VOGONS


First post, by havli

User metadata
Rank Oldbie
Rank
Oldbie

Hello,

I discovered this board in storage and realized it possibly could make great little late DOS and win9x PC. I used this board in 2015 I think for some benchmarks, it definitely worked fine back then. It still posts now, even loaded overclock of 6% from the old times 😁

The problem is it doesn't recognise any keyboard - PS/2 and neither USB. I suspected the overclock to mess up stuff (although it was perfectly stable in the past). So I cleared the CMOS.... didn't help. The board is definitely alive as it even booted DOS from attached SSD before I cleared the CMOS. Now it is stuck at the message "defaults loaded, press f1 to continue".

I measured the voltages across the board and it is more or less as expected (1.1V for CPU, 1.8V for DDR2) and PSU also seems to be fine (12V, 5V, 3.3V, 5V SB). But all usb ports (there are 8 of them) and both PS/2 ports only read 3.4V. There are fuses to protect these ports from over current. But they read 3.4V on both sides, so they seem to be fine, the problem must be before them. The PSU is microATX FSP200-50GSB, it is relatively new and works fine with different boards.

I always thought these ports are powered either from 5V or 5V SB rail directly. Some boards even have jumper to confugure this (not this one). I am sure there are people experienced with repairs of PC hardware. 😀 Any ideas what could be wrong here?

Thanks

HW museum.cz - my collection of PC hardware

Reply 1 of 5, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

I'd also expect they would be direct from PSU 5V, with maybe the PS/2 on 5VSB for turn on via keypress.

Maybe for some reason they generate 5V on the board from the 12V supply. With the board off, see whether the PS/2 5V is shorted to the USB 5V (so are they on the same circuit, given you measured the same voltage). Then see if maybe the USB 5V connects to the tab of any of the power FETs on the board. May also test if any current can actually be drawn on that 3.4V when the board is on by connecting a 1k resistor from USB +5 to USB Ground. That should only draw a maximum of 3.4mA (3.4V / 1k), so if the voltage on USB +5 drops further then you know there's a high resistance from wherever that 3.4V is coming from.

Reply 2 of 5, by havli

User metadata
Rank Oldbie
Rank
Oldbie

Thank you for the tips. I will try that.

In the meantime I discovered suspisious mosfet on the board with partially cracked package. It seems to be this one https://www.vishay.com/docs/67441/si4501bd.pdf
It is possible it is related to this problem.

HW museum.cz - my collection of PC hardware

Reply 3 of 5, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie
havli wrote on 2022-01-16, 19:42:

Thank you for the tips. I will try that.

In the meantime I discovered suspisious mosfet on the board with partially cracked package. It seems to be this one https://www.vishay.com/docs/67441/si4501bd.pdf
It is possible it is related to this problem.

Seems plausible, that can be used to switch supplies on and off. Did see something similar on a board that allowed the RAM to be switched between two different 3.3V supplies, one generated from 5VSB for use when the board was sleeping. Either way, cracked chips are rarely a good sign.

Reply 4 of 5, by havli

User metadata
Rank Oldbie
Rank
Oldbie

Most likely the blown mosfet switches USB power between +5V and +5VSB, measurement confirms this. I have some spare mosfets, not the same one but should be pin compatible. I'll try to replace it later. With some luck it might do the trick.

HW museum.cz - my collection of PC hardware

Reply 5 of 5, by havli

User metadata
Rank Oldbie
Rank
Oldbie

The board is alive! 😀

The replacement mosfet seems to work properly. My friend resoldered it on the board... as I don't have the right equipment and skills to (de)solder SO8 package myself.

I will use different board for my mini PC project in the end - but it is good to have this one as a spare.

HW museum.cz - my collection of PC hardware