BQ25750: - AC IN Power Path on AC removal

Hello Richard,

Thanks for working with this

I can't see attached picture and the schematic is a bit blurry. Can you resend these pictures?

Can you tell me the system's input voltage, output voltage, charge current, and max system current?

Best Regards,
Ethan Galloway

Input voltage is 12V. The output voltage is input voltage or Vbat (12- 21V(max)) Charge current 1.5A  Max system current is about 1A steady state. 

/resized-image/__size/320x240/__key/communityserver-discussions-components-files/196/6204.pastedimage1730264427344v2.png

I have not seen the charger actually charge at 1.5A  yet. It is usually charging at 500mA. 

I can't seem to get the pictures to upload to here that is visible.  

PowerBoardSchmatic.pdf

Hello Richard,

Thanks for being patient with this. I'll get back to you tomorrow.

Best Regards,
Ethan Galloway

Hello Richard,

Thanks for sending me a clearer schematic. I'm looking over it and I have a few suggestions:

• R3,R4,R9, R10 need to be 10ohm resistors.
• I'd recommend making sure C1, C2, C11, C12 are ceramic capacitors.
• We recommend a minimum of 80uF across both VSYS/VIN_PWREN and VBAT for the charger to be stable.
• I recommend installing an RC fitler in parallel with ICHG and ILIM_HIZ resistors. We recommend this when those resistors are greater than 5kOhm.

With a max VIN of 20V and a max VBAT of 21V, I don't think you'll need gate resistors or gate capacitors.

What's being measured in the oscilloscope image by the way?  I can't quite understand it.

Best Regards,
Ethan Galloway

VIN max is 12V, not 20V. 

C1, C2, C11,C12 are ceramics. 

For the 80uF, I need to add more at Vsys? But Vbat is OK? I don't know what you mean by VSYS/PWR_EN

I can add a filter there, what do you recommend the time constant? 

The scope is measuring the CS resistor during VIN (AC) removal, and the current is flowing out of the battery, into R2 (scope shoot), through Q2/Q1, to keep up Vsys. (computer) That is when the Ibat OC error happens. 

Also, I do get an BAT_OV error from time to time. Would the 80uF help? (tho present on Vbat already) 

Hello Richard,

Thanks for the clarification on the waveform. I calculated about 6A of current flowing through R2 which is normal when the VAC is removed and the BATFETs turn on. How are you taking the differential measurement here by the way?

I think the most helpful change would be making sure R3,R4,R9, R10 are 10Ω resistors. This might be why the charger is giving a BAT_OV error.

The capacitance on VBAT is fine, but you'll need about 40µF capacitance on the ACP side of the input sense resistor. This node is marked as VIN_PWREN in your schematic. I'd also recommend installing a small ceramic capacitor on VSYS close to the switching FET Q3. This will help filter out the high frequency noise.

Let me know if changing those resistors fixes your issue and let me know if you have any more questions.

Best Regards,
Ethan Galloway

Probe right across the resistor. The scope is floating. Convert the 100 ohm resistor to 10 ohm resistors? 

How will that help with the Bat OCP? What is the threshold for this? 

Hello Richard,

The SRN/SRP pins measure the current through the sense resistor to determine the battery current. The chip internals were built with 10Ω resistors in mind. I highly recommend changing these resistors from 100Ω to 10Ω.

The charger triggers BAT OCP by measuring the difference in voltage across the SRN to SRP resistors. IBAT OCP triggers when the difference between SRN and SRP is greater than 120mV.

Best Regards,
Ethan Galloway

So how will that help with BatOCP?

Hello Richard,

The battery current is measured by the SRN/SRP pins and the SRN/SRP pins are designed with a specific filter in mind. Let me know if changing those resistors to 10Ω from 100Ω improves the circuit behavior.

Best Regards,
Ethan Galloway

The board goes into flickering on and off.  I can hear the inductor cycle on and off and all the status light flash really quickly in unison. So, it made it worse. This is no load, no battery, just AC applied. 

I got it to stop flickering, could have been a bad sense resistor. Now the board is allowing Vbat (Vsys) to go out and show up at AC input. 28V is present at AC fet gate. Fets check out OK. First time seeing this. But Bat fet gate is 0V. 

Hello Richard,

When you measure 28V at the ACFET gate, are you measuring with respect to GND or the FETs' source pin?

Did you check if VSYS is shorted to VBAT? That's the only thing I can think of here when VSYS=VBAT while BATDRV=0V.

Best Regards,
Ethan Galloway

Yes, the AC fet has 10V across D10. Bat fet has 0V at gate pin. But Q5 gate pin has 26V, 9V across Q5 gate pin and SW2. Why is it letting this reverse current path to occur? I also read 3.9V across Q3 gate pin and SW1. Almost like it is in boost mode? AC should be off in that mode right? I thought boost mode is not on by default. 

Q3 gate has a PWM on it, almost in a 99% DC, for what I assume is for keeping up the boost cap. 

Hello Richard,

I have a few questions to help debug this further:

• Can you send me the register file of the BQ25756?
• Can you remind me what the voltage of VBAT, VSYS and VAC are in this condition?
• What is the ACUV setting of the BQ25756?
• With board powered OFF, can you measure the resistance of VBAT to VSYS?

Best Regards,
Ethan Galloway

Can you send me the register file of the BQ25756? - Default, I don't have SW control on it. 
Can you remind me what the voltage of VBAT, VSYS and VAC are in this condition? Vbat = Vsys = Vac
What is the ACUV setting of the BQ25756? (10k/3.24k/7.2k)
With board powered OFF, can you measure the resistance of VBAT to VSYS? - 4.27k

Hello Richard,

richard Johnson said:

With board powered OFF, can you measure the resistance of VBAT to VSYS? - 4.27k

4.27kΩ is really low here. What's this value if your remove the FETs? Once you replace the FETs, does the value change?

Also, can you send me the part number for the BATFETs?

Best Regards,
Ethan Galloway

So, why is the is device allowing the current to pass through the dc/dc power supply?  

Do I need to abandoned this device and pick something else? It worked better with the 100 ohm resistors on the current sensor inputs.

Hello Richard,

richard Johnson said:

So, why is the is device allowing the current to pass through the dc/dc power supply?  

I think the BATFETs went bad in your circuit here. The BATFETs shouldn't allow current to pass through them with a VGS of 0V.

Can you replace these FETs with new ones and see if that helps? I would recommend using AONS6276 FETs. These FETs have a high SOA and are more durable.

You probably already have this, but we have an FAQ for the power path switchover. Here's the general FAQ for the part family.

Best Regards,
Ethan Galloway

Why do you think it is going through Q1/Q2? I told you its coming through Q5->Q3.

Hello Richard,

Thanks for being patient with this. I misread your original comment.

I need a bit more information to help us debug this.

richard Johnson said:

With board powered OFF, can you measure the resistance of VBAT to VSYS? - 4.27k

I still think you might have a damaged FET in your circuit. 4.27kΩ is really high here.

richard Johnson said:

Bat fet has 0V at gate pin. But Q5 gate pin has 26V, 9V across Q5 gate pin and SW2. Why is it letting this reverse current path to occur? I also read 3.9V across Q3 gate pin and SW1

Do you have an oscilloscope capture of SW1 and SW2? Does this behavior stop if you pull the CE pin HIGH?

richard Johnson said:

Can you remind me what the voltage of VBAT, VSYS and VAC are in this condition? Vbat = Vsys = Vac

Do you have a battery plugged in at this condition or is VBAT floating? Can you remind me what the voltage on VAC was and is the VAC voltage near the ACUV voltage?

Best Regards,
Ethan Galloway