Testing lead acid batteries

[Audiorepair]

OK, so I have pulled a sealed lead acid battery from a portable powered speaker (busker type amp).
It's 12v 5Ahr similar to:
https://batteryclerk.co.uk/products/long-way-lw-6fm5-5aj-sealed-lead-acid-agm-vrla-battery

What is the best way to quickly test it?  (It reads 13v at the moment)
The owner says it doesn't keep much charge any more, the unit puts out 14 or so volts of charging voltage with it disconnected, not that that might indicate much.

I'm guessing I could connect a load and then determine its internal resistance, though I don't know what numbers I should be looking for.
I was going to just take it to a mechanic friend of mine on the way home, as he will have a CCA tester etc, but thought I should really be able to do this myself without having to buy such a meter for very very occasional use.


Cheers.

[Audiorepair]

Tried an 8 ohm dummy load, and the voltage visibly drops continuously, clearly showing a bad battery.

I was kind of wondering what these CCA testers do.
I can't see them passing hundreds of amps through themsleves, so thought maybe there was some test approximation method you could simulate quickly on a bench.

But then I think you have to manually enter the CCA rating marked on the battery, and it will then test somehow and tell you what ballpark it's actually in.



Thanks.

[bdunham7]

Tried an 8 ohm dummy load, and the voltage visibly drops continuously, clearly showing a bad battery.

That's about a 1.5A load, so I'd expect the voltage to drop continuously, just not very fast.  If it drops to under 10.5V in less than an hour, then I'd agree it is bad, or at least reduced in capacity--assuming it was fully charged to begin with.

[Alti]

I can't see them passing hundreds of amps through themsleves

CCA is a cold craning amp rating specified in some normalized way (I do not remember which norm). IIRC it says that at -18oC and at CCA amps the voltage across terminals should stay at or above 9.6V. So this CCA just specifies internal resistance of a cold cell. Don't quote me with exact values but lets suppose 300CCA FLA has 12.6V open circuit voltage at -18oC and 3V are lost during 300A which means the battery internal resistance is 3V/300A = 10mR or less.

So the meter measures this impedance (briefly puts some load and measures voltage drop) and then does some chemistry correction to room temperature so that you did not have to perform those measurements in the freezer. Whether it pulls 1A, 100A or 1000A during CCA test - IDK.

[IanB]

I was going to just take it to a mechanic friend of mine on the way home, as he will have a CCA tester etc, but thought I should really be able to do this myself without having to buy such a meter for very very occasional use.

CCA is "cold cranking amps", it is a test mainly intended for car batteries which have to provide hundreds of amps to turn over the engine.

It is not really an appropriate test for a small portable battery.

The best way to test it would be with a small load as others have mentioned. Try a 1 amp load (like a 12 V, 12 W bulb) and see how many hours it runs before the voltage gets too low.

[Audiorepair]

Tried an 8 ohm dummy load, and the voltage visibly drops continuously, clearly showing a bad battery.

That's about a 1.5A load, so I'd expect the voltage to drop continuously, just not very fast.  If it drops to under 10.5V in less than an hour, then I'd agree it is bad, or at least reduced in capacity--assuming it was fully charged to begin with.

Sorry, forgot to include a time frame. 
It went from 13v to 11v in less than 10 minutes, it seemed obvious just watching it was not a good battery.
Though I have no way of telling how charged it was in the first place.

[floobydust]

Typically with old SLA I find one weak cell, so the battery can appear fine with good terminal voltage at rest - like 13V but even mild loads and it drops right away too much, under 12V and then stays there for a long time. Acting like a 5-cell battery during discharge.
The weak cell quickly sees reverse polarity so it doesn't last very long at all and the battery has pseudo high resistance. Charging is also wimpy, it won't take high-rate charging.
I have not tried the newer method of measuring the battery's ESR.

