Mac Air 2012 - A1465 820 3208A

[jpadie]

Thanks. I found a dead cap. Removed it and the mosfet chip and now have another rail back and the cap is behaving itself with normal increasing resistance when measured.

But still no fan spin or vcore. Time for sleep and I will investigate more tomorrow .

The chime was there after I finished checking a couple of weeks ago, alhougg the board was gettng super hot. The board has not been touched since. So something has degraded in the meantime .

 

[jpadie]

current status is that I have my PP1v05_S0 rail back now.
I'm getting 1.5v on PP1v5_S0 but only 1.35v on PP1V5_S3. which is confusing.

and still low voltage on ppvttddr_S3, zip on PP0v75_SO_DDRVTT
and nothing on PPVCORE_S0_CPU/AXG

I guess the next thing to investigate is the 150mv drop on pp1v5_S3?

ideas?

 

[jpadie]

in fact I'm going to hunt on the ppvcore instead.

no high on the en line to U7400. putting 5v in to that got fan spin.

so no all_sys_pwrgood.

the slight flummox now is that per my schematic I am missing both R7962 and … U7960 (ISL). pads are present for both; and the resistor ladder for pwr_goods is there and half populated.

I guess I must have the wrong schematic and/or boardview (although everything else is right so far...). how annoying.

 

[jpadie]

I have a board that was pre ISL and uses BJT to provide all_sys_pwrgd. not enough coffee.

but still no all_sys_pwrgd!

 

[dukefawks]

If there is no ALL_SYS there will not be Vcore. ALL_SYS must be present first. Check all rails and see which one is missing. Vcore should come on as soon as ALL_SYS is present.

 

[jpadie]

all_sys is made from Q7950 which takes input from:

1. PP3v3_S0_MON via a resistor divider. I get 3.3v coming in but the resistor divider is a different ratio to what's in the schematic. I get 0.77v at the divider and 0.7v at the pin. From the schematic I should be getting more than 1.05v at the divider.

2. PP1V5_S3RS0_VMON. I get 1.35v here and 0.78v at the pin. This is higher than the stated worst threshold.

3. PP1V05_SO_VMON. I get 1.05v where it should be and 0.78v at the pin again. Higher than the worst case threshold.

so the odd one out is pp3v3_SO_MON via the resistor divider.

For the resistor divider I measure
R7951 at 9,4kOhm

R7952 at 6.1kOhms

so a divider ration of 0.39

whereas it should be 0.33.

on the collector of Q1 I am seeing on 200mV despite having 3.3v on PP3V3_S5.

so I removed Q7950 and I'm now seeing the right voltages on the pads. So I guess that Q7950 was fried? is that a reasonable assumption?

 

[jpadie]

Does anyone know a European source of ASMCC0179? I can't even find a datasheet for it, let alone a vendor.

 

[dukefawks]

Donor board, this part is used on pretty much every Apple board. Do I think it is really bad....no. I have seen it once.

 

[jpadie]

Thanks . I wasnt able to find the same part on the donor boards I checked. I will look again.

 

[jpadie]

Going back to your comment about it being an unlikely culprit; these are the symptoms:

1. no all_sys_pwr gd.
2. dummy all_sys_pwrgd and the board spins up.
3. voltages on all the input pins to the chip are pulled down from where they should be. Remove the chip and all the voltages are correct at the pins.

given the chip design, doesn't this suggest an internal short across some of the bases? that would also stop the bases from rising high enough to turn on the transistors and thus the mosfet. If there were a short on all_sys_pwd gd after the mosfet then we shouldn't see the issues across the bases of the transistors would we?

If this is not the culprit, can you point me towards others? the all_sys_pwrgd line on this board (the BJT version) is quite simple so I couldn't think of anything else!

 

[dukefawks]

Every board has ASMCC0179: 3115, 2915, 3437, 00165, 3536, all of them.