Best way to determine if your batteries are still good?

[shaniac]

I have purchased a home that uses 12 L16 batteries to power the place. They are charged by a mix of PV panels and an assortment of good quality components via a 24V system.

My charge controller seems to never get above 28.6, and I fear that it is never going in to FLOAT mode to top off the batteries. I feel confident that my settings are correct and I think my problem is that I do not have enough panels to fully charge the batteries each day since I use them every day.

So my question for you is:

Before I spend more money on additional panels how can I determine if my storage is holding the power that is made each day?

I am feeling that something may be wrong, as 12 L16s should have an AH rating of 4500 (375 AH x 12 batteries.) Specs were taken from http://www.interstatebatteries.com/www_2001/content/products/product_6volt.asp

But I certainly do NOT have even 30 hours of reserves, as if I have a cloudy day in the winter, the batteries are below 24.7 by the end of the day.

So other than taking a battery out of the array and doing a load test to see how long it takes to drain, how can i determine if I need new batteries or not?

I have measured the SG on most of the cells and they are all around 1.275 near the end of the charging cycle. The batteries themselves are about 7 years old, and do not seemed to have been abused that I am aware of. Fluids have always been topped off and they have only gone without a charge if the days are cloudy, and even then the backup generator would have topped them off. The inverter shuts down if the batteries get low, so i do not think they have been deeply drawn down, and all last summer they would have seen 28.5 for many hours day after day.

They all seem to be working fine, other than I have no reserve. One would think that 12 L16 would give several days or reserves when only CFM and 2 laptops and a few 12V appliances (Sat modem, router, etc..) some wall warts for the fire dept pager and a 110V water pressure tank are used each day.

Any suggestions friends?

And also, if I do need to do a discharge test next, what is the best way to do it? I have a 24V DC bulb I could use as a draw.

Recommendations? Advice?

Thanks for your time.

[zeusmorg]

 First off, you don't state your temp(specific gravity readings are affected by temperature) get a specific gravity chart. Things I'd check, resistance across all your terminals to see if any faults lie there. Ideal specific gravity is 1.265@ 75-85F. That indicates a 100% charge.

 So if your specific gravity is the same in all cells, and voltage readings are equal on all batteries, I'd say you just have too big a load on the system. Reserve ratings are not for hours of TIME but amp/hrs used. Also keep your battery bank immensely clean, dirt draws moisture and can drain your system.

 Also most L16s are rated for a useful life of 4-8 years, so it is possible you're nearing the end.

[Volvo farmer]

A big system like yours should have a Trimetric. An Amp/hour meter is pretty critical in determining what is actually going on with your system. If you have a Trimetric, and you are using only solar charging sources, what is the Ahr reading since charged (hold select button three seconds) just before sun up?

Some other bits of information that I would like to see.

Solar panel capacity in watts.

Physical location for solar insolation calculation

Make and model of charge controller.

Other good quality bits used for charging? What?

Battery location (inside, outside, or just temperature would work)

I'm planning on gettting 15 years out of my L16s. Only fourteen years to go! I think yours could still have life if not previously abused. My neighbors got seven years out of Sam's club golf cart batteries in way too many parallel strings, and pretty seriously abused at times.

[SamoaPower]

I think you should count your blessings. If you have obtained seven years service and still have somewhat healthy batteries, you are doing well.

In a 24 volt system, four 6V batteries are required in series so you probably have three such strings in parallel. The 24V capacity of a four-battery string would be the same as a single battery at 6V or 375Ah. Three strings in parallel would yield 3 x 375 = 1125Ah at 24V, WHEN NEW.

It's reasonable to expect to lose roughly 10% capacity per year for cycled, well-cared for batteries of this type so I wouldn't be too surprised to find your total capacity down to 350Ah or so.

I've also found that that the terminal voltage at a given charge current at maximum state of charge will tend to decrease with age.

The fact that the SpG is well up there and reasonably matched to each other attests to a good state of health but tells you nothing about remaining capacity. You didn't mention about an equalization routine but perhaps you don't have the charge source capability.

I'm glad to see that you are aware that the only definative way to determine existing capacity is to do a timed discharge test. Most people seem to resist doing this, perhaps because of the time required and the necessary equipment.

