Horsefly
Solar Wizard
That's what I meant to say. Full resting voltage. I'll edit my post. Thanks Frum.
Best LifePo4 charge controller settings known to man for Maximum Service life and Minimum battery stress!!! 5,000-10,000+ cycles?
- Thread starter Go2Guy
- Start date
Pete@SolarBatteryBarn
New Member
It appears that most of us with practical experience agree that float is necessary and ok when set below rest voltage.
What sucks is all of the charge controllers I have tried disable float if you select a Li profile.
What sucks is all of the charge controllers I have tried disable float if you select a Li profile.
Goboatingnow
Solar Enthusiast
- Joined
- Jul 3, 2022
- Messages
- 1,325
Again for maximum life time do not float Li
Any impressed terminal voltage cause ion migration and sei layer growth and li plating effects
Hence after charging is finished a Li charger shoujd stop
Cheap 4-life
My body is 2.63 trillion volts, .07v per cell
How is that accomplished with a SCC?
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
Many charge controllers have a user profile which allows more control of the charge parameters.
Most controllers have more or less sophisticated charge termination logic.
This document will give a birds eye view of how charging works for LFP.
One of the assumptions in the document is a base load.
That means as soon as charging is terminated the base load will immediately start drawing down the battery.
You can also set your float lower if you want the battery to discharge a bit more before floating.
Cheap 4-life
My body is 2.63 trillion volts, .07v per cell
Right but what about when the charge controller begins to float the battery. Then solar is fully covering the load. voltage is still being impressed on the batteries terminals.. “after charge is finished the charger should stop” is what @Goboatingnow said..
The battery needs to go into float to allow solar to cover the load
John Frum
Tell me your problems
- Joined
- Nov 30, 2019
- Messages
- 15,233
When the charge source goes into float its no different than having 2 batteries in parallel.
Are you ok with having 2 batteries in parallel?
Ampster
Renewable Energy Hobbyist
Or when in series there is also voltage on the terminals. I wonder if it is current that @Goboatingnow may have meant when he used the phrase "any impressed voltage on the terminals"?
Choose an SCC that allows it. Even a Morningstar TS-45 PWM controller will do it.
It doesn’t.
Cheap 4-life
My body is 2.63 trillion volts, .07v per cell
What exactly does your controller allow? Mine floats my battery. That is what @Goboatingnow is saying not to do.. my controller always impresses voltage on the battery terminals and always allows solar to cover the load and keep the battery topped off and that can’t hapoen unless there’s voltage at the batteries terminals..
Cheap 4-life
My body is 2.63 trillion volts, .07v per cell
I am of course ok with that. I’m not saying to not float our batteries. I see no way not to float them if they are being used for excess solar storage
I guess the real question is what float voltage to use. Theoretically you can use 3.37 volts per cell indefinitely, so 53.92 volts in a 16S battery. I'm OK with setting that down a bit and not insisting on getting the cells stuffed completely full. Above that voltage, you probably want the SCC to stop charging at some current limit.
Since I am not present at my installation all the time, I don't want the cells to stay charged too long. I plan to set the float voltage just above the "rebulk" voltage on my Outback Flex 80 controller. Something around 3.325 volts per cell (53.2v for 16S), with rebulk set at 3.320 volts per cell (53.12v for 16S). That should leave me no worse than about 70% SOC, so even if I show up in the evening I should be able to run the A/C all night.
The plan would then be to set the "equalize" voltage to something a bit higher, e.g. 3.425 volts per cell (54.8 volts for 16S), then turn it on manually when I am there all day. I can monitor the current and stop the charge when it gets down to 0.01C. For my cells, that is 230Ah * 0.01 * 2 parallel = 4.6 Amps. Once or twice a year doing that should allow the BMS to reset the SOC calculation and balance the cells for a while.
Cheap 4-life
My body is 2.63 trillion volts, .07v per cell
Yes, that’s what most do if using their batteries for solar storage, they float their batteries at 3.37 and under which constantly impresses a voltage from charge controller on the batteries terminals. which as far as I understand is unavoidable..
I’m not understanding why you are wanting to use equalize. Couldn’t you set absorb set point higher than float and let the current tapper off on its own while absorbing and allow the bms to balance when needed?
Last edited:
I don't see the need to overcharge the cells every day while I am not there, so absorb will be just under 3.37 volts per cell and float will be well below that.
It is far easier to just start an equalize cycle when I want to than try to diddle the absorb settings. The cells should not need balancing on a frequent basis.
Simplify !
1) Buy 20% more capacity than you need, because we are going to stay out of the upper and lower knees.
2) Cycle at 3.375v / cell as your cv. This ensures that you will never achieve a full charge - and - stays away from the knife-edge crossover voltage of 3.4v. A top-balance "sanity check" done at higher voltages should be done at least once. Then drop back to 3.375v/ cell cv cycling.
3) Use a conservative LVD of 3.1v. Maybe 3.2v. Staying well out of the lower knee.
4) If you *must* float, then do so at 3.375v per cell, just like #2. Not 3.4v.
