Problems with battery power system

[MikeMl]

I know a lot about care and feeding of individual 12V flooded cell batteries, but not much about living off grid. I have a feeling that it would be difficult to keep a long series string of batteries matched, so that all the batteries have equal capacities, especially as they age. As batteries age, the failure mode even in a series string of 6 cells (i.e. a 12V battery) is always that one bad cell develops, which necessitates replacement of the other five cells. I believe that from a battery utilization that a parallel connection of batteries would be better than connecting them in series. I understand that inverters and wind generators might be more efficient at higher voltages, but that requires great effort to keep the batteries matched...

 

[vielle568]

Hello Mike,

I understand what you're saying but I'm blocked by the design of the system. The generator puts out a mean voltage of 240 volts and so I assume it requires a compatible bank of batteries. If this output voltage can be dropped to (say) 12 volts then the 20 batteries can be connected in series... ...but the inverter design will need to be modified to accept the low voltage input.

Meanwhile, I have some readings from the hydrometer for some (not all) of my battery bank. There are 8 batteries listed below; 6 cells each. Is there anything here that gives an indication of a bad cell? I noted for example on battery D4 that two cells were much lower than the others; is this the sort of thing we're looking for? Also, D1 and E4 are at very different levels; is this a sign of a problem?

Hydrometer Readings:
Battery D1: 1.15 1.12 1.13 1.14 1.13 1.13
Battery D2: 1.16 1.18 1.18 1.18 1.18 1.19
Battery D3: 1.19 1.19 1.18 1.17 1.18 1.16
Battery D4: 1.18 1.16 1.17 1.16 1.12 1.10
Battery E1: 1.15 1.16 1.18 1.16 1.15 1.18
Battery E2: 1.14 1.15 1.16 1.14 1.14 1.16
Battery E3: 1.20 1.19 1.20 1.19 1.19 1.18
Battery E4: 1.23 1.22 1.22 1.22 1.23 1.23

I shall finish off the readings later this evening; there are 120 cells to check; it takes a little time. The batteries are relatively young. They haven't yet been in service for a year and during that time the system has often been disconnected for work/modification. They should be in good shape.

 

[RODALCO]

In the power utility industry we use often battery banks of 110 Volts 100 or 200 Ah which consists of 9 x 12 Volts batteries in series.
The continuous float charging voltage is 121.5 Volts for a bank of 9.

In your case of 20 you need to go for a charging voltage of 20/9 * 121.5 is
267.3 Volts as you normal charging voltage.

When discharge tests are done on those banks ( yearly) the charger is turned of and a 60 % load is applied for 1 hour ( this is quite an arduous task for the batteries).
For a 100 Ah bank this is 60 Amps continuous for 1 hour!

At the start of the test, each battery is measured then in 5 or 10 minute intervals, readings are taken and recorded.

If batteries are faulty very soon gassing becomes obvious and voltage of faulty batteries will drop usually rapidly.

Ok start your tests as follows:

Firstly start with the charger running (it will show up overvoltage on suspect batteries)

OK, above tests are maybe to arduous in your system, but try a 10 Amp load and monitor each battery and record the voltages.


And as already mentioned by other members, check for loose connections between batteries.

 

[MikeMl]

...
Meanwhile, I have some readings from the hydrometer for some (not all) of my battery bank. There are 8 batteries listed below; 6 cells each. Is there anything here that gives an indication of a bad cell? I noted for example on battery D4 that two cells were much lower than the others; is this the sort of thing we're looking for? Also, D1 and E4 are at very different levels; is this a sign of a problem?

Hydrometer Readings:
Battery D1: 1.15 1.12 1.13 1.14 1.13 1.13
Battery D2: 1.16 1.18 1.18 1.18 1.18 1.19
Battery D3: 1.19 1.19 1.18 1.17 1.18 1.16
Battery D4: 1.18 1.16 1.17 1.16 1.12 1.10
Battery E1: 1.15 1.16 1.18 1.16 1.15 1.18
Battery E2: 1.14 1.15 1.16 1.14 1.14 1.16
Battery E3: 1.20 1.19 1.20 1.19 1.19 1.18
Battery E4: 1.23 1.22 1.22 1.22 1.23 1.23

I shall finish off the readings later this evening; there are 120 cells to check; it takes a little time. The batteries are relatively young. They haven't yet been in service for a year and during that time the system has often been disconnected for work/modification. They should be in good shape.

Attached is the table that I referred you to from Trojan Battery Co. Make note of the temperature corrections, but it appears from your measured SGs, your batteries are all over the place, with 1.10 being only 20% charged, and 1.23 being 80% charged. If I saw this much variation in SGs in my aircraft or boat batteries, I would conclude that they need to be equalized.

Letting batteries sit for months without periodically recharging them to near full charge is the worst thing you can do to them. Flooded lead-acid batteries loose 10 to 15% of their capacity per month if left idle. Partially discharged batteries form hard suphates which are almost impossible to remove. The only thing that prevents formation of sulphates is to keep the SG of the acid at 1.25 or higher. One way to keep the SG high is to periodically (~every week) charge them to where they begin gassing. Another way is to float charge at <2.15VPC, which doesn't add charge, but prevents self-discharge. Float charged batteries should be equalized once every two or three months.

