PDA

View Full Version : Solar panels not keeping batteries charged up. Need assistance to diagnose.



spudboy
19th May 2016, 03:27 PM
Hi - I need the brains trust to assist with the best way to diagnose why my solar panels are not keeping my batteries up to charge in my camper, please.

The basic components involved are:
-6 x 100W ultra thin solar panels, connected in pairs, so feeding in as 3 separate inputs to the MPPT charger at around 30V under load
- A 30A GSL MPPT charger (Australian made)
- 3 x Optima D31M Blue Top batteries connected in parallel. They are 75AH / 900CCA. So 225AH in total.

We haven't used the camper for 3 months, so nothing is turned on (like fridge, lights, pump, whatever). It is stored outside in shady sun, so I was expecting that with practically nill usage it would only need a trickle to keep it topped up. The voltage alarm went off when it dropped to 11.2V, and the auto cut-out turned everything off.

The people who supplied the solar panels and the MPPT charger have been very helpful. They asked me to do some tests to try and work out if it is the batteries or the panels which are the issue. They don't think it is the MPPT charger. Because the batteries are about 3 or 4 (or 5?) years old, they are suspecting that one battery may be faulty of have some faulty cells.

I've started by fully charging the batteries from 3 240V chargers. It took over a day to get them back up to fully charged, from being at 11.2 volts!

This is what I did to test the batteries:
o Separated the 3 batteries
o Charged them all, using 3 battery chargers.
o After 3 or 4 hours I would 'rotate' the chargers to make sure each battery got charged with each different charger (they were different capacity chargers) and 2 chargers had an AGM setting but the 3rd one is old and only had one setting
o Let the batteries 'rest' for 4 hours
o Recorded the unloaded voltage
o Hooked up 3 identical LED spotlights to the batteries
o I measured the Amps on each light when I connected them (as I thought that might be relevant) and put these details onto the graph below for each light.
o Recorded the voltages on each battery over 20 hours

This is the result of the testing - a graph!:
https://www.aulro.com/afvb/images/imported/2016/05/396.jpg

Any thoughts on the goodness of these results? Was the test load (about 0.8A) sufficient for testing?

Next thing is to work out how to test the panels....

Thanks
David

spudboy
19th May 2016, 03:31 PM
Here's a block diagram of how it is built:

http://www.aulro.com/afvb/attachments/general-chat/109419d1463639486-solar-panels-not-keeping-batteries-charged-up-need-assistance-diagnose-solarwiringdiagram.jpg

Blknight.aus
19th May 2016, 05:48 PM
ok the testing isnt really valid, you did your best but there is a better way.

hook all the batteries in parallel, plug the biggest charger into the bank and let it have at it..

you must hook up in a star configuration on the positive so you have one point that feeds all the batteries you can then measure the amps going into each battery. let them sit on the biggest charger you've got for 24 hours, then disconnect them from each other. Now repeat the 4 hour sit and then test the volts and amps into a known draw.

with the way you have charged the battery they may have not all come up the same.

Assuming they have all received the same basic charge profile then the green battery has at least one cell that is close to dead or 2 or more cells that are being lazy.

depending on how exactly your panels are wired up an inputted one just one panel not putting out charge will limit the whole arrays out put, in reality Id expect you to have to have at least one in each pair down to really cut you out but if half of each panel is shaded and you're wired like I think you are then you loose all of your charging ability.

what you need to measure on the batteries is

1. volts and amps while charging
2. volts once charging stops
3. volts after 4 hours
4. volts and amps while under load

you also want a slightly bigger load. 5-10 Amps will be closer to what you want bu t the rough rule of Thumb I use for a test load is 10% of the capacity of the battery. so 7.5A in your case and you should hit 10 hours worth of life.

ozeraser
19th May 2016, 06:03 PM
A lot of solar panels don't work in partial shadow like monocrystaline, if a panel is in half shadow it can often stop producing.

Also a lot of chargers actually power themselves via the battery and if there is little to no power coming in, it will vampire the batteries. If your usage outweighs the incoming you will slowly flatten your batteries.

Just connect a single battery to the MPPT charger, measure the voltage across the battery when you expect it to be charging (full sun for a good test). If its less than 13.8 you have a bit of an issue. Amperage is a something to take note of but dont as most standard multimeters max out at 10A.

