My Batteries and/or Charge Controller(s) Acting Up

[Canine]

I have two 3024i connected to two 285 watt, 60 cell, Solarworld panels topped off with a ProRemote with shunt. Everything has been working superbly except for the last 2 or 3 weeks. The charge controllers weren't charging properly, so I turned off the solar panels, then turned off the charge controllers via inline switches. That helped at first, but not now. All the connections are tight and have no corrosion. Now the charge controllers won't go into float mode at all. Always in bulk. The four batteries in the battery bank are 7-year-old, CD technology, AGM, 12 volt, 140 a/h batteries I had purchased used from a UPS from a cell phone tower. The ProRemote also shows a 99% charge efficiency, which is not correct. I can change it to 94% or any number, but it will adapt itself to the batteries showing 99%. The batteries aren't getting a full charge now or if they do, they aren't keeping up voltage like they used to. They were fading before, but nothing severe, just normal wear. Let's say I use 45 a/h through the night. When it charges during the day, the number will slowly drop as it's being charged just like normal, but when it hits about 30 a/h, it jumps to 0 a/h left to charge and displays a full battery bank. It will go to 12.7 volts, but instead of me waking up to 12.6 or 12.5 volts like normal, it will show 12.4 to 12.2 volts. I can't think of any more information that may help.

With this information, can you diagnose the problem? Would old batteries or a faulty battery cause the charge controller to wig out? Can old batteries fail during a 3 weak time period of time instead of failing slowly over several months or instead of overnight?

 

[tx2sturgis]

Question, and admittedly out of left field:

If both controllers started acting up and doing the same thing at the same time, then, by any chance, are they connected thru or by an app on a smartphone, and/or, updated 'over the air'?

Are they interconnected to the internet, in other words?

:huh:

 

[John61CT]

Yes yes and yes.

Test with a known good bank, I'm betting time to replace.

Check all your connections too of course.

 

[Canine]

txsturgis, No. It is a completely a closed circuit system. They are networked together (daisy chained together via a twisted-pair wire) and act as one unit using the option ProRemote. By being networked together, they are in effect one charge controller.

 

[Canine]

John, that's what I was afraid of since they are old. Every day I get use out of them is a bonus for me since they are now living beyond their life expectancy. What I'm thinking of is taking the batteries down to a proper battery tester for the specific type of battery I have and maybe eliminating the bad one if the problem lies with just one bad battery. I can easily live with a 3-battery bank for a few months until I make a bit more money to purchase a bank of batteries. I already used my emergency fund to buy a truck a couple months ago.

 

[Weight]

Used batteries are the first thing to question. You may be able to salvage the bank by removing one bad battery. If it is just one. In previous posts, you present as an expert, why do you even ask this?

 

[Canine]

Weight, I don't know everything. I know a lot and share what I know, but ask when needed. In this case I haven't experienced this type of failure and am not able to diagnose well enough to know if the problem lies with the batteries or charge controller or both. Since batteries and charge controllers are expensive, I definitely want to be as informed as possible before I started buying new items. And I'm way out in the boonies for work, so contacting someone with the right diagnostic tools is not really an option. I sure hope it is one battery. That would be cool. I could use another 5 or 6 weeks so that I have the money. I wonder if I'll make it? Maybe I should start taking bets!

 

[Weight]

Sorry if I came across wrong. I'm one of those old fiicks. I hope you have at least a multimeter to check voltages and help with trouble shooting. Your controllers will have to stay in bulk until the batteries come up to voltage. That could take a long time starting from a low SOC. Or after reading again. Could you try dropping one battery out and see how the controllers behave. Use the controllers to trouble-shoot one battery at a time.

 

[Trebor English]

contacting someone with the right diagnostic tools is not really an option.

Since they are AGM a hydrometer can't be used.  That would be the right diagnostic tool.  

What you could do is charge them up as much as your solar controller will allow then separate them.  If there is one bad one it might stand out with a lower voltage in 2 hours.  If there is one good one it might stand out in 2 hours with a higher voltage.  If two hours of resting doesn't show any differences then 4 identical brake light bulbs (not LED) might be useful loads for testing.

 

[Canine]

I snooped around Amazon for a tester that is 'strong' enough to test big batteries, is a reasonable price, and seems to be decent quality. I found this:

https://smile.amazon.com/dp/B01M0AR...olid=1M88K8VCZV4ST&psc=0&ref_=lv_ov_lig_dp_it

The price has me wary, but I'm not looking for all the bells and whistles of a $3000 commercial tester.

 

[Weight]

I don't think those testers relate much to deep cycle batteries. Check with the expert, Rod said do not buy one. https://marinehowto.com/are-battery-conductance-testers-worth-it/

 

[Trebor English]

When I was school bus driving and working in the shop I was the guy who got all the old bus batteries.  When they had new ones in the stock room buses got new pairs as needed.  When they ran out I provided used pairs that were the best of the bunch.  Each pair that came out of a bus typically had one real bad, maybe both.  Sometimes both were good.  I charged them, let them sit a week, then load tested them.  A good battery can take a charge, hold it, then give it back.  

The shop bought a meter similar to the one linked.  The manual had a chapter on how to use the meter to improve sales.  

The meter displays a measured CCA, cold cranking amps.  That's the current the battery can deliver for 30 seconds without going below 7.2 volts at zero degrees F.  How can you measure that without chilling the battery to zero degrees F?  How do you measure 600 amps with those skinny little wires?  How do you complete the test in two seconds?

