Thermistor sharing BQ24251

Hello Martin,

I would recommend placing a reverse conduction diode from the LDO to the thermistor resistor divider. This will eliminate the reverse conduction and any issues that it would create. This would diode OR the bias voltage much like what you are doing to protect the rail that you are pulling the bias from.

However one thing that I would be concerned about is that there is going to be a non-negligable amount of variation in the biasing voltage with the LDO output 4.9V and the VTT being closer to 4V or possibly less a large amount of error will be introduced between the two bias sources. This variation in the bias voltage along with potentially incongruent sensing thresholds could create a recipe for havoc in your system. However after looking at the Maxim gauge it looks like you can work around this bias variation by changing the temp offset as discussed on Pg. 39 as the bias voltage changes.

The sensing thresholds for the BQ24251 are clearly provided on Pg. 8 relative to the voltage from the LDO:

The safety thresholds threshold for the Maxim part are user programmable as discussed on Pg. 32.

Out of curiosity have you looked at any of the TI gauges such as the BQ27425-G2A or BQ27545-G1?

Dear Ryan,

thanks for your answer! Actually, there is no problem with the reverse protection diode for MAX17047, as MAX17047 eliminates diode drop by measuring ratio of AIN to THRM producing temperature output independent of bias voltage magnitude.

However, I guess, placing reverse protection diode in LDO rail will significantly move all the temperature thresholds of BQ24251 and I think there is no way to compensate it, so NTC monitor of BQ24251 will be unusable.

We like MAX17047's low power consumption and ease of use, what we don't like about it is the need of characterization of battery pack during development and missing non volatile memory.

However, we haven't found proper TI fuel gauge able to gauge 16 000+ mAh liion cell. Even BQ27425-G2A specifies max 4000 mAh cell. Is that correct value? Because BQ27425-G2A contains some capacity registers with 32 000 max value (unit is mAh). Otherwise BQ27425-G2A would probably fit our needs if we can live with higher power consumption.

Hello Martin,

Regarding your statement about the shift in the TS thresholds. The thresholds with the device stay a function of the LDO voltage but you would just need to adjust the voltage divider calculations to match those threshold voltages at the proper temperatures using the slightly lower post diode rail. With that in mind I do not see that being a problem and would allow you to use both the charger and gauge NTC monitors.

Regarding the gauge capacity did you have a chance to look at the BQ27545-G1? This gauge is entirely capable of supporting up to 32000mAh capacities (See the datasheet for accurate capacity numbers).

Regarding the current consumption, the bq27545-G1 has some functional sleep modes that will help reduce the power as well.

Let me know if you have any questions and I will do my best to answer them but for deeper questions about any gauging products I would recommend posting in the dedicated gauging forum here:

http://e2e.ti.com/support/power_management/battery_management/f/180.aspx

Hello,

thanks. I'm trying to select proper resistor values that will work both for fuel gauge and charger. I found out that the parallel resistor to the thermistor (R3 on page 29 in BQ24251 datasheet) mustn't be used - fuel gauge isn't able to compensate its influence.

What is the beta and R0 of termistor that doesn't need R3 resistor? I tried beta = 3500 and R0 = 10k along with R2 = 10000 but obviously I'm not able to get proper TS thresholds for all temperatures.

Martin

Hello Martin,

Have you tried using the TS calaculator tool available on the EVM page and in the link below?

http://www.ti.com/lit/zip/sluc436

I have started to play with it placing an arbitrarily large value for R3 in H4. I would recommend looking at the suggested R2 value from the gauge and using this tool to compare some of the thermistors that you are considering.

Thanks for that calculator. However, the Solve button doesn't work for me in LibreOffice and Excel 2010 (macros enabled).

I'm able to get reasonable small error for HOT and COLD thresholds with R0 = 10000, beta = 1800, R2 = 12000 (and R3 = infinity).

Is there an error in datasheet? You write there that R3 is not neccessary when beta near 3500 (page 29) which I think is not truth.

So the problem now is that NTC's with beta = 2000 are not available. The conclusion might be that thermistor sharing between charger and fuel gauge is not possible.