularity. But it seems that the trend is towards more monolithic, use-it-and-then-throw-it-away-and-buy-a-new-one systems.

One of my coworkers was a crew chief in the early 90's for a B-52 that was built in 1962. He said, even though the plane may be fourty years old, so many parts (nose, wings, tail, fuselage sections, navcomp, weapons, et cetera) have been replaced over time that the build date of the plane is more like 1980-something.

Cheers,
LV

Re:Shuttles until 2020 (or beyond), B-52s until 20

[Billly Gates]

I disagree.

B-52's did a fine job in the gulf war of 1991 and kosov. Why upgrade? I believe the b-52 bombers and the more modern aircrafts still use the same old early 20th century technology. They are more expensive but only slightly more efficient. Something newer may not be a whole lot better but would cost alot more. As a taxpayer I do not want a more modern jet. I pay too much as it is for military operations and for more b-52's. The b-52 was made to be modular and upgradable. You can just upgrade the computers and put more modern gps guided missiles on them as tine goes on. The stealth jets just use some fancy materials and designs to knock radar beams away from it but its still based on the same concept.

Sounds very familiar

[morcheeba]

Up until about 4 years ago, I used to work for a major/small satellite company. One of my last tasks was to update the processor board used in many satellites... you guessed it, it was based on the 8086. What was my upgrade? I added an 8087 to it! These math coprocessors are even rarer - we bought the last 50 bare die in existence to eventually custom-package in a special high-density radiation shielding ceramic package.

It was about that time that I decided that the company was going in the hole. It's not that the 8086's were particularily good processors... True, they are made with a bigger geometry and suck more power -- things that make them generally more radiation resistant than anything produced in the 1990's. But, they were never designed to tolerate radiation. (NASA isn't stupid - they have high-performance radiation tolerant parts like the RAD6000).

Since my company wasn't making even minimal internal investments (they had a '386 based system that they built but never applied power), I decided that, for my career, I should leave. I notice now that they are hiring people with 5 years of PowerPC experience -- eventually they must have decided to get with the times, but since they didn't keep their employees current, they shot themselves in the foot and now have to hire outsiders.

p.s. I'm back on the job market - anyone need a kick-ass PowerPC engineer? :)

Chip Testing

[kninja]

The reason that they are using such old chips, is probably due to stability. These old chips probably do not produce much heat, and therefore are less likely to fail. I remember something about a satellite getting upgraded to a 486, because the 486 had passed the x years radiation test. The pentium had not, due to not being around long enough.

In this article. I think it is just probably cheaper to find the replacement parts than to redesign the system again, it was probably designed well, robust and stable, unlike most modern systems. The engineers trust it, and perhaps are working on something to replace it eventually, but anything new would need a lot of testing, probably about 10 years. Maybe they'll upgrade to a 386 or a 486 soon.

Re:The reason for open protocols

[Tokerat]

The real problem with this is that is just now, within say the last 5 years or so, becomming a common practice. Sure, we've had open protocols for many more years than that, but c'mon, what about all the businesses who where convinced by vendors that their proprietary stuff was "more well suited for your particular application"? They're gonna have a hell of a time.

I cite as an example the local hospital here about 5-6 years ago began updating their old VAX-style mainfame and VT100 terminals (from probably the 70s sometime) to shiny "new" Windows 95 and NT machines. They spent about $1 million on it, incuding new networking and employee training. A few weeks ago, they upgraded again, for a cost of $3 million this time, to another Windows NT/XP solution, again with training for fewer people this time and they had to hire about 30 people to trnasfer all records form the old system to the new one because the new system didn't use any standards. Wouldn't it be easier to have a server with some flavor of SQL (Oracle to keep the suits happy?) and a few programmers to create and maintain a front-end app that would work well on whatever types of PCs the hospital uses (their quite partial to IBM)? I'm sure it would save a ton of money in the long run since they seem to liek to upgrade every few years. If the data just worked or the server/clients could be upgraded seperately as needed, they' woudl save millions in efficiency.

Of course, that's money that the medical computing systems' vendors dont' get to make. So it'll never happen. Methinks it's unfortunate how capitalism and technological advances have a tendancy to choke each other to death.

Re:Shuttles until 2020 (or beyond), B-52s until 20

[ncc74656]

One of my coworkers was a crew chief in the early 90's for a B-52 that was built in 1962. He said, even though the plane may be fourty years old, so many parts (nose, wings, tail, fuselage sections, navcomp, weapons, et cetera) have been replaced over time that the build date of the plane is more like 1980-something.

Every few years, they basically take the plane apart and put it together. Since many of the parts are no longer available from the original manufacturers, the facility that does this work has the ability to build pretty much any needed replacement part from scratch by measuring/analyzing/reverse-engineering the originals. (There was a story linked by /. a few months ago about the Air Force sending a B-52H to OCALC to be refitted and turned over to NASA to replace its B-52B launch aircraft, but I can't seem to locate it.)