• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 778 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Inquiry on bq25504 with VSTOR < VBAT

JP Norair
Intellectual 630 points
Other Parts Discussed in Thread: BQ25504, BQ25505

Hello: I'm trying to figure out the cheapest and lowest-component-count way of working with the bq25504 on a system where the load voltage must be <3.6V, but the battery should be 4.2V.  The load voltage does not need to be especially stable, it just needs to be 2V < Vload < 3.6V.  For the record, 1.5uA < Iload < 20mA.

I would love to hear if anyone has experimented with this.  I'm going to post my hypotheses below, and if you have comments, please post!

I suspect one way to do it is just to put a diode between VSTOR and the external load, such that there is a forward drop enough to bring voltage below 3.6V.  In this case, I am *hoping* that at loads below the diode Vf, 4.2V isn't going to hurt any of the components.

It seems that bq25504 switches between VSTOR and VBAT as the source of Vdd (as it needs to charge each of these), so putting a diode between VSTOR and CSTOR would not seem to work.  This would be preferable to the first solution if it did work, since it would provide better limiting on the voltage of CSTOR.  It may be possible to use two diodes in reverse directions, though.

In any case, if TI is listening, a lot of batteries like 4.2V charge voltages, and an increasing number of ultra-low-voltage digital parts do not like Vdd higher than 3.6V.  If you make an update to the bq25504 (maybe bq25504b or whatnot), it would be much appreciated if Vstor could be set less than Vbat.  It seems to me that this would be a relatively simple change to the boost converter logic.

  • 0
    TI__Guru**** 166626 points

    The soon to release bq25505 has an integrated buck converter to do exactly what you want.  It will soon be posted on www.ti.com.

    Regarding your idea about the diode, you might want two diodes otherwise you might not have enough drop at light loads.

    There is already a FET with body diode between VBAT and VSTOR.

  • 0 in reply to Jeff F
    Intellectual 630 points

    bq25505 certainly sounds interesting, but I bet it has a much bigger component count with the buck regulator -- maybe not, I'll be interested to see.

    I noticed the FET with diode between VSTOR and VBAT, but based on the information I can get on the bq25504, there doesn't seem to be a way to have VSTOR < VBAT without additional components.  Simply put, there isn't a lot of information on how this PFET between VSTOR and VBAT is utilized -- modulating this internally would be sufficient to have systems where VSTOR < VBAT.

    However, I had an idea after I posted the message.  I can set the VBAT_OK voltage range to something tight like 3.3-3.6V, with OV = 4.2 and UV = 2.2 as needed by my thin-film battery.  Then, I can put an external NFET+diode across VSTOR-CSTOR, with VBAT_OK controlling the gate.  This should prevent the voltage in CSTOR from ever getting above 3.6V, because no current can be applied to the CSTOR above 3.6V.

    This or something like it *should* do the job -- I'll be experimenting soon.  The uncertain element is the internal PFET between VBAT and VSTOR, for which the turn-on rules are not evident in supplied materials.  Here is a schematic of the business-end of the circuit I'm considering.  The rest is bog-normal application circuit for bq25504.  If it works, it will be vastly cheaper and smaller than the MAX17710 solution is, and basically just as good.  I expect that I can find a way to put this NFET somewhere to make it work.

    [Edit, revised circuit]

  • 0
    Intellectual 630 points

    OK, after more research on the bq25504 logic, I realized the previous design would not work because VSTOR would be tied to VBAT during most of the usage, and CSTOR would get discharged by the load without the bq25504 realizing it.

    However, I've come up with another hack.

    In this circuit, the VBAT_OK signal will go high when the VSTOR voltage rises to 3.5V, and it will drop when VSTOR sinks to 3.2V.  Overvoltage is set to 4.2, and Undervoltage to 2.0V -- this is consistent with the weird type of battery I'm using.

    The trick is to use VBAT_OK to make the Undervoltage circuitry divert current to VBAT (and not VSTOR) in all cases where the VSTOR voltage would otherwise be above 3.5V.  The charging logic in the bq25504 seems to allow Overvoltage and undervoltage at the same time -- and they are directly attached to VBAT rather than VSTOR -- so this should work.  If it does indeed work, it is a hack I'm proud of.  :)  I'm having the protoboards made now -- I'll follow up with results once I get the boards and test them.

  • 0 in reply to Jeff F
    Intellectual 630 points

    Is there any news on the bq25505?  I'm able to use the bq25504 as described above, although I wouldn't mind having-back the VBAT_OK pin to use for its normal purpose.  Thanks.

  • 0 in reply to JP Norair
    TI__Guru**** 166626 points

    RTM planned for end of April.