Didn't top-balance...will try anything before disassembling.

[Jesshaines]

Ok, so I'm a relative newb. Living totally off-grid in the desert, full time student with 2 cats (who need to stay cool in the summer). Basically, I have a lot on my plate. When I built my system (16S 280AH) almost a year ago, I didn't top balance based on recommendation of a friend with experience. He said with quality cells, its not necessary....

Well, here I am with one cell consistently running (it usually reaches 3.4 when the rest of the pack is 3.375-3.38). Sooooooo....I'm willing to try anything before I go to the step of disassembling the battery, packing it up to a friend's house an hour away with grid power, doing my best to baby sit the process but I won't be able to be there all the time, trying to manage having only limited power at my house (I have a friend who could share solar power with me, but a much smaller system). you get the picture - not enticing, but I'll do it if necessary.

Things to try before I go there (and looking/hoping for comments feedback suggestions)

I just installed Victron 150V60A SCC. I've only had 2 days to mess with it. So far loving the control. The first day, I set Absorption too high (only 55.0V but my runner started running). Today, I was experimenting with much lower Absorption settings, like 54.2V and then slowing stepping up to 54.6V

It seemed to allow the cells to come closer together - delta went from 0.025V to 0.007-9.

But, at the step from 54.4 to 54.5, the pack seemed to stop coming up to meet the SCC. Or - I could be totally misunderstanding what I was observing. Ran out of sunlight, and I know that at these low voltages, delta isn't all that meaningful. Still - 0.007 is lower than 25....?

1) is this a horrible strategy? Should I continue with this in coming days and see what happens? I know (or think I know) the big difference here is the cells are in series not parallel like in true top balancing.

My other idea to try to avoid dissassembly - manually discharging the cell that's the runner. From what I can gather, this process is very unscientific. basically, connect a load (resistor, light bult) across the high cell and let it discharge some to meet the pack. With only V reading on DVOM as my guide.

Ok, like I mentioned - relative newb here. Forgive me if silly questions.

2) In order to do this, I'm thinking the safest thing is to disconnect the cell from the rest of the pack. Just wondering if anyone can say if that's necessary? What would happen if I tried to discharge the cell while still in series (just trying to understand).

Thanks for reading and thanks in advance for any help

 

[Ampster]

I was experimenting with much lower Absorption settings, like 54.2V and then slowing stepping up to 54.6V

Yes, as you found out that will allow the balancing to reduce the delta. Remember with the low balancing currents and the size of your cells it may take some time. But at least you are seeing progress.

 

[Jesshaines]

Cool!! Thanks so much

 

[Daddy Tanuki]

Get an active 5 amp balancer (heltec for instance) use it to balance then remove and save until needed again

 

[Jesshaines]

Get an active 5 amp balancer (heltec for instance) use it to balance then remove and save until needed again

Thanks.

Have you had success with this? I do have an active balancer. I watched OGG video about this. It seemed his conclusion was not to balance while charging. But if I only use the active balancer while not charging, it's only about 1-5 minutes before the runner sell settles back down to below 3.4

And when I tried using the balancer while charging it seemed to make it worse not better

 

[Ampster]

But if I only use the active balancer while not charging, it's only about 1-5 minutes before the runner sell settles back down to below 3.4

That is because at that voltage the curve is very flat and voltage is not an indication whether that cell is balanced. You have to balance further up the curve when charging. That is why it is called "top balancing".

 

[Hedges]

Assuming you have BMS which will protect individual cells at about 3.65V,
You should be able to charge the pack until BMS disconnects.
Then, either bleed off the high cell with a resistive load, and repeat charge/bleed until all around 3.65V (as you mention),
Or, use voltage regulated 3.65V charger to bring up the low cells individually.

But maybe you don't really have a problem. At 3.4V +/-, I'm not sure the cells have particularly different SoC.
You haven't said you capacity tested and found insufficient capacity before one cell hit low voltage disconnect.

So long as your SCC limits to a voltage where BMS does not disconnect, could be able to simply use the system. Passive balancer of BMS may eventually get them closer and let you increase charge voltage slightly. Maybe just bleed the high cell, then adjust voltage and live with it for a while.

 

[Jesshaines]

True - I haven't brought the pack down all the way to LVP to see what happens. I've been down to around 30% state of charge as calculated by the BMS. If I remember correctly cell div was quite high, so I took that as a indication.

My old SCC I believe was malfunctioning. It wasn't entering absorption / float when I told it to. So maybe patience patience patience is the answer. Get the victron with settings where no OVP is going to happen and wait and see.

