Abusing an SLI battery so you don't have to!

[PNjunction]

Using an SLI (starter-lights-ignition) battery for long-term power draw is probably the #1 rule of things NOT TO DO.

They are designed for short power bursts only, typically taking only 5% of the capacity to start a vehicle. Their grid-pastes, electrolyte acidity and such are not designed for long term power draws. SLI batteries are the WORST CHOICE in our application.

But, from a self-reliance standpoint, I'm going to put one to the test to see how long it will last. And how fast I'm going to waste my money. And how I won't hold the manufacturer responsible for this type of abuse.

My quest was to answer the question "what if this was the ONLY thing around?"

I'm going to follow some rules however. What I'd call the 40/10 rule. I hope it works.

That is, the battery will be discharged no higher current than the 20hr rate divided by 2. And will only be discharged to no more than 10% of it's capacity to stretch practicality a little. So that means low power stuff for limited amounts of time.

The victim is a brand new Optima Red-Top which is a STARTER battery. Mine is rated at 44ah. Obviously I don't care about CCA, RC and the like.

The load will be a simple inverter / LED bulb as the load, pulling no more than 2.2a. For only 2 hours at a stretch. Recharge will be FAST, but an allowance for an extended absorb or long float will follow each cycle to ensure it will be recharged well.

The overall idea is to conserve power even more, yet still abiding by the max of 10% capacity withdrawn.

Poor thing. We'll see how long it lasts, but won't be blaming the manufacturer for my misuse of the product. This is just a test for an extreme / worst case scenario.

I don't advise anyone else do this. We'll see what happens, but hopefully this will take some time. I'm on the second cycle now.

[solar pete]

Using an SLI (starter-lights-ignition) battery for long-term power draw is probably the #1 rule of things NOT TO DO.

They are designed for short power bursts only, typically taking only 5% of the capacity to start a vehicle. Their grid-pastes, electrolyte acidity and such are not designed for long term power draws. SLI batteries are the WORST CHOICE in our application.

But, from a self-reliance standpoint, I'm going to put one to the test to see how long it will last. And how fast I'm going to waste my money. And how I won't hold the manufacturer responsible for this type of abuse.

My quest was to answer the question "what if this was the ONLY thing around?"

I'm going to follow some rules however. What I'd call the 40/10 rule. I hope it works.

That is, the battery will be discharged no higher current than the 20hr rate divided by 2. And will only be discharged to no more than 10% of it's capacity to stretch practicality a little. So that means low power stuff for limited amounts of time.

The victim is a brand new Optima Red-Top which is a STARTER battery. Mine is rated at 44ah. Obviously I don't care about CCA, RC and the like.

The load will be a simple inverter / LED bulb as the load, pulling no more than 2.2a. For only 2 hours at a stretch. Recharge will be FAST, but an allowance for an extended absorb or long float will follow each cycle to ensure it will be recharged well.

The overall idea is to conserve power even more, yet still abiding by the max of 10% capacity withdrawn.

Poor thing. We'll see how long it lasts, but won't be blaming the manufacturer for my misuse of the product. This is just a test for an extreme / worst case scenario.

I don't advise anyone else do this. We'll see what happens, but hopefully this will take some time. I'm on the second cycle now.

Your quest for knowledge PNjunction knows no bounds, thanks for doing what you do, cheers

[PNjunction]

Hopefully nobody gets the impression that this is a cheapskate work-around for getting the right type of battery in the first place.

I know how to hurt them easily and have done so in the past. Then again, one can tickle them to the point of being impractical, like using 40 lbs of lead to charge a cellphone.

Thus I'm going to extend the DOD to a max of 20%. In my case, that would be around 8-9 ah total per cycle. My concern over the max discharge current limit is trying to protect the grid paste designed for short bursts. Long draws may change it chemically even if at the .1C rate, so I'm really going to take it easy during discharge with no more than .05C.

Honestly, I'm just guessing at this point but the goal is to drive a low-current load perhaps through most of the night if need be, knowingly with the wrong battery type, and recharge asap the next day with plenty of time to finish absorb and even extend it - solar permitting of course.