This is not some sort of 80/20 fight. It's about compensating for the lack of lab-perfect conditions of cycle-testing done at factories. Ie, high-heat levels are less degrading if you are not in the upper knee. Likewise, a very conservative dod is used. Because you are using this in the real-world, not the lab.
Manufacturer lab-perfect cycle-testing has the problem of temperature-controlled environments, and when using back-to-back-to-back cycling, "beats the clock" in regards to secondary reaction growth, because they simply do not have the time to cycle them once a day for 10 years before putting them on the market.
1) Buy 20% more capacity than you need, because we are going to stay out of the upper and lower knees.
2) Cycle at 3.375v / cell as your cv. This ensures that you will never achieve a full charge - and - stays away from the knife-edge crossover voltage of 3.4v. A top-balance "sanity check" done at higher voltages should be done at least once. Then drop back to 3.375v/ cell cv cycling.
3) Use a conservative LVD of 3.1v. Maybe 3.2v. Staying well out of the lower knee.
4) If you *must* float, then do so at 3.375v per cell, just like #2. Not 3.4v.
This is not some sort of 80/20 fight. It's about compensating for the lack of lab-perfect conditions of cycle-testing done at factories. Ie, high-heat levels are less degrading if you are not in the upper knee. Likewise, a very conservative dod is used. Because you are using this in the real-world, not the lab.
Manufacturer lab-perfect cycle-testing has the problem of temperature-controlled environments, and when using back-to-back-to-back cycling, "beats the clock" in regards to secondary reaction growth, because they simply do not have the time to cycle them once a day for 10 years before putting them on the market.
Last edited:
RandyP
Solar Enthusiast
- Joined
- Sep 21, 2019
- Messages
- 742
I'll bet you never imagined the can of worms you opened up when you made this post. This subject has been hashed over, kicked around, discussed, cussed, preached about and it will never be resolved to everyones satisfaction. Some just want to argue, some believe they have the right answer. At any rate, there is no one answer to this enquiry that will fit all LiFePO4 cells manufactured and their end uses.
Then there's the fact that Prysmatic LiFePO4 cells are not all made to the same spec for the same use. Most we DYO solar guys buy are failed LiFePO4 prysmatic cells built originally for the auto industry. We have no idea exactly how good the cells are, but they are cheaper than the automotive grade cells. And the chinese failed cell auction buyers/resellers of failed cells just tell us what we want to hear, reselling the failed cells as 'Grade A' cells. There is nothing absolutely guaranteed about any combo of LiFePO4 cells assembled as a battery. It's a crap shoot for most. But if its cheap and 'Grade A', we Americas will buy it. The Chinese Grey market loves us.
Last edited:
Yikes - I hope msg 376 counts. I have an addendum:
Note: By cycling at 3.375v to ensure that you'll never achieve a full charge, with a battery capacity you have purposely oversized in capacity by 20% or more than you need, you can also conveniently ignore any "tail-current" or "absorption time" issues. You can absorb down to zero amps and not worry, because you are under 3.4v where this is not damaging.
This is especially important to do if you are using LFP in a "standby" or ups-type mode. AND very helpful if you are running so-called grade-B. Just purchase 20% more capacity than you need, do a sanity-check balance, and then drop back down for daily use.
Therefore the best charge controller - if you run conservatively - is:
ANY SCC that can set the CV/Absorb/Boost (whatever the manufacturer calls it) down to 3.375/ cell (13.5v for nominal 12v batt).
Buy a Victron. Or a $20 cheapie. Whatever fits your budget and desired build quality. The only thing it has to do when running conservatively is the bold text.
Note: By cycling at 3.375v to ensure that you'll never achieve a full charge, with a battery capacity you have purposely oversized in capacity by 20% or more than you need, you can also conveniently ignore any "tail-current" or "absorption time" issues. You can absorb down to zero amps and not worry, because you are under 3.4v where this is not damaging.
This is especially important to do if you are using LFP in a "standby" or ups-type mode. AND very helpful if you are running so-called grade-B. Just purchase 20% more capacity than you need, do a sanity-check balance, and then drop back down for daily use.
Therefore the best charge controller - if you run conservatively - is:
ANY SCC that can set the CV/Absorb/Boost (whatever the manufacturer calls it) down to 3.375/ cell (13.5v for nominal 12v batt).
Buy a Victron. Or a $20 cheapie. Whatever fits your budget and desired build quality. The only thing it has to do when running conservatively is the bold text.
Last edited:
I float mine at 13.0 volts. Battery will supply the loads for close to 70% of capacity before the float provides a floor.
That's cool. Anything 13.5v or under for float. And if one is running conservative, the same 13.5v value for CV too. (Just to get the most capacity out of this purposely capacity-limiting voltage)
It's almost too simple to believe. Takes all the fun out of choosing an SCC, because that's the only thing required. Tail-current, absorb time - all thrown out the window since we're not charging to full.
It's almost too simple to believe. Takes all the fun out of choosing an SCC, because that's the only thing required. Tail-current, absorb time - all thrown out the window since we're not charging to full.
Similar threads
