Battery Charging and Charge Controller Setup

I'll try to do this quick and may have to break it up into smaller replies ...

The number 1 cause of agm failure is UNDERcharging. Sensing your inexperience, he may have dumbed it down a bit just to try and make sure you charge it frequently. It is not necessary to take them down that far, although doing so once in awhile is a good proof-of-performance test. Normally in solar we don't take them below 50% DOD in order to get good cycle life out of them. Only going 25% DOD is even better, and may be impractical and costly doing anything shallower.

Heh, they are familiar with those terms, but you may have been speaking to a sales-droid or trainee or someone who just didn't have time to go into it all and just hand out go/no-go values.

You'll want to add these two documents to your maintenance library about Deka / East Penn. Search for the following documents in pdf form:

Valve-Regulated Lead-Acid (VRLA): Gelled electrolyte (gel) and Absorbed Glass Matt (AGM) Batteries. Document 0139.pdf
and
1913 Renewable Energy Charging Parameters Document 1913.pdf

More to come - have to break it up ...

I'll try to keep this short as well, but at least get you started. Also to keep it simple I'll just assume a 12v battery. The charts in that 1913 (not the year it was made!) are the ultimate guide.

BULK Voltage - In a cyclic service, you just set this the same as your absorb voltage (more on that later). Some controllers may not even have a bulk voltage setting, just absorb and float.

BULK CURRENT - basically what current you deliver to a battery when it is in a discharged state. This is usually limited by the charger itself, or in the case of solar panels, what your solar panel array is capable of. Thing is, one must pay attention to the MAX current rate the battery can handle. In the case of the Deka's, they are NOT "pure-lead" types, and have a limitation - if you exceed it with the Deka's, you create "hot spots", that can lead to thermal runaway.

Most "conventional" agm's like the Deka's can only handle about 0.25 to 0.3C max charge current without creating hot spots. So for a 100A battery, your charger or solar array should not exceed this 25A.

In bulk as the battery is being charged, the terminal voltage rises.

ABSORB Voltage:
This is critical. For the Deka's in that chart, they show about 14.4v for absorb. What this means is that once a battery has charged enough to reach about 80% of charge, it starts entering the absorb stage. This stage means that the controller will NOT let the battery terminal voltage rise any higher than 14.4v. As the terminal voltage rises, the ratio between the battery voltage and the absorb stage setpoing gets smaller and smaller, and because of that, less current flows. The end of this stage is typically when a C/20 ratio is reached. That is, when a 100ah battery being charged reaches an absorb current level of 5A, that is considered the end of the absorb stage.

The most interesting takeaway from this is that while the controller merely limits the voltage (aka "CV"), it is THE BATTERY ITSELF that is limiting or ramping down the current! It is not the controller. (yes, some might but for this sake of discussion lets leave it at this)

FLOAT voltage.
AGM's are not fully charged when the absorb phase is over. AGM's *need about 8 hours or more* of float to actually finish the job. The consequence of not doing so is "walking down" your capacity little by little over time as that .01% of uncharged material turns into hard-sulfate.

Dentists will love this analogy - without a good float on an agm, it is like brushing your teeth without flossing. Little by little under the gumline, plaque (um, hard sulfate for us), builds up, lowering overal tooth capacity. You gotta' floss as they always harp. Here, we say you gotta float! But there is a BIG CATCH.

more to follow ...

FLOAT continued .....

The problem with not being able to float an agm for 8 hours or more is simply that in a solar system, we don't have the luxury of time to do so daily.

So that means work-arounds to compensate.

The two most common are

1) apply a weekly preventative-maintenance charge on an AC charger that will do the required float. The longer you space out this PM charging, the faster your capacity reduces from hard sulfation.

2) Set both your absorb AND float to the same value. Ie, a common measure would be to set both of these values to 14.4v. That helps to ensure that you get in as much as you can, as fast as you can before nature's light-switch turns off your solar output. Ordinarily this would fry your battery on an ac charger that did this, but in cyclic service, the loss of the sun helps prevent this work-around from destroying the battery. Again, we are assuming cyclic service. If you are NOT cycling it, then yes, go ahead and set float to something recommended, ie usually about 13.5.

This is very true, ESPECIALLY with "conventional" agm's like the Deka's. Some high-end agm's can take an EQ, but should only be done when reduced capacity is noted. If you do the work-arounds above, you won't have to do this, and you can't with the Deka anyway.

But then why does their chart show "EQ" values?
That is because this is not the traditional EQ that those coming from the flooded world tend to think of. That is, a "traditional" EQ is basically changing your float value to something like 15.5v, and this fries agm's not engineered to do so.

