Objet Eden 330 3D printer embedded PC not POST'ing

[prototonic]

I thought I would post something here since I've enjoyed EEVblog videos and content for a while, and there doesn't appear to be any information about this fault on the net.

So, long story short, I acquired an Objet Eden 330 3D printer right around the time my home town of Melbourne went into Covid19 lockdown, around April or May 2020. So, it sat there doing nothing for a long time.

First time I powered it on, a few months ago, I heard a pop and smelled the magic smoke, to my thorough disappointment. First check of the 3D printer power supply seemed OK. Inside there is an embedded PC and this was dead.

The embedded PC that runs the machine features an IEI PCI-6S2 backplane with ISA and PCI slots. There is an ACS Electronics SB214PC4-F Multi-Axis Motion Control Card on the ISA bus, an Objet PCI card with Spartan FPGA, and the host PC motherboard is an Advantech SBC-81612VE with 1GHz Pentium 3 and VIA chipset, this connects to the backplane via PCI and ISA.

Anyway, back to troubleshooting - I replaced the power supply, got lights on motherboard, CPU fan spinning up, but no POST. Clear CMOS, no help. Bought some PC133 SDRAM to swap with the old stick, still nothing. Double check all jumpers, visual inspection of motherboard to see if anything's obviously broken, no help.

The backplane has LEDs for all voltage rails. Noticed the -5V LED isn't illuminated. Replacement power supply I bought is of course modern and doesn't have -5V rail. Was pretty skeptical that this would prevent POST but as luck would have it a mate had a spare old PSU with -5V output that I could borrow. Put it in, no change!

Pulled out the other two cards connected to the back plane to see if it would help. Nada. Disconnected CD-ROM, nothing. Getting pretty despondent at this point, leaving a few days between attempts to alleviate the butthurt.

A few days ago I inspected the motherboard a bit more closely, and noticed a little bit of leaked capacitor goop jammed in the tiny space between the offending cap and the CPU fan header, see attachments. On closer inspection of the four 1000uF 10V caps in this area, another one was bulging at the bottom. After replacing these, it boots! I'm yet to put it back into the printer chassis and reconnect everything, that's coming up.

In retrospect, this probably would have been a lot easier to find if I'd pulled everything out and checked voltages, but just didn't really have the time to allocate to it!