820-00165 Not turning on

[bjf]

I have an 820-00165 that was not spilled on. It does not turn on. Main voltages are all normal. ALL_SYS_PWRGD is at 0V. CPU Vcore is at 6Ω to ground rather than the 15 or 16Ω I would normally expect. Any idea what I should make of that?

 

[dukefawks]

What are "main voltages"? 5V pulsing? PPVRTC?

 

[bjf]

No pulse on any 5V lines. PPVRTC is good at 3.3V. No shorts anywhere that I can see.

For main voltages I'm referring to the things I would normally expect to be in place for the board to switch out of an S5 state. PPBUS, 3V42, 3V3_S5, 5V_S5, voltage to the SMC. Nothing seems off which is why I was wondering if that slightly lower measurement on the resistance to ground for vcore might be an indicator of what the problem might be. Not sure where to look next.

 

[Gurmon]

So to clarify, the board is stuck in S5 state ?

Do you have a scope to measure clock signal ?

 

[bjf]

Yes, not moving beyond S5. And no clock signal.

 

[Gurmon]

Replace u1900 to restore 32k clock.

 

[bjf]

I replaced it. No change. SYSCLK_CLK25M_X1 has 0.6V and SYSCLK_CLK25M_X2_R has 0V. Any thoughts on where to go next?

 

[dukefawks]

Check for 32kHz output on U1900 pin 12. Check all rails for short to GND. Check 5V rail again for pulses, they will be short and hard to catch on multimeter, better use scope.

 

[bjf]

I was borrowing an oscilloscope but had to give it back. I have a Fluke multimeter that can measure frequency so if I can just use that I would rather not spend money on an oscilloscope. Any idea how to use a counter to measure frequency? I can't figure it out and can't find any info anywhere.

 

[dukefawks]

You put the meter on freq. mode and neg probe on GND and pos. on the 32kHz output on U1900. My meter says 32 then. No idea about Fluke though, compare with a known good board. Could of course be any board with an identical clock chip, so anything 2012 and up.

 

[bjf]

Thanks, I tried that on a fully functional board and must be missing something. I'm watching this guy use my exact meter and I can't get anything other than a default 60Hz https://www.youtube.com/watch?v=5EUCeY-U26U I'll have to figure out what I'm doing wrong.

 

[2informaticos]

0.45V is a normal DC voltage at pin12 when has 32KHz output clock.
That DC voltage will not assure 100% working chip; but 0V DC points to no oscillation 100%.
Output pins 8&15 will not work before the board starts.
Check voltages at pins 5, 13 and 1.

 

[bjf]

Ok, thanks, yeah, that's normally what I've been doing - just checking for voltage and comparing to a good board. I would love to figure out this frequency counter but for now I'll do it the old way.
Pin 1: 3.3V
Pin 5: 3.3V
Pin12: .45V
Pin 13: 3.4V

Also:
Pin 3: 0V
Pin 4: .5V

 

[dukefawks]

Lets check some basics here. S5_PWRGD, DSW_PWRGD, PM_RSMRST_L, ADAPTER_EN?

 

[bjf]

S5_PWRGD: 3.4V
DSW_PWRGD: 3.4V
PM_RSMRST_L: 0V
ADAPTER_EN: 3.4V

I noticed that PP1V05_TBT is almost shorted. It's at about 0.007 (8Ω) to ground in diode mode. I removed everything in the circuit short of U2800. No change.

 

[smiba]

Not sure what the measurement is supposed to be, but keep in mind that 8Ohms on 1.05V is not a lot. (Its 138mW or 131mA)

 

[dukefawks]

PM_RSMRST_L should be present. Is PP3V3_SUS present?
Compare the 1V05_TBT resistance to a known good board. U2800 can short out more rails if it is fucked.

 

[bjf]

On this line it's around .150 (400) to ground on a good board.

 

[bjf]

No 3V3_SUS. I was already thinking about pulling U2800. Can I pull it and still have a chance at a functional board (w/o tbolt)?

 

[dukefawks]

Sure, just rip it off.