Battery bank mid-point problems

[incagarcilaso]

Hi. I desperately need to get some expert advice on this problem as I cannot get a solution from my supplier or the manufacture.

I have the following set-up: 

2 x Axpert 5Kva, 48V system

24 x 250 W Yuraku panels

24 x Enersys 12 OPzV 1500 ah

 BMV 702 monitor

Victron CCGX

Dedicated W7 vm on home server running AICC.

The problem is with a serious midpoint deviation on the Enersys batteries. These were a very serious investment and I want to take care of them to protect this investment but cannot get any sound advice on how to address the problem of different voltages on the top and bottom 12 of the whole 24-battery bank. This is what happens.

When the batteries are in discharge or charge the mid-point deviation is slight and within, I think, acceptable levels - my BMV reads about 0.7% and a few tenths of a volt. But when they get to near charged (with about 40 amp hours to go to full charge) this deviation grows to 3.5% and remains at this level all the time that they are fully charged  or nearly fully charged. If there is intermittent cloud it means that the upper bank swings between 25 V and 29 V many times. This cannot be good. This has been going on for about a year now, while I have been trying to get a response from the supplier or manufacture, both of whom say that it is not an issue. I have the values for charge set to float at 54 and bulk up to 56.4. The bulk charge only lasts for about 15 minutes. The bank that does not perform as well will never get above 27.2V, which means that to compensate the other bank is at above 29 V, which I think is far too high.

What I am going to experiment with, as a workaround but not a solution, is to reduce the bulk charge value to, say 55.2V, in the hope that the lower bank will still reach 27.2 and so the upper bank will only reach 28V. 

I don't know exactly what is causing the problem or how to solve it. The whole reason I bought the BMV 700 was to protect this investment but even with this info I can't do anything about it. Is it that the upper bank are accepting the charge too easily or that the lower bank has faulty batteries? The supplier came to read voltages and said that none to the cells were seriously faulty. He measured them during charge but not at the moment of charge that this suddenly happens (near full charge at around 40 amp hours).

What steps should I follow? I am not going to get any help from supplier or manufacturer so have to solve this myself. Even if I have to buy one or two new cells I will do this if it means I can save the rest. Thanks for any input.

[Guest]

Did you replace the fuse of the bmv ? Mine was also out then @Jaco de Jongh sorted it out for me

[Jaco De Jongh]

Hi Sir.

1 hour ago, incagarcilaso said:

24 x Enersys 12 OPzV 1500 ah

I see this is Tubular Gel Batteries so that will rule out the possibility of a Equalization charge if I understand things correctly. (If i am wrong about this someone else will jump in soon to correct me.)

1 hour ago, incagarcilaso said:

Even if I have to buy one or two new cells I will do this if it means I can save the rest.

I know that in a Lead Acid bank this can not be done if you have used your bank for a period of one year as you have used yours, the difference of the internal resistance of the old and new batteries will cause failure of the rest of the batteries. I am not sure if it can be done with Gel banks.

57 minutes ago, Manie said:

Did you replace the fuse of the bmv ?

Just try this. please? Just replace the 2 fuses on the BMV as we found the fuses to be the reason for a high midpoint reading. I am referring to the 2 x 5x20mm glass fuses supplied with the bmv.

 

[incagarcilaso]

3 hours ago, Manie said:

Did you replace the fuse of the bmv ? Mine was also out then @Jaco de Jongh sorted it out for me

No, I had no idea this could cause the issue. I will try changing the fuses, although not sure where to find them. If it were the fuses could the mid-point reading get progressively worse because that is what has happened over the past 12 months. Initially the deviation was a maximum of 2% and now it is 3.5%. Thanks also @Jaco de Jongh I see you make the same suggestion. I'll try that and report back. It would be great if it was a simple as that.

[Chris Hobson]

I think a problem with fuses is rare but cannot be discounted. Check the BMV reading against a good multimeter. I would recommend a good balancing system. You will need probably 7 HA02 plus one HA01 to effectively balance your battery string. One could pair up two 2V cells for a nominal 4V balancing system and you would need half the number of battery balancers.

 

[incagarcilaso]

25 minutes ago, Chris Hobson said:

I think a problem with fuses is rare but cannot be discounted. Check the BMV reading against a good multimeter. I would recommend a good balancing system. You will need probably 7 HA02 plus one HA01 to effectively balance your battery string. One could pair up two 2V cells for a nominal 4V balancing system and you would need half the number of battery balancers.