The meter displays state of charge.  There are three ways that I know of to measure the state of charge.  A) measure the current in and the current out over time.  B) the Balmar Smartgauge method measuring the voltage 1000 times per second over time when charging and discharging is observed.  C) measuring the specific gravity of the electrolyte to determine how much sulfuric acid is left and how much is just water because the sulfur is on the plates.  How does this meter measure state of charge?  If it really does that then everyone should have one of these testers.  Don't mess with shunts and counters and such, just clip on the tester and two seconds later it tells you the state of charge.

I can understand why a shop would spend the money to buy one of these meters.  It lets them show a customer with apparent authority that a battery replacement is necessary.  The battery sales profit will pay for the meter in no time at all.  For a person, a battery owner, I recommend a volt meter and a load device like a pair of headlights.  For a quick state of charge check, a hydrometer works.

 

[Canine]

I don't think those testers relate much to deep cycle batteries. Check with the expert, Rod said do not buy one. https://marinehowto.com/are-battery-conductance-testers-worth-it/

Excellent link! That does a great job of explaining the difference between capacitance testers and testing actual a/h capacity of a deep cycle battery. Ideally I would do a 20 hour-long test of each battery in a water bath that keeps the batteries at a perfect 77 degrees while the batteries is hooked up to a specific, consistent load to get the best possible test results. That isn't going to happen.

My charge controller with the shunt appears to be basic capacitance tester (not unlike the link I had provided) that gives fairly accurate information when the battery is tested repeatedly over a period of time. It does have value and is surprisingly accurate. However, assuming one battery went bad in the bank of batteries, that is throwing off all the previous information that had been relatively accurate. One capacitance test is not necessarily accurate, but several over a period of time is surprisingly accurate. It appears the charge controller is still healthy, but the bad battery(ies) are throwing them off.

Now that I have a better grip on how testing a battery works, I suspect one battery had an acute failure. If so, I can take each battery out, put the mulitmeter on it, and should see a reading of 10 volts or lower on one or more batteries. That is the hypothesis anyway. If it is one battery, it will have obviously failed in a dramatic fashion and a proper 20 hour-long test wouldn't be necessary to identify the acute failure. I need to do this soon, though, because if I have one bad battery and 3 OK batteries, the 3 OK batteries won't be OK for long.

Am I on the right track so far? Or have I stumbled off the trail?

 

[John61CT]

Yes a load test is the only decent way to gauge State of Health, aka residual capacity

without huge lab grade gear.

All it takes is

an ammeter (or AH counter) and a load device that lets you adjust the amps drawn.

a smartphone for logging

ten hours and patience.

 

[Canine]

OK, looks like I have a plan. It won't happen right away as I'm working right now. Will give an update when am able to test the batteries.

 

[Weight]

Yes. I would at least check each battery with a voltmeter. Even if you can only do it with the controller's built in meter. One battery at a time. Yes. You have to get that bad battery out of the bank. Urgent.

 

[Riverman]

I've used this type of manual load tester for over 50-years.

It puts a good load on a battery and will tell you in a few seconds if it's good or bad.

I don't need anything better....

https://www.harborfreight.com/100-amp-612v-battery-load-tester-61747.html

 

[Canine]

Will be taking the batteries out shortly. How long should I let them settle before putting the multitester on it? They are now as charged as they are going to get.

 

[John61CT]

2-4 hours may be enough, sometimes 72 for real precision.

Completely isolated, including from each other

 

[Canine]

Update. I disconnected out one battery at a time, let it sit for 4 hours, then checked voltage. (I didn't physically separate them because they weigh 100 pounds a piece.) One is at 12.62 (awesome!), one at 12.51 (meh), one at 12.50 (meh), and one at 12.39 (gah). They used to read 12.7 in the morning when lightly used, then it went to 12.6, now if I'm very careful, the bank will read 12.5, but most of the time it reads 12.4 in the morning. It's nothing to get it to read 12.2. It may not even be able to read 12.5 in the morning anymore. The charge controllers aren't wigging out now. The will consistently go into absorption mode, but will not go into float mode anymore. The batteries are being normal for their age and the CCs appear to be reflecting that.

As for the charge controllers acting a bit weird, they have the ProRemote with a shunt so they adapt to the changing battery conditions and adjust the "fuel gauge" and charging times. A great feature of the ProRemote with a shunt is that it will go into float mode when the batteries hit a certain electrical reading instead of going into float after a specified amount of time like most charge controller set ups. (Absorption is good, but too much of it is not as good.) It appears the computer in the ProRemote was having problems adapting to the relatively sudden change in battery condition. That's a guess, but seems reasonable.

I bumped up the absorption rate from 14.2 volts to 14.5 volts to maybe get rid of some of the sulfation and to help push the voltage through the layer of sulfation. The batteries undoubtedly have some sulfation. 14.5 is still within manufacturer's recommendation. I may bump up the float setting a bit, too, if it will ever go into float again. Even with 30 amps at 14.2 volts for several hours it won't go into float. The fuel gauge says full, but the batteries aren't taking a charge like they used-to-could.

When I get new batteries, I may equalize these old ones to see if they can be revived; they might give a few more good months. I've heard of people doing that with AGMs and having success. Would be an easy experiment.

Well, this was a learning/relearning experience. I had forgotten the name for surface charge and was having a hard time Googling that information. Bottom line: time to get new batteries. I appreciate you guys listening to me talk this out and getting this squared away. Easier to do this with your help rather than entirely on my own. I notice I get a lot smarter, faster when I listen to other people.