To compare existing capacity to original capacity, the test would have to be done using the same parameters as the factory. Battery capacity is a function of discharge current, the higher the current, the less the capacity. Batteries of this type are often rated at their 20 hour rate, meaning a current that will deliver the rated capacity in a 20 hour period with an end point of 5.25V for a 6V type.

For one of 375Ah rated capacity at a 20 hour rate, the discharge current would be 375/20 = 18.75A. Ideally, this needs to be a constant current throughout the test period which rules out a fixed resistance load since the voltage under load will decrease with time during the test, decreasing the current. A constant current load could be built. I'm afraid an arbitrary 24V bulb wouldn't give you very useful information.

Testing a single battery out of a bank of 12 would give you an approximation of total capacity but would be less than accurate. Again, ideally, you would test each battery separately. The capacity of a series string would be that of the lowest capacity battery in the string and the total capacity would be the sum of the lowest in each series string.

As I said, seven years ain't bad.

[zeusmorg]

 Guess i should have put this in my first post.

You can also test a battery by performing a heavy load test,load test battery at 1/2 the CCA rating, or 3X the amp-hour rating, for 15 seconds. The battery must be above a 75% state of charge.(1.225 specific gravity @ 75-85F). A voltage below 4.8 volts for a 6 v battery indicates it's time to replace.(9.6v for a 12v battery, 19.6 for a 24 volt batery).

  These tests are performed by a lot of automotive parts stores at no charge with a SUN VAT 40, (or similar) however if you do take them in to be tested , watch the guy! Some of them have no clue what load or for how long to test..tell him!

their job is to sell you a new battery after all.

[tecker]

I would add to the comments individual charging and watch the voltage build up and how much the voltage returns after a 30 min 10 amp charge . If it shows voltage increase fast a drifts down to the voltage you started with could be sulpf=hated . If it pegs the the needle and possiblly  warms up and drops very fast you have a shorted cell .

[SparWeb]

I don't mean to contradict you SamoaPower, but in this case, a discharge test with a fixed resistance can be used.  Granted, it puts some undertainty into the test results, but a little bit of mathematics will make up for most of that.

It is important to use a resistor whose resistance isn't going to change a lot.  Lights aren't as good a choice in this case because their resistance changes drastically with temperature.  If you get some big edge-wound resistors from a surplus store or electrical contractor, you can use that.  It must shed about 2000 Watts to work, so a few old space heaters may be the ticket.

Say your resistor has R=1.5 Ohm

At the start of the test: 28V/1.5 Ohm = 18.7 Amperes  This is equivalent to the 20-hr rate for one battery.  If you want to test the whole set, I guess it must be around 0.5 Ohm.  

Make regular measurements as the test progresses, say, once every hour.  At each hourly increment, check both the voltage and the current.

When the voltage has dropped to the minimum, make one last current measurement and stop the test.

Plot the current values on a graph with the times on the x-axis.  Calculate a "mini" Ah for each average current between two readings.  Add all the pieces up and you will be in the ballpark.

Because the current has dropped off during the test, the obvious conclusion is that your test over-estimates the capacity, because although you started at the 20-hour current, you likely end at the 30-hour current.  This isn't as bad as it seems, because you are trying to use the batteries at a 20-30 hour rate anyway.

In my case, with Absolytes in my backyard, that's not a problem, because GNB publishes Ah rates in great detail for many rates, not just one.  Maybe you can wheedle a few charts out of Trojan.  In fact I think I've seen a table for L16's with different rates on it on the net...

Even if you can't find the data, just knowing the result of the test will tell you if the batteries are adequate - judge them by what you need at home, not just what the mfr. specified.  If the Ah rate you measure won't keep your house running for as long as you need it on cloudy days, then it's time for new batteries.

[shaniac]

WOW! Thanks so much for the many responses. I had to read the posts several times as most of that was over my head but I will do everything I can to understand and catch up.

Based on what you all wrote: I have the following questions.

1. How did you ever get so knowledgeable? : ) But seriously, no need to reply, it is fairly obvious, and I do really appreciate your help.

Now on to REAL questions.

2. The batteries are Interstate UL16s. One of you mentioned TROJANS, are my Interstates just Trojan batteries with an Interstate sticker on them?
   
