Not sure why my images went missing all the sudden.

Since I have a BIOS chip programmer, if it were me at this point I would try reflashing the BIOS. But if you don't already have one, it would be a bit much to expect you to buy one on the somewhat slim chance that it would solve this issue. If you enjoy this sort of thing (messing about with old computer hardware), they are a worthwhile investment. I received a dead Tyan S1854 in a parts lot I purchased and I reflashed the BIOS and it then worked perfectly.

Hello, it's been a long time (a few years), that I write here about the problems and solutions of the motherboards, unfortunately this symptom that you describe, seems very similar to an ASUS that I have in repair, in the MB ASUS I identified a Mosfet (or regulator?) Fault, it is located near the AGP slot, I think that when the post code (by the way an ISA-PCI Post Card, displays the last error code), reaches the video display, the Mosfet activates, but being short it turns off the MB.
If you want you can give him a check, but if that's ok, it could be something else around.

check for shorts on all rails (without any CPU, RAM or addon cards installed).
Still, with just a bare board, check for voltages.

A thermal cam would show something as well, if you have access to one.

Just yesterday I recapped a SS7 board that showed absolutely no life until a recap.

Repo Man11 wrote on 2024-12-07, 06:12:

Since I have a BIOS chip programmer, if it were me at this point I would try reflashing the BIOS. But if you don't already have one, it would be a bit much to expect you to buy one on the somewhat slim chance that it would solve this issue. If you enjoy this sort of thing (messing about with old computer hardware), they are a worthwhile investment. I received a dead Tyan S1854 in a parts lot I purchased and I reflashed the BIOS and it then worked perfectly.

Good idea, I do have a TL866II. I pulled the chip, peeled back the Award label and it's an Amtel chip (Amtel AT49F002NT / 90PC / 9928). I chose that chip in the software and did a read. The read seems to have worked without issues as I can see the ASCII text, etc. I saved it to a .bin file just to have. At this point, if the read went ok does that mean it's probably ok? If not, should I try reflashing and with with BIOS? (from the ASCII text it looks like it was 4.51PG 12/20/99)

Thank you all for your responses! I'm tackling this step by step, so I'll address each suggestion and hopefully we'll get this fixed.

Correct me if wrong,

the board comes back on by itlsef after a while?
Do you have an oscilloscope?

Can you post a pics with readable ICs of the half where the PSU connector is?

Check if this one works as expected:
https://www.alldatasheet.com/datasheet-pdf/vi … M/US3004CW.html

That does sound like for example short on CPU power rail.
Check continuity/diode reading on all 4 big mosfets next to CPU slot, also measure all voltage rails on ATX connector against ground.
Abit was infamous for going with cheapest most counterfeit capacitors, pretty much all Abit boards leaked, might be worthwhile to check if capacitor juice didnt eat something.

Nexxen wrote on 2024-12-07, 17:10:

Correct me if wrong, […]
Show full quote

Correct me if wrong,

the board comes back on by itlsef after a while?
Do you have an oscilloscope?

Can you post a pics with readable ICs of the half where the PSU connector is?

Check if this one works as expected:
https://www.alldatasheet.com/datasheet-pdf/vi … M/US3004CW.html

Ok, after I read your comment I tested a bit more to confirm. Oddly this is what happens. Flip on PSU from the back, Push the power button on the harness, computer comes on for a count of 3, turns off. There's some type of 'cool down period' where it will not turn on again right away. I figured out, it doesn't turn on by itself like I thought, but it's something very sensitive. Ex. after about 30 seconds of me not touching it after it powered itself off, if I just barely touch the pad of my finger to even one side / one wire of the the power button it will turn itself back on, this time for like a count of 60 before turning itself off again. No BIOS beeps either. I repeated this 3 times in a row to make sure.

