Asus P2B | Slot 1 + Pentium II | no power

[AidenX]

Originally Posted by Per Hansson

You're not going to get anything without a CPU installed.
If you leave the memory out the system will stop very soon at memory init, better to have that installed too.
It would be interesting to see if the board does work without the +12v.
None of the devices connected to +12v seemed critical IMO.

Yes, without cpu 0 beep is normal. xD
I only wrote how i tested if the power supply turns on/off yesterday when still wasn't sure that the short was ONLY on the +12v line.
So all the other components was disconnected just for keep them safe....

Sorry if i didn't write it clearly before! ^^"

I will try again with more in-depth tests and I'll keep you updated!

EDIT:

Tested the board again, this time with all the components.
CPU, GPU, RAM, FANs and power led.
Seems that only the power button works (and turns on / off the power supply). Nothing else.
All dead, no beeps (also without the ram), no signal, no power led.

[Per Hansson]

Ok, it is possible that the hardware monitor chip (U6) prevents the system to boot with +12v missing.
So start looking for your short, I would suspect Q13 first as I wrote in post #18.

[AidenX]

So....if I understand the boardview correctly, the +12v PIN #10 ATX power connector is linked to:

C66 U1 U2 U3 CB81 R15 AGP PCI1 PCI2 PCI3 PCI4 SLOT1 SLOT2 R93 CB72 and the famous Q13.

PIN #10 to C66 has 000.0 ohm, like to U1 U2 U3 etc...

The last 3 are different...R93 - CB72 - Q13.
Between them and the PIN #10 the resistance is NOT completely zero, but remains at 000.2 ohms even after a while.

Also in my board the CB72 component (a capacitor?) is missing (i just tested the pin), maybe because it's a different revision...

Is it time to desolder Q13 and test the resistance without it?

[Per Hansson]

From the ATX connector pin 10 to each component listed it will be a direct connection, the further away the higher the resistance will become, this is what I said at the beginning and is a way to "home in" where a shorted component is.

That said there is no point in measuring resistance to the +12v side of each component, you already know it is ok because the ATX power supply shuts down due to a short!

What you need to look for is a component that is shorted to ground, like C66 in your example, if you measure across it and get 0.0ohm then it is shorted to ground!

But the short can also be to another rail, you said that you had 3.84v on the +12v rail with pin 10 disconnected from the ATX connector, well, what was the +3.3v voltage at the same time? If it would be the same you know it is shorted to the 3.3v rail...

Originally Posted by AidenX

Is it time to desolder Q13 and test the resistance without it?

Yes, but I would just fire up the board again, with the +12v reconnected...

[AidenX]

Originally Posted by Per Hansson

From the ATX connector pin 10 to each component listed it will be a direct connection, the further away the higher the resistance will become, this is what I said at the beginning and is a way to "home in" where a shorted component is.

That said there is no point in measuring resistance to the +12v side of each component, you already know it is ok because the ATX power supply shuts down due to a short!

Uhm ok...

Originally Posted by Per Hansson

What you need to look for is a component that is shorted to ground, like C66 in your example, if you measure across it and get 0.0ohm then it is shorted to ground!

C66 has one pin with 0 ohm to ground and another with 38.02k ohms to ground, so it's fine right?
All the other pins in the +12v line have the same 38.02k resistance to ground.

R93 has one pin with 38.02k and the other with 42.70k to ground.
Q13 has pin1 with 43.30k, pin2 (the tab) with 3.955M and pin3 (12v line) with again 38.02k ohms to ground.

Originally Posted by Per Hansson

But the short can also be to another rail, you said that you had 3.84v on the +12v rail with pin 10 disconnected from the ATX connector, well, what was the +3.3v voltage at the same time? If it would be the same you know it is shorted to the 3.3v rail...

Yes, there is that 3.84v on pin 10 of the board also without the yellow wire of the power supply connected but i don't know if it's normal or should be completely 0v.

As i wrote in the other post (when i compared this board with a different one), the three +3.3v pins on the board are NOT linked with anything, not even between them or with the ground.
In the boardview 1.10 revision they are linked between them and with a diode that is missing in my board (1.02).

About the ATX voltages on the board (Scheme here https://en.wikipedia.org/wiki/Atx#Power_supply):

Wire-PIN  |      Signal      |  Power OFF |  Power ON

Orange       +3.3v                 0v         3.35v
Red          +5v                   0.107v     5.007v
Grey         Power good            0v         4.355v
Purple       +5v standby           4.979v     4.977v
Yellow       +12v                  0v         3.84v (not connected to the power supply!)
Blue         -12v                  0v        11.90v (my multimeter does not indicate the sign)
Green        Power on              3.64v      0.001v
White        Reserved/old -5v      0v         0.042v (my power supply doesn't have the -5v mark on the label)
Black        Ground                0v         0v

Originally Posted by Per Hansson

Yes, but I would just fire up the board again, with the +12v reconnected...

