4S LiFePO4 as car battery without BMS?
 

So I’m considering going to a lithium battery both to increase charge efficiency and to reduce weight for my car. My first thought is LiFePO4 for its stability and energy density, and Headway cells (40152S) seem a good fit given their capacity and available discharge rate. Given that their nominal voltage is close to the resting voltage of a lead acid battery, would that mean I wouldn’t need a BMS? Just don’t charge in below freezing contditions?

You want a BMS. It's cheap insurance.

These things do not do well with over(the most likely problem) charging, or being drained down to nothing. They're not forgiving like lead-acid.

Under ideal circumstances, you might get away with using an active balancer instead...but the real world is anything but ideal.

You can always wire up the starter so it and it alone bypasses the BMS, if it's going to draw too much current for one.

Spend the $20 and save yourself the headache of ruining your cells.

I've had a LiFePO4 with no BMS in the Prius for 5 years now. They don't do well getting deeply discharged, and can't be charged while below freezing, so it's important they live inside the cabin if it freezes where you are.

I've got a small 4Ah pack in the glovebox of my Acura at the moment. Decided to add a balancer to that one, which I recommend.

https://www.amazon.com/gp/product/B0...?ie=UTF8&psc=1

I have built 3 12 cell battery packs using 12 Ah LiFePO4 cells and a BMS for each.
All of my 36 cells came out of the same battery factory box and are closely, but not exactly, matched in capacity.
What you need to understand is that you can destroy a LiFePO4 cell by overcharging, and that the voltage over a cell rises very rapidly indeed when it is full.

I charge my cells at 3A.
During the 4 hours that it takes to charge the packs the voltage per cell rises from 3.20 Volt when nearly empty to 3.35 Volt when nearly full, then within seconds it jumps to 3.85 Volt where the BMS cuts off the pack.
It will then top balance the cell down to 3.65 Volt, which takes a minute or so. The other cells will still be around 3.35 Volt.

If the BMS had not cut off the charge the full cell would have gotten over 4 Volt, which would start to damage it; the higher, the worse. Still the other cells would not be full so the total voltage over the pack would not show you it is about to get damaged...

The chargers cut off when the BMS cut off so I need to reset each charger to get it charging again. The full cells will quickly hit 3.85 Volt again and kill the charge, but the other cells will one by one also reach 3.85 Volt. With every cycle the total voltage rises a bit. Then finally all cells are balanced and full.

You really need a BMS, especially if you do not constantly measure voltage at the cell level yourself.

...and I disagree, with my Prius battery as my anecdote.

The battery isn't being cycled, it's briefly supplying power to start the car. 14.4v is only 3.6v per cell, and most cars will output only 13.6v if the battery is topped up, which is 3.4v per cell.

Now, I've only run it for 5 years and the battery could be on its last leg, and a BMS might have allowed me to go another 5 years, or perhaps my battery is still doing just fine. If I get my hands on the car again I might pull it out and test to see how its holding up, and perhaps add a balance board. since it's cheap and easy.

That's alright, if the cells are very closely matched and if your car actually does output no more than 13.6 Volt. I'm skeptical about that, as my Insight will charge the battery at up to 13.95 Volt (verified) with load, so it would be even higher unloaded. IIRC most alternators produce about 14.2 to 14.4 Volt.

Then, if your cells are ever so slightly out of line you might get 3.3 + 3.3 + 3.3 + 4.5 Volt. The high cell may start to dissipate some current which may rebalance the pack, but up there you're already damaging it.
A BMS prevents that. Mine also cuts off if the temperature rises too much, at deep discharge, at overcurrent, you name it.

There's a downside on using a BMS too though. There's cost, it may limit the current below what you'd need for cranking, and the BMS itself may be faulty and cause imbalance.

Yet if you're willing to pay for a BMS, have the specs right and check whether it does in fact balance the cells right (like I did with all 3 of my packs) then it is the difference between hoping everything keeps working and knowing that it does.

So, from advice given, I should relocate the pack inside the cabin and run a balancer on it. Seems easy enough. Are Headway cells usually good or should I order more than needed in anticipation of getting a bad one? You guys think a 15Ah pack would be sufficient? The cells are rated at continuous 10C discharge with bursts up to 15C possible

Been running the same LiFePO4 with no BMS since 2011...

I found the specs of the Headway cells I wanna use. It seems they’ve come down in price since last I checked (duh, winning).

Further research leads me to think that a 4S1P pack using the Headway 40152S cells will start the car okay, most data I’ve found points to most smaller cars taking ~200A to start which is well within the capability of the battery cells. Potentially in the winter time the car will pull more amperage than the pack is rated for (possible damage?) and I won’t be able to do much EOC without killing the pack but it should work okay, and weigh ~5lb. Relocating it to where the passenger airbag was should also afford me to remove about 4-5 feet of positive wire and all the brackets for mounting the original 42lb battery