Generally when the runner sell hits 3.4 the rest of the pack is still at 3.37 to 3.38.

Thanks for your input!

 

[Ampster]

Generally when the runner sell hits 3.4 the rest of the pack is still at 3.37 to 3.38

And 3.4 volts is when balancing usually starts depending on settings.

 

[Jesshaines]

Then, either bleed off the high cell with a resistive load, and repeat charge/bleed until all around 3.65V (as you mention),

So I just want to confirm that I'm understanding the correct procedure for that.

Is panels disconnected and SCC off enough to do this procedure? Or should the battery be disconnected from the rest of the system at that point?

 

[Hedges]

Usually when someone balances everything would be disconnected, but could try in situ if you want to keep loads powered.
Just leads clipped onto busbars of that cell and brought out to a resistor ought to do it regardless of charge/discharge. Difficulty would be harder to compare states of cells. Do some math and time it to remove 1 Ah, then a couple Ah, pausing to observe behavior as it gets back to full charge before doing some more.

The most important thing of course is make sure nothing can short cells. No metal jewelry and any metal clips insulated so they can't bridge between terminals.

 

[mikefitz]

As suggested you may not actually have a serious issue and the BMS, given time will balance. Most BMS are set to balance, typically, once the cell is over 3.40 volts, under charge and when the delta exceeds 15mv.

Bleed off the high cell when it's over 3.40, under charge, and in the completed battery that's charging.
A car filiment headlamp bulb, around 50 watts, is an ideal load, the bulb lights up significantly at 3.4 volts and indicates you are making a connection. Apply across the cell for several seconds whilst monitoring the voltage to get a idea of the duration required.
Make up some probes with insulated shafts to prevent accedantal shorting, and apply manually to the buss bars on the high cell.
If things happen too fast stop charging and asses the situation. The objective is to get all cells in the range 3.45 to 3.50 under charge.

 

[sunshine_eggo]

Lotsa good advice already, but I'm going to brain dump:

Ok, so I'm a relative newb.

And you may be suffering from paranoia.

Well, here I am with one cell consistently running (it usually reaches 3.4 when the rest of the pack is 3.375-3.38).

What the shit? We're calling cells hitting 3.40V while the others are at 3.37-3.38 a runner?

This sounds like you're just looking to find a problem even when there isn't evidence of one.

Things to try before I go there (and looking/hoping for comments feedback suggestions)

Set absorption to 55.2V and float to 54V. If the battery can attain absorption voltage without triggering charge protection because a cell has hit 3.65, you're done. If it can't reduce absorption until it doesn't trigger charge protection and set float to 0.1V below that.

Leave it that way for a week. Try 55.2V/54.0V again. Rinse and repeat.

I just installed Victron 150V60A SCC. I've only had 2 days to mess with it. So far loving the control. The first day, I set Absorption too high (only 55.0V but my runner started running).

No you didn't.

Today, I was experimenting with much lower Absorption settings, like 54.2V and then slowing stepping up to 54.6V

It seemed to allow the cells to come closer together - delta went from 0.025V to 0.007-9.

But, at the step from 54.4 to 54.5, the pack seemed to stop coming up to meet the SCC. Or - I could be totally misunderstanding what I was observing. Ran out of sunlight, and I know that at these low voltages, delta isn't all that meaningful. Still - 0.007 is lower than 25....?

Good you know this, but you seem to forget it in the next sentence.

1) is this a horrible strategy? Should I continue with this in coming days and see what happens? I know (or think I know) the big difference here is the cells are in series not parallel like in true top balancing.

My other idea to try to avoid dissassembly - manually discharging the cell that's the runner. From what I can gather, this process is very unscientific. basically, connect a load (resistor, light bult) across the high cell and let it discharge some to meet the pack. With only V reading on DVOM as my guide.

Ok, like I mentioned - relative newb here. Forgive me if silly questions.

It's not a horrible strategy, but you haven't presented data indicating a significant problem.

I suspect that you've regularly failed to charge into the balancing range, and your individual cell SoCs have drifted.

You need to regularly charge to 3.45V to ensure the batteries are allowed bo balance.

2) In order to do this, I'm thinking the safest thing is to disconnect the cell from the rest of the pack. Just wondering if anyone can say if that's necessary? What would happen if I tried to discharge the cell while still in series (just trying to understand).

God no. Simply just hit the cell with the resistor. 12V automotive lightbulbs are good options.

