acdoctor
Solar Enthusiast
On my 280s from 3.55 to 3.65 is only a few minutes, virtually no capacity difference. So if most were at 3.55 and 1 hit 3.65 I don’t think it would bother me.
Unusual cell voltage (HVD and LVD) but not always the same cell(s) - 16S 48v
- Thread starter UV-PWRD
- Start date
I don't charge that high, I have bulk set to 55.4 or so I think, Used to be higher but backed it off a while back for longevity (hopefully).
The alligator clip leads from the power supply might explain that. Numerous reports of people having problems that are easily resolved with a larger gauge wire and ring terminals.
I upgraded those after reading the same which is how I got it up to about 6a, ran double leads to test also, I think it was just the supply.
Anyway, it balanced eventually.
I’ve gone with 10mv, the sooner the balancer can start the easier it is to try and sort things out. Also only one balancer at a time should implemented.
And yes the haltec active balancer should only be “on” or run when cell voltage is above 3.4v. I have a switch on mine I plan to wire to my Victron SCC output load to a relay to toggle on at 13.6v.
Last edited:
houseofancients
Solar Wizard
i would set this may be a bit lower as the heltec balancer, and make it stop at the point where the bms active balancer comes on ..
EG.. if you set the bms balancer to start at 3.45 , i'd start the heltec at 3.4 and stop at 3.45..
what would also help the process along a bit better if you can lower the AMPS going into the pack from 3.4 onwards .. just to give both balancer time to get the balancing done, given the inbalance the pack has ( this is where bms , charge controller comms come in handy
still i would expect it to take a few cycles to get back into balance
Bulk/Float voltage should help that but the reality is there's not enough sun currently for me to fill the pack anyway but I "think" we're getting it near 80%
For the moment I'll leave it and see how it behaves and if need be as above will re-check all the connections and do a top balance.
I don't think it's crazy to need to do full proper balance once a year or so as a maintenance item.
Also, why bring the Heltec in first given it's the more powerful balancer?
For the moment I'll leave it and see how it behaves and if need be as above will re-check all the connections and do a top balance.
I don't think it's crazy to need to do full proper balance once a year or so as a maintenance item.
Also, why bring the Heltec in first given it's the more powerful balancer?
Updates.
I did not top balance as I would've had to label/mark all the balance leads to ensure they went back in the right spot.
I removed each connection, cleaned, used copper grease on the surface of all battery terminals and the stud threads, removed 1 bus bar (it was double stacked) and any washer that was not essential.
Let the pack charge slowly, with the heltec running as it does bugger all unless the pack voltages vary Let it charge the top 10% back and forth a few times and keep balancing - Balance in general looks very good.
The unusually large drop/gain in voltage for random cells still remains though it's more obvious with drop rather than gain, now it's cell 12 that will reportedly drop to 2.9v under a 2kw load and cell 4 that will hit HVD before anything else.
I'm starting to wonder if the BMS repairs were done properly or the balance leads/connections are just no good. Ideally, I'd love to try another BMS, I'm reluctant to pull the whole pack down and top balance as the voltage drop behaviour doesn't lend itself to the cells being widely out of balance given the same cell is hitting higher voltages evenly with the rest.
Seems more like a capacity/connections/BMS reading issue.
I did not top balance as I would've had to label/mark all the balance leads to ensure they went back in the right spot.
I removed each connection, cleaned, used copper grease on the surface of all battery terminals and the stud threads, removed 1 bus bar (it was double stacked) and any washer that was not essential.
Let the pack charge slowly, with the heltec running as it does bugger all unless the pack voltages vary Let it charge the top 10% back and forth a few times and keep balancing - Balance in general looks very good.
The unusually large drop/gain in voltage for random cells still remains though it's more obvious with drop rather than gain, now it's cell 12 that will reportedly drop to 2.9v under a 2kw load and cell 4 that will hit HVD before anything else.
I'm starting to wonder if the BMS repairs were done properly or the balance leads/connections are just no good. Ideally, I'd love to try another BMS, I'm reluctant to pull the whole pack down and top balance as the voltage drop behaviour doesn't lend itself to the cells being widely out of balance given the same cell is hitting higher voltages evenly with the rest.
Seems more like a capacity/connections/BMS reading issue.