Obviously I would have tried it only with the power off just to see if there was any difference in the ohm values.

Anyway, say me if there is something useful this time...thanks.

[Per Hansson]

Originally Posted by Per Hansson

From the ATX connector pin 10 to each component listed it will be a direct connection, the further away the higher the resistance will become, this is what I said at the beginning and is a way to "home in" where a shorted component is.

That said there is no point in measuring resistance to the +12v side of each component, you already know it is ok because the ATX power supply shuts down due to a short!

Originally Posted by AidenX

Uhm ok...

Let me know if anything was unclear about that, it is important you understand the measurements, otherwise there is little point in doing them...

Originally Posted by AidenX

Anyway, say me if there is something useful this time...thanks.

None of these measurements point to a fault, except maybe the presence of voltage on +12v with the cable disconnected that you had already mentioned.
What I meant in my previous post with that was that if the voltage you see there is the exact same as another voltage then you can infer that those two rails are shorted together...

At this stage I would lift both legs of Q13, reconnect the +12v line and see if the board works.

[AidenX]

Originally Posted by Per Hansson

Let me know if anything was unclear about that, it is important you understand the measurements, otherwise there is little point in doing them...

Well...at the beginning I thought I had to follow the boardview lines and see if there was any "break" in the connection between where it starts and each component... or at least find a different resistance value than the pure 0 ohm.

These things are quite new to me and not everything is clear.... sorry... ^^"
For example, there are times when things shorted or almost shorted to ground are "normal" and it doesn't help to understand what is really wrong. XD
But I'm trying to learn whenever I find something broken instead of just throwing it away!

Originally Posted by Per Hansson

None of these measurements point to a fault, except maybe the presence of voltage on +12v with the cable disconnected that you had already mentioned.
What I meant in my previous post with that was that if the voltage you see there is the exact same as another voltage then you can infer that those two rails are shorted together...

In the power connector all the voltages are different, or do you mean that i have to test the voltages between the components and not from the initial start point?
Like what voltages has Q13 on pin1-2-3 while the board is on?

I did the tests fine for your last question on post #24 (but nothing is really useful) or i did them wrong again?

Originally Posted by Per Hansson

At this stage I would lift both legs of Q13, reconnect the +12v line and see if the board works.

Sure that nothing will explode? xD

[Per Hansson]

Originally Posted by AidenX

Well...at the beginning I thought I had to follow the boardview lines and see if there was any "break" in the connection between where it starts and each component... or at least find a different resistance value than the pure 0 ohm.

These things are quite new to me and not everything is clear.... sorry... ^^"
For example, there are times when things shorted or almost shorted to ground are "normal" and it doesn't help to understand what is really wrong. XD
But I'm trying to learn whenever I find something broken instead of just throwing it away!

Good, I just wanted to make sure you understood what we are doing.
Exactly we are not looking for a break in a wire since the PSU shuts down for overload we are looking for the opposite

Originally Posted by AidenX

In the power connector all the voltages are different, or do you mean that i have to test the voltages between the components and not from the initial start point?
Like what voltages has Q13 on pin1-2-3 while the board is on?

I mean, i did the tests fine for your question (but nothing useful) or i did them wrong again?

No the tests are fine, what I meant was that if you saw 3.333v on both the +3.3v line and the +12v line it would mean something.
Now that the voltages are different it does not really point in any direction...

Originally Posted by AidenX

Sure that nothing will explode? xD

Yes Q13 is just for the fans, and the board did not explode with +12v missing so it will not explode now
Infact the whole purpose of Q13 is strange at best, because the board can't control the fan speed anyway.
Maybe it is for protection of the PCB traces (Q13 burns up before the traces if you put a too big fan or short the fan, it is just my guess...)

[AidenX]

Originally Posted by Per Hansson

Yes Q13 is just for the fans, and the board did not explode with +12v missing so it will not explode now
Infact the whole purpose of Q13 is strange at best, because the board can't control the fan speed anyway.
Maybe it is for protection of the PCB traces (Q13 burns up before the traces if you put a too big fan or short the fan, it is just my guess...)

So....i disconnected pin 1-3 of Q13.

The power supply turns on / off, the voltage is 11.80v (a bit low?) on pin 10 but it's still all dead.
No led, no fans, no beep, no signal like before.

