A1990 Board no boot

[Stephen]

No power no boot,
Connected to battery:
20V
.59amps climbs up to .90 then resets back to zero and repeats, CPU and GPU does not get hot.

Not connected to battery:
20V
.07 Amps and steady no climb or drop

Voltage readings:
PPBUS_G3H 12Volts
PPBUS_DCIN_CHGR 20Volts

Also the board can get into DFU but it will not restore. Gives off errors.

Board is not water damaged and is mint condition with no previous work or repair.

[Stephen]

Re: A1990 Board no boot

Bump

[mon2]

Re: A1990 Board no boot

Hi.

You have no current draw when operating off the external power adapter but ok with the battery.

Focus on U7000.

Post the voltages on:

I2C_PWR_SDA
I2C_PWR_SCL

also on the related interface lines @

U6950

SMBUS_3V3_BATT_SCL
SMBUS_3V3_BATT_SDA

[Stephen]

Re: A1990 Board no boot

Will post my readings Monday thanks mon2!!

[Stephen]

Re: A1990 Board no boot

I2C_PWR_SDA - 1.80V
I2C_PWR_SCL - 1.80V

U6950

SMBUS_3v3_BATT_SCL - 3.40V
SMBUS_3v3_BATT_SDA - 3.40V


Thoughts?

[mon2]

Re: A1990 Board no boot

These are good readings for the I2C / SMBUS line. There is a logic level translation (mosfets) to allow for interfacing the 1v8 GPIO pins and other side is 3v3. They do not appear to be stuck.

See attached. While it may seem early, suspecting that U7000 is defective based on the fact that you have 20V but low current draw. So you do not have a faulty cap on the INPUT to U7000 otherwise the 20V would sink down to a very low value.

The output rail is ok since the battery works.

The I2C bus is ok so it is not the root cause - although we did have this case and was linked to the ESD diodes camped on the I2C lines. We had bad voltages on the SCL or SDA line. Removed the ESD device and that fixed this logic board.

You can check if the mosfet is ok @ Q7030, Q7040 and Q7065. It may be tricky to measure these mosfets due to the density of the board.

One suggestion is to use wirewrap or a single strand of wire wrapped around the meter probe and use the flying single strand of wire as your probe.

Then remove all power (including battery), meter in resistance mode and check the resistance of each mosfet:

source & drain
source & gate
gate & drain

If you have a very low reading on the meter, then the mosfet is bad and will need to be replaced. In our case, we went ballistic and replaced each mosfet, U7000 (twice) and still did not work - till the ESD diode was removed.

Consider to check the mosfets (Q7030, Q7040 and the battery mosfet @ Q7065) and then I would replace U7000. I believe that Mobilesentrix (USA) is carrying the ISL9240 devices now. We brought in a batch from Aliexpress, if you are stuck.

We want to be sure that the battery mosfet is not damaged (shorted to be always ON) and possibly causing U7000 to shutdown. This is another possible scenario.

https://www.mobilesentrix.com/isl924...179-bga-40-pin