MC33771 ISense configuration for the RShunt resistor

[m_karthikeyan]

Hi NXP,

      We are using a S32K144 with a MC33771 in a BMS[Battery Management System] solution, with the Battery Cell Controller Software Driver v1.1 provided from NXP. and we are using the following macro in particular from the driver to calculate the current.

/*!
* @brief This macro calculates ISENSE value in [mA]. Resolution is
* (600/R_SHUNT) mA/LSB (V2Res / Rshunt = 1000 * 0.6 uV / rShunt uOhm).
*
* @param rShunt Resistance of Shunt resistor in [uOhm].
* @param iSense1 Content of register MEAS_ISENSE1.
* @param iSense2 Content of register MEAS_ISENSE2.
* @return ISENSE current in [mA].
*/
#define BCC_GET_ISENSE_AMP(rShunt, iSense1, iSense2) ( \
(BCC_GET_ISENSE_RAW_SIGN(BCC_GET_ISENSE_RAW(iSense1, iSense2)) * 600) / (int32_t)(rShunt)

The Actual used part is a pair of 8 SMD component in parallel with each having a resistance of 1mOhm or 1000uOhm. but we are getting a higher value of current when converting to mA. but if we use the value of 2mOhm we are getting the current accurate to a mA while measuring with a multi meter. Why is there a Gain multiplication of 600 in the Macro and is there any possibility of production resistance, that might be added while using a SMD Does the process add extra resistance, is there a way to compensate in the driver/part.  Or Is there any user recommendation  to try, or steps while calibrating the ISense values. when using MC33771.

Thanks,

Regards,

Karthikeyan Mitran 

[JozefKozon]

Hi m_karthikeyan,

please see below for a reply from an application engineer I have contacted.

DESCRIPTION

The input voltage range for ISense is +/-150 mV (and not +/-300 mV). The result is stored on two registers for a total of 19 bits as a signed two's complement. This matches with the V2RES resolution of 0.6 µV/LSB.

Typically, high precision shunt will have 2 power outputs and 2 sense outputs. In your case, since you seem to be using SMD components in parallel, the position of the sensing line on the PCB traces will have a strong influence on the resistance value seen at the IC. Most likely, the resistance will be higher due to the PCB parasitic resistance.

You can check the actual voltage at the pins of the IC (between Isense+ and Isense-) with an accurate voltage meter, and make sure that the IC is delivering the right voltage measurement.

Bottom line, we don't recommend the usage of shunt resistance in parallel, this will lead to too much manufacturing variation. We recommend to use a single shunt with dedicated sense terminals or where the PCB layout allows an accurate sense position. For example WAF-M-R001 from IsabellenHuette in the 1mOhm range or BAS-M-R0001 in the 100µOhm range.

With Best Regards,

Jozef

[m_karthikeyan]

        Thanks a lot for the reply, I measured the voltage across a multi-meter and it gave 0.2mV for a current of 0.998A and the resistors are all measuring the same 0.2mV and the ISense tap was from the closest resistor. The resistors should give 125uOhm But measures close to 200uOhm. Is there any way to cross check the same with the coulomb counter to re-calibrate the resistors or the measure the current better? Thanks a lot for your efforts.

Thanks,

Regards,

Kathikeyan Mitran

[JozefKozon]

Hi Kathikeyan Mitran,

your measurements are correct. If you calculate the resistance of the resistors including the parasitic resistance of the PCB traces, it gives out about 200uOhm. R=V/I=0.2mV/0.998A=0.0002/0.998=0.0002004Ohm. 

If you want to achieve the 125uOhm you can decrease the resistors value slightly, measure the voltage on the Isense+ and Isense- and calculate the resistance again. 

Other way would be to use only one shunt resistance and put it as close as possible to the IC. As recommended the application engineer.

Please consider the options. Which is more suitable for you.

With Best Regards,

Jozef