820-00840 wont boot

[estelanim416]

Hi

Unit stopped turning on, no visible liquid or other damage

Getting 5v then goes up to 20v and stays there forever but pulling only 0.01A
checked a few things

PPBUS_G3H 12.99v
3V3_G3H 3.39v

L6995 to ground
Pin 1 277
Pin 2 276 (is it too low?)

U7000
P_IN 20.1v
AUX_DET 5v
VDDA 12.9v
VDDP 5v
SMC_RST* 3.3v
AUX_OK 3.3v
AMON 0v
BMON 0v


BUF_SMC_RESET_L 3.3v

Any advice appreciated

 

[2informaticos]

Forget U7000 and 3V3_G3H, if already got 13V at PPBUS_G3H.

Do you get 3V3_S5?

 

[estelanim416]

Thanks

Nothing on 3V3_S5

Please advise, I cant seem to work out where 3v3_s5 originates from

 

[2informaticos]

Check U7800's (7 OF 10) LDO voltages.

 

[estelanim416]

Thank you

I hope I checked the right ones

Pin 16 PP5V_LDO5V 4.9v
Pin 39 PP3V3_LDO3V 3.2V
Pin 46 BANJO_FBLDO1 0v
Pin 73 BANJO_VINLDO1S 0v

The rest is just copies of the LDO5V or LDO3V
You have asked me to check 7 of 10 if I understand correctly so I am obviously missing something.

 

[2informaticos]

(7 OF 10) is the U7800 block in schematic.

SMC_PM_G2_EN comes on?
Check PMIC_SHUTDOWN_L.
Also check U7890, pins 2, 4, 6.

 

[estelanim416]

Ok, thanks

SMC_PM_G2_EN 3.3v
PMIC_SHUTDOWN_L 3.2V

U7890

Pin2 0v
Pin4 0v
Pin6 3.3V

 

[2informaticos]

Be sure R7822 is good.

PP3V1_RTC is there?

Test starting in SMC bypass mode.
Check for short on all big coils.

 

[estelanim416]

PP3V1_RTC 3v

R7822 no matches found C7822 doesnt exist either I hope its a typo

Test starting in SMC bypass mode. - I've tried that with/without battery - doesnt seem to make any difference. I dont have a test chassis thou so maybe keyboard is bad but thats a long shot. Lots of shorts on coils so I am guessing this is something to work out

Checking coils (motherboard disconnected, red on ground, COM on coil terminal)

L7370 short to ground on pin 1(SAVR_SW) not sure if that's normal
( checked around L7370 - all seems fine, the small cap by the cpu short on both ends C10C5. PPVCCSA_S0G short to ground everywhere I check)
L7220 short to ground on both terminals (tested caps around the cpu PPVCCGT_S0G is shorted to ground)
L7210 short to ground on both terminals
L7410/20/30 short on both terminals
L7730 no shorts
L7030 - no shorts
L8030 - no shorts
L7960 - no shorts
L8410 no shorts
L7690 - short on both ends
L7670 - no shorts

 

[2informaticos]

Sorry, check R7862 instead.

If you associate beep with short, is a big mistake!
Multimeters beep until 0.050-0.075, depending by model.

Please post diode mode to ground readings on the coils.
Switch to ohm scale for values lower than 0.010 in diode mode.

 

[estelanim416]

R7862 3.2v on both pins

Yes, it was the beeps that I considered shorts
2k scale

L7900 Pin 1 .046 Pin 2 .046
L8060 Pin 1 .061 Pin 2 .061
L7370 Pin 1 .009 Pin 2 .009
L7220 Pin 1 .006 Pin 2 .006
L7210 Pin 1 .006 Pin 2 .006
L7410 Pin 1 .002 Pin 2 .002
L7420 Pin 1 .002 Pin 2 .003
L8030 Pin 1 .561 Pin 2 .561
L7960 Pin 1 .065 Pin 2 .066
L7730 Pin 1 .484 Pin 2 .484

L7430 Pin 1 very jumpy to infinity and down to .003 Pin 2 .003

L7030 this one 20k scale Pin 1 13.43 Pin 2 13.43

L8410 this one 200M scale Pin 1 1.1 Pin 2 1.1

L7690 this one 200 scale Pin 1 0.5 Pin 2 0.5

L7670 this one 20k Pin 1 14.25 Pin 2 14.25

I hope it makes sense

 

[2informaticos]

Not too much sense; did you read my last post?
"post diode mode to ground readings"

More information here, in case didn't read all thread before.

HOW TO collection

I hope this "How to" collection will help our members. I will possibly modify this post in time, if updates available. Trigger SMC_ONOFF_L (PMU_ONOFF_L): It reffers to simulate power button function, when the keyboard is not connected. We need to do this when we work with the board...

boards.rossmanngroup.com

"Test in diode mode:
Connect positive probe of the multimeter (red) to ground and touch the test point with black probe.
Sometime (rarely) we need to test in reverse direction."

BTW, don't waste time checking on both pins of a coil.

 

[estelanim416]

I did just that. All readings are diode mode to ground, red to ground
Some of the coils would show out of range so I had to change the scale
Attached picture of my multimeter

 

[2informaticos]

L7410/20/30 are tied together.
Any reason to check any of them?
Good enough to check on common (output) pin.
Same applies for L7210/20.
Try to understand a schematic, instead of wasting time.
And stop to check on both pins; do you really expect to get different readings?

You can't get out of range reading in diode mode, when red probe is connectd to ground!
I already said to switch to ohm (not K/M) scale for very low values only.
I'm sure you will understand the reason in time.

However, I don't see any strange value, but should be better to compare with good board.

Remove U7890 and join together pads 4 & 6.
Then check for voltage at L7690.

 

[estelanim416]

Hi
I simply don't know what I don't know. I've signed up for electronic engineering course as I am getting more and more of such jobs. Thank you for bearing with my pure ignorance.

Removed U7890, joined 4&6

L7690 1.08v
as it was plugged C7933 went red hot - removed it, line is not shorted, I have another cap from 820-4924, original is X6S and the replacement is X5R, does it matter if other specs match?

Edit: well replaced it and it didnt burn so seems fine, now L7690 has 3.28v

 

[2informaticos]

Is important to be 10uF/6.3V.
But you can also use one rated for higher nominal voltage, like10V; just make space for it.

 

[estelanim416]

Ok, C7933 replaced,
With U7890 removed and pins 4&6 connected, PP3v3S5 is 3.28v
When I put U7890 back, PP3v3S5 is 0.01v
Is it U7890 then? or does it mean something else?

 

[2informaticos]

So 3V3_S5 power supply is good now.
But still not enabled by PMIC.

Keep U7890 soldered and check again voltage on its pins 2, 4, 6.

 

[estelanim416]

Ok
Pin 2 0v
Pin 4 0v
Pin 6 3.28v

 

[2informaticos]

Try changing U7800...