Active balancer that is active below 3.4v will unbalance your cells. Stop already with the Heltec. Make sure your JK does not balance below 3.4v. You need to top balance again (just the JK will work, but patience is required). Get all of the cells above 3.6v, let the active balancer portion of the BMS work and you will be good. The Heltec tries to balance your cells all of the time and has undone your original top balance. Looks also like you have some minor connection problems from the readings on the resistance of your BMS wiring.- That pic is the first day it went in to service, so there was no additional balancer then. the red wires are the BMS sense/balance leads.
- I used both balancers for a few months, but prior to that it has lived with 1 balancer it's whole life, it's back to 1 balancer now, the Heltec stand alone.
- Washers are only on the top of the last fitting between it and the nut, that comment is just to say they are where they should be.
- No rating was given on the bus bars, but I was advised that double stacking reduced any risk of overloading them, I guess it also allows the potential for increased and uneven resistances between the cells?
- The heltec specs imply it's always on above 3v, but current ramps up as the voltage differences increase.
3S-21S 5A Capacitive active equalization/Active Balancer
[vc_row][vc_column][/vc_column][/vc_row]heltec-bms.com
- I had the BMS balancer originally only working above 3.3v but with any cell delta above .05 it would be working, so basically all the time.
In my head though, to see a cell going HVD and LVD means it just has less capacity than those around it yes? Unless there is something to the resistance comment I made earlier, those cells have higher resistances and all have increased (bad internal resistance/failing cells etc).
Despite what Andy says, an active balancer is not needed for anyone who fully charges on a regular basis. I use one to help with the initial top balance and then take it off.
No need to let cells rest for more than a few hours when top balancing, letting them sit at 100% is one of the worst things you can do for longevity and causes degradation quickly.
Last edited:
Quite true. Once you get above 3.45v the power stored above that is at most 1 or 2 amp hours, and above 3.55v you are talking maybe 0.1 amp hours.
There might be more too it as the problem continues. Last night under a 2.7kw load I had cell 12 dipping noticeably so went down with the Fluke 325 and checked that cell and confirmed it was genuinly a solid .1 .2 v below the other cells under load 3-3.1v vs 3.2-3.3v in adjacent cells. So either the balance is WELL out or I do have some actual cell issues, the thing that gets me is if it was a cell problem why does the problem cell move around so much.
I am going to have to pull the pack and top balance it to be sure, at which point I'll remove the Heltec and leave the JK doing it;s thing above 3.4v.
The thing with the Heltec, and I've tested this it moves barely any current unless there's an actual cell difference. The 325 has a clamp meter so I've measured this. I've never seen anything more than 3 being shuffled into cell 12 and when there's no load on the pack it's doing basically nothing, .1a etc
Re connections, I noted the balance wire leads are not a perfect fit for the studs, so I was considering re terminating each wire with a new terminal that exactly fits the stud. Of course this also means biting the bullet and labelling all of the individual balance leads first so I know where they go back :'(
I'll have to check to be sure, I can remove it but it was a recommendation on here and given the the need for a washer between the bus bar and the balance lead (due to holes size on the buss bar) it seemed like a good idea.
BUT given the Fluke tells me that cell is actually low it's not just a BMS reading issue (though I understand that resistance values on those leads likely play into the measured voltage a the BMS being out).
No, you should not have a washer between the bus bar and the balance lead ring terminal. Use a better ring terminal on the balance lead.
Depending on how long you had the Heltec connected, they could be extremely out of balance, yes.
How to top balance (3 Pre-charging the cells to get them ‘mostly’ charged.):
Top Balancing LiFePo4 Cells using a low cost benchtop power supply.
To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...
diysolarforum.com
Excellent resource, I highly recommend the pre-charge process to help speed things up (i.e. using a BMS to charge until a cell hits HVD, keep the HVD set at 3.65v please, don't try and set it higher).
Explanation of why you want to top balance (sounds like you already know why):
https://diysolarforum.com/resources/explanation-for-beginners-of-top-and-bottom-balance.127/
A little bit of current adds up when done 24 hours a day, 7 days a week.
If you are in the USA, 1/4 inch will work fine, much easier to find than 6 mm ring terminals.
