LiFePO4 Batteries: Discussion Thread for Those Using Them as House Banks

[diugo]

Separate disconnects for the load and charge busses are ideal---but the single disconnect approach is less expensive, simpler to wire, and more idiot proof. For example, what would happen if a charger was inadvertently wired to the load buss?

[mcarling]

Originally Posted by diugo

Separate disconnects for the load and charge sources are ideal---but the single disconnect approach is less expensive, simpler to wire, and more idiot proof.

The problem with idiot-proof systems is that they only work well for idiots. Someone who doesn't know the difference between a load and a charging source shouldn't be wiring up an electrical system anyway -- no matter how idiot-proof the components might be.

Originally Posted by diugo

For example, what would happen if a charge source was inadvertently wired to the load source?

Obviously, then that charge source would be disconnected when there was no good reason for it to be disconnected (just as it would be with the overly-simplistic idiot-proof systems) and would not be disconnected when it might be the source of an over-voltage situation.

[diugo]

Another problem with the single disconnect approach is that high voltage charge sources might be connected directly to the loads, if the battery is suddenly isolated.

[mcarling]

Originally Posted by diugo

Another problem with the single disconnect approach is that high voltage charge sources might be connected directly to the loads, if the battery is suddenly isolated.

I think that's the best reason yet for avoiding these.

[Lagoon4us]

Originally Posted by goboatingnow

No me either. It's not suitable for boats.

Dave

That's true this stuff is typically EV's.

[mcarling]

Originally Posted by diugo

Another problem with the single disconnect approach is that high voltage charge sources might be connected directly to the loads, if the battery is suddenly isolated.

The more I think about this, the more I come to think that disconnecting the battery bank while leaving the charging systems connected to the loads may be worse than leaving the battery bank connected.

[deckofficer]

Originally Posted by mcarling

The more I think about this, the more I come to think that disconnecting the battery bank while leaving the charging systems connected to the loads may be worse than leaving the battery bank connected.

I'm not sure I'm understanding the concern. All charging sources should be regulated for the batteries, not exceeding 14.8 volts. Is the concern with PWM type of regulation to keep say the solar panel's 18+ volt output from damaging electronics if the battery (like a capacitor) isn't in line?

[mcarling]

Yes, the capacitance of the battery bank is important to attenuate voltage spikes that might damage electronics.

[daddle]

Originally Posted by deckofficer

I'm not sure I'm understanding the concern. All charging sources should be regulated for the batteries, not exceeding 14.8 volts. Is the concern with PWM type of regulation to keep say the solar panel's 18+ volt output from damaging electronics if the battery (like a capacitor) isn't in line?

This is a common problem in the marine "business." There is no serious attempt to specify the behavior if systems. One is left to heir own engineering to figure out what the various devices actually do. Perhaps if the battery monitor has a "disconnect" output one could use it as a signal to do the appropriate thing with each load and charge source. The alternators might be disabled. The solar circuit opened. The house loads shed while keeping the navigation powered up but with an alarm. Etc.

[mcarling]

Originally Posted by daddle

This is a common problem in the marine "business." There is no serious attempt to specify the behavior if systems. One is left to heir own engineering to figure out what the various devices actually do. Perhaps if the battery monitor has a "disconnect" output one could use it as a signal to do the appropriate thing with each load and charge source. The alternators might be disabled. The solar circuit opened. The house loads shed while keeping the navigation powered up but with an alarm. Etc.

This approach requires separate triggers for under-voltage and over-voltage.

[JayH]

Originally Posted by mcarling

This approach requires separate triggers for under-voltage and over-voltage.

I would also want triggers for high SOC and low SOC. I would prefer to turn on a diversion load, start a generator, or sound an alarm. Disconnect should be a last resort.

I'd really like a charger that integrates with the BMS, has a consistent charge profile, can regulate the various sources, manage diversion, etc. Instead of disconnect, the charger could just be turned off. I'm generally not a fan of complex integrated systems, but it seems to be warranted in this case.

[CharlieJ]

This is how I am approaching the design for a 2000Ahr LFP bank using both source and load busses:

Attachment 52645

[Maine Sail]

My cells arrived today. Have not yet measured the voltage of each cell but they look to be in good condition. They are clearly not in "new" condition but quite good for the price.

I happened to have some customer batteries next to my new cells..

200 Pounds and 280Ah of flooded group 24 batteries vs. 120 pounds and 400Ah of LiFePO4 !!

These 400Ah cells are taller than typical 12V batts but weigh less and take up less foot print than the group 24's do...

[deckofficer]

Maine Sail, I see your trusty Volteq on the left.

The major difference between those two groups of batteries in your picture is usable w-hr.

The lead acids = 1680 w-hr

The LiFePO4 = 4096 w-hr.

Nice comparison picture.


I had to wait for some more parts to arrive before I could keep balancing. While waiting I did a top balance using that variable PS you recommended and started discharging yesterday. Already a vast improvement, 34 hours at a 30 watt drain, all cells reading 3.27 volts. That is 1020 w-hr and the last time at 980 w-hr one cell had dropped to around 2.5 so I stopped. My bank is rated at 1280 w-hr, and I believe I will see that this time.

Here is my "colorful" way of a controlled 30 watt load.

MOV03020 - YouTube

[diugo]

Very psychodelic load you got there Bob! Maine Sail, your cells look much brighter yellow than mine. What does their date code say?

I just now finally got all my 260Ah cells down to around 3.07V each, after about 12 hours of discharging at 5A. That suggests they were shipped at around 12*5/260 = 23% SoC.

I have them all in parallel now, so will leave them overnight to complete the bottom balance, then fully charge them over the weekend to see how well matched they are.