Hi @aspiremr,
don't run the test until the battery is completely flat. This will harm the battery and reduce its life even further. Run it down to only 10V then take the load off, wait a few minutes, then measure the voltage with no load. Then put the battery back on charge.
The only real way of testing capacity is to put a load on the battery that is representative of the load you will have in practice and measuring how long it takes before the battery is at or around 10.5V. Once you have this information, you can make some estimates of how much charge you have left at any time based on how long you have been running and some idea of how much charge came in from the solar panels. Considering your load, 5A is a reasonable estimate, hence my suggestion of a 60W / 5A load. It was also fortunate that this matches a 20hr discharge that is used by manufacturers to rate the battery capacity.
You can also measure the no load voltage and that will also give you some indication of state of Charge (SOC).
Here are the figures I got from the internet and use for my Gel battery:
100% > 12.85V
75% > 12.65V
50% > 12.35V
25% > 12.00V
0% > 11.80V
I usually look at it in the morning when nothing is running and my battery is rarely below 12.6V.
But, these values are only correct if the battery is at 25C.
Can you point me at where you got the little table? This is quite different to what I use. Where 11.9V would be about 10-15% of capacity left, and your table states 40%. I would like to compare these results.
Remember, that this is the voltage measured with no load. The voltages you are measuring now with your battery are with a load connected. Hence the differences in the conclusions.
Great idea to take the battery to work to continue the testing. Once you know that the battery will last at least 8 or so hours, you can start the discharge test at say 9pm, take the first hourly measurement, and go to sleep, take a measurement in the morning, take the battery to work and continue measuring.
My quick guess is that the A battery is not going to last the full 20 hours, but hopefully more than 10 hours. If it was me, I would be happy to continue using them as long as I had more than 50%.
Also, a couple of good, recharge / discharge cycles may well increase the capacity some more.
cheers
Mike
Thanks @mikerezny ,
Ok, that all makes sense. Interesting to put a load on that is indicative of the load I usually expose my batteries to. Obviously with the parallel connection from one battery to the other being on the wrong side of the shunt, I'm not too sure how much I can rely on those measurements. But when I played with batteries this weekend and turned all internal LED lights on it was drawing apps 3 amps (according to drifter, fridge compressor cycled off). When I opened fridge door and got fridge to cycle on it was drawing apps 6.3 amps (according to drifter, all lights off). And when i had fridge cycling on and LED lights on it was drawing approx 8.7 amps. So 5 amps is probably a reasonable average.
Although, i think the fridge barely cycles on when i am out in the bush and the nighttime temperature is 2 or 3 degrees. Daytime is not much more than 15, and that is not for a very long time.
The table I was referring to I actually found on a thread in this forum somewhere. But may not have understood it or the context properly. Attached below.
View media item 290
@mikerezny,
Your table above says battery is empty at 11.8 volts, and less than 25% charge at 12.00 volts. But i should be running the load on the battery down to 10 volts, but not run it completely flat. Is this the difference again between a loaded and an unloaded reading?
Ta
MR
Hi @aspiremr,
this table seems more likely for a Lead Acid battery.
I think I got my table from here:
http://www.energymatters.com.au/components/battery-voltage-discharge/
That page has the table that has the same values as I use. It also has some supporting information that is worth reading while you are watching the clock.
cheers
Mike
Hi @aspiremr,
completely flat means that the voltage across the battery is absolutely zero. 0% capacity is when the battery, although not completely flat, at 10.0 or 10.5V has reached the end of its capacity to do any more useful work. Although I have not done this, I would expect that a battery that has been run down to 10V would recover to about 11.8V a few minutes after the load has been removed.
Yes, the difference is between the voltage with a load and determining the SOC which is always measured 30 or so minutes after the load has been removed.
cheers
Mike
Hi @aspiremr,
I found several sites with tables showing the same values as this table.