 

OK. I know very little about this battery technology. This is one of the problems. I have not been able to find equipment to balance them. It seems like the top bank that gets up to 29V takes the charge much better than the bottom bank, which never goes over 27.2 V but I can't work out why as a multimeter reading of all 24 didn't show any single cell with a poor reading - mystery. It also seems significant that this happens only when the batteries are near reaching a full charge and only if the input current is high enough. If the input current is not over 30 amps they stay at a more even voltage (but I suppose this is because they don't get up to the 56.4 V.

What exactly are the HA01s and HA02s? If you could let me know I'll see if I can use them to balance the cells out. Thanks.

[Chris Hobson]

9 hours ago, incagarcilaso said:

 It seems like the top bank that gets up to 29V takes the charge much better than the bottom bank, which never goes over 27.2 V but I can't work out why as a multimeter reading of all 24 didn't show any single cell with a poor reading - mystery.

Have a look at  each of the thin wires supplied with your BMV. Use your multimeter to check the resistance Ω. They should be the same. Each wire is fused (the lump in the middle). Have the fuse and its setting checked. If the cables are fine you are noticing the early stages of batteries going out of balance, which happens first in late absorb and then the time period gets longer and longer. I have also noticed it is worse when charging current is higher.

9 hours ago, incagarcilaso said:

What exactly are the HA01s and HA02s? If you could let me know I'll see if I can use them to balance the cells out. Thanks.

They are Chinese manufactured battery balancers.

http://www.ebay.com/itm/Battery-Equalizer-HA02-4-X-12V-Used-for-Lead-acid-Batteris-Balancer-Charger-/122138833727?hash=item1c700ab33f:g:ZIwAAOSwLF1X3ksV

If you are based in South Africa you can get them from @Chris Rossouw. They only start operating above 2.4V so with 56.4V as your bulk charge the won't work with individual cells. You could pair the cells up. I was mistaken so you will not be able to use a HA01 which is specifically designed for 12V.  You will get away with 4 HA 02 balancers. 

[superdiy]

Your batteries need balancing. I highly doubt that the fuses are problematic, since the readings are acceptable when the battery voltage is lower, but goes more out when the voltage rises - that is a typical indication of unbalanced cells.

As per the suggestions above you should invest in some sort of balancer, the HA02s should work and should definitely make a difference if you pair the batteries, but with 2V cells the best option is probably to balance each cell. There are many options available and a quick google might help. One of the suppliers, with a few different product options, is Balancell.

[___]

The reason for the mismatch is that the internal resistance of one side of the bank is higher than the other side, so on charging (since V=IR) the voltage swings up faster. If you could hold it there long enough the low cells might eventually catch up, but you will potentially be overcharging the others. This is where the balancers come in: They redistribute the charge a bit, usually bypassing some of it (ie it flows through the low batteries but past the high ones), thereby pulling the low ones up.

One trick you can do, which I have done myself in the past, is use a 12V or 6V battery charger. They usually have a transformer inside so they are completely isolated -- well they have to be if they are grid powered and sold off the shelf -- so there is no danger to using them on one battery in the string. What I did then was to put the charger on the low battery or batteries at night while the inverter was in bypass. This allows you to bring them up too. I won't say I was very successful with that process though, but in theory it should work.

[SilverNodashi]

What interconnects are you using?

Are all the bolts /  nuts secured properly? 

Can you take an ohm reading of each interconnect and compare them?

And, perhaps, take a multimeter and compere the Voltages with your BMV, especially when the readings are high - just to rule out a bad BMV / bad cables / bad fuses / etc. 

[superdiy]

21 minutes ago, SilverNodashi said:

What interconnects are you using?

Are all the bolts /  nuts secured properly? 

Can you take an ohm reading of each interconnect and compare them?

And, perhaps, take a multimeter and compere the Voltages with your BMV, especially when the readings are high - just to rule out a bad BMV / bad cables / bad fuses / etc. 

With one string of batteries, a bad connection somewhere will affect everything in the string, not only specific batteries / cells. Dissimilar interconnects, dissimilar cable lengths etc. etc will also not have any effect on one string. On parallel strings it is a different story though.

[SilverNodashi]

1 minute ago, superdiy said:

With one string of batteries, a bad connection somewhere will affect everything in the string, not only specific batteries / cells. Dissimilar interconnects, dissimilar cable lengths etc. etc will also not be an issue.

but does he have one set of interconnects, or two? I've seen guys running two sets of interconnects on those 2V cells. So, does it otherwise suggest a bad cell? Perhaps swapping the cells around, and monitoring the current good ones and current bad ones, to see if they change "roles"?

