BQ78350-R1: Wrong SOH after one discharge cycle

SOH = 100% * FCC/DesignCapacity.

If your FCC drops to 81% of DesignCapacity after one discharge, then your charge termination settings and EDV setting utilize only 81% of DesignCapacity (which is not unreasonable). The problem is with the definition of SOH for this gauge. It reflects the ratio of FCC (=how much charge you can get out of your cell) and DesignCapacity (the rated capacity of your cell).

If SOC reaches 100% well before the charger stops (and you didn't just start gauging after a reset, which means that the gauge took a first order guess on SOC based on voltage measurement) then FCC was calculated too low.

This can have many causes.

For example, FCC was measured for a very high load (which significantly diminished FCC) but you continued discharging past EDV2 with a low load.
Or FCC was measured for a significantly lower temperature.
Or FCC wasn't measured in a long time (or at all). Or you started charging right after resetting the gauge where the gauge has both inaccurate FCC and inaccurate RM.
Or you use compensation for end of discharge voltages based on load and temperature and you didn't follow the instructions from GPCCEDV exactly and programmed incorrect CEDV coefficients.

SOH will change if FCC changes. It generally doesn't increase from 81% to 100% while charging.

Thank you for your answer.

Is a qualified discharge needed to update the FCC?

What happens if a high load current is used till the batteries are fully drained? That means the following conditions for a qualified discharge never met:

* The battery voltage reaches the EDV2 threshold during the discharge period and the voltage is greater
than or equal to the EDV2 threshold minus 256 mV when the BQ78350-R1 detected EDV2.

* –Current() remains < Overload Current when EDV2 is reached.

depends the NEAR FULL to IPSCALE?

How can we update the FCC to the real value?

Yes, a qualified discharge is mandatory for an FCC update. From the TRM 9.1.5:

"The BQ78350-R1 updates FCC with an amount based on the value in DCR if a qualified discharge occurs." minus Reserve Capacity.

And 9.1.6;

"A qualified discharge occurs if the battery discharges from RC ≥ FCC – Near Full to the EDV2 voltage threshold"

EDV2, if CEDV is enabled, will change. See TRM 9.1.7:

"If the [EDV_CMP] bit in CEDV Gauging Configuration is set, automatic EDV compensation is enabled and the BQ78350-R1 computes the EDV0, EDV1, and EDV2 thresholds based on values stored in CEDV Cfg subclass of data flash and the battery's current discharge rate and temperature."

Therefore the EDV2 threshold should be met.

The gauge won't update FCC if the current exceeds the Overload Current (which you can change in Fuel Gauging->CEDV Cfg->Overload Current.

The gauge won't update FCC if the current exceeds the Overload Current (which you can change in Fuel Gauging->CEDV Cfg->Overload Current.:
Does this mean we have to move the threshold to a higher value? 4A is set now, but we discharge with about 20A till EDV2. Is it a good idea to set the value to 20A?

scales NEAR FULL with IPSCALE?

Hello,

Today's a national holiday. We will get back to you tomorrow.

Regards,

TI Apps Team

The gauge won't update FCC if the current exceeds the Overload Current (which you can change in Fuel Gauging->CEDV Cfg->Overload Current.:
Does this mean we have to move the threshold to a higher value? 4A is set now, but we discharge with about 20A till EDV2. Is it a good idea to set the value to 20A?

scales NEAR FULL with IPSCALE?

what happens if I set FC_for_VDQ to 1? I couldnt find any explanation.

The SOH is jumping low while discharging, then jumping the SOC low. I thought, the SOC should only jump for FCC learn Down.

The overload current is a safe-guard preventing the gauge from updating FCC if the load is too high. This is about excessive loads, not typical loads. The EDV compensation loses accuracy if the load gets too high (rule of thumb, 10C).

FC_for_VDQ = 1 requires that the gauge detects a valid full charge termination to qualify a discharge for FCC updates. Otherwise, RM has to be within "Near Full" mAh of FCC.

Near Full is in mAh. You have to set it to use the same scale as applied to FCC.

SOC will always be forced to BatLow% at EDV2. If FCC was correct for this discharge scenario (load, temperature), the jump will be minimal.

SOH only changes if FCC changes.

Thank you for your reply.

We checked again our discharge behaviour. The CMP is off. The Designed capacity is set to 12Ah. We drained about 16Ah from the battery. The SOH is despite the drained capacity is higher the designed capacity jumping down to 75%. EDV2 is fixed set to 3,15V and 4%. How can this happen? I expected, that SOH stays at 100%.

What is FCC and coulomb count during this discharge? Do you have a register log file that shows how much charge the gauge accumulated during the discharge?

The log file shows that the gauge held SOC for a while. I can't see what SOC that was but assume it was the BatLow% default 7% and it held it until voltage dropped below EDV2.

Why did you set Design Capacity to 12Ah?

The gauge simply divides FCC by DesignCapacity for SOH so in order to know what happened, I'd have to see what FCC was and what FCC became when SOH jumped.