Another question about Absorbtion time during charging...

EagleOne · July 2012

Hello again -

I have been trying to figure out the best 'length' for absorbtion on a battery bank that doesn't get daily use. I have read that a good indication is to watch the current while in Absorb, and if it gets to down 1%-2% of AH capacity of the bank then the batteries should be close to fully charged.

I am going to try to attach a screen capture of a log file from today's absorb cycle. It should show that the amps start low and steadily rise during the absorb stage (but staying around the 1% mark). The absorb stage is currently set for 4 hours. I would have thought that the current would have steadily decreased during the absorb stage. The MS controller is putting a UTC timestamp on the graph, but I am 7 hours behind UTC (just in case you look at the times). I only graphed from the start of the day to the end of the absorb stage but you can see it going from night-to-bulk-to-absorb-to-float modes.

My batteries are at least a decade old, and I am figuring they have at least 80-85% capacity left judging by the AH used and the voltage readings after letting them sit for a few hours. I can draw more than 400AH and still have above 12.2V after resting.

Does anybody have any input on my batteries/how they are accepting a charge? (Other than the fact that they should be 24/48V and have a heckuva lot more panel

) Would it be better to have a longer absorb time? I can set it to any amount of time, the factory default came at 180 minutes on the AGM setting I am using but I set it to 240 minutes just for an experiment.

Thanks for any input,

- Chris J.

Cariboocoot · July 2012

Re: Another question about Absorbtion time during charging...

What is wrong with this picture?

It should show that the amps start low and steadily rise during the absorb stage.

Absorb is a constant Voltage charge state: the current should start out high and gradually reduce as the batteries near full charge.

The trouble with charge controllers is that they show the current going out of the controller, which is not necessarily the same as the current going in to the batteries. Note the sharp current drop when Float is achieved, followed by fluctuating current during Float. Like most of us you leave your system running while charging; the loads will affect the current from the controller. This is the main reason why End Amps isn`t ideal for for determining Absorb time length.

You could try turning off the system one day and just let it recharge the batteries and see what you get.
You could buy an Outback or MidNite charge controller.

I can draw more than 400AH and still have above 12.2V after resting.

How do you know you have 400 Amp hours drawn? 12.2 Volts at rest is good.

EagleOne · July 2012

Re: Another question about Absorbtion time during charging...

Cariboocoot -

Why am I not surprised that you are here offering input?

When I said 'It should show that...' I meant the picture of the graph (just in case it didn't post correctly), and I was expecting pretty much what you mentioned about starting 'high' and tapering down on the current. I am curious if the 'steady increase' in current according to the graph/reality is telling me something about my batteries.

I have a Bogart TriMetric that I use to determine the loads drawn from the batteries. I am trying to set the parameters on the Bogart to match the actual charging from the MS MPPT controller so that the logic on the Bogart will be in sync with the 'reality' of the solar system. I am trying to see if I can get the Bogart to 'reset' to 'fully charged' at roughly the same time the MPPT goes into float, and today it did pretty close to that. I changed the 'efficiency' parameter to 91%, I think it defaults to 94% IIRC and it seems to be closer now.

My parasitic load is usually about an amp or less - I have a wireless router running 'directly' off of the 12V batteries so I can monitor the system from a PC in my shack in the backyard. Other than that, today's load was nothing. I put a load on them overnight a few days ago and it took two partly sunny days to get them to go into float yesterday, so today is when I figured it would be good to watch the absorb cycle a little closer.

Thanks again, I appeciate the input.

- Chris J.

Cariboocoot · July 2012

Re: Another question about Absorbtion time during charging...

Chris, you may be on to something here. 10 year old AGM batteries are likely suffering greatly by now. If these were flooded cells you could check the SG - and likely be disappointed.

Bogart battery meters are great, but you`re probably going nuts trying to get it to work on batteries that old. It`s much easier with new batteries that have a known capacity.
(For those wondering you start out programming the monitor to know the batteries have `X` Amp hour capacity and `Y`efficiency.)

You probably can not pull 250 Amp hours out of the batteries without the panels connected. But really some form of discharge test is in order, as the `reverse current behaviour`during Absorb looks a lot like some sulphation being knocked off causing a change in internal resistance (sulphor off plates, resistance goes down, current goes up). Most AGM batteries do not recommend equalization. Even if yours do it probably won`t do any good at this point.

That light of a load can be discounted, I think.

EagleOne · July 2012

Re: Another question about Absorbtion time during charging...

