LiFEPO4 retrofit

Robaroni

Robaroni

New member
Joined
Jun 22, 2018
Messages
17
Hello everyone,
Have been reading some of your builds - great work!

I thought I'd get some feedback and advice from those who have come before me.

Here's what I have and am doing:

I have a off grid/ intertie system in right now with 2 Outback 4836 inverters and charge controllers from Outback too. Total system is 4.4Kw. ( I do have another system in with 6.4Kw but it's just an intertie so it's not part of this design - yet anyway!) The system is presently running 8 Rolls L/A 530s batteries at the end of their life (~14 yrs)

This gives me ~80 to 90amps to work with for charging. I can reduce the current to the batteries but this also reduces my intertie output,. so I'll run the CC's to their max. or close to it.

I have 640 newTenergy LiFePO4 5.5aH batteries (32700) which I've done some prelimtesting on and they are up to specs.

What I think I'll do is set up something like Owitte did in Germany. I'll run 16s 38p. Because I already have upper and lower voltage setting ability all I'm planning on is balancing. Of course, as many of you know I can run several different 'S' and 'P' configurations and get the same results but I like the 38p because it keeps a pretty stable 3.2V pack (I'll fuse each battery). I am open to advice though.

I'd like a balancer that bluetooths or other wireless communications out to my PC so your thoughts on balancers that have worked for some of you is appreciated. I'm not against China if you guys haven't experienced excessive failures after time with certain ones although at 38p I question their ability to balance well. Hopefully the new Tenergy's won't need much!

Just as an aside, I usually charge my lithiums to 80% of max. This system isn't a full off grid, what I do is keep my batteries charged and when we have mains failures the batteries kick in for a day or two at most unless we have a major weather condition.

Thanks for the help,
Rob
 
I used those tenergy 5.5ah lifepo4 and got balance issues when fast charging to full. I built 2x 4s20p 110ah batteries and they both had balance problems. Eventually I parallel them together to form 1 220ah battery. To solve the balance problems I had to use 4s active balancers which cost me 92 dollars. Thats the only thing that solve my problem. Someone else here in the forum use the same brand tenergy but the 26650 cells and they had balancing problems. I didn't capacity test my cells because they were new out of the box and had no cell holder to test with. I only had 160 cells, you would have to test all your 640 cells and then use repackr to put them in the right place to prevent balance problems, just one weak cell will mess everything up. The active balancers will compensate for the weak cell(s) if you don't test them all.

Myself I would recommend active balancers if you plan on fast charging (at high amps) your battery pack. Fast charging always triggers the bms to stop the charge early, even slow charging eventually triggers the bms but it takes longer. The balancers I use would be very expensive on a 16s almost 400 dollars. I had my 4s balancers connected 24/7 almost a 1.5 years and they work excellent. They work in the background, you never have to bother with them.

Recently on ebay I saw some active balancers and bought 2 of the 4s (15 dollars each) I still haven't received them to test them. But they also sell a 16s board for 43 dollars. These balance at a more modest 1.2 amps (compared to the 6 amps that the ones I use is capable of). (Li-ion Lifepo4 LTO Battery Active Equalizer BMS 1.2A Balance 3S 4S 6S 7S 13S 16S)

image_ltcmey.jpg



4s 6 amp activebalancersI been using (cost 92 dollars)

image_mnxyie.jpg


I been using these isdt bc-8s to monitor the balancing, they have low/high voltage alarms (cost 18 dollars) for 16s you would need to connect 2 of them.

image_gshfzs.jpg
 
jonyjoe505 said:
I used those tenergy 5.5ah lifepo4 and got balance issues when fast charging to full. I built 2x 4s20p 110ah batteries and they both had balance problems. Eventually I parallel them together to form 1 220ah battery. To solve the balance problems I had to use 4s active balancers which cost me 92 dollars. Thats the only thing that solve my problem. Someone else here in the forum use the same brand tenergy but the 26650 cells and they had balancing problems. I didn't capacity test my cells because they were new out of the box and had no cell holder to test with. I only had 160 cells, you would have to test all your 640 cells and then use repackr to put them in the right place to prevent balance problems, just one weak cell will mess everything up. The active balancers will compensate for the weak cell(s) if you don't test them all.