Concerning your MPPT:

 

[Jesshaines]

And you may be suffering from paranoia

True!! I'm a work progress. Better too careful than too careless?

Regarding the runner.... Well yeah it hits 3.4 and quickly spikes to 3.65 triggering OVP if I don't manage through settings. As suggested.

Thanks for your input.

 

[Daddy Tanuki]

True!! I'm a work progress. Better too careful than too careless?

Regarding the runner.... Well yeah it hits 3.4 and quickly spikes to 3.65 triggering OVP if I don't manage through settings. As suggested.

Thanks for your input.

don't worry everybody on this forum's sphincters' tighten up when they are a few millivotls off trust me a couple of years ago Sunshine Eggo was the same. ha ha ha .

 

[Berseker]

Ok, so I'm a relative newb. Living totally off-grid in the desert, full time student with 2 cats (who need to stay cool in the summer). Basically, I have a lot on my plate. When I built my system (16S 280AH) almost a year ago, I didn't top balance based on recommendation of a friend with experience. He said with quality cells, its not necessary....

Well, here I am with one cell consistently running (it usually reaches 3.4 when the rest of the pack is 3.375-3.38). Sooooooo....I'm willing to try anything before I go to the step of disassembling the battery, packing it up to a friend's house an hour away with grid power, doing my best to baby sit the process but I won't be able to be there all the time, trying to manage having only limited power at my house (I have a friend who could share solar power with me, but a much smaller system). you get the picture - not enticing, but I'll do it if necessary.

Things to try before I go there (and looking/hoping for comments feedback suggestions)

I just installed Victron 150V60A SCC. I've only had 2 days to mess with it. So far loving the control. The first day, I set Absorption too high (only 55.0V but my runner started running). Today, I was experimenting with much lower Absorption settings, like 54.2V and then slowing stepping up to 54.6V

It seemed to allow the cells to come closer together - delta went from 0.025V to 0.007-9.

But, at the step from 54.4 to 54.5, the pack seemed to stop coming up to meet the SCC. Or - I could be totally misunderstanding what I was observing. Ran out of sunlight, and I know that at these low voltages, delta isn't all that meaningful. Still - 0.007 is lower than 25....?

1) is this a horrible strategy? Should I continue with this in coming days and see what happens? I know (or think I know) the big difference here is the cells are in series not parallel like in true top balancing.

My other idea to try to avoid dissassembly - manually discharging the cell that's the runner. From what I can gather, this process is very unscientific. basically, connect a load (resistor, light bult) across the high cell and let it discharge some to meet the pack. With only V reading on DVOM as my guide.

Ok, like I mentioned - relative newb here. Forgive me if silly questions.

2) In order to do this, I'm thinking the safest thing is to disconnect the cell from the rest of the pack. Just wondering if anyone can say if that's necessary? What would happen if I tried to discharge the cell while still in series (just trying to understand).

Thanks for reading and thanks in advance for any help

get the cheaper JK 2A balancer of the 4amp neey active balancer and slap on the pack...it will balance the pack over time

 

[Warpspeed]

As Hedges says in post #7 just discharge the runner back down to average voltage with a load resistor.

More details in this thread where this approach worked successfully:
https://diysolarforum.com/threads/top-balance-struggle.65000/

 

[Skypower]

You’d best get it done and over with. The longer you keep bringing it close at the knee and not balance, the more it’ll put a bad memory and relationship to the other cells. Harder to fix. The key is being able to reduce the current so the balancer can work on it before you move on towards bulk voltage. Even with a 2 amp smart balancer it’ll take time unless you step in and intervene with load on the high and boost on the low( I’m impatient), but this takes vigilance. You can’t put a load or boost on a cell, walk away and get a soda. Disconnect those before you walk away. Make sure the monitoring is active and not frozen and you’re looking at realtime data. Things go nuts very quickly.

A better way is time and low current. If you’re not in a rush and got better things to do than babysit a battery, you can just install a Heltec balancer, which is pretty cheap and if you also buy an A30 Controller, you can leave it hooked up and not have to turn it on and off manually at balance voltage. The Neey active balancer (black slab pictured) is also very good, has an app, but costs more. It’s quicker at getting the lower deltas down fast but the Heltec is better at getting the big difference down fast. Both are good. Pick your poison.
Pictured; Brute force method, what I had handy about 400’ of #10 as a resistor and a bench top power supply. The easier, slower perhaps smarter method with balancers and time. In any case you need to be able to control your current so balancing is manageable for what ever method you take.