Disconnect the batteries to anything other than the MPPT charger, then test again.

Finally just start disconnecting panels as a single 100W panel should easily charge a battery over a few days (depending on the charger).

Next is to read the details on the back of the panel and then measure them in full sun, again however check what the amperage is from them as you could blow your meter.

Could you just take it to an auto electrician?

Sitec
19th May 2016, 06:17 PM
Just thinking from a different perspective here.. I'd park the truck in a different location.. With the shade of the trees and the arrival of winter the light has been pretty average of late even without shade....

spudboy
19th May 2016, 10:07 PM
Thanks for all those thoughts. Some answers in a random order....

Firstly the easy one - Simon:

I've parked the camper out in the full sun (even though I know it is low in the sky at this time of year) to see what input I am geting.

In the midday sun, with all 6 panels going into the 12V charger, I was getting 4.6A at 12.9 volts going in, which seems really low for 6 x 100W panels.

I've installed a 100A shunt to measure what is going in from the solar panels, and a digital meter.

spudboy
19th May 2016, 10:13 PM
Then Dave:

The 12V company that sold me the gear asked me to separate the batteries, then charge, then measure, so that's why I did it that way. I can easily do it again with them joined together.

So the green battery could be a dud? It's only 0.1V different than the others. Which might be pulling down the other 2? It is getting on now. I was hoping it to not be a battery issue as 3 new ones are going to be about $1200 :(

If it's a problem with the panels, that is better for me as they are under a 10 year warranty!

Following your comments, I've been charging it with my biggest charger, a 15A Ctek,with all 3 now joined together. It was pumping in 14.8A at 13.4V for a while, and then it went down to 4A at 14.1V. It is still charging and the Amps are going down bit by bit.

I will try the load test again over the weekend, using some bigger loads. I have some 50W 12V globes, so that'll draw 4A. Can't think of anything that draws 7A that I have 3 of! Guess I could use 6 50W globes instead of 3....

spudboy
19th May 2016, 10:20 PM
Then OzEraser:

When the 3 batteries are low, the voltage creeps up slowly over the morning, from (say) 12.9 to 13.6V over a few hours. Even if it's nice and sunny.

I have another (separate) setup with the same MPPT controller, but a lot more panels and only 1 battery, and that starts out at 14.2V in the morning, for an hour or two, then drops back to 13.4V for the rest of the day.

I can disconnect the 'output' circuit easily, so no power can be used by the fridge or lights (not that they are on).

The details of the panels are:

Panel Type: Mono-crystalline

Code: ZM-9154

Maximum Power (Pmax) 100 watts

Open Circuit Voltage (Voc) 21.70 Volts

Open Circuit current (Isc) 6.10 Amps

Max Power Voltage 17.70 Volts

Maximum Power Current (Imp) 5.70A

Module Efficiency 21.30%

Normal Operating Cell Temp (NOCT) 48+-2*C



It's not so easy to take it to an auto electrician, as it's not registered at the moment...

Blknight.aus
20th May 2016, 05:30 AM
to test your panels, hook a pair of your 50W bulbs in parralell to each panel and measure volts and amps in full sunlight.

to make 7.5A put 2 50w bulbs in series and then put that in paralell with a singe 50w bulb, its not perfect but its close enough.

given that your whole voltage range from fully charged to effectively flat is about 2v .1v is a 10% variance.

your amps from the panels and regulator will be lower now as you have charged the batteries from the mains. To test that properly you need to have either flat batteries or a dummy load that can draw the full output of the regulator.

drivesafe
20th May 2016, 10:34 AM
Hi David and first off, your batteries should never have been allowed to get down to 11.2v and I assume were left there for some time.

It is quite OK to cycle Optima batteries down to 10.5v as long as they have been discharged down to 10.5v in a short time, like over a few days, NOT MONTHS, and are recharged shortly after being discharged down to 10.5v. ( no lower than 11.58v for all other batteries )

Any battery that is left with a terminal voltage of less than 12.35v or an SoC of 75%, will begin to sulphate.