The test is two-fold, in that I'm also putting GE / Sylvania led's to the test to see if they hold up with an MSW inverter waveform, or fail within a month.

I was kind of hesitant about announcing what I'm doing, but for the sake of science, and pointing out to newbies that this is basically WRONG from the start, I had to.

[thastinger]

In for the results, I'd bet you'll still be trying to kill that battery a year from now.

[SunEagle]

In for the results, I'd bet you'll still be trying to kill that battery a year from now.

I would second that estimation. Those Optima's are pretty stout and with only a 10% DOD I expect it to last a lot longer than a standard car starting battery.

[PNjunction]

Additional notes:

Missing from most anecdotal internet stories are full details. Here are a few more:

To make things a bit more practical, I'm using a 20% max DOD now.

Red top had a date sticker of 1/15, so it was already about 8 months old. OCV as measured on a Fluke 87V was 12.42v when I got it home.

The first 3 cycles were performed with a Tecmate-Optimate 6 charger. It passed the short term and long-term testing. I wanted to see if the Optimate would detect any issues right off the bat as it is no ordinary smart cc/cv charger. Supposedly the oscillating absorb helps balance the cells internally - well at least make sure each individual cell really got charged in addition to other charge-curve testing procedures. I proved that to myself with my homebrew 12v battery made out of 2v Hawker Cyclons which were deliberately unbalanced as a test, so I'm pretty confident I'm starting out on the right foot.

The last two cycles were charged with a 15A Samlex SEC-1215A, which dutifully put out 15A in bulk as measured by a clamp-on ammeter. Using an IR handheld thermometer, I inspected each cell up and down and did not notice any hot-spot trends, other than a minor increase in temp at the positive terminal. Thus, I'm pretty sure I got one that didn't suffer any retail / customer damage or vandalism.

I know I'm going to forget about it at some point, so I toyed with the idea of using a programmable Powerwerx ITS-12 LVD disconnect. But, since I'm using inverters with 10.6v lvd's anyway, that will do to prevent a catastrophic discharge since I'm pretty much on top of things. Kids - this is NOT what I use as my cycle-stopper - this is just a *catastrophic* discharge protection. I never plan to go that low.

Speaking of cycles, each cycle will have at least an hour between charge and discharge - when I have time. Knowing that just jack-hammering cycles out of it will just be beating the parasitic clock, this test will be under more real conditions - during weekdays I don't expect to be doing more than one cycle.

Solar consists of anything from an 80 to 120 watt panel. That should charge up just fine in my solar insolation area - even in December. I'll be using a Xantrex C12 pwm CC which has absorb and float BOTH set by me to 14.6v to help get in some extended absorb time daily.

My experience with Optimas in the past of all colors has been very good - even prior to my interest in solar. Thing is, I ALWAYS charged them completely once I got them - whether I did a parking-lot swapout or just brought one home they always went on a charger to ensure a proper "first birthday" so to speak. The charger was a standard linear cc/cv and not an agressive "speed charger". No maintainer was needed since I don't have a lot of parasitic drain, or the vehicle gets driven daily, but in cases of weeks of just doing starbucks runs, yep - I'd treat it as well to some tasty electrons with the charger.

This isn't really about mistreating an Optima as a brand choice, I merely chose that as an easily obtainable step up from the usual ups-style generic agm's.

I'm a little freaked out about the lower voltages under long continuous loads, as I have never really put an SLI battery to this test before.

[PNjunction]

IR measurements

As an additional precaution, I took a baseline of the IR of each of my Optimas that serve different purposes. They line up with the published specs more or less.

Although not a lab instrument, I used my iCharger 306b to check them while at full charge:

D25 Red Top (this is the one I'm going to really keep my eye on during this misapplication-test:
published .003 ohm measured .0035

D34M 75ah Blue Top
published: .0028 measured .0030

D51 38ah Yellow Top
published .0046 measured .0049

I've been wanting to this and the red-top test finally got me around to it for the others as well.

