LiFePO4 DIY options

[tony.latham]

Are you under that western heat dome?

We are... it's been a bit brutal. Where we're headed should be a bit cooler.

Tony

[Tom&Shelly]

^^^ ha ha, I had to look up how much a farad is. That’s a big capacitor for sure. Not sure it is bigger than a 1.21 gigawatt Flux Capacitor though! I’m a biologist not an engineer.

I once worked on an industry team where we built a system to correct the effects of atmospheric turbulence on laser communications systems. Such a curious combination of disciplines that I claimed we were a truly integrated lab: engineers, scientists, and even mathematicians were all welcome and treated with equal (dis) respect!

About the best fun in my career. I started that piece of it as an electrical engineer (communications) and ended up claiming to know something about atmospheric turbulence. Mostly, though, I was a kibitzer!

Tom

[tony.latham]

I once worked on an industry team where we built a system to correct the effects of atmospheric turbulence on laser communications systems.

That's about six feet above my pay grade. Probably more.

Tony

[lfhoward]

The 3.5 volt charging experiment overnight was a success. This morning we had 3.33 volts in the battery pack (+0.01 volts). Since some of the sources I read on top balancing go directly to charging at 3.65, I wanted to see what difference that makes. I’ve been sitting next to the battery for an hour checking its vital statistics. The deal with charging at 3.65 is that I need to pay vey close attention to the battery voltage because the battery can be damaged if its voltage goes higher than 3.65. When the battery hits the right side of the charging curve, voltage can skyrocket quickly. I won’t leave it at 3.65 if I am not able to keep checking it (like overnight when I am asleep). But we are now reading 3.34 volts so charging progress is much faster.

Constant voltage setting / Amps in
3.45 / 1-1.5
3.50 / 1.75-2.25
3.65 / 4-4.4

You can see that the battery charges much faster with the power supply set to 3.65 volts constant voltage. The trick will be not to overshoot, but keeping a close watch on the battery I think I will be fine. Once the battery’s voltage rises above 3.45 or 3.5 I will plan to sit with the battery and babysit it so it doesn’t go over 3.65 (which can happen quite quickly).

[tony.latham]

I will plan to sit with the battery and babysit it ...

Thanks for being detailed with this thread.

Tony

[Tom&Shelly]

I once worked on an industry team where we built a system to correct the effects of atmospheric turbulence on laser communications systems.

That's about six feet above my pay grade. Probably more.

Tony

Nah! Not above or below, just different!

Sometimes I wish I'd taken a career path that led to working outdoors. But, it was the Cold War and the Air Force offered great deals: kept sending me back to school. One BIG regret: I took a 20 year hiatus from camping and outdoor activities, lived in a city, and never was able to build my strength or endurance level back to where it was in high school.

But enough of this non-sense. Now I really do feel like I've hijacked the thread!

Tom

[lfhoward]

No worries about hijacking the thread!

I plan to take all the battery lessons from here and generate a concentrated “how-to” thread for building a LiFePO4 system after all is said and done. Or at least it’ll be a “what worked for me” thread that others can reference and take away what they want.

So feel free to kibitz away! Good to have you along for the ride.

[lfhoward]

Batteries just measured 3.35 volts. So that’s +0.01 volts in 2.5 hours at this charging rate (3.65 volts), rather than taking overnight or all day.

[lfhoward]

I got a delivery today with the parts I need to make a battery heater. LiFePO4 batteries cannot he charged below zero degrees C, so these will warm them up to the point where the BMS will allow charging.



The black is a silicone trivet, followed by the orange beehive heaters followed by an aluminum sheet that will be wrapped in Kapton tape to prevent short circuits. The heater sandwich will go under the batteries and take up approximately 1/4” of room at the bottom of the box. The Inkbird thermostat will turn the heating pads on and off at predetermined temperatures. I will also wire a shutoff switch so that it doesn’t waste energy when I don’t need it to heat the battery.

[bdosborn]

I've been running the same thermostat controlling four bee heaters and my BMS for a couple of months now. No charging, heating, or loads, I wanted to see how long it could sit without the battery getting discharged. I'm at 82% charge on my 400 a-hr battery. The thermostat itself doesn't use hardly any power.



Bruce

[lfhoward]

Bruce, that’s nice to know my thermostat isn’t an amp vampire. I think you were the one that recommended it initially, several pages ago. I’m glad it is a proven setup.

We are now at 3.36 volts in the battery. Making real progress now. I will reduce the power supply to 3.45 or 3.5 for the night to keep the battery voltage on the flat-ish plane and not the steep tail. I only want it getting above 3.5 volts when I am watching it like a hawk.

[lfhoward]

I made a bezel to hold the thermostat and the on/off switch for the BMS. The side of the battery will be accessible through the utility door so the controls face the side.


Behind view.

[lfhoward]

Top balancing continues…

I’m currently stuck at 3.35 volts but that’s the flat part of the charging curve and I expect the battery will stay there for a looooooooonnnnngggg time.

I had the power supply set to 3.5 volts overnight and by this morning only 0.09 amps were flowing. The closer the power supply and battery voltage are to each other, the less current will flow. Got to love physics. I’m now charging at 3.65 volts on the power supply and the current is about 2.75 amps. Again, I am checking the battery voltage often with my multimeter to make sure I am not above 3.45. Once it gets there, a full charge is imminent and I will need to sit next to the battery and disconnect when it hits 3.65 volts.

[lfhoward]

I just discovered something really important about my power supply. Even though it was a 10 amp unit, I have never seen anywhere close to that. I was charging at 3.65 volts and sending only about 2 amps. Well, the limiting factor was the little alligator clips. The machine knew to limit amps through those because the wires couldn’t handle it. They were maybe 16 gauge if I’m guessing. I replaced them with a set of 12 awg wires & clamps that came with my inverter and the difference is night and day.

I was able to ratchet the voltage back to a more comfortable 3.45 volts and am sending 5 amps. At 3.5 volts it was above 8 amps, but I don’t really want to sit here and watch it and 3.45 is well within the safety zone. We are talking about quadrupling the current or better, while keeping voltage down.

Now I can see how this top balance might be doable.

[lfhoward]

The battery reached 3.45 volts last night as the charger came down to zero amps. Now shooting for 3.5 volts, which shouldn’t take long. After that, I’ll go for the top balancing goal of 3.65 and keep it under a watchful eye.

Edit: hit 3.5 in under 20 minutes. Shooting for 3.65 now.

Edit: hit 3.65 in 15 more minutes. Now waiting an hour to make sure the charge stays above 3.5. If not, we charge this back up to 3.65 again to top it off.


Edit: after 30 mins sitting, voltage is still at 3.62.

Here is my cobbled together test rig for the capacity test.


This particular inverter and the welding cable I’m using for the positive today will not be making their way to the trailer. But the Victron BMV212 will tell me how many amps the battery holds as the inverter draws them down. I am shooting for about 30 amps of draw which would be 0.2 C on a 150 amp battery.

After an hour the battery is still holding 3.60 volts! That’s a go for the capacity test coming up next. Now to rearrange the battery cells into 2p4s and hook up the BMS.