Optimas are far less likely to sulphate as quickly as most other batteries but left for long periods, they will and I suspect yours have.

While your test is a basic one, it will give you a very reasonable indication of the condition and state of charge of your batteries.

From the data you posted up, I suspect your batteries are about 15 to 20% down on their full potential capacity.

If your batteries are suffering from the effects of sulphration, CONDITIONING charging of your batteries may recover some or even all of your lost capacity.

You can use any 2 ( or more ) stage charger. It must have a BULK and FLOAT stage and then you can try to recover lost capacity.

The best way to CONDITION charge a battery is to turn your battery charger on, and regardless of whether your batteries are fully charged or not, the battery charger will go BULK charge mode at first.

Then depending on how high a state of charge the batteries are in, the charger will eventually go into FLOAT mode.

It is the FLOAT mode part of the charge cycle that will VERY slowly recondition the battery.

Try to leave the battery charger on the batteries in FLOAT mode for around 8 to 10 hours and then turn the charge off for 10 or more hours, to allow the batteries to settle, and then continually repeat the cycle for a good few days.

After a number of CONDITIONING cycles, try repeating your discharge test you carried out above, and see if you have any improvement in battery capacity.

spudboy
20th May 2016, 12:48 PM
Well, a week or so prior the voltage was 12.2 or 12.3V, so it hasn't been sitting at 11.2V for months on end, luckily.

Just as well I had fitted a low voltage alarm and cut out, or I never would have noticed!

I have just put on my mid-range CTek charger, which is a 7A model that has a an "AGM Battery" mode and "Recondition" mode. This is following a 24 hour charge up with my bigger 15A CTek that doesn't have an AGM setting. It is "reconditioning" as we speak :)

I've just been down to my local electrical shop to get some fly leads to hook up 6 halogen 50W lights, so 3 pairs of lights to the 3 separate batteries. I'm going to hit the batteries with another test at much higher amps this time (am guessing about 8A this time instead of 0.8A)

spudboy
20th May 2016, 12:52 PM
...To test that properly you need to have either flat batteries or a dummy load that can draw the full output of the regulator.

What can I use for a dummy load? All of the batteries I have around here are charged up!

spudboy
20th May 2016, 01:33 PM
So, I have made a quick and dirty testing rig to draw more current. A couple of 50W halogen globes hooked up in parallel. In theory they should draw 8.2A.

http://www.aulro.com/afvb/attachments/alternate-energies/109457d1463718772-solar-panels-not-keeping-batteries-charged-up-need-assistance-diagnose-20160520_134548.jpg

And in practice they actually draw 8A!

http://www.aulro.com/afvb/attachments/alternate-energies/109458d1463718789-solar-panels-not-keeping-batteries-charged-up-need-assistance-diagnose-20160520_134118.jpg

The 7.2A going in is from my little CTek 7A charger.

spudboy
20th May 2016, 02:00 PM
Hey Dave,

You sure it's OK to run 10% of the capacity for a long term test? At 75AH a load of around 8A is close enough to 10%.

But I was doing some reading and found a company saying you shouldn't load up more than 5% of the capacity of the battery, which would make 4A a better test maybe?

Will I do any damage if I hit it with 8A for 20 hours? Or do I just stop testing when the voltage reaches a certain level, like 11.5V or something?

loanrangie
20th May 2016, 02:22 PM
This may or may not be relevant but is a 30a controller adequate for 600W of solar panel, obviously 600W is max and probably not realistic ?

Not_An_Abba_Fan
20th May 2016, 04:25 PM
What have you got in your camper that you need 3 batteries for? The best set up is to have the least amount of batteries that you need with the most amount of charge you can get going into them. Work out what you draw in a 24 hour period, then what you need to replace that in an 8 hour period. I have a 100a/h deep cycle with only a fridge and LED lights running. Occasionally I'll have the tap on for 10-15 second bursts. I have worked out I need a 120w solar set up but I'm getting a 200w one. At the moment I have a generator with a 4amp charger running during the day and that is adequate.

If you have a dead cell a or a couple of lazy ones, then that battery will suck the life out of the other two. I wouldn't run 3 together anyway, takes way too long to charge them. I'd run each separately and switch between them so that when one needs charging, it won't take as long and you have 2 fully charged ones on stand by all the time.

spudboy
20th May 2016, 04:38 PM
OK - I haven't thought about having separate batteries and rotating them!

