Bench testing on a cold Australian night

[mcart117]

I roll my eyes a bit when people warn of the dangers of charging LiFePO4 cells below 1º c because I live in Australia and I am more worried about keeping the cells cool when the ambient temperature goes over 40º. So I set up a bench test with the charging cycle to run over night, and before turning in, had a quick check of the temperature and forecast for the night.



The test bench is on an enclosed verandah, but the roof is tin, so it gets chilly. Although I thought it was unlikely the cell itself would go below 1º, I stopped the test, just in case. I wonder would the regulars here have let the test run, or would they also have stopped it?

The following morning there was a frost outside, but a glass of water on the bench showed no signs freezing. So I might have got away with it, but sometimes it's better to be safe than sorry.

 

[wattmatters]

You did the right thing.

Charging at too low of a temp will cause damage to an otherwise perfectly good (and expensive) battery. Normally overnight a battery is being discharged rather than charged, and there is more tolerance for low temps when discharging. The time it can catch people out is on cold sunny mornings when a PV array + charge controller starts soaking up the rays.

If a battery is going live outside then it really should have a BMS with temperature sensors and low and high temp protection settings. Despite us being in Oz, it can still get bloody cold in places, especially anywhere away from the coastal areas. Lowest temp on record in Australia is -23 °C (-9.4 °F). Inland cities and towns will regularly go below zero °C in Winter. The average overnight low in Canberra in July is 0 °C.

 

[mcart117]

If a battery is going live outside then it really should have a BMS with temperature sensors and low and high temp protection settings. Despite us being in Oz, it can still get bloody cold in places, especially anywhere away from the coastal areas. Lowest temp on record in Australia is -23 °C (-9.4 °F). Inland cities and towns will regularly go below zero °C in Winter. The average overnight low in Canberra in July is 0 °C.

Thanks @wattmatters. The battery will be in a shed, and the (Overkill) BMS will have a temperature sensor. But that is not in place while I am running capacity tests on individual cells.

I am running the tests with the ZKE Tech 5 amp tester, so even the charge phase on a 304Ah cell takes days. Twice now I have stopped this running overnight. At the back of my mind I am thinking that's a recharge cycle being used up every time I do it, but better to lose a few recharge cycles than the whole cell.

 

[wattmatters]

The battery will be in a shed

I guess the shed has no power to run the test in there. Hence the battery build.

I am running the tests with the ZKE Tech 5 amp tester,

How many cells you testing? I think I'd just do a random sample.

 

[mcart117]

How many cells you testing? I think I'd just do a random sample.

OK so this is all a learning process for me and there is one post here which wasn't even made when I ordered my first batch of 4 cells, which I am testing now. The post links to a youtube video by Sun Fun Kits LLC. The video talks a lot about grading and one of the things he said we should all ask for is the original test data spreadsheet from EVE:



He even said it should have the original colours. I think that's going a bit far. When I was ordering a second batch of 8 cells from Amy Wan at Shenzhen Luyuan, she sent me a CSV file:



It doesn't have the colours, but the columns seem to match up. The right hand column was added by me, and shows the manufacturing date implied by the cell code. Note how all the capacities on both spreadsheets exceed 316 Ah. So this all looks good, and I might randomly select just one of the Shenzhen Luyuan cells for testing when they arrive.

But I did not get this data from Shengen Basen before making my original order, and I shall test all four cells.

 

[wattmatters]

I do find cell capacities listed to 7 significant digits with 0.1 mAh precision a bit amusing!

 

[mcart117]

I do find cell capacities listed to 7 significant digits with 0.1 mAh precision a bit amusing!

Totally agree! It's the 3 digits before the decimal point that count.

But the key thing is that these cells have been through a rigourous testing process and passed, whereas many others sold on Alibaba haven't. Well, some of them have been through the process, but they've failed.

 

[wattmatters]

Have fun with the build. In the end I didn't bother and have some server rack batteries coming.