Battery Testing Methods

[Dockhead]

I have been posting about my battery agonies on another thread. My battery fun has now brought me to the point where I need to test my batteries somehow in order to determine whether I can bring my bank back to life, or whether I need to toss them all. Whether a bad battery or three could be replaced without replacing the whole bank of 8.

So far I have been using classical methods of voltage and specific gravity.

But voltage tells nothing about the condition of the battery, and not all that much about the state of charge. A well sulphated battery like mine will show high voltage after charging, but still have little capacity. The low specific gravity seems to show that the sulphation has reduced the effective surface area of the plates so much that there is not enough to interact with the electrolyte to bring it back up to high specific gravity (my amateurish guess; if anyone understands it better, please correct me).

So I'm looking at battery testers and have been reading some interesting materials from people who test large numbers of batteries in cell phone towers and other industrial installation.

There seem to be three methods:

1. Timed discharge under load. I remember my surveyor used this method to test batteries on the Oyster 485 I almost bought a couple of years ago. A big load is put on each battery for a certain time, and the voltage drop is measured.

2. Electronic tester using a micro-load. Same idea I think -- except that a very small load is used, and the response of the battery is analyzed digitally.

3. Conductance. Internal resistance of the battery is measured, which is supposed to be a highly correlated proxy for battery capacity. Described here: CONDUCTANCE TESTING QUESTIONS AND ANSWERS


Conductance testers are available for less than $200 or so, for example:

Amazon.com: Argus AA300 Basic Digital Battery Tester and System Analyzer: Automotive


Using one of these methods, it seems that we can tell not only state of charge, but something about the condition and capacity of our batteries. Has anyone done any of this kind of testing?

[Wotname]

Only ever used method 1, both professionally and privately. In essence, this method closely duplicates what you actually use the battery for in real life!

I have found it a reasonable method.

[Dockhead]

More information:

Large Pulse Resistance

Argus claim to have invented yet another method -- Large Pulse Resistance

[avelectro.org]

1 and 3 method used. The 1 method is suitable and enough for most cases. when checked out the used batteries, it makes sense to check the electrolyte level and density first, if the battery construction that allows, of course.
LPR method seems to be a computerized method number 1, but there is a likelihood of inaccuracy when checking high capacity batteries due to their chemical inertia, and limits of the check pulse current at 100 A - imho the current value of the check should depend on the capacity of the battery checked, maybe it will be taken into account in their new testers.

[GordMay]

Greetings and welcome aboard the CF, avelectro.org.

[Undadar]

I know several people that use the West Mountain "Computerized Battery Monitor" (West Mountain Radio - Computerized Battery Analyzers) and rate it very highly. There are probably several manufacturers of such a device. I agree that this type of load/time test is the only way to really know what shape your batteries are in.

[ggray]

Here's a test suggested in a 1993 Ample Power catalog:

Charge the battery to 14.4 v and hold it there for an hour. Let the battery rest overnight. Check voltage in morning.

12.8 v or more .... battery good
12.6v-12.7v ....marginal
12.4v-12.6v ....very poor
less than 12.4v .... worthless

This may predate gels, and AGM.

[Adamante]

Originally Posted by Dockhead

Using one of these methods, it seems that we can tell not only state of charge, but something about the condition and capacity of our batteries. Has anyone done any of this kind of testing?

If all else fails, you can always chew your way through the testing procedure outlined in http://www.cruisersforum.com/forums/...ies-12707.html

It's a sticky in the Batteries section of the forum, so someone apparently thought it might be of some use.

[Roy M]

This is going to piss off some folks, but check out: http://pulsetech.net/

I've been using and referring these products to people for over twenty years. Most recently, an engineer, for whom I was installing an electronics package, tried it out on his four-year old 8D House bank. It was a flooded cell battery that wasn't holding a charge. He put on a PulseTech and left it in place for two months, then retested the specific gravity and battery capacity. Both had improved substantially to the point he was actively considering not moving to a larger bank. He did, but he now has the unit on the new 740 amp hour bank.

Give it a try, or don't. It's only your money and your decision. There are a lot of folks who have them and swear by them. There are lots who would never try them and condemn them. Oh well......

[barnakiel]

My first test is the age. Oftentimes: an old battery = a bad battery. Because of use, or because of being left half-charged, or discharged, over long time.

Then I charge the battery till full, with a multi-stage, properly sized, charger.

Then I will apply a fixed load for a specific time, then I will let the battery rest for a time, then I will measure the voltage.

Bad battery will not accept the charge, hold the charge, deliver the expected amps without falling below the expected volts.

Older type batteries will also allow you to inspect the tops of the plates visually. The look of the electrolyte will also tell a story.

b.

[hellosailor]

Dockhead, the old load tester is "brute force" and requires basically a high power heating element in the tester, as the load. The new testers replace brute force with microelectronics and in theory give you a much faster answer while not further depleting your battery. But they cost gobs more.

Any battery distributor or warranty center will be using one or the other, these days usually the new electronic ones. For your personal use any of them will do.

As will the old simple method: Charge fully, apply a load for a few minutes, let stand overnight and measure the voltage. Applying the load and letting it stand takes off any surface charge so what you're measuring the next day IS an accurate picture of condition. If you really want to verify that, find a cheap load (12v heater or car headlights) of a known value, hook it up, and clock how long it takes before the load dims or the voltage drops.

All the newer better more expensive stuff? Just gives you the same answer in less time.

[daddle]

Originally Posted by ggray

12.8 v or more .... battery good
12.6v-12.7v ....marginal
12.4v-12.6v ....very poor
less than 12.4v .... worthless

This may predate gels, and AGM.

Yes, but no. These numbers must be adjusted for battery temperature to be useful.

Checking the specific gravity after a full charge provides really good info. If the sulfate is not in solution as H2SO4 acid the specific gravity will reflect that and the battery needs service or replacement.

Current tests are good. But is not that what we do by simply using the battery in service for a few days?

[chala]

I use a Projecta 12 volt carbon pile 500 amp load tester. It give me an idea, after consideration, of which of my 5 batteries is the weaker and may need replacing.