3. From what I understand I need to do a load test. I do not have any fancy capacitors or items such as that, but I do have a 1500 Watt portable heater I could use and an old electric iron if I need it as well,  So presuming the 1500W heater will be enough, when do I do this load test? At the end of the day after a nice long sunny day? Or early in the morning, and I just disconnect the panel array?
   
4. To perform the load test, I just turn on the heater and take voltage readings using my MPPT device? Or do I need to take SG readings as well? the ambient temp will be around 52 degrees or a bit cooler, seems like that is where the weather parks itself this time of year. What exactly do I need to do for this load test?
   
5. Volvo asked:
   
   

A. Do you have a Trimetric? Nope, sadly I do not. I do not have the change to pop for one right now, and my power shed is far from my home,(200 feet) but I do have an APT Powercenter that tells me my load, and an RV Products RV50 (older MPPT device) that tells me the voltage and what juice is coming from my array.

B. Solar panel capacity in watts? I have 4 new BP 160s, 2 BP 150s, on one array, and a bunch of mismatched 12V panels wired in series on the other array, but they all join in one set of wires inside my APT Powercenter. Sample Panels are 2 Kyrocera 120s, 4 53W panels and two 64W panels. Once I get the battery thing figured out, I will have some questions about this miss match of panels, but I digress.

C. Physical location for solar insolation calculation? No idea what that means?

D. Make and model of charge controller. My charge controller, I believe is my RV Products (now Blue Sky, but you knew that)SB50 (Solar Boost 50) I have read the manual on it and have been playing with the four adjustable settings inside the unit. After a few calls with the support guy, I feel my MPPT is in good shape, but I don't seem to be able to dial in the other three as well as i would like. But that is another thread. So if that is not my charge controller, than it must be in my Inverter, but I do think the SB50 is it.

E. Other good quality bits used for charging? What?

Inverter is the good old international harvester of boxes, a trace SW4024 from 1994. And my generator I use for when the batteries are getting low or need to be equalized.

F. Battery location (inside, outside, or just temperature would work?) Batteries are inside a standalone building built just for power. It has three rooms. The inverter and SB50 and other electronics are in the middle room. The generator is on one end,(west) and the batteries are on the other facing East. The panels are below the building on the slope just south of it, not more than 15 feet away. The battery room has NO power going to it, so night time work requires a headlight and is a bit more of a challenge. The room vents well and has sensors on the batteries to test the temperature. I am guessing the room stays around 65 degrees as it is never hot or cold feeling in there.

SP wrote: In a 24 volt system, four 6V batteries are required in series so you probably have three such strings in parallel.

This is correct, and it seems I do not have the AHs I thought I had, since they are in parallel, I guess I only have the AH of just three batteries, not all 12. Is that correct?

Sparweb writes:(about the load test) Make regular measurements as the test progresses, say, once every hour.  At each hourly increment, check both the voltage and the current.

I can check the voltage easily enough by looking at the SB50, but how do I test the current? I have an analog multi meter, but don't know much how to use it.

He also writes: If the Ah rate you measure won't keep your house running for as long as you need it on cloudy days, then it's time for new batteries.

But what I guess I do not know is... when I buy the new batteries, do I just buy the same number or how do I find out if I need one more or two more per group?

Again, thanks for all your help here, let me know the best way to perform the tests above, and I will be more than happy to post the results. Thanks again so very much for your time.

[Volvo farmer]

It appears to me that you have enough solar to efficiently charge those batteries, provided your loads are not excessive.

C. (translation) Where do you live?

I cannot help you load test your batteries because I feel that you are unwilling to spend $165 on the proper tool to do such a test (add $30 if you don't already have a shunt). A Trimetric will also keep track of your overnight loads. I think it is possible that your batteries are OK and your loads are excessive but there's no real way to know that without having the proper tool.

There may in fact be an easy way for you to load test your batteries, I don't know. Maybe someone else here will dream up something for you to try.

I don't mean to sound harsh but this is the way I see it. You have $6000 in panels, $2200 in an inverter, $500 in a charge controller and $2500 worth of batteries there. You seem to be willing to guess that you need to spend $2500 in replacing your batteries, but you don't have $165 to spend on a critical measuring tool used in nearly every single off-grid solar installation I have ever seen; A tool that would in fact give you the information you desire, and allow you to measure current, as well as current over time.  