Coot -

I am curious about what you think about a couple of things. First, you mentioned a 'discharge test' - so what can I do to get a good simulation of that? I currently use my friends RV to put a load on my batteries, and I have my Dometic icemaker that I can add to it that can put a pretty constant load on them, about 2KWh+ per 24Hrs running non-stop just for the Dometic (gotta have ice for the cold drinks, right?!)

Secondly, would it be advisable to check the performance if I set the absorbtion stage for, say, 12 hours and see what happens? I have the day off on Monday so I could monitor the batteries for absorbtion current and temperature, etc. to make sure nothing gets out of hand. We should have a partly sunny/cloudy day on Monday that 'should' be enough to maintain an absorbtion voltage level for most, if not all, of the day. I wouldn't expect the batteries to get much hotter than about 80F if even that (not counting any heat by keeping them in absorb for so long).

I would like to keep the batteries servicable for at least the rest of the summer and get what I can out of them. I am thinking about reconfiguring the system to switch to 24V, but I would need a new 24V inverter and also a 24V-12V supply to keep some existing 12V lights, etc. so that will most likely have to wait until the winter time. It is absolutely mind-boggling on how fast the costs can add up for 'free energy'!

- Chris J.

Cariboocoot · July 2012

Re: Another question about Absorbtion time during charging...

You've got the discharge idea right: steady current draw over time. Of course as battery Voltage goes down current goes up because the load wants a certain amount of Watts, not Amps. This is why it's hard to do. But if you can fix a Watt hour draw over a given time that's not too far off. Then you have to be able to measure the SOC at the end. This is another problem as you can't count on the Bogart being accurate under the circumstances. You may as well take a before and after resting Voltage and see what you get. That also won't be particularly accurate. This is one of the toughest tests to do without the right lab equipment. Nearly impossible, in fact. But you're more or less looking for relative results.

Looking at the utter lack of solar panels you have for 1000 Amp hours of battery, how on Earth are you charging them? The Morningstar's recording will only be accurate as far as it is concerned: if there's charging coming from another source (inverter charger for instance) the solar current can go up during Absorb because the other source is dropping its current (differences in Voltage set levels for instance). 630 Watts of panel won't do much for the combined battery bank; are they separate? I hope so.

Mostly it's hard to imagine batteries of any sort lasting 10 years and still being anyplace near their original condition, no matter what the operating circumstances.

EagleOne · July 2012

Re: Another question about Absorbtion time during charging...

I think I will wait until the system goes into 'night' mode and then change the absorb time to 12 hours. I can set the monitor to record the voltage to watch it overnight, and that should give me a fairly decent resting voltage in the morning when the sun comes up. I will continue to monitor it during the day during the bulk and absorbtion stages, as I don't think I will get into float with a 12 hour absorb. And let them rest for the night again to see what the voltage is before putting a load on them.

I will not put a load on overnight, with the exception of the wireless router so I can monitor the system. Just to further clarify, I have the router setup in bridge mode, and no PC in the 'RV'. I have this PC (that I am typing this response on) hard wired to my other router in the house. I will post the graph of the 12 hour absorbtion on Monday night to show the results of the extended absorb time. I am hoping to see the current taper off at some point during the absorb stage.

As mentioned, these are old batteries and any use I can get out of them is basically a bonus. What is that old phrase? You get what you pay for. And considering that these cost me exactly $0.00 then I look at it as a bargain. I have used a cheap 'store bought' so-called 'multi-stage' charger to maintain them over the years and then I bought a Samlex 60A three stage charger to keep them maintained for the last year or so. I still have the Samlex, but it has been disconnected since I added the solar panels at the end of May, first week of June. I have had the batteries installed in the van on a 'permanent' basis for a year+, that is why I got the Samlex - which is also installed in the van (but currently disconnected). Before that I had the batteries in the garage wired up 10 in parallel with the cheap charger plugged in 24/7. It did/does have an AGM mode that would maintain 'proper' voltage levels, but no temperature compensation, and I could only monitor Amps using the display on the cheap charger (not sure of it's accuracy). By the way, in the garage the batteries were hooked up to + and - busses that were 1/4" copper plates using 2 AWG wiring, but not exactly the same length for each battery - varied between 8" to ~14". Now, all my cables are the same length, still 2AWG going to 1/4"x3"x10" bussbars on each bank. 1/0 from the batterey bussbars to some more (+) and (-) bussbars where the panels and loads are connected (with a shunt for the Bogart in the (-) line to combine both banks before the second level of bussbar).

