Charging LFP below 32F - 0C freezing temperature - possible - with newer chemistry?

Hedges said:

I'd think a few inches of foam would make heat loss negligibly low. For summer, need a manual or automatic vent opened.

Click to expand...

very low. It's funny but last summer, the 2" of PIR around the batteries stopped them heating up as much as the garage air temp got to in the sunny afternoons!

Hedges said:

I don't think many, if any, of you have charge current adjustable based on temperature.

Click to expand...

I do I have temperature and SOC mappings... see...

Hedges said:

Or to lower it at night.

I'm not sure that maintaining battery temperature at night consumes much more than warming it back up in the morning. It is just a matter of that extra 5C delta T, and thermal resistance.

I'd think a few inches of foam would make heat loss negligibly low. For summer, need a manual or automatic vent opened.

While you're at it, raise not just 5 degrees but enough that the battery can comfortably accept the power you will have available. The curves for cold temperature charging dropped quite low, like 0.2C. I don't think many, if any, of you have charge current adjustable based on temperature.

Click to expand...

I have solar assistant configured + home assistant, so can adjust the max charge amperage based on temp. But it is sort of irrelevant since I can't get above .1C with my current production.

I'm thinking in this early portion of the morning, here is a screen shot of mine this morning:


So if heating begins at 6:45am, or so, right when solar just starts to come in (ignoring my home load, just focus on PV Charging the battery), the battery will begin heating because it detects charge current, then by 7:30 (maybe much sooner, depending on Cell temp, and insulation), the battery is warm, and ready to accept the 1kw of solar that is coming in. In that 45 minutes, you've lost maybe 4-500wh of possible solar, because it was only letting the heaters run, and not allowing any other charge current.

And by that time, if it starts charging the battery with a full 1kw, at 7:30am, and say you have a 100ah battery bank (abnormally small, but exaggerated for the example), then you are still only charging at .2C rate. If you have a 280ah bank, which I would consider minimum for a 48v bank, then its only a 0.07C rate. Should be no problem for a LiFePO4 battery at 2-5*C.

Another thing that popped into my head, is what if you don't have good sun in the morning/next day? You heated that battery all night to keep it ready for the next day, and wasted that energy because there is no charging the next day. PV might not even cover your loads, so its definitely not charging the battery. Now you lost that energy you could have used during the day because it was wasted in the night to heat.

I'm starting to question if you are reading my replies carefully. The example you provide is not reproducible with a Seplos BMS. You cannot "let the heaters run, and not allow any other charge current". The BMS does not allow for this to happen. There is simply no settings in the BMS that will allow you to have the heaters on and the cells not accepting a charge with a Seplos BMS.

If you don't keep the batteries above 0 deg C you have no choice but to charge the cells below 0deg C, at least until the heaters get them over 0deg C which can be several hours based on others who have tested this exact scenario. Which is why I originally asked what the impact of charging at an extremely low C rate might be.

beckkl said:

I'm starting to question if you are reading my replies carefully. The example you provide is not reproducible with a Seplos BMS. You cannot "let the heaters run, and not allow any other charge current". The BMS does not allow for this to happen. There is simply no settings in the BMS that will allow you to have the heaters on and the cells not accepting a charge with a Seplos BMS.

If you don't keep the batteries above 0 deg C you have no choice but to charge the cells below 0deg C, at least until the heaters get them over 0deg C which can be several hours based on others who have tested this exact scenario. Which is why I originally asked what the impact of charging at an extremely low C rate might be.

Click to expand...

My apologies, I assumed the Seplos activated the heaters in the manner that @SeaGal mentioned on post #7. If the Seplos BMS does not switch off the charge mosfets, and allow the heaters, then it is pretty worthless if you ask me. Maybe reach out to Seplos and ask for a firmware update? It should be capable of such with the current hardware if I remember correclty.

Lt.Dan said:

My apologies, I assumed the Seplos activated the heaters in the manner that @SeaGal mentioned on post #7. If the Seplos BMS does not switch off the charge mosfets, and allow the heaters, then it is pretty worthless if you ask me. Maybe reach out to Seplos and ask for a firmware update? It should be capable of such with the current hardware if I remember correclty.

Click to expand...

No worries. I'm hoping that is what they eventually do. It seems like they didn't think beyond adding the heaters to help keep battery warm. Which if you had grid access is fine I guess (although likely a waste of power as you mentioned). I don't have grid access, so I think I will likely have to try and keep them from freezing at night.

beckkl said:

No worries. I'm hoping that is what they eventually do. It seems like they didn't think beyond adding the heaters to help keep battery warm. Which if you had grid access is fine I guess (although likely a waste of power as you mentioned). I don't have grid access, so I think I will likely have to try and keep them from freezing at night.

Click to expand...

Hopefully the heater circuit isnt physically downstream of the charge circuit. In that case no amount of firmware updates may fix it.

Well, well well. After doing some youtube searching, it looks like someone recently discovered that the BMS CAN heat with the charge mosfets off. It turns out that the BMS waits 30 minutes before turning on the heating fet when under temp protection is triggered. Kind of strange.

Couldn't you just put an in line (circuit) stat that closes at 35F or 2C? It would keep the heater from running until it's cold.

Partimewages said:

Couldn't you just put an in line (circuit) stat that closes at 35F or 2C? It would keep the heater from running until it's cold.

Click to expand...

You can do a lot of things. Just was trying to figure it out using the BMS as-is, as it is supposed to be able to handle it. Turns out it was just some poor documentation on seplos's part.

Partimewages said:

Couldn't you just put an in line (circuit) stat that closes at 35F or 2C? It would keep the heater from running until it's cold.

Click to expand...

Perhaps to avoid an on edge trigger?

sunshine_eggo said:

Yttrium doping of the cathode can permit safe charging down to -20°C.

Click to expand...

the fact I have to quote Sunshine tells you something.

the only cells that are safe to charge below freezing are the ones where the maker says you can. so in the case of Winstons pay more, get Yttruium doping of the cathode and charge to way way below zero... other wise stop dancing and figure out a way out of the hole you dug.

Daddy Tanuki said:

the fact I have to quote Sunshine tells you something.

the only cells that are safe to charge below freezing are the ones where the maker says you can. so in the case of Winstons pay more, get Yttruium doping of the cathode and charge to way way below zero... other wise stop dancing and figure out a way out of the hole you dug.

Click to expand...

OH THE HUMANITY!!!!!!