I even made sure the wiring harness that goes to the pin headers was held down so as to rule out it moving and touching something on the header and starting it again - it's really just something very sensitive that re-activates the power. I have also tried pressing on the mobo, pressing caps, chips, slots, etc and it won't turn itself back on that way. Truly something odd is happening. Lastly, I even tried using two 1-pin dupont connector wires (like for Arduino's) connected to power-on header pin1 and pin2 and all I have to do to turn it on is touch the tip of one of the wires with my finger. I don't even have to touch the wires together. Some type of power-sensing issue?

I've attached a picture as requested. Let me know if you need me to read something on a chip that won't show up in the pic. I just replaced all those large caps

I do have a oscope, nothing professional, just an Adrian Black reviewed Zoyi ZT-703S that I don't know how to use yet. I'll look at the board to see if I can find that chip you mention above.

Mappy8088 wrote on 2024-12-07, 16:59:

Repo Man11 wrote on 2024-12-07, 06:12:

Since I have a BIOS chip programmer, if it were me at this point I would try reflashing the BIOS. But if you don't already have one, it would be a bit much to expect you to buy one on the somewhat slim chance that it would solve this issue. If you enjoy this sort of thing (messing about with old computer hardware), they are a worthwhile investment. I received a dead Tyan S1854 in a parts lot I purchased and I reflashed the BIOS and it then worked perfectly.

Good idea, I do have a TL866II. I pulled the chip, peeled back the Award label and it's an Amtel chip (Amtel AT49F002NT / 90PC / 9928). I chose that chip in the software and did a read. The read seems to have worked without issues as I can see the ASCII text, etc. I saved it to a .bin file just to have. At this point, if the read went ok does that mean it's probably ok? If not, should I try reflashing and with with BIOS? (from the ASCII text it looks like it was 4.51PG 12/20/99)

Thank you all for your responses! I'm tackling this step by step, so I'll address each suggestion and hopefully we'll get this fixed.

I'd say that at this point it is almost certain the problem is something else. If it were me, I'd probably make sure that I had the most up to date BIOS on the chip.

Mappy8088 wrote on 2024-12-07, 18:28:

Ok, after I read your comment I tested a bit more to confirm. Oddly this is what happens. Flip on PSU from the back, Push the […]
Show full quote

Nexxen wrote on 2024-12-07, 17:10:

Correct me if wrong, […]
Show full quote

Correct me if wrong,

the board comes back on by itlsef after a while?
Do you have an oscilloscope?

Can you post a pics with readable ICs of the half where the PSU connector is?

Check if this one works as expected:
https://www.alldatasheet.com/datasheet-pdf/vi … M/US3004CW.html

Ok, after I read your comment I tested a bit more to confirm. Oddly this is what happens. Flip on PSU from the back, Push the power button on the harness, computer comes on for a count of 3, turns off. There's some type of 'cool down period' where it will not turn on again right away. I figured out, it doesn't turn on by itself like I thought, but it's something very sensitive. Ex. after about 30 seconds of me not touching it after it powered itself off, if I just barely touch the pad of my finger to even one side / one wire of the the power button it will turn itself back on, this time for like a count of 60 before turning itself off again. No BIOS beeps either. I repeated this 3 times in a row to make sure.

I even made sure the wiring harness that goes to the pin headers was held down so as to rule out it moving and touching something on the header and starting it again - it's really just something very sensitive that re-activates the power. I have also tried pressing on the mobo, pressing caps, chips, slots, etc and it won't turn itself back on that way. Truly something odd is happening. Lastly, I even tried using two 1-pin dupont connector wires (like for Arduino's) connected to power-on header pin1 and pin2 and all I have to do to turn it on is touch the tip of one of the wires with my finger. I don't even have to touch the wires together. Some type of power-sensing issue?

I've attached a picture as requested. Let me know if you need me to read something on a chip that won't show up in the pic. I just replaced all those large caps

I do have a oscope, nothing professional, just an Adrian Black reviewed Zoyi ZT-703S that I don't know how to use yet. I'll look at the board to see if I can find that chip you mention above.

