Deksor wrote on 2021-11-06, 16:10:

My board is ver 2.0 and I didn't make any changes to it either (I'm definitely not the original owner though).

I believe your PCB and his PCB are rev 1.0, but some of the components may have changed.
Looking at the markings around your LIF socket and the fact we have a proper socket 3 makes me think that they might have had issues with availability of socket 3 ZIF so they simply went back to LIF socket instead. There are many boards with this kind of difference, and yet still the room to install a ZIF socket.

Ok. I didn't know manufacturers did that. I thought that if there was a change in the components, it was because there was a hardware revision. But this was my mistake. Always learning.

Sorry. Damn it! I can't bump my bios. I think my MB has some faults. My HDD IDE doesn't work properly anymore. When I tried using EZDrive or ONTrack DM it worked beautifully, but as time went by, it started to cause problems. For example: corrupt files and uppercase letters changed to lowercase. Very weird (CONFIG.SYS > CONFIG.SyS). Suddenly all files were modified/corrupted. MSDOS failed overall. Don't know what to do. I have to test some components to see what the hell is going on.

brunocrod wrote on 2021-11-06, 18:35:

Sorry. Damn it! I can't bump my bios. I think my MB has some faults. My HDD IDE doesn't work properly anymore. When I tried using EZDrive or ONTrack DM it worked beautifully, but as time went by, it started to cause problems. For example: corrupt files and uppercase letters changed to lowercase. Very weird (CONFIG.SYS > CONFIG.SyS). Suddenly all files were modified/corrupted. MSDOS failed overall. Don't know what to do. I have to test some components to see what the hell is going on.

y 01111001
Y 01011001
seems like a one bit error.

weedeewee wrote on 2021-11-06, 19:03:

y 01111001 Y 01011001 seems like a one bit error. […]
Show full quote

brunocrod wrote on 2021-11-06, 18:35:

Sorry. Damn it! I can't bump my bios. I think my MB has some faults. My HDD IDE doesn't work properly anymore. When I tried using EZDrive or ONTrack DM it worked beautifully, but as time went by, it started to cause problems. For example: corrupt files and uppercase letters changed to lowercase. Very weird (CONFIG.SYS > CONFIG.SyS). Suddenly all files were modified/corrupted. MSDOS failed overall. Don't know what to do. I have to test some components to see what the hell is going on.

y 01111001
Y 01011001
seems like a one bit error.

Can you tell me more or or indicate some reading?

Is it possible the Ontrack or Ezdrive the cause of my problem? The uppercase to lowercase corrupted files error? Has anyone heard something like that? Or maybe i should create a new post here?

just thinking it could be a memory error, since it's just one bit.
Can't recall what ontrack or ezdrive do exactly. compression or just disktranslation?
I'm just guessing.

Memtest 4 or maybe older version. Reseat your ram modules. My board suports LBA so never used an disk overlay.

Ok, update, finally arrived, my bag of resistors and the voltage regulator (MIC 29302WT).
I did my homework and looked among all the resistors for the ones that were closest to the necessary values.
R1 = 158.6K
R2 = 92.7K = 3.36V
R2 = 101K = 3.18V
R2 = 111.7K = 3.0V
R2 = 82.1K = 3.63V
Everything fine up to that point, I check what resistor connects with the ADJ pin of the regulator and with all the pins on one side of the JP8 jumper.
R20 = 158.6K
R21 = 92.7K
R69 = 101K
R68 = 111.7K
R67 = 82.1K
I check that the ADJ pin receives 1.24v, and, in the OUT pin, surprise, the voltage varies between 1.94v and 2v.
I start to check components and the one strange thing I think I find is a cross between EN, IN and GND pins. I do not think it is a problem with the regulator, because I bought two and checked the other and it does not have a short, and also it is still the problem that I had in the my first attempt.
The other strange thing is that there is a direct connection between one side of the JP8 and the ADJ pin of the MIC regulator.
Even if I did the maths wrong, this shouldn't happen, right.
Maybe this board is in some way incomplete because there isn't a short in the of the LT regulator and they are connected with some pins of the MIC regulator.