[Living Large]

For some reason, and don't ask me why, your post reminds me of the guy that got "some really good batteries" at WalMart, wanted to parallel 4 strands of wire for both sides of his PV to CC wiring to economize, and posted his diagram sketched on a napkin. His friends were helping him by buying parts, and his dad wouldn't let him return some or all of the stuff that had been piled up, when it was advised he return what he had so far.

Back on topic, this should be an interesting experiment. Good luck with your baby.

[PNjunction]

Heh, it really is a silly test using a battery not designed for the task right from the outset.

I'm learning something but can't put my finger on it yet. Well, other than $150 slipping through them. What I learn better be worth it!

[inMichigan]

format of the results

Hello,

What are you measuring that when plotted against time will be the 'show'? (example: kwh withdrawn between 100% to 80% SOC)

Thanks for sharing your experiment,
inMichigan

[PNjunction]

I'm just going by published specs at this point - 44ah total capacity and only taking 20% from it, ie only drawing 8.8ah (9ah for convenience) since I'm discharging well below the 0.1C rate.

While I do have a West Mountain Radio CBAIV battery analyzer, I could use that but that's overkill for this test. I just measured current at the battery terminals when my load consists of a simple LED bulb and a cheap msw inverter.

So at night when I get home from work, I run the little setup to provide kitchen lighting for many hours (but no more than 9ah total) before hitting the sack. (I'm testing the led/msw combo for mtbf at the same time). So at least it is providing a useful service to me while testing.

I'm familiar with how a healthy Optima should react over the years, and just made sure that it got a REAL good commissioning charge. (hint: I sleep with the Enersys Odyssey manual under the pillow, and use that, rather than the over-simplistic Optima consumer specs..)

Right now I'm looking for sudden-death trends, will keep an eye on the internal resistance, and at some point allow the CBA-IV to do a total ah test - but the catch is that to do that initially, means taking it below where I want to go! I'll save that more or less destructive test for last.

[PNjunction]

Odyssey was right - I *AM* undercharging my agm!

I knew it before, but had to see it before my own eyes. That is, we undercharge our agm's for the most part by being too conservative, especially in the cyclic mode.

I never really could tell because it is hard for me to purposely abuse a battery, and see the "walk down" in capacity with limited lab gear and most importantly a limited time-factor. I have to work to keep buying chargers, instead of tending to all my battery charging during the day. I do have some sort of life beyond batteries.

BUT, with 80-100F heat that won't quit, using an Optima SLI with moderate mis-application slow discharges, and following up with charging with at least an hour of rest, I am FINALLY seeing a downward trend when not FULLY charging! ( I do my charging when it's 80F)

Basically, Odyssey (and even Optima!) point out that just finishing absorb IS NOT ENOUGH. You either have to:

1) Do an IUI charge, which no commonly available consumer charger does for a few hours. (Deltran / Battery Tender may have some special order ones, you won't see them at most auto-parts stores!) This is a constant current of about 1-3A for 2 hours or so, AFTER absorb, with either a 15.5v cap, or perhaps higher. One has to be careful here, as applying this constant current needs to be done at the VERY END of absorb, and not before, otherwise you are just spinning your wheels getting nowhere gassing. This is basically to jam it in before a VERY short turnaround, like factory shift changes for motive power utility vehicles.

2) OR, failing that, following absorb with a 16-24 hour float which Odyssey mentions - which has to be HIGH ENOUGH to actually be effective. Ie, 13.5-13.8v.

Interestingly, back in 1998 or so, the IUI protocol for Odyssey was discussed in the EV/Cyclic application manual, but nowadays, you are directed to speak to an engineer about it. Makes sense since no consumer charger is designed for IUI that I know of. This also makes sense to prevent the unknowing to think that just cranking the voltage up without knowing what you are doing will destroy the battery. It has to be done carefully if you are going to do it at all, and with some sort of precision.

My recommendation is not to just "crank it up" if you have some other source, but do a long float instead. Do it at the wrong time, and welcome to the sound of rice-crispies. Warning - Enersys knows how to detect this abuse when they cut your return open.