[shay]

Without these, it is likely sulfation has caused reduced capacity. See my diary for info.

[jimjjnn]

Do you have a temperature sensor on the batteries? I saw that your temp is 54 degrees. Fahrenheit or celsius ? If F , your batteries may be undercharged because of low temp. Also, output would be lower from  cold batteries.

Just a thought

[Gary D]

 Congrats on choosing a home with an existing re setup!

 As Volvo farmer stated, you need some way to watch the amps in versus the amps out of your battery bank. You are now the "power company".

 Do you have a well pump that you aren't accounting for? Anything and everything counts. Even the inverter draw. Also dust buildup on the panels or shading will affect things. As will time of year, depending on your latitude ( the location question). Don't count on the panels ever producing their rated output, altho with mppt it "should" be close.

 It would seem that your usage is a bit higher than the previous owner. "take all of this with a grain of salt" just thoughts....

[erne]

There are some other things you might want to consider. Are you aware of phantom loads? Does your wall wart pager keep your inverter on all the time?  Your inverter is a electrical device and use 20 to 45 watts when operating. My l-16's are 15 yrs old and still fine. You need an e-meter or similar product. You may have a problem in your solar input. Box those batteries in to a close insulated enclosure and there working will help maintain a better working temperature....erne

[shaniac]

Wow, thanks gents, the feedback you are giving is truly exceptional, you are helping me out immensely and wish I could express my gratitude to each one of you. THANK YOU so very much!!!!

Now, to answer all your questions.

Volvo asked "Where do you live?" I live in Northern California not far from the coast. We get tons-o-son, and my panels are on the southern slope on part of our ridge. I get sun on them for most of the day, but the late afternoon they get shaded.

Volvo, I do want to get a Trimetric, and someday I will, I just have had a tough few months financially and it will need to wait till I have a little extra one month. If I do need to replace my batteries then I will be in deeper dodo as it will most likely take me several months to save up enough for a new bank. This is part of why I am trying to determine if I need new ones or not, so I can help budget for teh next 12 months.

You wrote "There may in fact be an easy way for you to load test your batteries, I don't know." I may have found something... I was installing four new batteries at one of our local fire stations today and a neighbor came by with a load tester and we hooked it up to the two older L16s I had just pulled. (Our entire community is off grid.) The tester was designed for 12 volts but it seemed to work on the big 6ers. We knew one of the L16s was bad because it had a bad center cell, but we did not know which one. This device placed a load on the battery and then had an analog meter that showed the condition. One was GOOD, the other was REPLACE. I am wondering if this tester was the same as zeusmorg was talking about higher in this thread.

Anyway I intend to borrow it from him and see what it says about my bank. And Volvo, don't worry about sounding harsh, it did not come across that way, I appreciate all your comments, I can see your intentions are good, and your advice has already proven helpful. So thank you, again.

Jim asked: "Do you have a temperature sensor on the batteries? I saw that your temp is 54 degrees. Fahrenheit or celsius ?" Yes, there is a temp sensor on the bank, or at least I believe there is. There is a red (sensor looking) device taped to the side of one of the batteries. I think it runs to the MPPT device, but I have no way of seeing a reading of what the temperature is. When I said it was 54 degrees, I was thinking ambient air temperature, not battery temp. I do not know what the battery temperature is. The 54 was Fahrenheit by the way. 50 - 70 is our typical Northern California temperatures during Fall/ early Winter.

Gary D writes: "...you need some way to watch the amps in versus the amps out of your battery bank. You are now the "power company". Do you have a well pump that you aren't accounting for? ...It would seem that your usage is a bit higher than the previous owner."

No well pump, but we do have two pumps. One is a slow pump that pushes our water up from our spring on the northern slope. I have that on a timer and run it during peak sun hours. It's load is minor. Our other pump is our pressure pump for domestic water. Now that bugger pulls quite a draw, but is only on for 60 seconds at a time, depending on water usage.

Our usage may be more than the previous owner but I would think not as we have no TV, or devices we use during the day other than the two laptops, router and satalite modem. But who knows what the previous owners had.