[Looking at the utter lack of solar panels you have for 1000 Amp hours of battery, how on Earth are you charging them?]

I am going to use that quote as a constant reminder that I NEED more panels. What the heck, it's only money right?

Maybe I should have put your quote/question before the previous paragraph? I would like to replace the batteries with new SunXtenders (or similar) after my 'learning curve' has somewhat leveled out (and my slush fund has gone up). Mo' money, mo' money! And yes, I will NOT wire them in a 12V configuration. See, I am learning!

- Chris J.

EagleOne · July 2012

Re: Another question about Absorbtion time during charging...

Well I shut off the panels overnight using the breaker between the panels and the charge controller. I waited until a little after 12 noon when we had full sun after the morning clouds cleared out before I turned the breaker back on. At that point I had the batteries at 12.7V and should have been fully charged still from yesterday, or at least as full as they could get.

Within 10 seconds of connecting the panels, the controller went into absorb mode at a high current and it tapered off quickly, like one would expect. It did start to slowly rise like it did yesterday, and I figured the experiment is pointless beyond this time so I turned the breaker off again so the controller thinks it is nighttime now.

Thanks again for your input.

- Chris J.

SolaRevolution · July 2012

Re: Another question about Absorbtion time during charging...

Cariboocoot wrote: »

... steady current draw over time. Of course as battery Voltage goes down current goes up because the load wants a certain amount of Watts, not Amps. This is why it's hard to do. But if you can fix a Watt hour draw over a given time that's not too far off. Then you have to be able to measure the SOC at the end. This is another problem as you can't count on the Bogart being accurate under the circumstances.

Splitting hairs for what it's worth-

Doing the load test to determine how many Watts you can draw seems like a more useful measure of your battery's health. Testing how many amps can be delivered is how you can measure a battery's health as compared to it's rating. This is what is typically used for warranty purposes.

When testing for Watts it is appropriate to use the loads you regularly use. After all, this is the "cart" your "horse" must pull.

When testing for amp hours you need to use a DC load sized bassed on the C-rate of the the batteries. That is, if your batteries are 1000ah @ C-20 your test load should be a 50 amp DC resistive load. You can then take out 25-30% of the rated capacity and see how it is reflected in the SG or at-rest-voltage. (Rolls Surrette has requested load tests based on C-10 for some of the warranty claims I've looked into) Some battery manufacturers provide charts of the length of time the batteries can be expected to support different load levels before the vpc (volts per cell) falls to a given point. Most of these charts list pretty deep DODs so I hesitate to really make use of them. The time it takes to do a full test is usually cost prohibitive while on the clock anyway.

The Trimetric is actually very useful for amp load testing. Press the select button for 3 seconds and then repeatedly until you get to the resettable cAH (Cumulative Amp Hours).

I've been meaning to set up a DC load tester. I've been thinking about using a diversion load air heater on a potentiometer so I can set the amps and a diversion controller which can track amp/hrs out. I wonder, would a Tristar PWM would be able to provide graphing for this? I've tried to use a grid tied inverter to "sell" the load test current so I would not just be wasting the precious energy but keeping a good steady current flow is not an easy task.

So far, for doing amp load testing, I've been using an AC heater with a 750 watt and a 1500 watt settings and a few clamp lights with 100w incandescent bulbs to keep the amps as close as I can to the desired range. Time consuming but effective.

-Alex

Vic · July 2012

Re: Another question about Absorbtion time during charging...

Hi Alex,

I built a fan-cooled DC load bank with selectable R values, but this selection is fairly coarse, but the Rs were selected for the largest banks here, and can do full bank discharges for 20-hr capacity. For me, the dificulty comes when needing to do a single 2-volt cell discharge, as the required R values are SO small. An electronic load would be nice. So I will wait for you to do all the work, and will ask you for the data so I can copy your work.

I, too, use selectable power level AC heaters, BUT, the "ceramic" (PTC) jobs do not burn their rated currents, at least not after a few seconds of running, and the Nichrome wire element jobs seem much worse than the PTC ones regarding the accuracy of rated power consumed on Low.

We also need a high current single-cell battery charger, perhaps selectable 2-4-6-12 Volts at 60 or so Amps. Please get to work on that as well ! Just kinna kidding. Vic

Another question about Absorbtion time during charging...

Comments

Categories