Maximum amps for testing AGM capacity?

[Naio]

After a winter of being very sick, and a spring of catching up on farm chores, I am finally back to being able to work on the van at least once in a blue moon. Next step, testing the 100aH Fiamm telecom batteries I got used. I've been topping them off periodically, but otherwise haven't done anything with them since hitting them with the meanwell to recover capacity last year.

My question: How many amps are ok at continuous load, if I want to run them each down to 12 volts for testing? Is 30 ok? 60?

I have the van in storage, so I can't hang out for too many hours during testing. And, since I finally got this giant heavy batteries to the van, I don't really want to move them back to my house :-). But I have killed AGM deep cycle batteries before with too many amps, and certainly don't want to do that.

In case it matters, things I have available for load are a couple of 300 watt heaters, 600 or 900 watt ones, lamps with 60 watt bulbs, and assorted minor loads. I don't have anything 12-volt, so I will be running this off an inverter.

[BFL13]

Discharge voltage and SG per SOC is fairly linear unlike recharging, so it is "good enough" for the first 10 hours though.

That article PT linked elsewhere about "partial tests" made the point that it is not so linear, so you had to do all 20 hours, for real accuracy, but I have done many 10 hour runs with Wet 12s and 6s and 12v AGMs and the numbers come out right, including when cross-checking SG, voltage, and AH count.

I did see that a 5 hour test to 75% is not so good. I think the reason is not linearity, but that the measuring tools' margins of error are too much for such a short time run.

BTW, I don't know if they mean to drop to 10.5 loaded in 20 hrs or if the 10.5 is after bounce back. If the load is by inverter, you can't get that low before the inverter quits from low voltage anyway.

[pianotuna]

Hi BFL13,

Discharge rate is not a straight line relationship, so probably the ten hour test suggested takes the battery a bit lower than 50%.

Ten hours is a good number to use,especially this time of year.

I do push my batteries harder than most folks (179 amp draw to run the microwave, this morning). I wish LI could be used in the cold. Now, where did I see firefly jars on sale? *grin*

[BFL13]

You are not being clear about that 12v. The voltage during the test is the loaded voltage. That bounces back after the test, and that voltage is the one that tells you your SOC--but only after time has passed so you are closer back up to a "resting" value.

Eg with Wet batts, where 12.2 is 50%, they will be at 11.6v under a 20 hr rate load at 50%, and will bounce back to 12.2 by next day. AGMs have higher voltages per SOC, but not all AGMs are the same for that either. Mine are about 12.4 at 50% and 13.0 full.

You need a way to tell what your batteries' SOCs are, to do a capacity test. An AH counter is the easiest way. You can use SG with Flooded but not with your AGMs. Voltage per SOC does not work unless you have a good voltage per SOC table for your type of battery in their spec sheets, and it is a next day thing to take the measurement.

IMO you cannot do a capacity test while popping by the storage yard for a few hours to run a test. You can do a test to see if the batteries are any good at all, but that might be all you want to know.

If you put a load on them and their voltage does not "collapse" but goes down in a regular way, you are good to go. Some time when you have the measuring equipment and the time, and shore power, you can do a proper capacity test. I do it with my Trimetric for AH counting, and only run from full down to 50% or 10 hours, whichever comes first. Then just double the AH. No need to do the 20 hours all the way down.

[wolfe10]

For useful life, you will not want to discharge (taking about in use, not testing) below 50%. Don't know the specifics on your batteries, but suspect that will be around 12.2 VDC.

And depending on what amp load you will be running, the batteries will give more hours if wired in parallel as one battery bank. Yes, Mr Peukert is alive and well.

[Naio]

Thank you Ed.

I'm not sure standardized testing is what I need, here. And taking them down to 10.5v sounds awfully rough on the batteries.

I think what I need to know is, how many usable hours can I get out of each battery, and I think going down to 12 volts will tell me that. Am I wrong?

[Ed_Gee]

Something very important here has not been mentioned. The standard method to measure a 12V Battery is to discharge the fully charged battery at a constant rate down to 10.5 volts. If you have a 100AH battery, to measure the standard 20 hour AH rating you would apply a constant discharge of 5 amps and if the battery is 100% up to spec it will not reach 10.5V until 20 hours have passed. If you are measuring a 2 volt cell, you would draw it down to 1.75 volts.

[mchero]

Keep us posted on your tests,
Thanks

[Naio]

I have four, and they are not connected to each other at this time. I will be testing them separately, partly to see how I want to connect them into, hopefully, two banks later on.

[pianotuna]

Mr Peukert is alive and well.

[wolfe10]

But the load for a proper load test WILL depend on whether they are being individually tested or are all wired in parallel.

Said another way, load is a dependent on the size (amp-hrs) of battery, whether in one jar or many in parallel.

[pianotuna]

Wolfe10,

Naio wishes to run an inverter for the draw down part of testing, as she has no 12 volt loads. I don't remember how many jars she has.

[wolfe10]

Naio,

Please see my post above.

Not sure I am the only one who doesn't know if you are asking about testing each battery individually or if they are in parallel to form a larger battery bank.

[pianotuna]

Hi Naio,

MrWizard and I have similar batteries, though mine specify 139 amp-hours at the 20 hour rate. I got the 1c discharge number from the Victron book.

[Naio]

Thank you, everyone!

MrWiz, it is great to have your experience with theze specific batteries. And Don, victron book looks way cool.

[Sam_Spade]

Naio wrote:

My question: How many amps are ok at continuous load, if I want to run them each down to 12 volts for testing?

How about something near to what you expect as a "normal" load ?

I see no point in finding out how long it takes to get down to 12 with a huge greater than normal load.