Erne asks about Phantom loads: Heck yes I got em', seems like everything has PLs these days, Most items I keep on power strips and just turn them on when needed. Laptops, fax / printer, modem, router, boom box, etc.... The Fire pager just stays on 24/7 because we are always on call. Our inverter does stay on 24/7, I do not have it on search mode. When this thread is over, I would love to hear your advice on that, I do recall reading in my inverters manual that I could set the box to search OVER a certain draw, but I have not looked in to that. I am sure there are folks whom are very familiar with the older trace 4024s out there. I do also try and keep the panels clean from crud, but that seems like a never ending battle. My rack adjusts slightly to match the suns angle, but honestly, I only adjust it about every three months.

Shay asked about "pulse charging? Desulfation?"

I know I am supposed to equalize the batteries once a month with the generator, but it seems my generator only gets my batteries as full as a good sunny day. On a sunny day the panels will charge the batteries up to 28.9 to 29V and that is all I can get from my generator as well (according to the voltage on my MPPT unit.

What is Pulse Charging? Is that just short charging times? If so we don't have that. Desulfation happens when you equalize correct? I am presuming that is also happening when O get the batteries above say 28.7. Your thoughts?

Thanks again to everyone, I really appreciate your time and your help on this. Let me know what you think about trying out the load tester. I can give it a try this weekend.

[Flux]

I haven't read all this discussion but I did notice this point.

"I know I am supposed to equalize the batteries once a month with the generator, but it seems my generator only gets my batteries as full as a good sunny day. On a sunny day the panels will charge the batteries up to 28.9 to 29V and that is all I can get from my generator as well (according to the voltage on my MPPT unit."

I an not sure what your MPPT unit is but for equalising you need to over ride any form of charge controller. I get the feeling that your controller is holding the generator charge at the normal value chosen by the controller. With the generator directly charging the battery with nothing limiting the volts then you should be able to get the voltage up. 29v may be what the controller normally considers its absorbtion voltage before it drops to float. This would be ok for normal charging but occasionally things need a good gassing and 29v is probably not enough. 32v may be needed to stir up the electrolyte perhaps every month or two. It will deal with any stubborn sulphate and mix the electrolyte properly.

If your specific gravity is well up then the 29v may be doing the job.

Flux

[zeusmorg]

 The tester i was referring to is used in automotive and truck battery testing, it shows both volts and amperage, and had a variable resistor.. big one.. made by SUN corporation.

It's not something your average RE guy would have laying around.

[shaniac]

The batteries do start to churn and bubble after 28.0, so I do know the fluid is getting mixed nicely. As for 32 V, I do not have any way to get the voltage that high unless I can see if there is a way to adjust that via the inverter or charge controller. As for equalization, there is a red EQUALIZE button (mmmm shiny red button) that is on the top of the charge controller. When it is "UP" it is supposed to be doing it's job, but when I let it go, the voltage does not really increase. I think I need to go back and reread the manual again.

As always thanks a million thanks for all your advice. Anyone else have any opinion on the 12Volt load tester and if I should give it a shot this weekend? I am thinking it certainly could not hurt.

[Volvo farmer]

My neighbors have the same problem with an older Trace inverter. It needs a $300 gadget to change the charge point settings. Go online and get a manual for your inverter. There is some way to make the Trace charge the batteries at 31V from the generator, but maybe it needs the same gadget as my neighbors.

You likely don't have enough solar to push the batteries up over 31V. I have over 1500W of solar, one less string of L16's than you, and I have a hard time pushing the voltage that high.

Try this someday when you know you will have a full day of sun. First thing in the morning, run the generator till the battery voltage stops increasing. Let the generator run two more hours. Then turn the generator off and push the equalize button on the charge controller and see what happens. You might get the batteries up over 30V in the middle of the day if you shut down most of your loads.

[erne]

look on the wall wart on your pager and see what voltage it is taking the a/c to it will have the info on it. you can but a voltage reducer or increaser for the monitor that is dc and that way you will bypass the inverter. the lower the voltage usage on your inverter the lower the eficienty is and the more juce it uses. you need to have it in serch most of the time. I would imagine your previous owners did this. 30 watts times 24 hours is a lot of electricity plus a 30 percent conversion factor. you can do this for very little money.. hope this helps  I can give you a name of a company that makes quality d/c to d/c converters if you need....erne