[incagarcilaso]

3 hours ago, Chris Hobson said:

Have a look at  each of the thin wires supplied with your BMV. Use your multimeter to check the resistance Ω. They should be the same. Each wire is fused (the lump in the middle). Have the fuse and its setting checked. If the cables are fine you are noticing the early stages of batteries going out of balance, which happens first in late absorb and then the time period gets longer and longer. I have also noticed it is worse when charging current is higher.

They are Chinese manufactured battery balancers.

http://www.ebay.com/itm/Battery-Equalizer-HA02-4-X-12V-Used-for-Lead-acid-Batteris-Balancer-Charger-/122138833727?hash=item1c700ab33f:g:ZIwAAOSwLF1X3ksV

If you are based in South Africa you can get them from @Chris Rossouw. They only start operating above 2.4V so with 56.4V as your bulk charge the won't work with individual cells. You could pair the cells up. I was mistaken so you will not be able to use a HA01 which is specifically designed for 12V.  You will get away with 4 HA 02 balancers. 

Thank you - this helps. I'll take a look at the BMV to see if I can do the checks you say. The imbalance does exactly start in late absorb but now after so much time, as you say, the period gets longer and it is indeed worse the higher the charge current, exactly as you say. Is this because of "bad cells" or is it just that they need balancing?

I looked at the HA02 but couldn't see in the description that they were for 2V cells, it only mentioned 12V batteries. If you are sure they will work with my 2V cells I will go ahead and get them. Don't understand why they won't work with the 56.4V bulk charge. If they start at 2.4V and 2.4 x 24 is 48.96V, what am I doing wrong with my calculations? At any rate I can change the bulk charge setting to whatever is needed. I think a bulk charge of 55 or so would also be fine with these elements.

[incagarcilaso]

46 minutes ago, SilverNodashi said:

but does he have one set of interconnects, or two? I've seen guys running two sets of interconnects on those 2V cells. So, does it otherwise suggest a bad cell? Perhaps swapping the cells around, and monitoring the current good ones and current bad ones, to see if they change "roles"?

I'm not sure if I have two sets of interconnects - are you referring to the interconnects for measuring the mid-point voltage or the interconnects between all the cells. I could take a photo of the bank of cells if that helps.

I thought about swapping 2 or 3 cells between the higher voltage bank and the lower voltage bank to see if that helps. Is that what you were referring to? 

[Chris Hobson]

7 minutes ago, incagarcilaso said:

Thank you - this helps. I'll take a look at the BMV to see if I can do the checks you say. The imbalance does exactly start in late absorb but now after so much time, as you say, the period gets longer and it is indeed worse the higher the charge current, exactly as you say. Is this because of "bad cells" or is it just that they need balancing?

 

No they should not be bad cells at this stage just that they are out of balance. Over time the could turn into bad cells.  It may take some time to get them back in balance with balancers just be patient. It took my 260Ah batteries about 5 weeks to get back into balance.

11 minutes ago, incagarcilaso said:

I looked at the HA02 but couldn't see in the description that they were for 2V cells, it only mentioned 12V batteries. If you are sure they will work with my 2V cells I will go ahead and get them. Don't understand why they won't work with the 56.4V bulk charge. If they start at 2.4V and 2.4 x 24 is 48.96V, what am I doing wrong with my calculations? At any rate I can change the bulk charge setting to whatever is needed. I think a bulk charge of 55 or so would also be fine with these elements.

They specify 2.4V and my calculator differs from yours (2 x 24 = 48 and 4 x 24 = 96 divided by 10 equals 9.6 so that is where I get 57.6 higher than your bulk charging rate. If however you pair cells you effectively have a 4 V battery. Have a look at the balancing system @superdiy suggested. I see that the HA02  minimum voltage lockout is 1.8V so I am sure they will work with 2V but I am not sure. Contact the manufacturer @Chris Rossouw has their details. If not you may have to raise your bulk voltage by 1.2V if you batteries will allow it. I am sure that HA02s will work you may have to pair cells but that may not be necessary.

[Chris-R]

I have just requested the details from the manufacturer. I am sure it will work on the 2v cells, but let's confirm.

[Chris Hobson]

11 minutes ago, Chris Rossouw said:

I have just requested the details from the manufacturer. I am sure it will work on the 2v cells, but let's confirm.

Thank-you

[SilverNodashi]

35 minutes ago, incagarcilaso said:

I'm not sure if I have two sets of interconnects - are you referring to the interconnects for measuring the mid-point voltage or the interconnects between all the cells. I could take a photo of the bank of cells if that helps.

I thought about swapping 2 or 3 cells between the higher voltage bank and the lower voltage bank to see if that helps. Is that what you were referring to? 

