BQ78350: Correct usage of CEDV parameters obtained from GPC

Pietro Vignali

Part Number: BQ78350
Other Parts Discussed in Thread: GAUGEPARCAL,

Hello

For an 8S family of battery packs I have just provided data to GAUGEPARCAL and obtained a set of parameters as follow:

EMF   3802
EDVC0   519
EDVC1   0
EDVR1   1
EDVR0   439
EDVT0   1
EDVTC   11

They have been obtained evaluating an 8S2P pack, but we also need to program them in the data flash of bq78350R2 gauges for 3 differnt type of packs: 8S1P, 8S2P, 8S3P.

The parameter file obtained from GPC says:

CEDV parameters resulting from the fit. If EDVV bit is set to 1, EMF and EDVR0 have to be multiplied by the number of serial cells when written to data flash

EDVV does not exists in 78350, I think this gauge by default has all paramters scaled at cell level, is it correct ?
How do I program EMF and EDVR0 in the DF, considering that they have been calculated on a 8S2P pack and I need to program also to 8S1P and 8S3P :
- just multiply by the series cells ?
- or I have to scale also by the parallel cells, and in which case how exactly ?

Thanks in advance
Pietro

3 months ago

0 Dominik Hartl11 3 months ago

TI__Guru 52956 points

The gauge uses this on a cell level. Did you specify the number of cells = 8 in the GPC input file?

The results from GPC should be for a 1s cell with the properties of 2P (because that's what you used). You don't need to scale anything for the 8s (if you received the 1s parameters from GPC (which it looks like because EMF is 3802mV)). Please use three different GPC requests to obtain values for 1P, 2P and 3P.

0 Pietro Vignali 3 months ago

Intellectual 905 points

Hello Dominik,

yes in the GPC Config file i specified Number of cells=8.

Having said so, do you mean that these parameters already good for an 8S 2P ?

And making different GPC requests for 1P and 3P, does this means to generate the 6 discharge profile for both the 1P and the 3P ?

Pietro

0 Dominik Hartl11 3 months ago in reply to Pietro Vignali

TI__Guru 52956 points

Yes, they are good for 8S2P. I checked the FW to make sure. The bq78350 doesn't have the pack voltage EDV option so it always works on a 1s level for CEDV voltage calculations. GPC returns the results for a 1s cell. GPC doesn't know the number of parallel cells. All that really changes is a reduction in cell resistance (approximately scaled with 1/p) and an increase in capacity (FCC will scale linearly with p). I'll check with the algorithm expert at TI if you can simply scale the Rx and Cx values for p.

+1 Dominik Hartl11 3 months ago in reply to Dominik Hartl11

TI__Guru 52956 points

You only have to scale R0 for parallel cells. So if you got R0 for a 2p battery, scale R0 by 2/p.

R0_1p = R0 * 2
R0_2p = R0
R0_3p = R0 * 2 / 3

0 Pietro Vignali 3 months ago in reply to Dominik Hartl11

Intellectual 905 points

Great, thanks very much for the clarification!

Pietro

0 Pietro Vignali 3 months ago in reply to Dominik Hartl11

Intellectual 905 points

Hello Dominik

a related questions: what if a pack has set current scaling in the Specification Info 0x1A ?

Are these settings unrelated to IPScaling ?

Pietro

0 Nick Richards 3 months ago in reply to Pietro Vignali

TI__Genius 17181 points

Hey Pietro,

Dominik is currently out of office, but he will get back to you when he returns.

Regards,

Nick Richards

+1 Dominik Hartl11 3 months ago in reply to Pietro Vignali

TI__Guru 52956 points

You don't have to compensate for IPSCALE in the CEDV parameters.

Power management

Power management forum

BQ78350: Correct usage of CEDV parameters obtained from GPC