Which exact Deskpro EN do you have? Quite a few different systems used that name.
It maybe worth telling us the Motherboard and PSU part numbers as well, someone might no of compatible components.

But in general you have 3 options with proprietary systems.

Find someone to repair the PSU The Business range of Compaqs used high quality parts (as proven by yours lasting 21 years)
Buy a replacement PSU of the same type. Compaq were good with their part numbers so you can search on this.
Decide it's not worth the cost and call it a day.

Also the fact that the motherboard power led is on could also mean the PSU is fine and something on the motherboard itself is faulty.
In either case finding the parts are usually not to difficult, it's finding them for a good price that's hard.

How handy are you with a multimeter? Someone smarter then me can probably at least help with some basic diagnostics.

Thanks for your reply! I'm not entirely sure of the model number (most of the stickers have fallen off the case), but there is a serial on the rear panel: 6023DD22A019. The PSU model number is PS-5201-4T2. Lastly, the motherboard bears the number 162857-001 and is identical to this.

I don't have a multimeter handy, which limits my ability to diagnose problems. However, my situation may not be as dire as it seemed. This morning, I pressed the power button yet again, not expecting anything to happen...but much to my surprise, the computer sprang to life and booted! 😳 So apparently there was a temporary fault, and the PSU isn't quite dead just yet. 😁 That said, it could fail again...so I'm frantically backing up whatever data I hadn't backed up already.

An explanation that comes to mind, though I'm sure there's many others, is that the air/dust movement built up enough static charge that it freaked out the power good circuit, then after it had discharged enough, it decided to work again.

This seems to have adiagram for the Compaq power connecter should you ever get your hands on a multimeter.
https://www.eevblog.com/forum/beginners/bench … supply-from-atx

But I'm the same, I've to computers that started having trouble turning on, Both lasted at least another 2 years before calling it quits.
I won't have the time/space/skill to do that kind of repair for component lever repairs for few more years so that's at least 2 extra years of enjoyment I got out of them before been put into storage until he kids are older.

--Update--

This morning, I tested the computer to see if it still worked.

No dice. I turned it on. There was no video. The speaker made 1 long beep and 2 short beeps, and the Caps Lock light was flashing.

This page says that this combination of beeps means "Video error." This thread also suggested this was a video problem, although the original user said that reseating the RAM solved the problem. So I reseated the video card, the RAM, some of the cables, and the CMOS battery while I was at it. I turned it on.

There was still no video. And now, the speaker made 1 short beep and 2 long beeps, and the Num Lock light was flashing. So now it has a RAM problem?!

"Our situation has not improved."

Well all kinds of weird errors can be caused by bad power. I would suggest measuring the power at the PSU leads and opening it up to take a look at the capacitors.

For what it's worth, I did a bit of elementary eliminatory troubleshooting: Running the system with no video card and one DIMM, no video card and all DIMMs, and a reseated normal configuration (with the video card and all DIMMs). All three times, it reverted to the original error sequence: One long and two short beeps, and a blinking Caps Lock light.

The video card doesn't have any visually-obvious problems, nor does it have electrolytic capacitors. Considering that its presence has no real bearing on the beeps, I kind of doubt that it's the root of my problems.

Oetker wrote on 2021-06-27, 15:53:

Well all kinds of weird errors can be caused by bad power. I would suggest measuring the power at the PSU leads and opening it up to take a look at the capacitors.

I've a hunch that the electrolytic capacitors in the PSU are to blame: After all, they've been slowly cooking themselves for 21 years! Or it might be the capacitors on the system board itself: They're also 21 years old, and they do smell a bit odd. But I'm not equipped to diagnose or repair either component. I don't have a cache of spare parts or testing instruments handy. I could go out and buy a multimeter right now, but I've never used one before...and even if I learned how to use it on the fly and narrowed the problem down to a specific circuit and capacitor, I wouldn't have the ability to solder in a replacement. For me, the practical options are probably limited to the following bitter pills:

a) Find a working and compatible replacement power supply (assuming it's the source of the problem).
b) Find and pay someone to recap the power supply and/or system board.
c) Give away or scrap the entire computer, and figure out a way to do the few things I relied on Windows 95 for in Windows 7 instead. (Virtualization?)

Sigh...

I pasted the PSU part number into ebay. There are three that show up on ebay.ca - the cheapest being from a seller in Germany.

debs3759 wrote on 2021-06-27, 19:35:

I pasted the PSU part number into ebay. There are three that show up on ebay.ca - the cheapest being from a seller in Germany.