In Deka's case, you'll note that their "EQ" is just about .2v above the normal absorb, and is what would be considered a "light" EQ for most of us, and ineffective for flooded anyway. THIS is what the sales guy was referring to about going no higher than 14.6v.

So again, DO NOT do a traditional EQ with the Deka agm. Disable any such setting in your controller.

The best thing to do is follow one of the two workarounds discussed above.

Note that these voltages are temperature compensated which a quality controller will handle. Either by ambient temp sensing (keep the batteries near the controller) or even better with an "on battery" temp probe.

TIP - the general consensus is that if one is going to make an error with an agm, it is better to OVERcharge it than under-charge it - within reason of course.

With a little experience, you may find a specialized setup such as setting your absorb to 14.6, and your float to 14.4 just the right formula for your setup. But I would first try the canonical

Bulk = absorb = float = 14.4v

as a starting setup. Kind of makes it much easier.

Thanks for taking the time to write out such a lengthy reply! You were spot on I'm just about as green as they get but am learning more and more! I'm sure my questions might seem dumb to most but I was told there are no dumb questions. Once again thanks! I've read this a couple times now and have looked over the pdf's you suggested as well. I also called deka and there customer service lady sent me anther pdf that said Deka 8A & 8G BATTERY INSTALLATION AND OPERATING INSTRUCTIONS.

One thing I picked up from it a section said, "For cyclic applications it is important that the battery(ies) be charged fully after each discharge. It is recommended that 108% to 115% of the Ah capacity removed from the battery(ies) be replaced after each discharge. This additional Ah is to compensate for any efficiency losses between the battery charger and battery(ies)

Question: How do I ensure that I get that additional 8 to 15% of Ah?

You don't. All you can do is set the charge controller voltage 14.4 volts and hope by the end of the day you get fully recharged. To do that requires a good careful design under the worse conditions. In other words the panel wattage, battery capacity, daily use, location, and time of year use are matched up under worse conditions. It also requires a properly sized generator and AC charger to CYA during long cloudy spells and routine maintenance. What the lady is telling you is the batteries are 85 to 92% efficient on the charge side. That is taken care of during the design immediately after you have concluded how many watt hours you need in a day. If you use 1 Kwh per day, your panels must generate 1.5 Kwh in a day under worse case. Part of that overhead includes battery efficiency along with all the other system losses.

To expand upon Sunking's good advice, it is just a way of telling people who don't know, that the electrochemical process of recharging is not 100% efficient. Nearly all batteries are like this, and this is not really a Deka-specific thing.

A common misconception might be "I only took out 10ah out of the battery, so I only need to put back 10ah". OR, "I only discharged it for 1 hour at a constant rate, and therefore only need to recharge at the same rate for 1 hour".

Not so. It is basically a warning telling you not to make false conclusions like the above, and it will take some more time or energy to put back in more than what you took out.

That's guys for the advice another question, I got a Classic 150 and set it up Friday but now need to tweak it a little. For the AGM batteries I currently have how long should my absorption time be set to?

By default the controller looks like its set to 2 hours.

Thanks

You are spinning your wheels. AGM does not have a Absorb Stage. For AGM you want a simple CC/CV algorithm. Here is the deal every charger worth having made is CC/CV aka Float Chargers. When you set your Classic up for Bulk, all you are really doing is setting the charge voltage to Infinite meaning your controller will keep pumping in maximum current until the sun goes down, or it hits a set point voltage which you select. ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ Once Bulk voltage set point, the charger switches to CV mode or Constant Fixed Voltage and that is called Absorb, Float, and Equalize. Only difference between Absorb. Float, and EQ is the voltage set point. Forget all about that with AGM. Look in your battery owners manual for thee Dialy Cycle Charge voltage. For example it might say 2.4 volt per cell. Multiply the number of cells to get your voltage set point. So for 12 cells or a 24 volt battery you set Bulk = Absorb = Float = EQ = 28.8 volts. As far as that goes apply the same applies to FLA because there are not enough hours in a day to ever get you through all 3 stages. With Solar you want to charge as fast as possible to get the most you can with limited solar power.--------------------------------------------------------------------------------------------------------------------------------------------------------------------------With AC chargers Absorb is not a timed event. Absord is when charge current tapers to .03C of a battery at 2.4 vpc or there abouts. That would be 3 amps for every 100 AH of battery. Time has nothing to do with it. For solar manufactures cannot do that. Not enough time in a day. With AC you have unlimited time and power on your side to work with. Solar cannot do that.