Does your battery have 1 "+" and 1 "-", or two sets? If there's 2x "+" and 2x "-", or groups of "+" and "-", does it have one cable / plate between each battery, or two?

[superdiy]

1 hour ago, SilverNodashi said:

but does he have one set of interconnects, or two? I've seen guys running two sets of interconnects on those 2V cells. So, does it otherwise suggest a bad cell? Perhaps swapping the cells around, and monitoring the current good ones and current bad ones, to see if they change "roles"?

2 Sets of interconnects will still not make any difference on one string of batteries/cells.  The interconnects in parallel between 2 terminals just halves the resistance or doubles the current rating of it, but since it is one string, if you have a problem with one of the interconnects it will affect the current flowing through the whole string.

Even if you have one bad cell, it will also affect the current flow through the whole string.

Bottom line is that they are not charged equally - nothing new - happens to the best of them.  You'll have to get them all charged to the same level and the best way to achieve that is by means of some form of balancer which will either take away from the fuller cells or take away from the fuller cells and add to the emptier cells.

 

[SilverNodashi]

12 minutes ago, superdiy said:

2 Sets of interconnects will still not make any difference on one string of batteries/cells.  The interconnects in parallel between 2 terminals just halves the resistance or doubles the current rating of it, but since it is one string, if you have a problem with one of the interconnects it will affect the current flowing through the whole string.

Even if you have one bad cell, it will also affect the current flow through the whole string.

Bottom line is that they are not charged equally - nothing new - happens to the best of them.  You'll have to get them all charged to the same level and the best way to achieve that is by means of some form of balancer which will either take away from the fuller cells or take away from the fuller cells and add to the emptier cells.

 

I don't agree with this, fully. IF it was two set of interconnects, and one of the interconnects are loose, the current doubles, as you say, and that cell charges differently from the rest. 

[Chris Hobson]

2 minutes ago, SilverNodashi said:

I don't agree with this, fully. IF it was two set of interconnects, and one of the interconnects are loose, the current doubles, as you say, and that cell charges differently from the rest. 

It is like an accident on the highway. It restricts flow and the whole highway slows down. Unfortunately the electrons cannot speed up like cars do once past the constriction since  they are bumper to bumper.

[incagarcilaso]

1 hour ago, Chris Rossouw said:

I have just requested the details from the manufacturer. I am sure it will work on the 2v cells, but let's confirm.

Thanks, will be good to confirm that. I've been hunting all over but this is about the only model that I can find so hope it will do the job.

[incagarcilaso]

36 minutes ago, superdiy said:

Bottom line is that they are not charged equally - nothing new - happens to the best of them.  You'll have to get them all charged to the same level and the best way to achieve that is by means of some form of balancer which will either take away from the fuller cells or take away from the fuller cells and add to the emptier cells.

 

Yes, and if balancing solves the problem than that's great. I was also trying to find out exactly why this happens, although it now seems that it is not uncommon. If it's just a question of balancing that's find but if it's a bad cell I certainly don't know which one because there are all showing a very similar voltage. It's true that the readings show one or two tenths of a voltage lower across the board on the 12 that are not charging so high, which adds up to the 1.4 volts difference between the two banks. What confused me is why it didn't always show this, only on the last bit of the charge and only if the charging amps were high.

Here is a snapshot of what's happening today. Not much voltage input when reaching almost fully charged so the difference between the top and bottom is not so high. But if there was sufficient charge to push them up to the 56.4 bulk charge level the mid-point deviation jumps to 3.5%. About 8 months ago the worst mid-point deviation reading was 2.3%, so they are getting worse.

[superdiy]

27 minutes ago, SilverNodashi said:

I don't agree with this, fully. IF it was two set of interconnects, and one of the interconnects are loose, the current doubles, as you say, and that cell charges differently from the rest. 

 

22 minutes ago, Chris Hobson said:

It is like an accident on the highway. It restricts flow and the whole highway slows down. Unfortunately the electrons cannot speed up like cars do once past the constriction since  they are bumper to bumper.

Like @Chris Hobson said, one cell in a series string won't charge differently because of a bad connection or alike. If the cells were in parallel, a bad connection etc. will affect the cell with the bad connection and it will be charged and discharged differently to the other parallel cells. Same rule applies to parallel batteries and battery strings - a battery is in essence just a string of cells.

[superdiy]

21 minutes ago, incagarcilaso said:

Thanks, will be good to confirm that. I've been hunting all over but this is about the only model that I can find so hope it will do the job.

Have you googled yet, I've already posted a link earlier and there are other products available as well.