Thanks for all the replies.
In order of replies,
@amyk, I did try searching for the schematics with the motherboard and the part number was more like, h000052360. The search hits on images very similar to the motherboard I have, but the schematics are almost always code-named 'Inventec Darkar', and seems very different from the motherboard that I have, comparing the power-supplies.
@Rasz, wow excellent find. Definitely, I have to work on my searching skills
. I should probably give you a little more background. The problem started as a continuous power-poweroff cycling sequence. I suspected some over-current situation (short) and tested the different power rails. Most of them (like DRAM voltage) showed a swinging sequence (on-off) but then there enable signals were also swinging. I also did a lot of analysis on the startup chip (NPCE795) looking at the signals and my current theory is that the PCH did not get a PWROK (referring to Intel manual) and reset the startup chip which in turn resets the PCH.
Except for this supply which I suspect is 0.9v (VCCSA) for the intel PCH. In VCCSA case, the output voltage showed a quick spike to the desired voltage and then it went off, even though the enable was ON for longer (of course till the next restart which has a period of roughly 1.8secs). The power good for VCCSA was also low, although there could be other PGs connected in this line. Since I dont have a schematic, I am not sure. I went around probing different signals on this chip (RS8204) and they all look like expected. For example, the feedback was required to be at 0.75v and I case see a short pulse of 0.75v before it goes down and the enable signals had a 100k to VCC etc. All this while there were no dead shorts and I was not very confident to remove the MOSFETs and check them and *thought* the MOSFETs were also good, as there were no shorts. I provided an external 0.8v supply current limited to 200mA and it was taking less than 100mA.
Thankfully there were solder jumpers for the output of this power rail to connect to the load, so I removed these jumpers and then tried powering up the board and sure enough, it showed the same tripping sequence. After all this finding I suspected that it is the chip and ordered one and replaced it. I have only limited experience with replacing a QFN with hot air, and I hope it went ok. Moment of truth, I decided to power on (still vccsa power rail not connected to load) and I suspect the same tripping sequence and additionally I should be able to test the vccsa now. But unfortunately now there is a short somewhere and the board wouldn't start, probably the startup chip is stopping it to start due to the high current situation. I overrided the 19v MOSFET and supplied a 12v current limited (1A) supply and narrowed the short (touch the component with my finger) to this MOSFET. I removed the MOSFET (2WAC) and the short was gone and the motherboard is back to the same restart sequence. I am not sure if I cooked the mosfet, but I did cover the surrounding with Kapton tape. and the MOSFETs are far from the site of the chip.
Another follow up question: Would you recommend using another MOSFET (like AON7502, which I have in my parts bin) to test if the circuit works. Or any other ideas to test out will be great.
@flolic, Yes it is the output capacitor. Not sure if I cooked that too
Thanks again.
Home
Help
Search
Login
Register