It's just the power supply all alone. xD

[Per Hansson]

So you mean the system now turns on and runs, i.e. the power supply is not shutting down due to a short any longer?
And again, you will not get any feedback if you don't have a CPU and RAM installed.
Obviously the fans will not work because that is the purpose of the removed Q13.

[AidenX]

Originally Posted by Per Hansson

So you mean the system now turns on and runs, i.e. the power supply is not shutting down due to a short any longer?
And again, you will not get any feedback if you don't have a CPU and RAM installed.

Yes, the power supply turns on and runs.
A long press of the power button turns it off. Seems all normal now.

CPU, RAM, GPU, led, fans, and a monitor was connected when i tried it and as i said nothing works, only the power supply.

[Per Hansson]

Ok, this comes back to the hardware monitor chip (U6) probably preventing the system to boot.
I think whatever shorted on the board killed that chip.
There does exist a version of the Asus P2B without the hardware monitor chip.
They have different BIOS versions labeled "No LM75/LM78" by Asus
If you have a dedicated programmer you can program one of those BIOS files and try the board with that:
https://www.ultimateretro.net/en/mot...9628#downloads


The "No LM75/LM78" is a bit of a misnomer btw.
Because the board does not use a LM75 but instead a Winbond W83781D as I mentioned earlier.
This is internally like two LM75's according to this lm_sensors entry:

For starters, those aren't LM75's. Your mainboard actually has the Winbond W83781D which emulates two LM75's, but many systems which use the Winbond chip (such as the Asus P2B) don't have the thermo-resisters connected to the chip resulting in these strange -48 degree readings.
Source: https://stuff.mit.edu/afs/sipb/user/...#Section%204.2

But before you try that check now with Q13 still lifted if you have a short on any of the fan headers +12v connector to ground.

[AidenX]

Originally Posted by Per Hansson

But before you try that check now with Q13 still lifted if you have a short on any of the fan headers +12v connector to ground.

Hello!
There is no short between the +12v fan headers pin and ground.
I didn't try to flash another bios but still have some news.

The power supply now turns on also with Q13 full connected.
Not sure what happened, i cleaned better the board with ipa, did some tests with the multimeter and at the end soldered again the Q13 legs to see if the power supply still turns off or not.

Anyway the motherboard is like before, still dead with no light and no fans.
Seems that nothing changed.

So i tried to test the voltages around some components of the board (with the power off and on) with the help of the boardview.

In this tests only the Power LED and 1 FAN are connected to the board:

        PIN#                               Voltage Power OFF      |     Power ON
CHA_FA  1(FSTP1#) - 2(FANPWR) - 3(GND)      0.227 - 0.070 -     0 |  5.024 - 0.298 -     0
Q13     1(1202BB) - 2(FANPWR) - 3(+12v)     0.149 - 0.070 - 0.149 |  11.77 - 0.287 - 11.77
Q4      1         - 2(SW+5V)  - 3(Phase1)   0.254 - 0.214 - 0.210 |  0.380 - 5.026 - 0.380
Q5      1         - 2(Phase1) - 3(GND)      0.145 - 0.207 -     0 |      0 - 0.395 -     0
Q10     1         - 2(SW+5v)  - 3(Phase2)   0.163 - 0.214 - 0.002 |  0.382 - 5.025 - 0.382
Q9      1(GND)    - 2(Phase2) - 3(GND)          0 - 0.002 -     0 |      0 - 0.382 -     0
Q8      1         - 2(+3v)    - 3(+1.5v)    0.021 - 0.002 -     0 |  0.043 - 0.382 -     0

U2      1(+12v)   - 20(+5v)                 0.214 - 0.148         |  5.026 - 11.78
U3      1(+12v)   - 20(+5v)                 0.214 - 0.148         |  5.026 - 11.78

CHA_FA 1(FSTP1#) is connected to U6 and RN5.
Q13 1(1202BB) to R116 and R117.

As you can see, the +5v and +12v seems to be there (at least in this components), Q13 has 11.77v on both the legs (normal?) but still missing the voltage on FANPWR...

Also Q8 has strange voltages (+3v and 1.5v, maybe for the cpu not connected)...

On the "working slot 1 board" that i have, also without cpu-ram-gpu, the power light and the fans turn on (with 11.77v on FANPWR).

So....now that the power supply doesn't turn off anymore also with Q13 connected, what could be the problem? I'm more confused than before...

Thank you!

EDIT: I tested AGAIN Q8 with the CPU connected, still same low voltages!