The first point of any repair is to measure voltages. Check input and output voltages from regulators. The fact your post card shows nothing may indicate a missing supply rail, although Im not certain on that.

New findings:

- Voltages are fine all across the board including the CPU socket
- I have 12V on the molex and the ISA bus
- There is 0V on the fan connectors
- "--" on the POST card and no beeps

The only possible cause I can think of anymore is bad BIOS.. I will remove and reseat the PLCC chip, and I'll also look for a PLCC to 866II-Pro converter to reflash it 🙁

I am also thinking of doing a recap but a preemptive recap with no visible bulging or leaking sounds like a bad idea for now?

If you don't have an ESR meter to check them, replacing the caps isn't a bad idea. Worst case you can just remove them if you cant get the board working and reuse them. BIOS is also an option. Also, if you have a method of measuring it, check the clock generator/main oscillator is working OK.

Hippo486 wrote on 2022-11-24, 07:55:

If you don't have an ESR meter to check them, replacing the caps isn't a bad idea. Worst case you can just remove them if you cant get the board working and reuse them. BIOS is also an option. Also, if you have a method of measuring it, check the clock generator/main oscillator is working OK.

I do not own an oscilloscope, it is on my list of things to buy. I've been following reviews on some of the smaller ones Adrian Black have been reviewing.. This one looks like worth biting the bullet for: https://www.youtube.com/watch?v=Uqrel5fQpK4

The reason I want to try the BIOS before the recap though, is: Electrically there are no shorts and I can't think of anything between the 12V rail and fan headers that could fail open short of a broken trace which I can not seem to find. That means something on the circuit is NOT opening to let the current flow, and considering these are three-pin PWM fan headers the only thing I can think of that will control the voltage to those is the BIOS.. Makes sense?

I wouldn´t bother about the fan headers at the moment, they will get power as soon as the system starts in a regular way again.

All I can say is that I recently had both a VA-503+ and a PA-2011 (same era as PA-2013) not starting at all after a longer storage time. In both cases all the electrolytics in the VRM circuitry (those with 1000µF or more) had to be replaced by new (quality-) ones to fix this.
But it won´t hurt to reflash BIOS before that.

majestyk wrote on 2022-11-24, 08:20:

I wouldn´t bother about the fan headers at the moment, they will get power as soon as the system starts in a regular way again.

All I can say is that I recently had both a VA-503+ and a PA-2011 (same era as PA-2013) not starting at all after a longer storage time. In both cases all the electrolytics in the VRM circuitry (those with 1000µF or more) had to be replaced by new (quality-) ones to fix this.
But it won´t hurt to reflash BIOS before that.

This is good feedback thanks. As far as I can see the ones installed on the board are S.I.S. branded, basically noname. Unfortunately all replacements I have are also similarly low quality high ESR ones by Jwco etc. I guess I will have to make an order tonight and replace all the electrolytics on this board then 😀

I can recommend Rubycon ZLH or ZLS or some Panasonic low (not "lowest") ESR series with a decent 105° lifetime of 2500 - 5000h.

First of all I would check all jumpers. Maybe some of them lost contact. I don't think faulty caps cause the system not running at all. Probably it would become instable. Also maybe some chip lost contact: try to check contacts of smd chip with a needle or a toothpick and contacts of bga chips pressing on them.

majestyk wrote on 2022-11-24, 08:34:

I can recommend Rubycon ZLH or ZLS or some Panasonic low (not "lowest") ESR series with a decent 105° lifetime of 2500 - 5000h.

Can also recommend the Rubycons/Panasonics. Think the Panasonics will be a little cheaper but still good quality.

Roman555 wrote on 2022-11-24, 08:58:

First of all I would check all jumpers. Maybe some of them lost contact. I don't think faulty caps cause the system not running at all. Probably it would become instable. Also maybe some chip lost contact: try to check contacts of smd chip with a needle or a toothpick and contacts of bga chips pressing on them.

I used to repair motherboards for a Kuka robot system which used off the shelf socket 7 motherboards. All the boards would refuse to do anything until the caps had been replaced.

Its not a 100% fix every time, but it happens so often because of normal failure and the capacitor fiasco, that its always a solid start.

appiah4 wrote on 2022-11-24, 08:07:

I do not own an oscilloscope, it is on my list of things to buy. I've been following reviews on some of the smaller ones Adrian Black have been reviewing.. This one looks like worth biting the bullet for: https://www.youtube.com/watch?v=Uqrel5fQpK4

There exist a ~$80 2 channel SigPeak DSO2512G version of above https://www.eevblog.com/forum/testgear/new-2c … gpeak-dso2512g/ using both channels cuts sampling rate to 100MHz (or overclocked 120MHz).

YT review https://www.youtube.com/watch?v=yh4s4-BT5-8

appiah4 wrote on 2022-11-24, 08:07:

The reason I want to try the BIOS before the recap though, is: Electrically there are no shorts and I can't think of anything between the 12V rail and fan headers that could fail open short of a broken trace which I can not seem to find. That means something on the circuit is NOT opening to let the current flow, and considering these are three-pin PWM fan headers the only thing I can think of that will control the voltage to those is the BIOS.. Makes sense?

even if bios was controlling fans, PWM controller would still be designed to start full ON?

Anyway contrary to previous posters I would recommend Cheapest low ESR caps you can get _from a reputable source_. That means mouser, newark/element14/avnet, arrow, digikey, farnell, TME. For example https://www.tme.com/us/en-us/katalog/tht-elec … 1632190;&page=1
No need for elitism (pun incoming) when the cheapest "Elite" branded caps are solid https://www.eevblog.com/forum/beginners/anyon … or-brand-elite/ and will easily outlive your usage of this board.

rasz_pl wrote on 2022-11-25, 04:00:

...
even if bios was controlling fans, PWM controller would still be designed to start full ON?
...

It can be like on the picture. The FAN-ON is wired to GPIO of a southbridge. I think similar schematics may be used also in PA-2013.

I reflashed the BIOS and that did not solve anything.

There are two mosfets on the top of the board near the CPU socket. For one, the voltages read (IIRC): S: 9V, D: 7V, G: 5V; for the other they look odd though, S: 5V, D: 2.8V, G: 0V. 2.8V is what I expect (and find) in the CPU socket so I guess they are fine?

I can't really find what is wrong with this board, I may have to go through with the recap..