Check the component I linked the datasheet to.
My guess is the power on circuit not working, I had it before and it behaved like in your case.

rasz_pl wrote on 2024-12-07, 18:02:

That does sound like for example short on CPU power rail.
Check continuity/diode reading on all 4 big mosfets next to CPU slot, also measure all voltage rails on ATX connector against ground.
Abit was infamous for going with cheapest most counterfeit capacitors, pretty much all Abit boards leaked, might be worthwhile to check if capacitor juice didnt eat something.

Ok, had to watch a video on how to test mosfets. I have 3 PHP 55N03LTs that I tested. Check me if I"m doing this right.

Fluke on Diode mode
Spec sheet for the 55N shows when facing the chip, left pin 1 is gate, center pin / pad is drain, last pin right is source
Power and everything else is removed from the board
Touch my fingers on all pins for a second
Red lead from fluke to center pin 2 (drain)
Touch the Black lead from fluke to the rightmost pin 3 (Source)
Fluke beeps like in continuity mode, then starts counting from 0.2v and keeps going up to 0.8v, etc.
All three of these respond the same.

I've included a picture of the topmost one w/ pins marked.

rasz_pl wrote on 2024-12-07, 18:02:

That does sound like for example short on CPU power rail.
Check continuity/diode reading on all 4 big mosfets next to CPU slot, also measure all voltage rails on ATX connector against ground.
Abit was infamous for going with cheapest most counterfeit capacitors, pretty much all Abit boards leaked, might be worthwhile to check if capacitor juice didnt eat something.

This is what I got, not sure if it's supposed to behave this way. The pins not connected to ground, some are solid 0 while others count up in ohms until it reaches 0 and some are solid ohm, like pin 1, 2, 12.

Repo Man11 wrote on 2024-12-07, 18:45:

Mappy8088 wrote on 2024-12-07, 16:59:

Repo Man11 wrote on 2024-12-07, 06:12:

Since I have a BIOS chip programmer, if it were me at this point I would try reflashing the BIOS. But if you don't already have one, it would be a bit much to expect you to buy one on the somewhat slim chance that it would solve this issue. If you enjoy this sort of thing (messing about with old computer hardware), they are a worthwhile investment. I received a dead Tyan S1854 in a parts lot I purchased and I reflashed the BIOS and it then worked perfectly.

Good idea, I do have a TL866II. I pulled the chip, peeled back the Award label and it's an Amtel chip (Amtel AT49F002NT / 90PC / 9928). I chose that chip in the software and did a read. The read seems to have worked without issues as I can see the ASCII text, etc. I saved it to a .bin file just to have. At this point, if the read went ok does that mean it's probably ok? If not, should I try reflashing and with with BIOS? (from the ASCII text it looks like it was 4.51PG 12/20/99)

Thank you all for your responses! I'm tackling this step by step, so I'll address each suggestion and hopefully we'll get this fixed.

I'd say that at this point it is almost certain the problem is something else. If it were me, I'd probably make sure that I had the most up to date BIOS on the chip.

Yep, agreed about having the latest. I found 4.51PG-7L on the Retro Web and flashed the existing BIOS with it. No change in behavior.

Mappy8088 wrote on 2024-12-07, 19:20:

>I have 3

arent there 4 big ones and 2 smaller ones near the cpu slot? with mosfet the most common failure mode is being shorted, imo no need for more elaborate testing

Put a scope on slow timebase on:
- pin6 powergood of US3004 https://www.alldatasheet.com/datasheet-pdf/vi … SEM/US3004.html
- atx connector pin ps_on and powergood
- check both sides of all four coils, one side is being switched and will have messy reading but other side should have clean stable voltage at least for those couple seconds its running
- poweron pins you are shorting

power it on and watch what happens.

No post codes means either corrupted bios or cpu not running at all.
Afaik VIA 686 controls powering on, does it even without bios and doesnt have any facility to turn off on its own without running bios. Board should not turn off even with BIOS chip removed. It will however turn off if it doesnt like ATX supply or US3004 powergood.

US3004

Check if all the input voltages are there; do the same for output voltages.
If the I/O aren't consistent, it'll shut down.

Pin14, what happens when you power on?
Pin6?

Btw, they cost 2€ for 2 pcs on Ali's Press.