[Audiorepair]

I can't see them passing hundreds of amps through themsleves

CCA is a cold craning amp rating specified in some normalized way (I do not remember which norm). IIRC it says that at -18oC and at CCA amps the voltage across terminals should stay at or above 9.6V. So this CCA just specifies internal resistance of a cold cell. Don't quote me with exact values but lets suppose 300CCA FLA has 12.6V open circuit voltage at -18oC and 3V are lost during 300A which means the battery internal resistance is 3V/300A = 10mR or less.

So the meter measures this impedance (briefly puts some load and measures voltage drop) and then does some chemistry correction to room temperature so that you did not have to perform those measurements in the freezer. Whether it pulls 1A, 100A or 1000A during CCA test - IDK.

OK, so this presumably pulse testing under load is a reasonably accurate way of measuring the battery's internal resistance.

Which could/should be applicable to other lead acid batteries such as the one I have, to determine its serviceability or not, without having to drain it for a few hours under a load.

Clearly the current CCA testers are only designed for car batteries where current/volts vs temperature is critical in determining whether it is good enough to start a car in cold weather.
But perhaps a PIC etc could be programmed to pulse test more generic batteries based upon sampling its internal resistance and comparing it with an assumed new one.

Probably something you couldn't really easily simulate on a bench with a load and a multimeter, though buying an automotive tester is not likely to give any meaningful data either.
Which is kind of where I am coming from on this.


Edit:  There must be thousands of such cells out there doing Solar Panel duties, is there not a quick test for these?

[SeanB]

I built a quick and cheap one, using a spare plastic case and some leads from a dead boost pack. The unit uses a cheap panel mount meter, showing battery voltage, and the test load is applied via a push button switch. A chunk of old heater element, to give around a 10A load to the SLA battery, so you put it on, look at the voltage, and press the switch for 10 seconds, or till voltage drops below 10V, whichever comes first. If it times out without dropping below 10V it is still useful, but generally if it dies in 5 seconds it is dead, and if the battery rattles not even worth bothering.

Did however test a spare angel eye lamp set, leaving it on a dying 18Ah battery for the last 5 days, and today it finally is barely lit, just lighting up all the dies still, but current is very low, and it barely provides light. Supposed to be a 320mA current per unit, but now probably under 10mA, though not going to put a meter in. Want to see if it will do a week plus, though the second lamp has a little leakage on the one segment of 16 LED dies, as those dropped out this morning, leaving the other 2 in series still lit. now only one left on it.

[floobydust]

I do use moderate load (light bulb) testing first but it's so subjective sometimes to determine "weak".
I think the problem with heavy load (CCA) testing is you don't know the battery's state-of-charge (or temperature or rated Ahr capacity corrections).
False fails and staring at a panel meter needle sinking is also hard to really to gauge. Where's the stopwatch lol. I think it's an old school method, mechanic's shops have changed over to something easier to use.

ESR testing - you enter the battery's size, state of charge does not matter as much, gives an estimate of available CCA as well. It's pulse-loading. You can use a decent capacitor tester to measure the battery's ESR (with blocking cap) as well, but you need some idea of what the ESR ballpark is. There's no tables out there yet.

[Audiorepair]

ESR testing - you enter the battery's size, state of charge does not matter as much, gives an estimate of available CCA as well. It's pulse-loading. You can use a decent capacitor tester to measure the battery's ESR (with blocking cap) as well, but you need some idea of what the ESR ballpark is. There's no tables out there yet.

Thanks for that, I just tried measuring the ESR with a blocking cap, and on my cheap and not very accurate ESR meter, it is showing around 0.5 ohms.
I suspect this is way higher than it should be, but will perform the same test on the (looks identical) battery I have ordered, and see how that compares.

I know from experience that measuring ESR on caps varies enormously, large PSU caps can be close to zero, and crappy SMD electro's closer to 10 ohms, so many orders of magnitudes of difference there.
But you soon get to know the right ballpark and can make good decisions based on your measurements of old caps.
Perhaps the same could be true of lead acid batteries in their various shapes and sizes.

[IanB]

I guess you could measure the ESR of the blocking cap by itself, and then in series with the battery, and then the difference would be attributable to the battery? I think the internal resistance of a healthy lead acid battery would typically be very low, probably in the milliohms range.

