BQ24735 - faulty?

[Jane]

Can you please give your opinion on this matter?

Laptop does not work. I checked voltages of charging circuit BQ24735 .

Values are :

pin 20: VCC 19V

pin 6: ACDET > 2.4V

pin 16: REGN 6V

pin 5: ACOK 2.62V

Are these values correct and (if BQ24735 is working in the correct way)

ACDRV (pin 4) should have more than 19V?

Or I must check any other values to be sure to decide that BQ24735 is faulty?


AS I have ACDRV about zero, so I guess BQ24735 is faulty.

Your opinion?
Thanks

[mon2]

Re: BQ24735 - faulty?

It looks good so far.

The boost voltage of ACDRV is based on CMSRC pin and REGN.

You should have a series resistor on ACDRV of ~4k (based on the TI schematics which everyone appears to follow as a ref design).

Remove this series resistor for a quick test and power up again.

What is the voltage of ACDRV at the IC side of this signal (without the resistor that is driving the gate of the 2 x DCin mosfets) ?

[mcplslg123]

Re: BQ24735 - faulty?

to me ACOK at 2.62V is not correct. Should be 3.3V. Without a valid value of acok,acdrv wont work.

[Jane]

Re: BQ24735 - faulty?

ACOK, should be HIGH, not LOW but what voltage here means HIGH?
HIGH is 3.3V and higher? Or even 2.6V higher?

[mon2]

Re: BQ24735 - faulty?

ACOK is a power good pin and will be open drain.

That is, the pin will be LOW (~0 volts) or open like a SPST switch (high impedance / floating). Therefore, this pin will contain an external pull-up resistor to a voltage selected by the designer.

Sometimes, the pull-up voltage is REGN but not always.

Since ACOK is NOT a logic low (~0 volts), this means ACOK is floating. The actual pull-up voltage is defined by another power rail.

See attached datasheet for more details of ACOK.

[Jane]

Re: BQ24735 - faulty?

@mon2
Thank you for your explanation.
I checked the BQ24735 datasheet and I can see that ACOK is connected through resistor R5 (10k) to 3.3V.
But in my laptop schematic Acer Swift SF314-51 it is connected via R8811
1KOhm only.
So, do you think ACOK at 2.62V is not correct - it is low , as mcplslg123 says above?
Or is it OK?

[mon2]

Re: BQ24735 - faulty?

1) post or link the schematic of your logic board so we can all review

2) on cmos logic levels:

The logic level thresholds for 3.3V CMOS are a known standard. For a 3.3V CMOS device to acknowledge a logic high or low, the required voltages are as follows:
Logic low - input must be between 0V and 0.8V.
Logic high - input must be between 2V and 3.3V.

3) since the ACOK pin is not ~0 volts, this charger IC is ok with the adapter detection based on the voltage present at the ACDET pin.

4) what is the voltage shown in the schematic for the pull-up rail on this 1k at R8811 ?

If the rail is ~3v then yes, agree that this voltage is lower than it should be.

Measure the voltage present on the other side of the pull-up 1k resistor. Post the measurement.

5) next, remove this resistor with care (flux, low air pressure, enough heat to melt the solder). Do not lose the part but the resistor value is not critical. Could be almost any value from 1k..10k, as an example.

After this resistor is removed, is the pull-up voltage different? If it is not different, then ACOK pin is not influencing the voltage on this pin. That is, the voltage value is low due to another factor on the logic board.

6) remove all power, meter in resistance mode. Check the resistance to ground of the pull-up voltage rail. If the resistance is low, then there is some fault on this line. It is not the ACOK pin since the pull-up resistor has been removed for testing.

[mcplslg123]

Re: BQ24735 - faulty?

I highlighted the acok voltage as there could be a potential problem If the pullup voltage of acok resistor is 3.3V.@mon2 has already advised on how to proceed.On many occassion i found this resistor to be bad pulling acok comparitively low. So it needs to be ruled out.

[Jane]

Re: BQ24735 - faulty?

Thank you for your emails
@mon2
Please find enclosed the schematic of the laptop I try to repair
(BQ24735 is on page 86)
I removed resistor R8802( 4.02k) going from ACDRV to Mosfet but voltage was still not changed ( about 0V)

Removing R8811( about 1k - going from ACOK) partly helped.
Now ACOK is 3.3V but ACDRV is still about 0v

So, can I suspect BQ24735 being faulty?

P.S.

I also find out that Q8800 was shorted so i removed it - not replaced yet

Thanks for help

[mon2]

Re: BQ24735 - faulty?

Yes, now replace the charger IC.

Still, leave the 4k resistor in series with ACDRV off the logic board. Test again to confirm that ACDRV is ~25v when powered on.

Be sure the dcin mosfets that is driven by the 4k resistor are not defective. Remove all power, meter in resistance mode, check the resistance across the mosfet pins.

Then solder back the 4k resistor and test again.

