BQ24610: Starting a high current battery charger design, is this the right part

Hi Jefff,

Thanks for the quick reply... I see that I had a typo but you caught it anyway (The pack is a 4S12P configuration). I am not sure if I can "pump" more than 10A into the batteries during charging so that might be fine but I will look at newer FETs as to reduce the possible heat generation using the older FETs. The 20A requirement is not for charging the battery but driving the load while charging the battery. I was thinking of using a discreet switching circuit that would disconnect the battery path to the load and connect the load to the external power supply, when the circuit detected an external power source. Would like to hear your thoughts on that. And finally when I have completed the design, does TI offer just a quick once over on the circuit to make sure I am in the ballpart.

Thanks,

Mike

Hi Jeff,

Here is a copy of my schematic... I did not compute the values for Iset1, Iset2 and ACSet because I wasn't too sure of the equation, though I listed the operating parameters of the circuit on the schematic. Please let me if there is anything else you need from me to complete your review.

4087.schematic.pdfThanks,

Mike

Hello Jeff,

Sorry for the long pause but we finally received our custom boards back and unfortunately their are not working as planned... Perhaps you can help begin to understand the issue. Here are the following steps

1. We brought the board up in a similar fashion as described in the TI app note sluu396a. We were successful in step 3.41 - AC Adapter detection threshold, where all the listed measurement points read correctly. When we increased the input voltage to 16.V, we noticed that VBAT was only reading 0.741 V. Below are some of the voltage values that might give insight into what might happening.

ACDRV: ~9.9V

REGN: 5.99V

VSYS: 16.4V

VREF: 3.29V

Any and all insight would be helpful once again.

Thanks,

Hi Mike,

Can you send the final schematic with resistors values filled in?  What were the voltages at ACSET, ISET1 and ISET2?

Regards,

Jeff

Hello Jeff,

With a VIN set at 18.6V and TS set at 1.8V, I measure the following ISET1 = 0.987V, ISET2 = 0.304V and ACSET = 0.988V. Also attached is our final schematic 

Regards,

Mike 

TheRunner_v5_pg2.pdf

Hi Mike,

I don't see anything wrong with the schematic or the DC values.  What is STAT doing?  Is the converter switching, i.e. is there a squarish waveform on the PH node like anything in the plot below:

I am assuming you don't have a battery attached so I would expect the following:

Regards,

Jeff

Hi Jeff,

I do not have a battery attached at the moment but I have CE hard-wired to Vref, so when I look at the circuit the waveforms (PH,LODRV) look more like Figure 2 (Charge enable) from the data sheet.  Could this be the reason why I don't see the representative voltage on VBatt (because I don't have a battery connected at this time). Also STAT1 and STAT2 are both low (battery not detected)

Hi Mike,

If battery not attached, you should see a waveform like Figure 18 above, not a low voltage.

Regards,

Jeff

So that waveform is VBatt, I wasn't sure what it was

Hi Mike,

Yes, if no other issues with the charger and the capacitance in parallel with missing battery isn't too large, then the output should cycle as shown above.

Regards,

Jeff

Hi Jeff,

I really appreciate you hanging in there with me while I try to figure this thing out. I took a snapshot of the power up sequence with probes on Vref, VBatt,PH and LODRV. A couple of observations 

1. The signal amplitude  of PH is 6.2V, while LODRV is 8.8V, VReg is 3.3V and VBatt is 1.34V, not sure what the others should be but VBatt is definitely lower than expected

2. It seems that the startup waveforms are correct (though the amplitude is not right

3. The capacitance across the battery terminals is 100uF, which I think is acceptable.

I hope this gives some insight into what could possibly wrong.

Thanks,

Mike

IMG_1541.pdf

Hi Mike,

Can you check HiDRV?  It should be same as PH except with a dc offset = REGN.

Regards,

Jeff

Here is a snapshot of PH(Magenta),HIDRV(Cyan) and REGN(Yellow)...As you can see the offset from PH to HIDRV is only 1.4V where REGN is 6V. IMG_1543.pdf

Hi Mike,

I think something is wrong with the BTST capacitor.  Can you remove and replace?

Regards,

Jeff

I am assuming you are referring to the capacitor between BTST and PH... I changed it and no change in the output but do you think it could be the Diode between BTST and REGN

Hi Mike,

Yes, it could be the diode.  

Regards,

Jeff

Jeff,

So that I know what I am looking for... you said under normal conditions, HIDRV should be offset from PH by a DC offset equal to REGN and that the capacitor and possibly the diode could be affecting it. Can you share some more insight into what could possibly be suppressing the voltage levels. I know that a big open ended question but I think I am missing something fundamental to my design.

Thanks,

Mike

Hi Mike,

BTST cap, with low side at REGN, has high side of REGN + PH so that it can turn on the high side NFET, in other words, the NFET gate should be REGN volts higher than the input voltage = top of the PH pulse.

Regards,

Jeff

Jeff,

Thanks for the explanation, that being said I am honestly at a lost on how to pursue next. Another datapoint is we powered up a second board and initially it demonstrated that same behavior as the first board till we inserted a battery and the board has taken a step backwards. The voltage on REGN is now 10.8V with a Vin of 16.5V (Go figure) Don't know if this helps or am I just confusing the issue.

I agree so I placed the order today for the EVM, it will get here in a couple of days so till then, I probably won't bother you. Thanks for all your help so far.

Regards,

Mike

Ok.  Keep me posted.  Good luck.

Jeff

Hi Jeff,

I was able to get the circuit operational (finally)..it ended up being the cap coming off BTST and PH, replaced it and now its working (somewhat). Which leads me to my next question. In my application, the AV input is a 18V30A power supply and the battery pack is 3S10P module. I would like to pump 8-10A into the battery during rapid charging and the system load will vary from 5 to 10A. so it seem to me that I need to disable the chip's DPM feature but I still want to keep the PowerPath control, is my assumption correct

Hi Mike,

Glad to hear it.  The IINDPM feature only limits the output current of the converter, i.e. limits the charge current.  So, you could disable the IINDPM by shorting the RAC resistor or you could increase it to 10A + 3*4.2V*10A/18V/efficiency.  The power path feature that turns on the BATFET when input power is removed works the same either way.

Regards,

Jeff

Hi Jeff,

So I am putting our custom board through its paces and a couple of observations/questions.

1. The final battery configuration is a 3P12S and we have updated the resistors accordingly.

2. When the unit is only charging the battery via the external power supply, the circuit performs as expected. The external supply's current ranges from ~8A down to less than 1 amp during the charge cycle. 

3. When the unit is in full operation, the unit requires ~15A - 20A, what we have observed is that the battery is supplying all the current necessary and the external supply is not supplying anything. My expectations is that the external supply would provide all the current necessary and use any remaining current to charge the battery. Is that a wrong assumption.

Regards,

Mike 

Hi Mike,

The battery should be at least a diode drop below the system = PWR_IN - I(SYS)*Rdson(ACFET).  As long as the VCC voltage is greater than the SRN voltage, BQ2461x enables the ACFET and disables BATFET.  So, is the VCC voltage dropping below the battery voltage and BATFET is turning on? 

Regards,

Jeff  

Jeff,

So PWR_IN = 18V, VCC = 17.4V, VBatt = 11.2V, Vsys = 10.9V and  BATTDRV = 5.99V

Hi Mike,

I don't understand how BATTDRV is low, meaning BATFET is on but the VCC>VBATT at SRN.  Any chance you can pull up STATx and PG so we can get more info on the exact fault?

Regards,

Jeff