Try reinserting memory modules and add-on cards. If this doesn't help. Then replace them with known good ones. I found out that memory modules are pretty sensitive and can go bad first. Then it depends on the motherboard to output error beeps - some just don't and you're in the dark.

Not getting warm? Is that on the original PSU or a known-good one. I'd be inclined to mistrust the PSUs more than the internal components under those conditions, at least once you've ruled out damage from leaking battery corrosion.

Apart from that, take it bit by bit. Even a completely dead system will usually have some salvageable pieces.

Any chance we could get some photos?

Do you have a multimeter? Spare power supply?

Is that "P200MMX" a system that has one or just the processor itself?

What era of a Deskpro is it?

The P200MMX is a loose CPU i tested in a known working Socket 7 system I built the other day and shows no signs of life.

I’ll check out the PSUs in them as that could well be the issue. No error beeps unfortunately, the one I’ve tested up to now does have a “diagnostic light” but it doesn’t show anything. Perhaps the other ones will reveal more.

I’ll get some photos when I get home- they are the P3 era compact Deskpro EN models

So I forgot to get pics of them before putting them away again.. d’oh. Nothing to see though other than lots of dust but I will try and get some.

There’s 4 identical Deskpros in there - I got one working after removing a faulty RAM stick so that leaves the other 3. All 3 exhibit the same behaviour, no video and no beeps, even after reseating everything, cleaning the RAM slots and using known good RAM.

I’ll get my multimeter out and check some voltages but I don’t think the PSU connections are standard ATX so I can’t use a known good one to test it 🙁

Can't you just use the PSU of the working one to test the others ?

It's really hit or miss with barn finds. Electronics (especially BGA solder junctions) don't do so well with a lot of thermal cycles. Humidity can also corrode stuff, so there's that point as well. The best thing to do is test all the components one by one, and rework any boards that exhibit problems or don't work. A hot air tool on the BGA chips can help bring components back to life. Also get a USB microscope and inspect all solder mounting, because some of them can crack due to thermal cycling. It would be best just to throw some flux on the suspect gull wing mounts (like surface mount memory) and go over it with a soldering iron. On RoHS components, use a USB microscope to look for tin whiskers and remove them with a soldering iron. That shit is a major problem since they went to lead free solder.

Deksor wrote:

Can't you just use the PSU of the working one to test the others ?

I can... but there’s no guarantee that one’s working 100% and if the others do then kick into life I’m kind of stuck as I wouldn’t have a clue where to get a replacement from. They do also all power up and stay on, including the mechanical drives, just with no other activity, which to me points away from the PSU?

Once I’ve tried everything else I might try some more in depth board repairs such as the ones you described - worth a shot before scrapping them. All the caps are of good quality and look fine, the BIOS chips on the other hand are SMD and won’t realistically be possible to try reflashing with the equipment I have available.

It may help to take the motherboards and components out and give them a good clean. Get all the dirt and corrosion off. There may be crud causing shorts and preventing post.

probably the psu or ram

i pulled an old iinspiron 531 out of my barn iafter it was in there for like 5 years it wouldnt even post
put in some new ram and it worked just fine

Thanks guys. They do need a good clean inside although it’s mainly dust - but I’ll give it a shot.

There are definitely bad RAM sticks in there, each of them has an identical Kingston stick in which must have been added as an upgrade at some point as well as another identical stick - the Kingston ones are all knackered whereas the others all seem to be fine

They’ve been stood a lot longer than 5 years from what the guy said. Some quite rusty looking ones but the Deskpros aren’t too bad - haven’t tried the others out yet.

I’ll try out the “good” PSU and memory in the other Deskpros and report back, might make a start on the others too. Also need to test and get rid of all the CRTs, I don’t have the space for 11 of them! I’ll probably keep one for myself, try and sell the ones that work and junk or give away the rest

Nice! Save all the lovely Trinitron and varient tubed CRT's ya can! They are getting stupidly hard to find any more.

Yup, they’re nothing special but they are CRT, I haven’t tested any yet but hopefully there’ll be a few usable ones 😁

I’ve tested the Deskpros out today. One of them is now fully working - with the exception of the CMOS battery, there is a corroded trace I think leads to it which may be the culprit but it’s physically nowhere near the battery?

Two of them power on but don’t show any error lights or video output, and the third one shows a flashing red light as soon as you press the power button which according to the chart inside the lid is for “CPU thermal shutdown” - except the heatsink is mounted properly and it happens the moment you press the power button. Any ideas on that one?

I’ve tested all the PSUs in the working board and they all work at least as well as each other, ditto the CPUs and memory and cleaned all the CPU socket and RAM slots with contact cleaner - there isn’t anything else replaceable on the boards and no jumpers either 😒

britain4 wrote:

the third one shows a flashing red light as soon as you press the power button which according to the chart inside the lid is for “CPU thermal shutdown” - except the heatsink is mounted properly and it happens the moment you press the power button. Any ideas on that one?

Does the CPU fan start spinning at all?
Maybe a damaged fan controller/sensor IC or a damaged trace going to or from one?
Look for a Winbond or ITC logo. There are other brands that I don't remember.

Writing from experience that with some effort most of that lot can be saved however the drives and power supplies are usually going to have problems but don't be surprised if some of it that despite the conditions they were stored turn up working. I've gotten my hands on hard drives that at one point or another were water damaged yet somehow survived and this is with heat exposure sitting in the sun for months or longer, the only reason was that the drives were not modern at all so the heads were much more resilient. The problem with power supplies is the caps going bad.

Damaged traces are a very real possibility given the issues with the working one - I’ll have to have a closer look at the boards when I get time. I’ve definitely isolated the issue to the board itself for each of them but that still means there’s about 1000 different things it could be!

There are 2 with rough cases, one OK and one clean one (now the working one) so if I could get a second board working I’d be happy enough using the others for spares. They came from a large school or office according to what the guy said so there were probably a LOT more of them - I suspect these ones ended up in the barn because they broke before taking them out of operation.

All the power supplies I’ve tried so far out of it have been working fine - I’ve noticed a lot of dead hard drives and RAM though. Lots of nice goodies in there, it was well worth the trip out 😀 you wouldn’t believe the state of the slot 1 PII system I’ve just tested out of it that seems to work! (Except the hard drive)