BQ78PL116: SAFT Smart Battery VL41Ah (3841MFE0400)

Hi Jeremy,
This is a really old legacy device and most of the folk who have a working knowledge of the device have left TI. We will ask around to see what we can find out to help
thanks
Onyx

The gauging issues are probably caused by not having the proper ChemID programmed and not having the gauge optimized. I am not familiar the ChemID 0008, but You will have to use bqWizard to upload new chemistry data to the gauge. You can also import new ChemIDs into the bqWizard tool. If you are having problems using bqWizard, then make sure that the USB-to-GPIO adapter has been updated to the proper firmware version. It can be downloaded from the PL114 website. Finally, the PL116 does not work very well with LI phosphate cells due to their low voltage profiles and the way that the cell voltages rest after charge. I hope that this helps.

Thank you both.

I suspect that the ChemID is actually OK, as the RSOC curve appears smooth and the FCC is fairly stable.

My main concern is with the AvgTimeFull as there appears to be 11 bits of the 2 byte parameter missing. Perhaps this (and potentially other reasons) could explain why the gauge has gone NRND? There doesn't appear to be an Errata document that explains this .

I'd be grateful if you could ask around and also for any recommendations on alternative large format LiFePO4 batteries.

Hi again.

Any further suggestions on this one?  Has anyone else had experience with the BQ78PL116, or alternative LiFePO4 batteries with impedance track fuel gauge and SMBus interface?

Regards,

Jeremy.

Jeremy,
I did not pay much attention to the AvgTime parameters when in worked with the PL116, so I am not sure about your issues with that. There are several people who have used the device with LFP cells with mixed performance. It usually required manually modifying the chemistry data to get decent performance. TI placed a warning on the product website to let people know not to use the device with these cells and this was one on the main reasons that the device was moved to NRND status. In short, it can be done but it is going to take a lot of effort and some support from TI.

TEC

Hi TEC,

Thank you for your message.  I'm planning to take a closer look at the SMBus to see if data is corrupted for just this one register (only the 5 least significant bits make it through), but I believe it is more likely to be a problem with the register itself, as I see the same result using three different hardware interfaces and I don't see this problem when reading other registers.

I'm not convinced that the problem relates to the use of LFP cells.  I suspect this concern relates to the lower voltage at end of discharge and applies more to applications using two cells, as described in SLUA465.  That being said, SAFT do warn that;

IMPORTANT: After the end of discharge,
the 12V Super-Phosphate battery can
remain in Under Voltage Protect Mode
for a maximum of 14 days. If re-charge
does not begin within this time, the 12V
Super Phosphate battery may be over
discharged. If over discharged, it will not
startup and will indicate a major alarm
that is not resettable. The unit will need
to be replaced.

As for the wrong chemistry file, I believe the one it has been programmed is OK, as the RSOC and RunTimeToEmpty change linearly as expected during discharge, with only minor kinks around the 80% and 20% mark.

Is an Errata note available?

How many gauge firmware versions are there and if there is more than one, what is the latest one?

Are there any development boards still kicking around in your office that I could purchase? 

Again, I'd appreciate suggestions for alternatives to the SAFT - I'm after a LiFePO4 battery with on-board impedance track fuel gauge with around 0.5 to 3kWh of storage.

Thank you,

Jeremy.

By way of an update, apparently the SAFT is no longer available as it has been discontinued. I'm still on the hunt for a LiFePO4 battery with on-board impedance track fuel gauge with around 0.5 to 3kWh of storage if anyone has any suggestions.