What would you do?

[grizzzman]

Combination of many things.
Cut off if total V is too low V.
Cut off if cell V is too low.
Cut off if total V is too high.
Cut off if cell V is too high.
Counting Ahr is more for state of charge. (SOC)

Thanks Bruce. I see I was not clear enough OK I will be more clear (hopefully) Yes cell level high and low cut off. What I was not clear enough with is during charging as an example cut charging at say 3.45 per cell or13.8 pac ( or there about ) as it is best to stay out of the knee's, when do you decide too turn the charging back on? Voltage? Amp hours down? To low and won't harvest enough? (ya never know what the next day will bring?)
Thanks

[native pepper]

My re-start charging begins when the pack drops to 13.6v, but that rarely happens as when the pack is full, the input is directed to the load. It's only low input days that the pack will get used when full, as the input doesn't correlate with the output and the pack compensates, then when it drops to 13.6v, on comes the charger.

That way at night, we almost always have a full pack and during the day when the pack is full, we use the direct input as much as we can. Others will be different and don't know if their charging system work that way, or take their energy always from the pack.

[T1 Terry]

Unfortunately there is no way of knowing for sure all cells will reach 3.45v at the same time, so the pack voltage might still be 13.8v but one cell could be 4v while the others are still around the 3.3v and 3.2v mark. Less likely to happen at 13.8v granted, 14.4v is very likely as all the remaining cells could be over the 3.45v mark and one cell at 4v, any one of the 3 remaining cell running a little behind and the 4v cell would be well into its death spiral.
Cell balancing alone does no help here as the cell voltage run away only occurs once a cell reaches fully charged, it can't store any more current so the voltage climb is near vertical. This is why all charging must be controlled at some point by monitoring cell voltage and stopped if a cell reaches 3.6v and no charging recommencing before that cell has dropped below 3.6v.
The time lapse between charge off and charge back on really depends on the load on the battery pack as well as the rate of charge. This part will probably have your eyes rolling back in your head and the risk of you dropping into a coma, so maybe best to get a coffee before reading any further.

Ok, when a cell is below the 90% charged mark it can accept a very high current rate up to 3CA (3 x the advertised capacity read as amps or 300 amps per 100Ah advertised capacity), well beyond what would normally be seen on a house battery pack. After the 90% full mark the voltage will rise rapidly if the current acceptance rate is not as high as the supply rate. Each cell will be different to the next in just which point this max acceptance rate is reached, all to do with internal resistance and actual capacity and how well the cell was condition charged but high rate charging will see a cell voltage run away happen where it is less likely at a lower charge rate...... I can see the heads rolling back already so I won't go any deeper on that bit.
At a trickle charge rate the run away doesn't occur until the cell reaches 100% full or saturation charged, then even at a 5 amp charge rate the cell voltage will climb rapidly, just not quite as rapidly at it would at a high charge rate. One cell only needs to be 0.1% ahead of the others to reach a voltage run away point while the others are still charging, a set voltage of 13.8v that turns back on at 13.7v will belt 5 amps in until the total pack voltage reaches 13.8v.... so what?
This is what happens, cell 1 reaches 3.45v, cell 2 = 3.36v, cell 3 = 3.6v, cell 4 = 3.39v. Now we have 3.45 + 3.36 + 3.6 + 3.39 = 13.8v, charging stops. cell 1 is nearly full so it will probably drop to 3.4v, cell 2 hasn't reached the nearly full mark so the charge is still spreading evenly through the cell, it will probably drop to 3.2v, cell 3 is saturation charged and has started the first steps to voltage run away, it is unlikely to drop below 3.5v, cell 4 has not reached fully charged so it will probably drop to 3.3v. Now we have 3.4 + 3.2 + 3.5 + 3.3v = 13.4v. This means charging has recommenced at 5 amps until a total of 13.8v is reached and the charge stops again. Not really a problem for cell 1 just yet, cell 2 will continue to accept charge, cell 3 is already full so voltage run away will be in full swing and cell 4 will be accepting charge. Now the make up will see all the other cell voltages lower and cell 3 will have run away to a voltage that will be causing damage. Why? Cell 3 could not accept any more charge so the voltage climbed immediately, the other cell were accepting charge so their voltage didn't climb much above their settled voltage. Effectively each time charge resumes the lower cells get lower and the high cell gets higher to equal the same 13.8v.

Unless the cell voltages are monitored and charging stops if a cell's voltage starts to run away the above scenario is a real possibility. Equalising can not shift this amount of out of balance as it only occurs at the top end of the charging cycle. Simply shifting capacity from one cell to another could make the problem worse, if the problem cell has lost capacity or has a high resistance connection it will show a lower voltage while discharging and the balancer will try to move capacity from the other cells to this lower voltage cell. Now the actual capacity of the cells has been altered making the cell that showed the lower voltage slightly fuller than the other cells, it will see saturation charged before the other cells.
Stop the charging at 13.8v or even 14v OR if a cell reaches 3.6v and not recommence charging until either high voltage has been cleared and you won't have a problem. Start balancing at this point and stop balancing once charging resumes and the out of balance problem will slowly disappear and the primary control of 13.8v or 14v will take over and everything will return to normal.

