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.

allowed design energy on Bq34z100EVM too low

Other Parts Discussed in Thread: BQ34Z100, BQ34Z100EVM

I am evaluating the bq34z100 using the bq34z100EVM and accompanying software and I find that the allowed design energy is very low.  The max allowed design capacity allowed by the software is 32767mAH and this is fine for my application since we have a 28 AH battery pack (14s10p configuration).  Our voltage is 48 V nominal and 59 V max.  This means our expected design energy is at least 48V * 28 AH = 1344 WH.  However the allowed range for the design energy is only 32767 mWH (i.e. same a as for design capacity).  Even if I assume x10 mode the design energy is far short the needed 1344 WH.

Is there a work-around for this? Can we ignore the design energy setting or will this affect the internal algorithms? I am still familiarizing myself with the functionality of the bq34z100 and want to make sure I don't get stuck later trying to make it work with a 14 serial -10 parallel pack. 

 

  • It appears that you need to do two things: (1) set the Design Energy Scale to 10, and (2) rescale the current.

    Setting the Design Energy Scale to 10 changes the mW and mWh parameters to cW and cWh for both dataflash settings and reported measurements. This affects the algorithm, so values of power and energy should be entered as such. As you have noted, by only making such a change, the max energy is 327 wH, not enough for your high power application.

    There is a useful property of this fuel gauge that allows you to rescale the current  by calibration. For example, you can use the "X10" idea, which doesn't change anything in the firmware, but just acts as a flag to remind the Host that the current, coulomb, power, and energy units should be interpreted with units ten times larger. Actually you can use any scale you want, but 10 is convenient. The multiplier you choose affects the dynamic range of the gauge. It is often driven by the amount of voltage drop or heat rise you can tolerate in your sense resistor.

    In your case, it looks like you will want to rescale the current by at least 5X to achieve your required 1344 Wh. This is done by simply scaling the calibration of the sense resistor by a factor of 5X. For example, if you calibrate with 5A, you would tell the calibration routine that you are applying 25A.

    Its important to realize what the actual reported values will become. Rescaling the current affects units of all current, coulomb, power, and energy reports. Setting the Design Energy Scale to 10 only affects the power and energy reports. So the effect is that power, enerrgy, current, and coulomb reports will be scaled up 5X. Make sure to go through all the dataflash settings and adjust down by 5X.

  •  

    Sounds like the rescaling option above may work but I was wondering if there is any way to simply not use the design energy to avoid the need for rescaling all the computations.

    Since we can provide the correct mAH setting for the design capacity is there anyway to simply ignore the design energy or will it affect the functionality of the gas gauge?  If we set the design energy to its maximum allowed setting without using 10X mode (32.7 wH) and set the capacity to 28 AH,  which setting will the fuel gauge chip use to determine the remaining capacity?

    Do you think that the limitation of the design energy should be corrected in future versions of the chip?  Since the chip allows setting AH to 32.7 and allows voltages up to 65 V shouldln't it be assumed that that number of what hours would have to be much greater than 32.7 wH?

     

     

    Prestor

  • Yes, as long as you set Load Mode = 0, then you can ignore power and energy values. However, if you have a constant power load, then gauging might be a little bit less accurate. Yes, expanding Design Energy Scale beyond 10X is a good idea for future enhancement - Thanks!