820-3435 Tries to turn on can't, quarter spin

JohnB8812

New member
Board came in previously ultrasonically cleaned but not super well. There was remanence of corrosion on the bios on pins 7 and 8. Touched up soldering and have continuity from pins 7 and 8 to their corresponding spots. Measured around a bit and noticed PPVRTC_G3H would kind of jump between 3.3 and 3.42. Replaced clock chip now only goes to 3.3 like normal, but it still refuses to start. It doesn't quickly quarter spin but rather it will do the on off on off on thing like 3437's normally do; however, it never stops going on off on off. Could this be a result of crap under a chip like the SMC or where should I look next? The termination resistors around the bios look pretty normal like no burning but have not measured them yet. JTAG looks clean as can be.
 

aprendiz

Moderator
As Duke always says, this could be anything, corrosion somewhere, under CPU, SMC,traces and/or resistors on SPI>SMC or SPI>CPU, U1900 (replaced)...
 

JohnB8812

New member
I kinda figured as much. Going to do what I can these people who sent it to us get a sweet deal (aka we don't make shit) so spending more than an hour on their stuff is really not the best idea. Going to check tho if CPU Vcore is cycling what would that mean?
 

JohnB8812

New member
All the rails kind of fluctuate once the fan comes on. For example if the fan goes off PPBUS goes to 8.38 then once the fan comes on it stabilizes to 8.59. CPU Vcore will go all the way to zero when the fan is off then pulse as high as 3.7 volts then stabilize when the fan spins at 1.75
 

dukefawks

Administrator
Corroded traces to U6100 are most likely at this point. Check all termination resistors and traces going to U6100.
 

JohnB8812

New member
Termination resistors and traces are good. HOWEVER, I just found that I never get 3v42 on U1950 at ALL. Does this mean U1950 is fucked or does this mean there is something up with 3v42? I do get a steady orange light at all times
 

JohnB8812

New member
Actually nevermind I got it scratch that. I did measure the resistors and traces and every single one of them connects :(
 

JohnB8812

New member
I'm going to jump pins 7 and 8 and see tho. The trace doesn't look burned at all, but that was where the original corrosion was when I got the board
 

JohnB8812

New member
No change. Every single trace on the BIOS connects to its corresponding spot. Same with resistors. They all measure in spec, and there are no obviously fucked probe points really on the entire board for that matter.
 

dukefawks

Administrator
I assume you already ripped off J6100? Pull the BIOS chip and do a diode mode reading on each pin. Try another known good BIOS for testing, can be from 3435 or 3437.
 

JohnB8812

New member
Swapped ROMs and still doing it. I'm gonna call this one because don't have a lot of time left with this one. Any other ides what could cause it? Rest of board looks very clean
 
Top