Tardy to the party, but I have a Deskpro EN SFF machine with the onboard TNT2 video and a P3 866Mhz. (Not sure if it's coppermine, tualatin, etc.) The system presents with SLOWNESS. It's running Windows ME (yes, it's buggy, yes it's bad, yes I should probably use something else.) The audio chops and hiccups on Windows startup, and the intro video (the one that shows ME's features) stops for a second or two then struggles to catch up, with choppy audio as well. Shouldn't be the case, as I had a 733Mhz P3 of the same model 20 years ago roughly, and that machine screamed.

The only other thing of note was a mangled RS232 plug on the back, looks like someone chewed on it. Turns out I don't think it was abused, but I think the machine took a nasty hit at some point. Even the lock lever for the 370 socket was broken with the piece flying around in the case. The movement on the socket when unlocking and locking offers resistance as well. Plus, the chip has to be mildly flexed to get all the pins to line up. It all adds up to likely trauma.

Whe I loaded the NVIDIA driver for the TNT2, the machine started turning off. I tried a different driver and got it to come back and stay on, for a little bit. After that, random power offs. I thought thermals because the PSU fan liked to tach up and stay there until the machine was idle for more than 15-25 seconds, so I redid the TIM between both CPU and GPU and their respective heat sinks. It came back for a while, but started powering off randomly again. Pulled all power sucking peripherals, it ran longer, but then shut off again. Tried a PCI video card (ATI Rage XL), and no improvement. I just today got a 24 pin to 14 pin adapter and plugged it in to the motherboard along with a different PSU, and got nothing. When I plugged the original PSU back in, it shuts off in the middle of the RAM count on POST.

Didn't realize the Compaq pinout was different than a standard 14 pin. I don't think I fried anything, as there was no indication of power to the board and it didn't let out any smoke. I have another EN with a good supply, but don't want to blow it now that I know a standard ATX with an adapter down to 14 pins will not work. Did you ever figure your issue out?

vintageGuy81 wrote on 2023-07-09, 16:49:

Tardy to the party, but I have a Deskpro EN SFF machine with the onboard TNT2 video and a P3 866Mhz. (Not sure if it's coppermin […]
Show full quote

debs3759 wrote on 2021-06-27, 19:35:

I pasted the PSU part number into ebay. There are three that show up on ebay.ca - the cheapest being from a seller in Germany.

Tardy to the party, but I have a Deskpro EN SFF machine with the onboard TNT2 video and a P3 866Mhz. (Not sure if it's coppermine, tualatin, etc.) The system presents with SLOWNESS. It's running Windows ME (yes, it's buggy, yes it's bad, yes I should probably use something else.) The audio chops and hiccups on Windows startup, and the intro video (the one that shows ME's features) stops for a second or two then struggles to catch up, with choppy audio as well. Shouldn't be the case, as I had a 733Mhz P3 of the same model 20 years ago roughly, and that machine screamed.

The only other thing of note was a mangled RS232 plug on the back, looks like someone chewed on it. Turns out I don't think it was abused, but I think the machine took a nasty hit at some point. Even the lock lever for the 370 socket was broken with the piece flying around in the case. The movement on the socket when unlocking and locking offers resistance as well. Plus, the chip has to be mildly flexed to get all the pins to line up. It all adds up to likely trauma.

Whe I loaded the NVIDIA driver for the TNT2, the machine started turning off. I tried a different driver and got it to come back and stay on, for a little bit. After that, random power offs. I thought thermals because the PSU fan liked to tach up and stay there until the machine was idle for more than 15-25 seconds, so I redid the TIM between both CPU and GPU and their respective heat sinks. It came back for a while, but started powering off randomly again. Pulled all power sucking peripherals, it ran longer, but then shut off again. Tried a PCI video card (ATI Rage XL), and no improvement. I just today got a 24 pin to 14 pin adapter and plugged it in to the motherboard along with a different PSU, and got nothing. When I plugged the original PSU back in, it shuts off in the middle of the RAM count on POST.

Didn't realize the Compaq pinout was different than a standard 14 pin. I don't think I fried anything, as there was no indication of power to the board and it didn't let out any smoke. I have another EN with a good supply, but don't want to blow it now that I know a standard ATX with an adapter down to 14 pins will not work. Did you ever figure your issue out?

In case anyone else comes up to this thread wondering about PSUs or slowness, the root cause of my slowness was something odd with the CPU cooling fan. I believe it was either not registering that a CPU fan was present, or the circuit that helped the machine determine the fan RPMs was wrong. It appeared to be shutting off to save itself from a thermal event. I had suspected thermal throttling, but was able to confirm it with a different Compaq CPU fan. Once I swapped the fan, it stopped shutting down on its own and the speed came back to where it should have been. No more choppiness or extreme slowness, assumingly because it doesn't feel the need to throttle anymore.