We have 3 batteries because I wanted about 200AH, so 75Ah x 3 = about the right amount. I can't remember how I got to the 200AH figure - it was 3 years ago!

We have a bit more than a fridge and lights; with a stereo and DVD player/TV, a 140L fridge, an 80L chest freezer and a lot of USB sockets to charge cameras and phones and laptops.

With 400W, I had hoped that things would be fully charged by lunch time! I don't reckon I am getting anywhere near 400W fed in though. The other 200W goes to the 24V truck batteries to keep them topped up.

spudboy
20th May 2016, 04:40 PM
This may or may not be relevant but is a 30a controller adequate for 600W of solar panel, obviously 600W is max and probably not realistic ?

Normally it is set to 400W on the 12V side, and 200W on the 24V side, but when it is very cloudy I can push all panels through the 12V side (as that is where the majority of the heavy use is - the 24V stuff recharges when we drive from the alternator, but not the 12V)

Not_An_Abba_Fan
20th May 2016, 04:59 PM
You need to work out your total draw over a 24h period to make sure you have enough recharge capacity.

I have a TV/DVD as well but it's 240v and only runs with the generator. I have an inverter but don't use it. I'd rather use the generator for the 240v stuff and leave the battery for the 12v.

drivesafe
20th May 2016, 05:25 PM
Hi David and the 8 amp load is fine, but as you have posted, you must not let the batteries be discharged below 10.5v for your Optimas ( and no lower than 11.58v for all other types of batteries ).

Also, if you have three house batteries, even if they are different sizes and/or types, keep them connected in parallel at all times.

This will allow the batteries to discharge and charge evenly, which means you don't have to spend half your holidays checking voltages and changing batteries.

Blknight.aus
20th May 2016, 10:51 PM
Hey Dave,

You sure it's OK to run 10% of the capacity for a long term test? At 75AH a load of around 8A is close enough to 10%.

But I was doing some reading and found a company saying you shouldn't load up more than 5% of the capacity of the battery, which would make 4A a better test maybe?

Will I do any damage if I hit it with 8A for 20 hours? Or do I just stop testing when the voltage reaches a certain level, like 11.5V or something?

it depends on how your batteries are rated, yes some batteries only like a 5% discharge rate but so far as I know optimas arent that type of battery.

run the test till you hit the nominal low voltage limit for your batteries. keeping an eye on the amps and the volts on a regular basis so you can graph out what each battery does.
discharge them all to the same terminal voltage after the load has been off for 5 minutes and then parallel them and hook them up to a mains charger. when it cycles off check all the batteries again.


as for your dummy load, those halogen lights are perfect.

DeeJay
21st May 2016, 10:14 AM
Quote: In the midday sun, with all 6 panels going into the 12V charger, I was getting 4.6A at 12.9 volts going in, which seems really low for 6 x 100W panels.




I have been mucking around with solar for about 12 months now - mostly household, but recently set up 2x 100w foldable panels & a 15 amp mppt controller for my camping trips. One thing I have noticed is that the output varies with the charge already in the battery. IE if conditions are optimum & they are feeding in to a battery that is 85% charged they will feed in 3 or 4 amps & if I then put the clips on a 50% (12.1v) battery it will go to 8 amps, so I would run the test again with "flatter" batteries, as 12.9V is 100% charged.
Cheers, David
PS I think it was you I purchased a 15 amp PMP controller off its going great..:)

Blknight.aus
21st May 2016, 02:08 PM
Quote: In the midday sun, with all 6 panels going into the 12V charger, I was getting 4.6A at 12.9 volts going in, which seems really low for 6 x 100W panels.