 

[vielle568]

Here's the full list of SG readings and voltages from the battery bank. Yes, the figures are all over the place. I'm not sure what it's indicating but it certainly isnt correct. There are three batteries now showing a low voltage reading (B4, C3 & E2), however these three were all up above 13 volts when the system was set running a couple of days ago. In other words, there was no obvious sign of any faulty cell at that moment in time.

MikeMl, you mentioned "equalising" the batteries. As the batteries are far from being "equal" I suppose this is what should now be done. Could you please explain how do I go about it? Thank you.

Vielle568

P.S. The temperature today (for the battery readings) was 22°C or 72°F

 

[Sceadwian]

Equalizing is giving the battery more current even after it's fully charged, just past the point where it starts to gas for a short period of time.

 

[MikeMl]

Here is a good description of Equalizing from **broken link removed**. I have found that by floating to prevent self-discharge on "floated" flooded lead-acid batteries, and doing an equalization on "floated" batteries about every six months gets a five year life on batteries that I was previously replacing every two years.

I keep a boat and a truck at Lake Powell, where they sit unused for months at a time at a location where there is no AC power. The summer temperatures there get to 110degF which is death to batteries due to accelerated self discharge at high temperatures. Batteries would die in less than two years.

I bought a 1/2A solar panel in 2002, an use it to float charge three batteries at 13.2V while the sun is shining. Since installing the float system, I have been getting five years out of the batteries. To equalize the batteries, I modified the alternator voltage regulator on the boat to put out 15.5V when I flip a toggle switch. Running on the lake for a hour or so each, the 65A alternator will equalize the two boat batteries. I use a marine deep cycle battery in the truck used to launch the boat, and can swap it into to the boat to equalize it.

 

[vielle568]

Thanks MikeMl and Sceadwian for the information.

I am going to have to get myself a charger capable of equalizing; the two chargers I have aren't capable of performing this task. Once I have the new charger I'll have to equalize the batteries one by one, but I'll do this leaving them all connected in series and connected to the generator. It will take quite a while to charge 20 batteries individually and if I disconnect everything the first charged batteries could start degrading due to the long delay. If all is left connected there'll be a charge coming in from the generator to keep the batteries in a healthy state.

I can see how you used the solar panel for the float charge for your boat batteries; it sounds like a very good idea. The alternator modification too seems a very practical way of maintaining the batteries in good condition. Thanks again for your information and for your help in this matter.

Regards, Vielle568

 

[MikeMl]

To equalize your entire string at the same time, you would need a current-limited (at about 3A) power supply with an open-circuit voltage of 310V (2.583VPC). That would not be that hard to do. Finding the transformer would be the toughest part of the project. Rectifiers are readily available.

I have built a current-limited unregulated power supply for equalizing 12V batteries. I used a 120V to 24V transformer, and then wired a 120V 200W lamp in series with the primary of the transformer as a ballast resistor. I used a 20A full-wave rectifier bridge between the transformer secondary and the battery. There is no filter capacitor; the current into the battery is full-wave rectified pulses at 120Hz (twice the AC line frequency of 60Hz).

The open circuit output voltage is about 25Vdc, but once connected to the battery, the battery sets the voltage as the supply is current-limited. During equalization, I see the battery voltage climb from 12.6V to about 15.5V.

The attached picture shows the charger. It is part of a computer-controlled battery tester I use to determine the AH capacity of aircraft batteries. There is also a 120Vac Solid State Relay in series with the transformer primary so that I can turn the charger on/off using a bit in the computer's parallel port.

The ballast lamp glows (less than full brightness) as long as the supply is connected to the battery. I put a 3A full-scale ammeter in series with the positive lead to the battery. I can get equalizing currents of 1A to 3A depending on the Wattage of the ballast lamp. I wired in a standard house light socket, so it is easy to screw in different ballast lamps.

If you wanted to equalize the entire string at the same time, you could start with a 240V to 480V industrial control transformer rated at least 1000VA. Such transformers are readily available here. The rectifier would need to be rated at more than 1KV at 10 to 20A. A 240V-50W to 200W lamp would be used as the ballast resistor in series with the transformer primary.

 

[vielle568]

OK Mike, if you can equalize the whole string in one go then the obvious choice is to try to build a suitable charger to do the job. I'm not sure what I can find over here in europe; I'll have to search on the web and see what I come up with. I'll be over in London for next week so maybe I'll dig something up there. I'll have more opportunity there than here in the French countryside!

Thanks once again for all your help and the information. I might even get this project up and running properly one day!

Regards, Vielle568

 

[MikeMl]

Reread my previous post again. I was editing while you were responding...

 

[vielle568]

OK Mike, I'll take another look

 

[vielle568]

I see the layout of your charger/equalizer. It seems fairly straight forward. Basically a high powered charger to bring the battery chain above maximum voltage for a while to create gassing in order to help clean the lead plates.
I'll have to get busy.....