We can easily change that to 12 hours. It is very unlikely you'll have 12 hours in a solar day!

Consult Deka's renewable energy charging paramaters guide (1913.pdf) and you'll see this:

"Charge until change in current is <0.10A per Hr / Max Time: 12hr"

Note the max time!! On an AC charger, this would be bad, but because you have nature's light switch working for you, you can let it run all day if you wish.

The 2-hour limit earlier was a conservative rating because they don't know that you might be in a standby/float situation, rather than a cyclic solar setup.

In other words, IF your system was just in standby mode all day every day without any use for 2 weeks minimum, that that would be a standby scenario. In that case, then yes, go ahead and use the 2 hour timer. Ideally, in a standby setup, you wouldn't set absorb at all, and just set everything to 13.5v float.

But if you ARE cyclic, that is using your system frequently, then it REALLY takes well more than 2 hours of absorb to get the most life out of agm. Either that, or very long floats which we just aren't allowed to do because of the sun's own light switch.


BTW, did the Midnite come with on-battery temperature compensation probes? If not, do it right and get them if they don't come with the kit and are an accessory.

Users trying to get the absolute most out of their batteries can also add the battery current shunt and Whizbang Jr. remote sensor to a Midnite Solar installation. With that you will get much better battery capacity monitoring AND you can set the CC to end absorb at a particular current without being confused by current going from the CC to loads at the same time.

Again, I'm pointing out the importance of temperature-compensation.

Most people freak out about "drying out your agm" by using the extended-absorb workaround to compensate for a lack of 8 hours or more of unavailable float time. Much of their experience was solely with the damage of an ac-powered system doing this 24/7. With solar, we HAVE to take into account the unreliability of solar-insolation where in many cases, we don't even reach the start or middle of absorb on a regular basis!

Invariably, they are NOT using temperature-compensation, starting out with some pos dried-out junk already, or are comfortable with walking-down their capacity by switching to float too soon, and for too short a time in a solar day. And in many cases are not using high-quality batteries like Deka, but some little agm pulled from an old ups.

Get your temp-comp working!

Maybe my thoughts should be put into a different thread, but as nifty as this circuit is, and while it makes sense for an AC powered system, it still leaves the agm going to float too soon. That is, after absorb is over we STILL need 8 hours or more of float. With solar, we just don't have that luxury (cyclic that is), and need to use an extended absorb technique to compensate. So setting your absorb > float value based on a particular current level doesn't really accomplish this.

I know it sounds crazy, and most balk at it thinking they are killing the battery by doing so, but forget that we are in a solar environment, and have to use workarounds for agm that scare most ac-powered system owners - and rightly so.

You could at least partially compensate for that characteristic of AGMs by setting the End Amps figure lower than the recommended value.

That would work too under solar where you may never reach that goal and the loss of sun every day prevents severe overcharge, but the warning here is that for stable ac-powered system owners lurking, that would actually be BAD, and instead should set it higher to take into account real-world aging affects.

For an ac-powered system, the way to do it without drying out the agm prematurely would be for the controller to be smart enough to notice a "stall" or no change in absorb current for about an hour as the trigger to proceed to float. As the battery ages, that absorb-current trigger level goes higher and higher over time. Thus if they called for .01C as the trigger, that is when the battery is first-born. Two years later, one may never reach that no matter what they do, and will overcharge the agm when in reality a stall at .025C is called for. Instead, set your end-absorb trigger to a higher value than initially called for, and possibly change it as time passes if you have no way of monitoring for a stall - basically trying to stay ahead of the aging to prevent dangling a carrot on a string always out of reach. And of course some good hours of float afterwards.

But this is primarily for an agm under ac-powered charging. Solar considerations like with frequent poor solar-insolation, operating at sub-par psoc levels for lengthy periods where the agm never truly gets back to 100% charge calls for a different outlook in operations, where slight overcharge is much preferable to slight undercharge. That slight overcharge is hinted at in the Deka manual with the "light eq" corrective measure of only .2V above nominal, as opposed to the flooded eq which of course is much much higher.

So instead of doing a corrective measure with cyclic solar, the easiest way to do all this without going into severe overcharge as one would with an ac-powered 24/7 system, is to keep it simple where everything is set to the absorb value - that is for say a 12v battery, bulk=absorb=float= 14.4v.

Admittedly the simplicity of it all seems incredulous, and when looked at from an ac-powered standpoint it is. Feed the vagaries of non-stable solar, and it makes a good workaround without having to resort to even the "light eq" (only .2v above nominal) technique.