This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ25895EVM-664: Can't draw 3.1A from PMID without causing boost fault when powered by battery

Part Number: BQ25895EVM-664
Other Parts Discussed in Thread: BQSTUDIO, BQ25895, EV2400, PMP4451, PMP11536


My application requires 3.1A (or close to it) at 5V and I'm hoping to draw it from the boost converter built in to the BQ25895.  I have a BQ25895EVM, EV2400, and bqStudio.  Here are the test conditions:

*  Power supplied from battery only (fully-charged 18650 Li-Ion pack, 4.1V, 4 x 3,000 mAh balanced cells in parallel, 12,000 mAh total, each cell capable of high discharge rate ~ 20A)

*  Battery pack protected with BQ29700 EVM.  Output of BQ29700 EVM attached to BAT and GND of BQ25895EVM.

*  BQ25895EVM jumpers all in factory default settings

*  OTG mode enabled

*  Boost voltage set to 4.998V

*  All other registers in system startup defaults

I'm measuring the draw from the battery, and up to 1.5A everything works fine.  At 1.6A and above, the BQ25895 EVM throws a boost fault and stops providing boost power.  I have also seen the BQ25895 EVM enter a hiccup state.

If I attach external power using the USB input, the system continues to work fine past 1.6A.  The failure only occurs when power is drawn exclusively from the battery.

So my question is, can I draw 3.1A from PMID in OTG / boost mode when batteries are the exclusive source of power?  Do you have any ideas why the BQ25895 EVM doesn't seem to be able to provide that much without faulting?  What other information might you need from me (registers, logs, etc.) to help make suggestions?

Thanks very much - I really appreciate your help!


  • Graham,

    The device may support 3.1-A boost operation when necessary conditions are met.

    Are you using resistive load at PMID pin as OTG load? Have you tried to connect the battery to BAT pin of the BQ25895 directly?



  • Hi Ning,

    Thanks for your reply.  Yes, the load at PMID is resistive.  At your suggestion, I tried connecting the battery pack to the BAT pin of the BQ25895 directly.  This did not fix the problem.  I also tried lowering the minimum system voltage to 3V (from 3.5V previously) in case the BQ25895 was prioritizing a different power path, but that didn't help either.

    Maybe you could help me understand the problem a different way.  Some of TI's reference designs that include the BQ25895 (like PMP4451 and PMP11536) use a separate boost converter (TPS61235) instead of using the built-in boost converter in the BQ25895 (PMID not connected) even though these designs appear to only require 3A.  Can you briefly explain why you might decide to use an external boost converter instead of the one built-in to the BQ25895?  If that's the recommended or preferred approach, I'm happy to use the external boost converter, but didn't want to have unnecessary parts if the built-in converter is supposed to be sufficient.

    One thing that may be important to note - I need 5V @ 3A continuously (without interruption).  So if the boost converter in the BQ25895 is intended more for battery charging or something similar (where a brief dip isn't a big deal), maybe that's a good reason to use an external boost converter?

    Thanks again for your help - I really appreciate it!


  • Graham,

    Did you go through the EVM User's guide step-by-step? Especially 2.3.3 Boost Mode Verification. Just use a lab power supply as "battery" connected to the BAT and GND pins, put an adjustable resistor box from the PMID pin to GND. Starting from very light load. Apply VBAT=4.5V, then increase the PMID load slowly. Please record all the register values during the test. Please also watch VBAT pin voltage and make sure VBAT pin voltage stays above 3V all the time. You should be able to get 3A from PMID.




  • Hi Ning,

    I didn't see this post until just now, my apologies for the delay in responding and for another question I asked (which you also answered) that covers some of the same information that you provided here.

    I'll run the test you proposed and reply with my results.