[Audiorepair]

I guess you could measure the ESR of the blocking cap by itself, and then in series with the battery, and then the difference would be attributable to the battery? I think the internal resistance of a healthy lead acid battery would typically be very low, probably in the milliohms range.

That's what I did.

I use a Peak ESR meter, I don't trust it to be particularly accurate, but it's easily good enough to find bad caps, which is all I need it for.

So the 0.5 ohms I measured will be compared with that from the good new battery when it arrives, and then at least I'll have an idea of what good and bad looks like on a battery this size.

[floobydust]

I measured ESR on some old SLA lying around: 12V 7Ah 0.04Ω, 12V 7Ah 0.03Ω
These two are OK but sag and (in series) cannot power a UPS for more than a few seconds: 6V 7Ah 0.10Ω, 6V 7Ah 0.08Ω

[Audiorepair]

Well the new battery arrived today, and it measures 0.36 ohms ESR.
The old one still shows 0.5 ohms.

Not quite the difference I was hoping for to use as a definitive quick test in the CCA style for a dead battery.

Unless my tester has issues testing a battery in series with a cap.
It seems quite happy measuring cap ESR down to almost zero, 0.36 ohms is a lot higher than I was expecting.

Oh well.

[shapirus]

Well the new battery arrived today, and it measures 0.36 ohms ESR.

are you sure your meter measures it correctly?
using TR1035/YR1035, which is designed exactly for measuring internal resistance of batteries, I'm seeing numbers around 3-6 mOhm on bigger (70-100 Ah) batteries, and internal resistance is, roughly, inversely proportional to capacity, so I'd expect something of the order of 60-80 mOhm max in a 7 Ah battery.

in any case, 360 mOhm in a new battery sounds too much.

p.s. I've just measured a pack made of two 7 Ah SLA batteries that were used in a UPS, they are half-dead and have seen much better days. Connected in parallel, they measure 13 mOhm, meaning 26 mOhm average per battery.

[Audiorepair]

Well the new battery arrived today, and it measures 0.36 ohms ESR.

are you sure your meter measures it correctly?

No,I don't think it does.

[IanB]

No,I don't think it does.

Out of curiosity, what happens if you put both batteries in series with the blocking capacitor and measure the ESR of all three combined? Specifically, put the two batteries in inverse polarity so that their voltages cancel out, i.e. connect plus to plus or minus to minus. I'm wondering if the imposed voltage of the battery is upsetting the ESR meter?

[Audiorepair]

No,I don't think it does.

Out of curiosity, what happens if you put both batteries in series with the blocking capacitor and measure the ESR of all three combined? Specifically, put the two batteries in inverse polarity so that their voltages cancel out, i.e. connect plus to plus or minus to minus. I'm wondering if the imposed voltage of the battery is upsetting the ESR meter?

That measures 0.74 ohms, the cap is 0.02 ohms tested separately.  So roughly the 0.5 old battery plus what it thinks the new one is.
The meter reads the capacitor value correctly using the series battery method, it thinks its seeing the cap.

I just don't think think this unit is capable of doing the battery ESR thing.
It has a 12v internal battery if that means anything.

https://uk.rs-online.com/web/p/lcr-meters/6858791?cm_mmc=UK-PPC-DS3A-_-google-_-3_UK_EN_Test+%26+Measurement_LCR+Meters_Exact-_-Peak+Electronic+Design+-+6858791+-+ESR70-_-esr70&matchtype=e&aud-827186183686:kwd-299736340931&cq_src=google_ads&cq_cmp=12689582349&cq_term=esr70&cq_plac=&cq_net=g&cq_plt=gp&gclid=CjwKCAiA85efBhBbEiwAD7oLQDznRXkkjtwBISa2mBwrZ4A2nER1kmncDBqbXAqQyLyt-wGCVH7QJBoCdekQAvD_BwE&gclsrc=aw.ds

[Audiorepair]

It thinks a 9v PP3 battery is 8.8 ohms, but then measures the cap as half what it should be.
(1000uF 50v electrolytic)

I'll just use it for testing caps, its quite good at that.