[Jane]

Acer SF314-51 - BQ24735 problem

This is part 2 of this thread
https://www.badcaps.net/forum/showthread.php?t=111377
To sum it up:
Acer SF314-51 is dead. I found out that ACDRV pin of BQ24735 is zero
while some of other values are correct - see the given thread above - so I replaced the BQ24735..
But the problem remained. Finally I found out there is a short between
pin 1( ACN) and pin2( ACP) of BQ24735.
I tried to find the culprit but no success.
Is it possible to inject a voltage to the rail and check with thermal camera?
If so, what voltage/current to use so that it can be possible to see the short in the camera?
Or which parts to check?
(I checked C8814,C8806,C8805,Q8801,Q8802 and probably C8828,C8813 too - not have boardview- and also BQ24735)
I enclosed complete schematic and also only a page of the charger.
Thank you for your hint/help

[Sephir0th]

Re: Acer SF314-51 - BQ24735 problem

This "short" is normal. There is nothing more than the Current sense resistor between these rails which has almost no value itself. You'll notice the same design at the sense resistor of the charging rail. So the issue is elsewhere. REGN is still present?

[mon2]

Re: Acer SF314-51 - BQ24735 problem

As per above comments - the current sense loop will appear to be a short but it is not. Pins 1 & 2 are connected by a very low resistance current sense resistor so it appears like a 'short'.

It is important to review the replaced charger IC for possible other faults / soldering.

Remove the ACDRV 4k resistor and power up again.

Confirm the voltage of:

REGN (pin #16) ; must be 6 volts
ACDRV (pin #4) ; should be DCin adapter + 6 volts (REGN)
ACOK (pin #5) ; should be a logic '1' to denote the adapter in is OK

[Jane]

Re: Acer SF314-51 - BQ24735 problem

Thank you for the replies.
I am sorry I did not describe my problem clearly. There is a short between
pin 1( ACN) and Ground and pin2( ACP) and Ground of BQ24735.
(I did not mean short between ACN and ACP.)
Yes, there is very low resistance between pin 1( ACN) and pin2( ACP)
but my problem seems the short circuit to Ground .

REGN (pin #16) ; must be 6 volts
ACDRV (pin #4) ; should be DCin adapter + 6 volts (REGN)
ACOK (pin #5) ; should be a logic '1' to denote the adapter in is OK

Yes, above values are correct. (but to have ACOK logic '1' I have to remove
1k resistor from that ACOK.If not removed, ACOK voltage is about 1.4V

I also removed ACDRV 4k resistor but no proper output still.
But I think the main problem is the short to Ground between ACN ( ACP).
What do you think?
Thank you for help

[mon2]

Re: Acer SF314-51 - BQ24735 problem

Apply lots of flux and then hot air with enough air pressure to melt the solder under this charger IC.

Spin the hot air tool in circles on the top of the IC.

The ic will sink in the liquid solder on its own.

Then gently nudge the IC to shift the part. The part will shift but return to proper alignment on its own. This is needed to remove any shorts.

Very likely the ic is shorting from the soldering process.

Then measure for the short again. Repeat if needed till the short condition is removed.

[Jane]

Re: Acer SF314-51 - BQ24735 problem

But I removed that charging IC as well (and tested for short) and there was still a short circuit - between Ground and ACN and between Ground to ACP.
Is it possible to use thermal camera? But what current/voltage to use?
Thank you for reply

[mon2]

Re: Acer SF314-51 - BQ24735 problem

Is JP8805 a solder blob on the board?

If yes, remove it and check if the short is still present or not.

There are only a few parts linked to the ACN pin. One of them may be causing the short. Review the caps on this line.

[Jane]

Re: Acer SF314-51 - BQ24735 problem

I removed JP8804 ( JP8805 is connected in PCB itself). Now there is no short on C8814 and on C8806( pin 1 on the schematic). So, I think BQ24735 should be OK. Do you agree?
I tried to find out R8808 but I did not.(no boardview ). Now there is not a short circuit on drain Q8801( there was when there was JP8804 connected).
Short circuit remains on C8805, C8806( pin2) Q8802 drain, C8828 and C8813.
I measured Q8802 and also cappacitors and they were not shorted.
Do you have any hint, please?

[mon2]

Re: Acer SF314-51 - BQ24735 problem

You are getting closer but we still cannot rule out this charger IC.

Next (in order):

1) remove the solder blob @ JP8805. Check the resistance to ground again.

If you do NOT see a short on C8805 after this removal, then the charger IC is NOT the fault.

2) After the above, review the caps near Q8804. You may have a shorted cap there.

3) Do you have a lab adjustable power supply ? I think you can locate the bad part without one but a variable power supply will help.

[Sephir0th]

Re: BQ24735 - faulty?

Haha, I have re-read the thread just to be sure and we indeed missed a simple short to GND of the main power rail...

As described above I wouldn't waste more time with this and start injecting if a high side short can be ruled out before. So what is the exact resistance to GND at the current sense resistor with decimal place?