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.

Calibration Accuracy

Other Parts Discussed in Thread: BQ20Z65

Hi,

I'm doing voltage and current calibration remotely using a power supply.
I'm referring to SLUA379E for the automation.

The only question is the accuracy of the voltage and the current during the calibration.
In the SLUA409A it is written that the the voltage and the current must be +-1mV or +-1mA.

But in commercial power supply products the power supplies have bad accuracy compared to the desired one.

For example, this one http://www.tek.com/datasheet/pws4000-dc-power-supply-programmable-dc-power-supplies
is ±0.02% + 3 mV for the voltage and ±0.05% + 2 mA for the current. If I set 12V and 2Amp for the calibration I obtain +- 5.4mV and +- 3mA !

Is this accuracy still OK, because I don't find any power supplies that fulfill the requirements.

Thanks for your answer,

Alex

  • Hi Alex,

    Just make sure that the multimeter that you use for voltage and current measurement during calibration  has +-1mV and +-1mA accuracy and you should be ok.

    thanks

    Onyx

  • Hi Onyx,

    Thanks for your answer, I will search for very accurate multimeter then. Do you have tips for me about some good and accurate multimeters ?

    Thanks

    Alex

  • Hi Alex,

    Fluke makes some pretty good multimeters.

    thanks

    Onyx

  • Hi Onyx,

    I use the bq20z65 chip and the sense resistor used is 5mOhm.
    I have seen in the SLUA529 that the sense resistor must have a temperature coefficient no greater than 75 ppm in order to minimize current measurement drift with temperature. Is that 75ppm/°C or 75ppm/range of temperature ?

    Thanks

    Alex

  • Hi Alex,

    It is 75ppm/oC

    thanks

    Onyx

  • Hi Onyx, thanks for all your very valuable answers,

    I'm using a bq20z65 and a 5mOhm sense resistor and I'm wondering about the necessity of the calibration. Is it compulsory to calibrate each new PCB if we use the sense resistor of 5mOhm 1% tol. ? In SLUA391, the flow diagram presents the step of calibration for each product. But is that not possible to flash each PCB with the calibration data of a 'golden sample'? If we can assure that the sense resistor is 1% tolerance, is the calibration still obligatory ? (I mean the current, voltage and temperature calibration and the board offset calibration)

    If we don't do any calibration again on each pcb, what is the effect on the state of charge ?

    Thanks,

    Alex

  • Hi Alex,

    Calibrating every board is recommended to avoid measurement accuracies that could be induced due to variations in manufacturing process, board offsets etc. If you must avoid calibration, then you will have to take about ten sample boards, calibrate each of them, average the calibration parameters and use the averages in your golden file. Only use this method for  calibration if calibrating each board is inevitable.

    If there is no calibration, one of the effects is that the cc deadband could be off and that could make the part see a non existent current when the device is in relaxation mode. This will could cause RSOC measurements to be off.

    thanks
    Onyx

  • Hi Onyx,

    Thanks for your precise answers, they are very helpful.

    I'm using bq20z65 with a 3-cells LiPo battery (4500mAh).
    I use the two external thermistors of the bq20z65 but I don't know where to place them.

    I hesitate between putting them in-between the cells, or paste them with a thermal adhesive paste on the little metal blades of the cells, or just putting the thermistors on the PCB where the little metal blades are welded (but the temperature of the components of the PCB can wrongly influence).

    What is the best solution ?

    Best regards,

    Alex

  • Hello Alex,

    The main purpose of the thermistors is to measure the temperatures of the cells so the protection features of the IC can kick in in the event of an over temperature condition thus protecting your cells as well as take into consideration the temperature effects on state of charge of your cells . Li-ion tend to be volatile at high temperatures so it is necessary to protect your circuit to prevent a fire. That being said, it is preferable to place the thermistors as close as possible to your cells. In between the cells in my opinion is best.

    thanks

    Onyx