EN and IN should probably be shorted together so the output is always on. IN shouldn't be shorted to GND or there's a power supply short. There should be some resistance, although it'll be fairly low (5V to GND). A direct connection from one side of JP8 to ADJ doesn't immediately sound odd (although I haven't drawn this out) as one side of both R1 and R2 needs to connect to ADJ.
[edit: I'm wondering if this PCB revision has R1 and R2 the other way around. If I bung your resistor values in, but flip R1 and R2, then the output voltage would range from 1.9V to 2.1V]
[Just to check, I've sketched up what I think the wiring is:

]

Can you post a photo?

I'm ashamed of my soldering job, I'm physical impaired and soldering with only one good hand is difficult.
I've tried to draw the different connections of the circuit, the lines represent direct connection tested with the multimeter, the voltages are average because they oscillate around 0,01. The crossed line represents the line that the jumper cap breaks.
In the photo I already swapped R20 and R21, R20 was supposed to be 158,6k and R21 92,7k, now R20 is 92,7 and R21 is 158,6.

Edit, forgot the photo.

Only thing that really matters with soldering is if the joint is good. I don't know if it'd make life easier, but it should be ok to solder the resistors from the back of the board which might save having to work around the other components.

Just on the resistors, from the colour bands it looks like they are 160k (brown-1, blue-6, black-0, orange- *1000), 82k, 90k, 100k and 110k, 1% tolerances. So, e.g. the 160k with the 90k ought to produce 1.24*(1+16/9)=3.4V.

From your measurements things look ok, so I was wrong the R20 and R21 might need to be swapped, so it should have been ok. I've sketched out how it looks like the circuit is connected. Put the 160k back in R20 and the 90k in R21, then check the resistance on the regulator between pin 4&5 (Vout-Adj, should be ~160k) and 3&5 (Gnd-Adj, should be ~90k).

Yes, the resistor have the values you said, I've tested then looking for the ones that read close to the desired values.
Measuring the resistance on those pins results in a weird reading for me, it starts low, around 70k and increases slowly in 0,1 increments and they don't reach the expected values, I ended de soldering the MIC and testing the soldering points it gives the same weird resistance.
C32 is connected with R20 on the back of the board and to the 3v/5v jumpers on the top.

Hoping wrote on 2021-11-25, 22:02:

Yes, the resistor have the values you said, I've tested then looking for the ones that read close to the desired values.
Measuring the resistance on those pins results in a weird reading for me, it starts low, around 70k and increases slowly in 0,1 increments and they don't reach the expected values, I ended de soldering the MIC and testing the soldering points it gives the same weird resistance.

Ok, that should help get the value you want.

The weird readings are probably some capacitors charging. When empty they have (ideally) 0 resistance, which will then increase. Eventually it'll stabilise at what ever the actual resistance is. With no CPU fitted it should eventually (time will depend on the meter you use and how much capacitance there is to charge up, give it a minute maybe) read close to the resistor values.

[edit: adding the following...]
Thanks for including the photo of the back of the board. As you pointed out, something seems a bit odd around the 5V--3V voltage select block. Could you remove all the jumpers on that block and check what connects where? We know from your previous measurement that the top pin of R20 connects to Vout/Pin 4. So I think that means that the top-right and top-left pins in that block are both Vout. But I can't see where they connect. Maybe the trace runs on the top of the board, under the jumper block. In which case that top jumper doesn't do anything.

I've marked up what I can see of the PCB traces (I had to guess a bit for the green traces showing the connections from JP8 to the 4 resistors):

I'm very bad in doing this drawings but I've tried to do it right, I know the soldering points of the MIC regulator look bad, but they are OK,they lost two copper rings that flew when desoldering the MIC.

Ok, thanks for doing that, I've corrected my diagram based on your measurements, and it now makes sense. So I'm not sure why you had the wrong voltage. Could you put it back together, but leave the 5V-3V jumpers off and put JP8 on the first position? Then measure the resistance (put the probes on the legs, not on the solder holes) between 3&5 and 4&5. Give the meter time to stabilise. Also, move JP8 to the 4th position and re-check the resistances.