Judging by some of the readings you have shown, cutting off your existing connections and putting on new ones would help. It certainly wouldn't hurt as long as you verify they are good. A decent meter is your friend in this case to measure the resistance of each connection.
These are what I've been using for balance connections, although if memory serves me correctly, the JK uses larger wires than most due to having an active balancer.
26-24 Ga. Heat-Shrink Ring Terminals, 1/4" Stud
26-24 Ga. Heat-Shrink Ring Terminals, 1/4" Stud For use in harsh environments, these terminals feature a heat shrink connection that also has an internal sealant. After crimping the terminal onto the wire, simply apply heat. Keeps moisture out while ensur
Depending on how long you had the Heltec connected, they could be extremely out of balance, yes.
How to top balance (3 Pre-charging the cells to get them ‘mostly’ charged.):
Top Balancing LiFePo4 Cells using a low cost benchtop power supply.
To get the paper, click on the orange at the top of this page. This is a tutorial with detailed steps on how I top-balance my LiFePO4 Cells using a low-cost benchtop power supply. To get the tutorial, click on the orange "Download" button...
Excellent resource, I highly recommend the pre-charge process to help speed things up (i.e. using a BMS to charge until a cell hits HVD, keep the HVD set at 3.65v please, don't try and set it higher).
Explanation of why you want to top balance (sounds like you already know why):
https://diysolarforum.com/resources/explanation-for-beginners-of-top-and-bottom-balance.127/
A little bit of current adds up when done 24 hours a day, 7 days a week.
If you are in the USA, 1/4 inch will work fine, much easier to find than 6 mm ring terminals.
Judging by some of the readings you have shown, cutting off your existing connections and putting on new ones would help. It certainly wouldn't hurt as long as you verify they are good. A decent meter is your friend in this case to measure the resistance of each connection.
These are what I've been using for balance connections, although if memory serves me correctly, the JK uses larger wires than most due to having an active balancer.
26-24 Ga. Heat-Shrink Ring Terminals, 1/4" Stud
26-24 Ga. Heat-Shrink Ring Terminals, 1/4" Stud For use in harsh environments, these terminals feature a heat shrink connection that also has an internal sealant. After crimping the terminal onto the wire, simply apply heat. Keeps moisture out while ensurcrimpsupply.com
More info, there's clearly connection issues at play here.
Last night I get and find that cell 12 is showing 3v at only 10a load yet was hitting HVD early in the day during peak charging phase (approx 60a).
I put the house back on grid so I could switch off the BMS and have a fiddle. I find cell 12 neg connection is barely finger tight, and these were only just recently re-tightened after me going through and removing extra bus bars/washers etc. I went over the entire pack twice during that session.
I went over the entire pack terminal connections again, all seem OK, though I note cell 12s terminal posts are higher than the rest, implying they are not bottoming out in the battery terminal holes (my terminal posts are screw in, not directly part of the cell).
I'm starting to wonder if the carbon grease is allowing these connections to back off/move with hot/cold cycles.
Reboot the BMS, cell 12 voltages are not perfect and resistance measured is back in spec with the rest, Cell 1 is now my low/problem cell.
Re the terminal posts, they are only M4 studs and the busbar holes are actually slots to allow for sideways movement meaning the balance need connections, when tightened down can deform into the slot a little which is why the washers were there.
I considered putting the balance leads underneath but then you have busbar making contact with the post of the terminal only and not the rest of the cell terminal surface area.
Appreciate everyone's continued assistance and interest in this also.
acdoctor
Solar Enthusiast
Do you have grease hydrolicly compressed in the bottom of the terminal preventing the stud from bottoming out?
I'm thinking almost certainly on cell 12 so I'll remove those studs and chase the thread to clean it out.
This has been such a good learning exercise and it's a real departure from building race cars :D
You don't want any type of grease on any of the threads at all. Thread locker is OK, but no grease. I used the screws that came with my batteries and bus bars. I've had no loosening of the screws and my batteries are in an RV trailer that sees some fairly rough roads.
A couple alternative solutions for your balance lead connection:
1. Put the washer in top of the ring terminal. That way the ring terminal won't deform into the bus bar slot.
2. Drill and tap a hole in the bus bar to fasten the ring terminal to. This is acceptable but not implemented by many.
The balance lead ring terminal should always be on top of the bus bar, never between the cell terminal and the bus bar.
