HP ELITE X2 1013 G3 weird battery issue

[huzo]

Hi, so the old battery was getting error stated it's capacity is starting too get low and should be replaced, so the customer bought a new one and replaced it, but the new battery was also getting this notice that it's capacity is getting low and should be replaced, so i took the laptop for testing and the battery is charging at the terminal at 8.03V , the voltage of the battery is 7.7V, and drawing around 600mA when charger is put and the laptop is off...so that should be ok... but the bios system test doesn't recognize the battery and can't do the battery test and so is the hp battery manager, it doesn't recognize the battery...BUT the battery is showing 0% charge and the laptop is working fine for like 3h playing youtube videos yesterday...so the conclusion is the battery is 0%,laptop doesn't see it, but the laptop works fine on battery! I've tried BIOS update,tried reseting the bios...Nothing was done to the board by the customer,he simply took the old battery out and put the new one ,and now both of them don't work properly...I've tried to put the old battery back and it acts the same as new one...Did the HP put some software on the board that you can't use a second hand battery or what?Like apple used to?

[mon2]

If the battery is not recognized then the smbus interface on the bms may be defective or one of the smbus lines is stuck.

Carefully measure the voltage to ground of each pin on the battery pack. Post each measurement.

[huzo]

That's the weird thing, it is recognized as the laptop sees the battery is 0% and should be put to charger immediately but the battery works for hours like that..., it's not like it says that it isn't present...Ok i will measure the lines for start later when i'm home...

[mon2]

Detection of a battery can be through a single line on the battery pack that forces the active low state to the battery charger. The actual charge value of the battery cells is always through the 2 wire SMBUS (I2C) interface used to extract details such as the serial #; battery manufacturer; # of charge cycles, etc.

[huzo]

Yeah you are right as always...But what could have happen if the battery was unplugged and another was put in? nothing else happened....i wil try to measure the lines...

[huzo]

voltages are :

GND
GND
GND
GND
0V
3.43V
3.03V
0V
BATT+
BATT+
BATT+

[pc_okay]

Yeah you are right as always...But what could have happen if the battery was unplugged and another was put in? nothing else happened....i wil try to measure the lines...

The curstomer may have use a screwdriver to replace the battery and short data line with BATT+... or he is very unlucky.
I don't think laptop is charging correctly "drawing around 600mA" depend on you system but laptop may use like 2Amp.
I have a few experience with broken data line and laptop were still working on battery but no charge and no info on battery.

[huzo]

Yes,that is a possibility...there is no schematics for this laptop, or boardview for that matter unfortunately...i changed a random chip near the battery connector, it has sda and clk pins, so i thought data lines go through him, but it' only a temperature sensor manager... G781.TOo bad ,the laptop is still beautifull and usefull...but the battery is working,just not properly...i will wait for mon2 to say is there anything to do to fix it...

[rediii]

did you flash the bios/ec?

[mon2]

3.43V
3.03V

These are good measurements. This confirms that the SMBUS (I2C) lines are not shorted and not being shunted to ground.

Can you hunt for any silk screening on the logic board that may help us to locate a schematic? Search for 'LA' or 'NM' type of markings.

[huzo]

Did an official update yes...something is wrong with communication and i need schematics or board view...

[huzo]

These are good measurements. This confirms that the SMBUS (I2C) lines are not shorted and not being shunted to ground.

Can you hunt for any silk screening on the logic board that may help us to locate a schematic? Search for 'LA' or 'NM' type of markings.

it's this one exactly...

[mon2]

No hits on this board in my database. Hunt for the charger IC that is onboard. Most likely it will be the BQ (Texas Instruments) or ISL (Intersil) series device and will be a square QFN package. Share the full details of what you find. From this detail, we can reverse engineer the board and how it works. If the IC is BQ, the DCin mosfets will be N-channel type.

[rediii]

schematics are unavailable atm. i have had a scenario where the ec rom was broken and that was why the battery wasn't charging right. but first do what mon says... do you have a scope ready?

[huzo]

No hits on this board in my database. Hunt for the charger IC that is onboard. Most likely it will be the BQ (Texas Instruments) or ISL (Intersil) series device and will be a square QFN package. Share the full details of what you find. From this detail, we can reverse engineer the board and how it works. If the IC is BQ, the DCin mosfets will be N-channel type.

IC is intersil 9538h

[huzo]

i changed it with 9238H, and it acts the same...There is soemthing else going on...

[mon2]

Did not know this but are you using the official HP USB-C charger?

Apparently these models demand a 12V PD contract which may not be available on other power delivery bricks. Should not affect the battery detection but interesting to learn.

https://www.reddit.com/r/Hewlett_Pac..._on_even_with/

[huzo]

yes it's original, and i have a Lenovo one also,they act the same...

[Sephir0th]

These are good measurements. This confirms that the SMBUS (I2C) lines are not shorted and not being shunted to ground.

Can you hunt for any silk screening on the logic board that may help us to locate a schematic? Search for 'LA' or 'NM' type of markings.

In would neither call those good measurements nor take this as proof that the Battery Clock and Data are fine. Voltages are not close enough to each other.

To finally get something like a clue i strongly suggest to move to the charging controller directly and measure CLK/DAT there. Not only Voltage, but resistance to GND too. Share the result

[huzo]

pin 22 and 21 are sda and clk, i have 3.46V on the chip pins,but on battery connector 3.05V on one pin.