[edit: apologies for the confusing colour scheme and not labeling them: Yellow=5V, Red=Vout, Blue=Adj, Green=R2 Select, Black=GND]

[2nd edit: meant to say, good to see the heatsink on the regulator. Do make sure that the legs of the regulator don't touch the pad on the motherboard, which is probably connected to GND]

OK, take measurements, no jumpers, 4-5 = 100; 3-5 = 162
JP8 first position, 4-5 = 52.4; 3-5 = 54
JP8 fourth position, 4-5 = 54.8.4; 3-5 = 52.5
MIC pins, 1 & 2 = 5.10v; 4 = 1.83v; 5 = 1.24v
Checked the resistance before re-soldering the MIC;
4-5 = 91.6; 3-5 = 160.4.
From these tests the jumpers do not seem to have any effect, only the 3v-5v block that changes the voltage in the socket, 5.10v or 1.83v, the changes in JP8 are barely noticeable.
The voltage on pin 4 of the MIC fluctuates a lot and that of pin 5 is very stable.
I may have damaged the motherboard myself on the first try without really knowing what I was doing?
The voltages on the LT are 1 = 1.24v; 2 = 1.25v, 3 = 5.10v, in case it is relevant.
I think it is time to summarize what has been done so far; Is it possible that the motherboard has some design error and cannot work properly?
It is possible that if I use an LT instead of the MIC it will only work with 3.3v?
Maybe the MIC is defective?
I'm totally lost.
Is there a n equivalent to the MIC 29302?

Maybe I can remove everything and inject the 3.3v somewhere using a step-down DC-dc converter? They are common on eBay and very cheap.

Hoping wrote on 2021-11-28, 18:22:

Checked the resistance before re-soldering the MIC;
4-5 = 91.6; 3-5 = 160.4.

Is that with the 160k resistor soldered as R20 and the 90k resistor as R21, no jumpers fitted to the 5V-3V block, and a jumper on 1-2 of JP8? Because they look odd. I though we had traced pin 5 (adj) going to the nearest end of R20 and pin 4 (Vout) going to the far end of R20. So with no regulator fitted and JP8 1-2 fitted, then between 4-5 we should see ~160k. Equally, pin 3 (gnd) goes to the near end of R21 and pin 5 goes through JP8 1-2 to the far end of R21. So with JP8 1-2 fitted we should see ~90k between 3-5.

So somehow, something looks back to front here. Could you recheck that there is a short between:
pin 5 to nearest end of R20
pin 5 to far end of R21
pin 4 to far end of R20
pin 3 to near end of R21

Have JP8 1-2 fitted, it shouldn't matter if anything else is fitted as this is just to double check what's connected to what.

It is possible that if I use an LT instead of the MIC it will only work with 3.3v?

LT1083/4/5 (7.5A/5A/3A) should do, and can be made so you can select voltages in the same way. But let's try and work out what's going on here first.

OK, did the test, three times.
Jp8 1-2, 3v block fitted.
pin 5 to nearest end of R20-short
pin 5 to far end of R21-short
pin 4 to far end of R20-short
pin 3 to near end of R21-short
Maybe the board has an inner layer of tracks.

Hoping wrote on 2021-11-28, 18:22:

Checked the resistance before re-soldering the MIC;
4-5 = 91.6; 3-5 = 160.4.

Hoping wrote on 2021-11-29, 21:46:

OK, did the test, three times. Jp8 1-2, 3v block fitted. pin 5 to nearest end of R20-short pin 5 to far end of R21-short pin 4 t […]
Show full quote

OK, did the test, three times.
Jp8 1-2, 3v block fitted.
pin 5 to nearest end of R20-short
pin 5 to far end of R21-short
pin 4 to far end of R20-short
pin 3 to near end of R21-short
Maybe the board has an inner layer of tracks.

Hmm. So it is wired up as I sketched before. But that says R20 goes between pin 4 and 5, and R21 goes between pin 3 and 5. From your last photos R20 is the 160k and R21 is the 90k. But you've also measured the resistance from 4-5 as ~90k and 3-5 as ~160k. So something's odd somewhere. Is it possible you accidentally measured those resistance back to front?