Myself I would recommend active balancers if you plan on fast charging (at high amps) your battery pack. Fast charging always triggers the bms to stop the charge early, even slow charging eventually triggers the bms but it takes longer. The balancers I use would be very expensive on a 16s almost 400 dollars. I had my 4s balancers connected 24/7 almost a 1.5 years and they work excellent. They work in the background, you never have to bother with them.

Recently on ebay I saw some active balancers and bought 2 of the 4s (15 dollars each) I still haven't received them to test them. But they also sell a 16s board for 43 dollars. These balance at a more modest 1.2 amps (compared to the 6 amps that the ones I use is capable of). (Li-ion Lifepo4 LTO Battery Active Equalizer BMS 1.2A Balance 3S 4S 6S 7S 13S 16S)

image_ltcmey.jpg



4s 6 amp activebalancersI been using (cost 92 dollars)

image_mnxyie.jpg


I been using these isdt bc-8s to monitor the balancing, they have low/high voltage alarms (cost 18 dollars) for 16s you would need to connect 2 of them.

image_gshfzs.jpg



Thanks JonyJoe, very helpful.
How did the weak cells show up? Let's say we charge two legs only and one of the legs has a weak cell. It would seem to me that the weak leg would drop faster on load than the leg without the weak cell. That would create a problem when the batteries dropped on load. The P string with the weak cell would fall below critical voltage before the others and that P bank would be damaged if the inverter circuit sensed hi and low total battery bank levels.
One way I see to remedy this is to monitor each of the 16 legs and shut off the entire battery if one leg dropped below critical. Or bypass the weak leg when it drops below critical voltage.
The problem is the level of weakness of the badcell(s)because if they are very bad then the system is compromised with a very short run time.

I wonder if the more cells in P that a leg has the less one weak cell would effect the bank? Of course the best way to address this is to check each cell which is a pain.

So now you solved the problem with active balancers but let's say a condition exists where the bank isn't charged for a long enough period for the active balancer to 'save' the bad cell. Doesn't that run into the same problem? Doesn't that P leg risk the chance of running below critical voltage?
Click to expand...
 
I recently got a 7s of the cheap 1.2amp balancer pictured and I believe it is working well. It has very low standby current, and the leds only light when balancing. Seems to be quite good.
 
camthecam said:
I recently got a 7s of the cheap 1.2amp balancer pictured and I believe it is working well. It has very low standby current, and the leds only light when balancing. Seems to be quite good.
Click to expand...

Thanks, Have a link?

Robaroni said:
camthecam said:
I recently got a 7s of the cheap 1.2amp balancer pictured and I believe it is working well. It has very low standby current, and the leds only light when balancing. Seems to be quite good.
Click to expand...

Thanks, Have a link?
Click to expand...

OK, the saga continues. I built a small jig to insert a single cell into to find the internal resistance. Ill try to make a thread on how to do this but to give builders and idea of how it's done here's what I did.

I inserted the cell into the holder and connected the plus and minus accordingly to an electronic load and a minimum/maximum bench meter. I programmed the load for 10 seconds at 1 amp. After the ten seconds I took the minimum (drop voltage) and subtracted it from the open voltage (maximum) and then used ohms law to figure the battery internal resistance.

So let's say the difference between min and max is .035 volts. Since the current is one amp the formula to calculate the internal resistance is:

E=IR; R=E/I = .035 volts /1amp = .035 ohms. Using one amp makes it easy you don't need to do any math, the voltage is the same as the resistance.

What I found was that the 12 or so batteries I tested all were very close. I'll test all the batteries but so far so good.

Rob
 
You must log in or register to reply here.
Back
Top