I am fairly well convinced the table is for a lead acid battery:
Here is one site that clearly refers to a lead acid battery fitted to a car:
http://carbatteryworld.com/car-battery-voltage/
So, for the time being, I will stick to the values I have been using.
cheers
Mike
Battery A summary.
Connected to 4.5 amp draw. Fully charged according to Ctek charger at start of test. Voltage reading at start of test 13.42 volts.
Battery seemed to lose charge in a fairly linear fashion until the last two hours, and it seemed to drop substantially faster in those last two hours.
Total time it managed before going under 10 volts (under load) was 8 hours.
55 watt globe at 12 volts equals 4.5 amps. 4.5 amps by 8 hours equals 36 amps.
Does this mean that this battery now has only 36 amps of capacity. Given it is a 96 Ah battery, that is only about 37% capacity. Is battery dead?
On removing load, the battery immediately read 12.00 volts. Will take another reading once it has been off load for a little while and see if this rises at all.
Thanks all again for the advice.
Will do the battery two load test tomorrow from work.
Connected to 4.5 amp draw. Fully charged according to Ctek charger at start of test. Voltage reading at start of test 13.42 volts.
Battery seemed to lose charge in a fairly linear fashion until the last two hours, and it seemed to drop substantially faster in those last two hours.
Total time it managed before going under 10 volts (under load) was 8 hours.
55 watt globe at 12 volts equals 4.5 amps. 4.5 amps by 8 hours equals 36 amps.
Does this mean that this battery now has only 36 amps of capacity. Given it is a 96 Ah battery, that is only about 37% capacity. Is battery dead?
On removing load, the battery immediately read 12.00 volts. Will take another reading once it has been off load for a little while and see if this rises at all.
Thanks all again for the advice.
Will do the battery two load test tomorrow from work.
Hi @aspirmr,
exactly! So battery A is delivering about 35% of rated capacity, 36Ah against 96h. But, as @Bellbirdweb has written, you may be able to improve on that by doing a set of repeated charge / discharge cycles. He may be able to give you some more information on how he has rescued batteries.
cheers
Mike
Basically doing what you have been doing.
Load the battery with a small load until flat (10.5v) and then fully charge and repeat.
If yours is at 35%, you may only get it back to 50% or so, but that can still buy you some time until you need to lash out and replace it.
Thanks @mikerezny ,
So if that means the battery is dead, I get back to the initial concerns. How does a battery die after so few uses? Bought brand new April 2015, used approx 25 to 30 times for 2 to 3 day free camps since then.
Clearly I must have been unwittingly abusing the batteries, or just got unlucky with a bad battery - we'll know that after I test battery 2. If it's the same then it would have to be really unlikely it is two bad batteries surely?
To some extent, the possible conclusions then are
- regardless of what my system is actually drawing when i am off grid, despite my inability to measure it, I am drawing more than these two batteries are capable of delivering, ; or
- The batteries have not been recharged properly by the Setec in between uses in between uses and hence they have never been at full capacity and hence the abuse is even worse than it might have been;
- The solar is not putting anything or anywhere enough back into the batteries when out in the bush.
1) batteries likely need replacing;
2) Battery cabling needs correctional work - thicker cable, paralleled correctly, shunt isolated on outside of parallel connections.
But, prior to doing either of these I now move on to trying to isolate any other problems - either with solar or other?
Sound about right?
On a positive note, my last weekend away the batteries coped without a low voltage alarm for 2 days and 2 nights, so if my battery capacity is as low as this test just demonstrated, that at least makes me feel that i can't be drawing too much of a load. Has to be inadequate re-charge between uses, and negligible benefit from solar, just need to isolate which one or possibly if both.
MR
Last edited:
Thanks @Bellbirdweb,
If all that effort is only still likely to get them back to 50% capacity, which I think is still pretty low, then I will probably replace them. However, I will persevere with the troubleshooting as suggested by everyone on here first. I will try to understand each one of the issues i have and correct each issue first so I don't ruin another set of batteries. But once I have confidence that the problems are each individually identified and a solution is at hand, then I'll replace the batteries.