I have been mucking around with solar for about 12 months now - mostly household, but recently set up 2x 100w foldable panels & a 15 amp mppt controller for my camping trips. One thing I have noticed is that the output varies with the charge already in the battery. IE if conditions are optimum & they are feeding in to a battery that is 85% charged they will feed in 3 or 4 amps & if I then put the clips on a 50% (12.1v) battery it will go to 8 amps, so I would run the test again with "flatter" batteries, as 12.9V is 100% charged.
Cheers, David
PS I think it was you I purchased a 15 amp PMP controller off its going great..:)

this only applies to regulated panels.

an unregulated panel will dump all the amps it can produce into a battery until it raises the battery voltage to its peak power voltage( which in spudboys case, is 17.7V ) this is clearly a not good voltage for his batteries.

the regulation of solar panels works (for the purpose of general charging and rough concept) exactly the same as an old school alternator. it will try to produce all the amps it can/needs to from the power available to it to maintain a fixed voltage.

whats very cool is to watch the volts and amps going into a solar regulator and compare it to whats going out.

spudboy
23rd May 2016, 08:48 PM
OK - sorry for the delay getting on to this. Spent the weekend away camping on the Murray. Just a 2 man tent and a station wagon - like the good old days! Purnong-Campground-Murray-River-SA (http://daviddeere.net.au/dnnd2/en-us/Blog/Post/1353/Purnong-Campground-Murray-River-SA)

Progress on my solar issue:

I've recharged the 3 batteries, all at once with one charger
I'm now load testing them with the 50W halogen globes, as 3 separate batteries


Here's something interesting though. These Optima batteries have 2 posts: one traditional round (automotive) post and another M8 threaded post with a wingnut. So each battery has effectively 2 positive terminals and 2 negative terminals.

The interesting thing is, if I measure the voltage on the pair of round (automotive) terminals I get a higher reading that if I measure on the M8 threaded terminals :confused:

On 2 batteries, it is a small difference, like 0.05V, but on one battery is it around a 0.11V. For example, the round terminals are reading 12.00V but the threaded posts are reading 11.89V, on the same battery.

That can't be right!

bee utey
23rd May 2016, 09:02 PM
Sounds like you need to get serious about cleaning, polishing and tightening your battery terminals. Clean metal to clean metal, no oxides, no paint, no sloppiness and maybe the current will willingly flow. ;)

spudboy
23rd May 2016, 09:23 PM
Hmmm - but I am probing them with multi-meter prongs, which have a very sharp point on the end. Am pressing with equal hardness on each post!

One is stainless (threaded one) and the other is lead...

spudboy
23rd May 2016, 09:54 PM
Here's some more results of the 3 batteries, this time under a 4.3A load instead of a 0.8A load. Only recorded 8 hours of stats instead of 20 hours this time.

https://www.aulro.com/afvb/images/imported/2016/05/275.jpg

drivesafe
23rd May 2016, 10:26 PM
Hi David and your batteries are still showing they are down on full capacity.

But a ROUGH calculation is that it is only about a little under 10%, which is not bad.

spudboy
23rd May 2016, 11:06 PM
OK - thanks for that info. So not time to chuck them out just yet?

Looks like the next thing to test is the solar panels.

Ferret
23rd May 2016, 11:53 PM
But a ROUGH calculation is ...

Can you explain how that calculation is done please.

Blknight.aus
24th May 2016, 01:03 AM
very roughly the back of the napkin version of the calculation is

a fixed amp draw for a given amount of time gives you the AH rating of the battery. its terminal voltage is determined ahead and the test stops when it hits that value

if the battery is rated for a 5 amp draw (just to make this example easy) for 8 hours untill the terminal disconnect voltage its a 40AH battery.

in reality you only usually get to about 80% of that with a dummy load like lights with a fridge the batteries get to rest and cycle and you will get closer to say a 90% discharge (see that little kick up at the end of spudboys graphs when the voltage rises?) if you let the battery sit for half an hour you can then use the amps in the battery until the voltage drops down again until it doesnt come back up over the disconnect voltage but this is torturing the battery. back to the math.

so with my example 40 AH battery and a 5 amp load I want to hit a little over 6 hours worth of draw (6.4 or near enough 6 hours 20 minutes lets call it 6)

so my expected 100% value for my example batteries as per the way spudboy has conducted the test is 6 hours. If he batteries only get to say 5 and a half hours then they are down 10% of their life.

this is the complicated version of it and if I did it right if you click it it will lead you to the wiki.


https://www.aulro.com/afvb/images/imported/2016/05/272.jpg (https://en.wikipedia.org/wiki/Peukert's_law)

if you understand all of that you just unlocked the secret that lets drivesafes traxide units white man magic amps out of nowhere on your batteries.

on edit.