[trobbins]

Audiorepair - perhaps take a step back and not assume that test methods for one type of battery apply to other types, or that your perception of health of the battery is valid.  You appear to have a small capacity VRLA monobloc used in an application that likely is designed to discharge the battery over a few hours.  It is not appropriate to try and test that battery under high discharge (<< 1hr) conditions, or with ESR type measurement, and expect to get a quick and ready assessment of condition.

Imho, you should separately charge the battery at rated float voltage until its float current subsides (which may take a few hours) - as float current is one indicator of health.  Then discharge the battery - and do that at constant current to make it easier on yourself to interpret the capacity - but that test has to be done to a lower limit voltage.  Then recharge and repeat.  This all takes a lot of time, but provides the appropriate data for battery health.  Taking shortcuts just means you are making a poorer assessment of health, and perhaps misleading yourself and the battery owner, especially as you have no experience in this area.    This type of health testing is typically written up in manufacturer app notes and service sheets, which you should track down for your battery (or equivalent model/make).  If you don't follow the manufacturer levels, especially for max discharge rate and lower voltage limit then you will likely be damaging the battery yourself without appreciating it - not a great outcome for the owner of the battery.

[Audiorepair]

Audiorepair - perhaps take a step back and not assume that test methods for one type of battery apply to other types, or that your perception of health of the battery is valid.  You appear to have a small capacity VRLA monobloc used in an application that likely is designed to discharge the battery over a few hours.  It is not appropriate to try and test that battery under high discharge (<< 1hr) conditions, or with ESR type measurement, and expect to get a quick and ready assessment of condition.

Imho, you should separately charge the battery at rated float voltage until its float current subsides (which may take a few hours) - as float current is one indicator of health.  Then discharge the battery - and do that at constant current to make it easier on yourself to interpret the capacity - but that test has to be done to a lower limit voltage.  Then recharge and repeat.  This all takes a lot of time, but provides the appropriate data for battery health.  Taking shortcuts just means you are making a poorer assessment of health, and perhaps misleading yourself and the battery owner, especially as you have no experience in this area.    This type of health testing is typically written up in manufacturer app notes and service sheets, which you should track down for your battery (or equivalent model/make).  If you don't follow the manufacturer levels, especially for max discharge rate and lower voltage limit then you will likely be damaging the battery yourself without appreciating it - not a great outcome for the owner of the battery.

I think you are missing the point.


It should surely be possible to determine the health of a lead acid battery without spending hours doing so.
Not all of us have that time to spare.

The automotive industry seems to have made strides on this by developing CCA tests that can be done via a small handheld unit in less than 10 seconds.

[james_s]

It should surely be possible to determine the health of a lead acid battery without spending hours doing so.
Not all of us have that time to spare.

The automotive industry seems to have made strides on this by developing CCA tests that can be done via a small handheld unit in less than 10 seconds.

It's not difficult. They can put a large load on the battery and measure the voltage drop to measure the internal resistance, the duty cycle can be very low allowing the testing device to be compact.

[trobbins]

An SLI lead acid battery is not the same beast as a small VRLA monobloc - you are making that assumption with little to no appreciation or insight into VRLA batteries.

Did you even track down manufacturer documentation on the battery you were given?

[floobydust]

Unfortunately ESR does not seem to show an order of magnitude difference with bad battery packs? It's actually important in EV's - who wants to piss around testing 100's of lithium cells. My patience limit is 6 cells lol.
https://www.eevblog.com/forum/renewable-energy/battery-impedance-measurement-esr/msg3299610/#msg3299610

I believe some ESR testers are whimpy in that they have a low test current, perhaps not to make the chemistry work enough to show problems. Or the test frequency is not ideal.
I use the Blue ESR Meter which can do 50mA pulses. But car battery ESR testers are 6-10A discharge pulses, around 100Hz though. Then they estimate CCA based on that.
Hioki uses a Nyquist plot.
So I 'm not sure why SLA are fooling the ESR test.