T1 Terry

[JohnM]

How do you decide when the DC to DC charger needs to be cut off? Or for that matter back on? Using voltage? Counting amp hours?

In my case my main DC DC Charger is connected via a relay from the ignition, so only comes into play while motor is running. I relay on the inbuilt charger algorithms to control the charge to the battery, but then I am working with AGM Batteries.

[native pepper]

Interesting Terry, having watched my packs charging constantly over a number if years, would agree with you, regarding using bms, active balancers (bleeders) and relays. Until we got the cell equalisers, now every cell is exactly the same from low voltage to full. Yesterday when we had mostly sun all day, the house pack went from 13.21v to 14v and according to the meters, every cell sat at the same voltage during the whole process. The balancers can move 8amps from cell to cell and as I was working on installing the A/c, was right beside the pack. It seems to me with my meager knowledge that the cells actually get a memory (which have been told is the case in a very small way) and the longer they are on the type of systems used, the more they react to that system in the same way. Have checked all the cells in each line on a number of occasions to see if any of them differed, yet still gave the right cell line voltage and none were different. That makes me happy as to me it shows my system is working as it should and i don;t really have to watch it, if any cell line does go astray, the cell alarms will let me know and they never have since setting this up with equalisers.

Could be terribly wrong, but before getting the equalisers the cells reacted the same way, the same cells would be over the other cell voltages, or under. When first installing the equaliser, they continued to react as before for the first couple of chargers and even thought they were draining the cells, but was told to persevere. Then they changed over another couple of charges and now the cells are all the same voltages during charge and all end up exactly the same when the charger switches off at 14v, 3.5v per cell.

Bit off topic this, but just shows what a buying group can achieve. Our group has been offered brand new Samsung UN65JS9500 Curved 65-Inch 4K Ultra HD 3D Smart LED TV, for less than $1000 delivered with our next order, 3 month replacement or refund and 3 years warranty. So it looks like our next order will be moved forward to this week so we can get some of those, just hoped the others want one as we have to take a few to get the price. The say it's a xmas gift for our long term business, I think it's more to do with moving stock as the chinese financial year end 31st December. So will have to get my act into gear to get it done before the 31st, they also offered us ipod6s for unbelievable prices, but wouldn't buy apple if someone paid me, but the others may.

[T1 Terry]

If you are confident the cells are now balanced and it is not the balancer working in the background, disconnect the balancer and see what happens over the next week or mth. This will tell you if the equaliser is hiding a problem or if the problem was that cells were never equalised

T1 Terry

[native pepper]

Done that Terry, the only time the cell lines get out of balance is when they are at 3.45v and the difference between them all is less than .3v. I charge the pack from the 4 points of it, not just 2 (+ -), was told this would make it much easier to keep them in balance. On my portable pack, it's only 2 points as it is only 3 cells per line and it shows the biggest differences when the equaliser is taken off, but that is on one cell line at one end. Haven't changed any cell sin that pack and it';s been chugging along for over 8 years, yet the equaliser have only been on a year or so and the difference is great. I should make the connections for all 4 points and see what happens, will do that today and let you know what happens over a week with the equalisers on and a week with them off, once the heavy rain stops.

To my mind, when the equaliser is removed and no charge going in, wouldn't that show up any imbalance, if a cell was losing energy. Considering how long I've been off grid and on the road, getting got this stage is so good, when you consider the past hassles in keeping storage working and with lifepo4 balanced.

Even with the equaliser attached, and no charge going it, if there was a bad cell, wouldn't the pack lose voltage as the equaliser tried to compensate for the bad cell. Yet to see that in any of my systems, except when first connecting the equalisers for the first time and couple of chargers with them on.

[T1 Terry]

No, the equalised could not carry a dying cell, but it will cover a connection resistance problem and if this is left without being rectified it will eventually lead to a permanent cell damage problem. If there is nothing wrong then the pack will remain in balance.
No idea what you mean by charging from all 4 points, any chance of a sketch to show how your batteries are connected up... or is there one already on another thread?

T1 Terry

EDIT: Fixed a typo that changes the meaning of the first bit

[native pepper]

ok here's my quick sketch of charging from all points, hope it makes sense and maybe this is how everyone does it.

[T1 Terry]

I think I see what may be causing your out of balance. The charging current and load current must flow across all cells in parallel before the series link to the next parallel string. If the cells are wired any other way the cell with the lowest internal resistance will carry all the current as that is the easiest path and the other cells in the parallel string will be left tagging along rather than sharing the load.

Battery bank 360Ah size comparison.jpg

I hope this explains what I mean by the full length path through all the cells in parallel before the series link to the next string in parallel.

For those interested, the battery with A22 marked on top is a 125Ah deep cycle AGM 12v battery, the pair on the other side are 125Ah 6v batteries in parallel, the lithium battery in the middle is built up with 90Ah cells so a total of 360Ah @ 12v.

T1 Terry

Had a second look and that picture shows how not to do it as the current will flow through the 4 cells in series and not through the full pack.
The cell links between cell group one (the negative end) and cell group 2 should be at the other end, so should the link between cell group 3 and group 4, the only one in the correct place is the link between group 2 and group 3.
I'll add a photo of a 200Ah 12v pack that has the links in the correct places

8_cell_12v_200ah_2.jpg