Spudboy is disconnecting his batteries at a higher than the nominal "flat" voltage, at a guess maybe 90 minutes (on the 5 amp chart) worth based on his graphs and when I would disconnect probably closer to 2.5 if he wanted to push. In doing this he is protecting and extending the life of the batteries.

I think hes lost a little more than 10% because his post charge voltage is a little lower than I would have expected which mean sulphation is starting to occur. If I had to project a life of the battery till they drop to 50% useful life I'd say maybe 2 more years.

the voltage difference between the posts will be because of lead oxide if you want to prove that measure the resistance between the stainless terminal and the lead one. I've also seen them where a fracture has developed in the mount inside the battery going to one of the terminals.

drivesafe
24th May 2016, 03:59 AM
Can you explain how that calculation is done please.


Hi Ferret, I state ROUGH because there are a number of factors that come into play when measuring varying DC voltages.

The current load would most likely have reduced as the battery?s terminal voltage dropped.

How much the load reduced may vary from next to nothing to a noticeable difference.

So as above, the calculation was a rough one, but it still gives a good comparative idea of the overall capacity of the batteries.

Using the data David posted up, and based on amperes used over a given time, with a 75Ah battery, you would have expected to get close to 9 hours use by the time the battery voltage reached that same 11.95v or approximately 55% of a 75Ah battery in top condition.

As both Dave and I have pointed out, there is likely to be some reduction it total capacity, due to the effects of some sulphation taking place.

As posted earlier, a number of conditioning chargers may revers some or all of the sulphating, but at a capacity lose of only around 10%, David, I sure as hell would not be replacing those batteries just yet.

spudboy
24th May 2016, 10:16 AM
Ahhh - here's something interesting :eek:

I mentioned above that I'm getting slightly different voltage readings, depending upon if I use the threaded posts to measure the voltage, or the big round terminals.

So, I went out and bought a Battery Analyser to see what readings I'd get, and the results are striking. Top row of photos is the threaded post reading. Bottom row is the round automotive posts. All readings taken within 30 seconds of each other:
https://www.aulro.com/afvb/images/imported/2016/05/260.jpg

What do you think is going on there? You'd think they'd be welded together to each other under the plastic, and therefore give identical readings.

Battery 3 has readings where one says replace the battery, and the other says it just needs a recharge!

drivesafe
24th May 2016, 11:52 AM
Hi again David and an interesting reply.

When you use the battery terminal posts ( those used for connecting battery clamps ), you are getting exactly what you need, to be able to see if the battery is in good condition.

The threaded studs are NOT intended for use for supplying motor cranking high currents, but are solely intended for low to medium RV current use.

So your readings taken from the threaded studs are pretty well irrelevant.

The three bottom readings clearly indicate all three batteries are fine.

A point of clarity, CCA readings have nothing to do with either SoC or a battery's remain/available Ah capacity.

If a battery is in good condition, it should be able to deliver it's rated CCA when discharged as low as 40% of it's original capacity.

Confusing but the fact is CCA and Ah are unrelated and you can not use one to determine the other.

You can have a battery with a very large Ah but a low CCA rating, or like with Optima batteries, you have small Ah capacity but very high CCA ratings

Blknight.aus
27th May 2016, 09:17 PM
the lead in the posts can oxidise and pick up contaminates. if you scrub the dark lead off of the battery post and then retry it I bet you get better readings.

you could also have had some monkey whack the lead post and thats caused it to fracture away internally. These internal fractures then build up corrosion and become internal resistances that mess with the readings on those battery testers.

spudboy
30th May 2016, 02:46 PM
The threaded posts are stainless steel, so no problems with oxidising there.

I've bought 6 of these
https://www.aulro.com/afvb/images/imported/2016/05/50.jpg

So I'm now running everything through the big round posts, not the threaded posts.

Not_An_Abba_Fan
1st June 2016, 11:59 AM
The way I have always treated a deep cycle battery for 12V auxiliary power is that the threaded posts are for power out, the big round posts are for power in.