From what I understand, if they are maintained properly with a good charger; if i have sufficient capacity to start with so they are not discharged too low; and if they are being charged adequately with solar during use, then they should last years. Especially given I only use the van maybe 8 to 10 times a year at most, always for short time frames.
But greatly appreciate the guidance on how to go about trying to recover them.
Load test battery 2 tomorrow. Then back to van next week to try and get to bottom of solar - heavy blankets and ladder at the ready.
MR
Hi @aspirmr,
still a bit of testing to go. But, IF both batteries are around 35-50% after a few recharge / discharge cycles. It is unlikely you were unlucky enough to have two faulty batteries. But it might be worth pursuing whether the batteries will be replaced under warranty.
It is worth a few phone calls.
Since the batteries still have some charge, it would be worth considering using them to investigate the rest of the system to find out what is happening. From these investigations, you will be in a very good position to make the next decisions.
When you get the batteries back in the van and wired as per the previous posts., I would suggest making absolutely sure the Setec is charging your batteries. You will be able to check that from the drifter panel (voltage and charging Amps) and checking with your multimeter. I would suggest it is still important to determine what is causing the problem.
Once that is sorted, you may well decide to not use the Setec to charge, and instead use your Ctek.
Then finally, I would suggest investigating your existing solar setup. Are the panels clean? What is the total charge in good sunlight. The charging rate on the solar controller should be about the same as shown on the drifter panel (with the fridge turned off). If it is only 7.5A or so, check that each panel is actually contributing by putting a blanket over each cell in turn and seeing how thatt affects both charging readings. Also look at the Drifter battery voltage reading during theses tests.
In any event, it would be worth getting a better controller, preferably an MPPT controller, and definitely one that has a more sophisticated battery charging system. At least a 3-stage charger.
cheers
Mike
Thanks @mikerezny
With regards to whether the Setec is charging the batteries or not, I had the Multi meter on them, and also compared with the Drifter panel this weekend.
When the 240 volt supply is turned on, and the battery bank is turned on at the drifter panel, the volt reading on the Drifter is 13.60 volts. The multi meter shows 13.49. However, the multi meter shows 13.49 when tested in one direction, and only 12.70 when tested the other way (e.g. +ve on B to -ve on A, versus +ve on A to -ve on B).
When I then turn off the battery bank and the 240 volt power, the Drifter reads 13.4 and the multi meter reads 13.34.
So there is definitely a voltage drop when 240 volt is removed.
Does this imply that Setec is charging, or is that not a definitive enough test until I get the batteries re-cabled correctly?
Thanks
MR
It's a little bit of an inexact science @aspiremr.
The link that @mikerezny put up has a really good explanation of the discharge total vs capacity for a deep cycle gel cell.
I have had 4 years from some and 12 months from others
Hi @aspiremr,
the really good part of this post is how really quickly you are getting a handle on your setup.
here are a few questions:
Did you charge both batteries using your Ctek before you went away?
Last weekend, do you you have any estimate from your solar controller, how much solar charge went into your batteries?
What gives you the low voltage alarm?
Does this mean you got through the weekend without using the generator?
Just quick estimate. If your batteries are only 36Ah, total 72Ah, and you got negligible solar, and did not use the generator for 2 days then a reasonable estimate of your power usage is about 36Ah per day. What do you think?
cheers
Mike
Thanks @Bellbirdweb ,
I hear you re inexact science.
With regards to the 4 years versus 12 months from batteries, I am assuming that is with a much more frequent use pattern than I have? Or was that with regards to how long they may last after recovering them?
I had thought that the longevity of the battery was measured with reference to the number of charge and discharge cycles, of which my batteries would have had no more than 30 (guess). Even if I'm out by 100% and it is 60, that's still really low yes? Shouldn't it be in the hundreds?
Thanks
MR