Test a lead-acid battery "in-the-field"

Tell us what the charger is, your description is entirely inadequate. Measuring battery capacity is well established practice.

NT

Il 24/08/2017 11:58, snipped-for-privacy@gmail.com ha scritto:

That was a description of an actual circuit. However I'd like to change it to improve.

Actual implementation is:

AC/DC ------|>-----------*---- to the load | BAT --*---|>-----------* | | --/\/\/\-- Measuring battery capacity is well established practice.

How to measure the battery capacity?

----------------------

** Simple, put the battery through one full charge and discharge cycle. Compare performance with new battery under same test.

Is the battery an SLA type or other?

..... Phil

To measure it accurately, but most field test devices simply check the voltage drop when the battery is put under a moderate to heavy load. If it drops too far or fails to recover quickly the battery is shot.

It is a quick and dirty guide to battery health or impending failure.

Adding an indicator LED, resistor and a zener in parallel to the test load so that the LED is lit if the terminal voltage is adequate and out if it drops too far. Some crude testers were basically just a crude 12v filament bulb glows brightly = OK dimly = NBG.

The charging circuit needs to be smarter too if battery longevity is to be maximised on float charge for immediate use at variable temperature.

--
Regards, 
Martin Brown

No, because you're reinventing the wheel. You also didn't supply enough information on what you're working with. Little details like the type and size of the battery is always helpful. The quality of the answers you receive are directly related to the quality of the information you supply.

So, I'll limit myself to a warning and some general advice: Those are 7 commercial battery chargers normally sold at automobile parts stores. I retrieved them from various mountain top weather stations after some of them killed a few lead-acid flooded cell batteries. It's fairly easy to design something that will charge a battery. It is NOT easy to design one that will operate over a range of temperatures, varying battery conditions, varying loads, deal with cell balancing, intermittent sources of AC power, power factor correction, grounding issues, and still produce a reasonable battery lifetime. This cannot be accomplished with a simplistic battery charger and usually requires a uP to record and monitor the situation as well as decide how to charge and when to stop charging.

You might want to look at some of the better battery chargers such as:

Is for testing the lead-acid flooded-cell battery SoC (state of charge), that is most accurately done with a thermometer and battery hydrometer. The specific gravity of the electrolyte can be converted to SoC.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Yeah. A single fixed-voltage "float" charger like this probably won't bring the battery up to full-charge after it's partially run down (it won't do the "saturation-phase" charging). And, without temperature compensation of some sort, it might accidentally overcharge.

Practical suggestion: buy a Deltron "Battery Tender" (full-bore or "Junior" depending on how much current you think you need during charging). They aren't expensive, they have a good reputation, and the tests I did of one using my deep-cycle Optima battery indicated that it performed just as advertised (constant-current bulk charging, constant-voltage saturation charging until the current level drops, and then switch to a lower-voltage float/trickle charge).

You can do two things to test:

(1) Hit the battery with a heavy load for a few minutes, and see how much the voltage drops. Sort of the equivalent of a "cold cranking amps" test for a car battery. This is a quick test which will catch batteries that are truly failing.

(2) Hit the battery with a moderate load (say, 10% of its rated ampere-hour capacity), for some hours, and discharge it down to the safe limit (I wouldn't go below 10.5 volts), while monitoring the voltage.

The latter is what the West Mountain Radio battery analyzer does. A deep-discharge test like this is probably the only real way to tell you what the battery is actually capable of delivering when it's called upon.

In either case you'll need to recharge afterwards, and if an emergency occurs during your test cycle then the battery won't be able to deliver all of what it could.

Also, you're going to need an effective DC load having a predictable resistance. Don't omit the cooling - the load is going to get hot!

Harbor Freight has a cheapo tester that is basically a voltmeter, a resister and a switch. Test procedure is to check the voltage and, if suitably charged, turn the switch to short the battery through the resister and read the result on the meter. A simple & effective test and a good starting point for planning.

Hul

pozz wrote:

Maybe I didn't explain my goals.

I don't need a third-party battery charger/tester... I need to design a simple battery charger/tester in my electronic system. Most important, the test of the battery must be done automatically (every day, for example) in the field, without user intervention.

As an example, take a look at this product [1]. It's a power supply for KNX systems, i.e. it's an AC/DC with an integrated charger and tester for sealed lead-acid (SLA) batteries. So they name it as "uninterruptable".

They have two points for batteries connection: one for lower total batteries capacity/charging current, the second for higher total batteries capacity/charging current.

They say a battery test is automatically perfomed every 15 minutes (the user can manually start the test with a button), so I argue the total duration of the test is around 1-5 minutes (otherwise the battery is always in test and not in charge). A potential-free contact shows the status of batteries: ok or fault.

This is what I'm trying to do.

This is a possibility. Suppose I have a 7Ah SLA battery. Could you suggest a value for the "heavy load", duration of the test and the criteria for the battery fault condition?

I can't discharge the battery so much, because I need an efficient battery when the mains goes away, and this happens at random times.

Oh yes, maybe it's better way. But I think it's not possible "in field", while the battery is connected to the backed-up system. For example, I don't think UPS performs this type of test.

However in the first test, the battery is discharged only a little. So I don't expect the "mains failure back-up time" is different if the "mains failure" happens during battery test or not.

[1]

So first you need to design a competent charger for lead acid batteries. Information on this is widespread, google is your friend. What you had was just wrong for a lead acid.

that rules out full discharge of course

then do it.

we don't know at what point you consider your battery to be inadequate for your task.

NT

It's an add-on to some of their battery charger models: "The Charge Wizard is a microprocessor-controlled unit that constantly monitors the RV battery voltage and then selects one of three charging voltages and one of four operating modes to properly re-charge and maintain the RV battery."

Power converter with desulfation mode

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558