Thus, those chargers that just drop to float but only monitor, or do so at too low a voltage are not actually finishing the absolute job of charging. Maintenance is another matter, although I'd like to avoid the 13.2v vs 13.5v "float wars" seen a few years ago. If you are a large commercial ups user, you aren't using auto-parts store chargers anyway, and count ROI down to the last penny, so I don't lose too much sleep over it, although I prefer 13.5v. Note that I use Deltran chargers too when applicable, especially in my hot-weather environment, so I'm not a hater....

What did this look like? It meant that when I applied even an 8 hour float of at least 13.5v, and then measured the OCV after a 12 hour rest, I see about a .03v improvement each and every time.

In other words, when I charge without a long float on the red-top: 12.92v
With a long float: 12.95v

I did this enough times already to try and be as accurate as I can when taking the measurement at nearly the same ambient temp for both me and the battery with the IR thermometer.

It may not seem like much, but it did confirm my suspicions that the engineers were right.

So now what about solar? Yep - for cyclic agm use, set both absorb AND float to 14.7/14.8 and fugeddaboutit. You'll never get the 8 hours of absorb time in that you really need, nevermind the additional float that is needed yet again beyond that!

What it comes down to is a practical measure for solar. Unless you can program your controller to do IUI, you are stuck doing the best you can with IUU, and therefore just jam in as much as you can before the sun goes down with BULK=ABSORB=FLOAT=14.8v or so. I like temp-comp too.

Maybe those too afraid to set both absorb and float to the same value, or those CC's that drop to float too early, could gingerly test the waters by at least setting their float to the highest voltage available, like 13.8 or whatever is the top limit.

So I *am* learning something first hand, even though it was already pointed out by the quality agm engineers. Guess I'm the type that has to learn the hard way.

SIDE NOTE for a different chemistry:
Interestingly enough, even AA rechargeables using the most advanced charger on the planet, the Maha C9000 does much the same thing! That is, for maximum capacity, when absorb is finished, two hours of lower voltage, 60ma/per hour current is applied before dropping to a TRUE maintenance trickle of only 10ma per cell. This isn't mentioned in the manual, but us AA geeks got the scoop from the Maha engineers. The improvement is seen when we do our discharge capacity tests.

At some point, and endless absorb won't accomplish the full job here either. Good enough for most consumers, but not us fanatics that have the gold standard in AA engineering.

It was interesting to see that from a general battery principle, you still need a good float to actually finish the job. Whether that is practical depends on the consumer.

[PNjunction]

Caution against bench-racing specs!

Thought I'd point out that the previous entry about not charging our agm's to the fullest was aimed at improving our solar charging techniques, and not fodder for bench-racing specs amongst charger manufacturers.

Unless you are a commercial user who needs ROI down to the last cent, ALL of the small / medium chargers, with perhaps the exception of speed-chargers not really designed for "normally healthy" batteries, do a GOOD job, and is certainly better than nothing when you apply them on a regular or even semi-regular basis when you look at the whole picture of your environment for improvement.

Or do as I do - like potato chips - you can't own just one charger!

[SunEagle]

Great information. Thanks for doing this testing.

I know I will be doing the charging of my Optima a little differently. Hopefully others will learn how to extend the life of their batteries with your help.

[PNjunction]

Thanks!

I'm trying to keep it real, and not be dancing on the head of a needle. I'll leave that to the marketing folk.

For instance, with a fully charged D31M at 75ah, if you place the common 800 milliamp "tender" on it, even after a few days rest, it should finish the absorb and go back to float within an hour or less. If not, you may have not been truly fully charged in the first place. We are assuming of course that this isn't some badly sulphated battery, but a new one gotten from a reputable dealer within a reasonable amount of retail storage.

That is the kind of non-lab, garage-gear testing tips I'm gathering with all this stuff. Talk is good, but only goes so far. Things that the average guy like me can repeat and duplicate inexpensively with real-world availability are even better.

Note that I quickly added a control back to my original quest: I'm taking a 20ah CSB evx12200 ups-style agm down to 50%, while the Red-Top only down to 25% max with the same charging regimen on each regardless of solar or ac charge. Will the mis-application of the red-top beat a ups-style small agm? We'll see.

Admittedly this is getting a bit strange and out of hand.