BQ78350-R1: Radiated RF Immunity BQ76930 BQ78350-R1. Immunity testing requires that the EUT (equipment under test) operates satisfactorily when subject to a strong radiated electromagnetic field.

Hi Adam,

We haven't seen this specific issue before, but each system is different when it comes to EMC testing. The cell leads on many battery packs can be long which make them vulnerable, so that is why I think it is a common concern as highlighted in this app note I believe you may have seen: http://www.ti.com/lit/an/slua368/slua368.pdf

I think the 100pF caps on VCx should not be a problem and would be worth trying first. Adding a filtering cap on REGSRC may also help and should be okay.

Best regards,

Matt

Hi Adam,

Have you had any success with resolving the EMC testing issue?

Thanks,

Matt

Hi Matt,

Thank you for your interesting.

I dont have any new about RF Immunity test because we can not reproduce this here in office. I am traveling today to the city where it will be retry.

Some prototypes are prepared where the inputs of BQ76930 are filtered with 100pF cap. (VCx, REGSRC, CAP1)

I hope It will be successfully.

P.S.:

The problem is that we have to discharge the battery with a big load to set a nominal discharge current. This resistor connects to PACKs with a longer cable whitch receives the RF transients. The load can not be on the table but the transients can not be filtered next to PACKs(out of battery) with a ferrit because it is not the norm/standard. It is crazy.

The common mode transient flows from PACK to the battery and causes DSG off at ~200MHz. 

Best regards,

Adam

Thanks for the update Adam.

I hope all goes well today!

Hi Matt,

I am again in the office.:)  I have news about RF Immunity.

Our system has an STM32L031K6T6 to control other functions. On our board there is an additionally MOSFET in series with a resistor(R90+Q9) to supplies the main device.It is controlled by SDSG. DSG,CHG are controlled by STM too. (HCFET)

VDD,VDDA of STM will be noisy at RF Immunity test and causes a reset. After this reset with load had the system a mailfunction. BOR (brown out reset)was completly off. (reset if < 1.6V)

After that we have set BOR level 1 (1.8V) to do reset if the REGOUT will be noisy.

Luckily it helps for us but we dont know exactly where is the noise from. It is coupled from PACK (common mode)

If you may have any idea how can we filter it please feedback. Probably we should filter REGOUT with common mode choke.

Best regards,

Adam

Hi Adam,

I'm not sure. I talked with my colleague to see if he had ideas - he pointed out that interference on REGOUT would affect the current measurement and protection circuit which may result in SCD or OCD, but that should affect CHG as well as DSG. Interference to REGSRC would likely mess up CHG level also.

DSG probably has lower resistance from the gate to the IC vs. the CHG path. 

Best regards,

Matt

Hi Matt,

does it mean that the STM receives the noise from an input and not directly from the VDD(REGOUT) ? 

Best regads,

Adam

Hi Adam,

I think the noise is most likely through REGOUT. I'm sorry, I misunderstood your description of the issue the first time I read it and I thought that the BQ76930 was shutting off the DSG FET - we were thinking of where the noise would need to come from to cause that. 

Now I understand the DSG FET is also being controlled by the SCM which is why REGOUT becomes an issue. I'm not sure what the best way to filter this would be. Have you tried adding a smaller cap on the VDD (REGOUT) connection? From what I've read, a common mode choke should be very effective, but I don't have experience myself.

Best regards,

Matt