Log in

View Full Version : Battery health tesing


January 22nd 06, 03:17 PM
I'm wondering if there's a simple or straightforward way to test an SLA
battery for general condition. Or a device I can buy that sucks a
charge out of a battery over a few hours and reports its actual
amp-hour capacity?

~ted/2NO

bumper
January 22nd 06, 03:53 PM
Google <ELK> and <battery tester>. Elk makes one that measures the internal
conductance of the battery in mhoes. My company uses them to test customer's
batteries with good results (alarm company).

bumper
> wrote in message
ups.com...
> I'm wondering if there's a simple or straightforward way to test an SLA
> battery for general condition. Or a device I can buy that sucks a
> charge out of a battery over a few hours and reports its actual
> amp-hour capacity?
>
> ~ted/2NO
>

COLIN LAMB
January 22nd 06, 04:02 PM
The cheapest way to measure the battery health is to put a load similar to
what you will be using and time how long the voltage is adequate to operate
your equipment. you do not have to be a techie to do this - just a simple
understanding of ohms law.

Depending upon the size of the battery and the battery current you wish to
draw, an example that we will use would be a Radio Shack 20 watt, 8 ohm
resistor catalog number 271-120. Current draw is calculated by using ohms
law: I (current in amperes+ = E (voltage) / R (resistance). In our example,
I = 12 / 8 = 1.5 amps. Power consumed would be P = EI = 12 x 1.5 = 18
watts. That means the load resistor will get hot under test, so keep it
away from things that will melt. Make sure there is good circulation around
it.

Take a cheap digital voltmeter (expensive ones will work) and monitor the
voltage from time to time. Be sure to note the time you start the test.
Stop the test when the voltage gets down to the lowest voltage you can use
in your aircraft.

Let us say you start the test and 4 hours later the voltage gets down to
10.25 volts and you terminate the test. You will have drawn 6 amps out, so
you now have a 6 amp battery. Often batteries are rated to go down to a
lower voltage than you wish to operate them at, so the new amp rating may
not be as much as advertised.

If you are really cheap, you can use an automobile headlamp or taillamp and
hang it on the battery and just watch how long it stays bright. You will
have to know how much current is drawn, which you can sometimes look up in a
database, or measure with an ammeter.

When doing tests like this, a little math gets you a long ways. This is a
somewhat simplified explanation, since the actual current drawn depends upon
the voltage and as the voltage drops down, the current draw will decrease -
but that is the same thing that will happen in your aircraft.

Colin N12HS

Andy
January 22nd 06, 04:12 PM
I did a lot of battery testing years ago. My PC was set up to sample
and plot voltage and current as the battery discharged into a resitive
load. It was easy to see a battery was good. It was easy to see a
battery was bad. It did not help me predict when a good battery would
become a bad one.

Try searching this group, it's been discussed before.


Andy

COLIN LAMB
January 22nd 06, 04:38 PM
Atthough I mentioned a simplified method of battery test, once you
understand the procedure, you can measure the battery voltage under load and
after the initial few seconds, you can plot the discharge voltage vs time.
So, if in five minutes, the voltage drops .03 volts, we know the voltage
drop is linear, so in 60 minutes the voltage will drop 12 x .03 or .36
volts.

As Andy mentioned, it can be computerized.

Interestingly, the ability to hold a charge also means the resistance to
change. So, if you see a battery on the shelf and put a charger to it and
the voltage soars instantly to full voltage, it is bad.

Once you understand what is happening, you can usually tell if a battery is
good within a few seconds, and know the projected life in a minute or less -
all with minimum equipment. A lttle knowledge gets you a long way. In
fact, you can often hang a auto lamp across the battery with a digital
voltmeter and watch the voltage drop. The less drop, the better the
battery.

Colin

Eric Greenwell
January 22nd 06, 04:46 PM
wrote:
> I'm wondering if there's a simple or straightforward way to test an SLA
> battery for general condition. Or a device I can buy that sucks a
> charge out of a battery over a few hours and reports its actual
> amp-hour capacity?

The model car/airplane folks have battery testers/chargers that will do
this, but I haven't seriously looked for one.

The other posters have mentioned load testing, which is what I do with
my batteries, but I'm a geeky former electrical engineer and enjoy that.

If you are talking about the typical $15-$20 12 volt, 7 ah glider
battery, there are other approaches that take advantage of their low cost:

* if you think it's down on capacity, get a new one.
* if it's more than 3 years old, get a new one.
* and for the truly frugal, wait until the first time it fails/almost
fails in flight, and get a new one. Not a bad technique for someone with
two batteries in the glider.

--
Change "netto" to "net" to email me directly

Eric Greenwell
Washington State
USA

jke
January 22nd 06, 10:42 PM
Eric Greenwell wrote:
> wrote:
> > I'm wondering if there's a simple or straightforward way to test an SLA
> > battery for general condition. Or a device I can buy that sucks a
> > charge out of a battery over a few hours and reports its actual
> > amp-hour capacity?
>
> The model car/airplane folks have battery testers/chargers that will do
> this, but I haven't seriously looked for one.
>
> The other posters have mentioned load testing, which is what I do with
> my batteries, but I'm a geeky former electrical engineer and enjoy that.
>
> If you are talking about the typical $15-$20 12 volt, 7 ah glider
> battery, there are other approaches that take advantage of their low cost:
>
> * if you think it's down on capacity, get a new one.
> * if it's more than 3 years old, get a new one.
> * and for the truly frugal, wait until the first time it fails/almost
> fails in flight, and get a new one. Not a bad technique for someone with
> two batteries in the glider.
>
> --
> Change "netto" to "net" to email me directly
>
> Eric Greenwell
> Washington State
> USA

jke
January 22nd 06, 10:51 PM
I have had been very happy with the CBAII computerized battery analyzer
from Westmountainradio.com. Click the RC Model Products button on the
left of the page.

It runs abou $100 and hooks to a PC to plot discharge curves.

They also offer the whatthour meter which provides load information.

More expensive solutions than some others suggested here, but very
convenient and easy to use.

Vaughn
January 23rd 06, 01:48 AM
"COLIN LAMB" > wrote in message
ink.net...
> In fact, you can often hang a auto lamp across the battery with a digital
> voltmeter and watch the voltage drop. The less drop, the better the battery.

That is exactly how I have been doing it for years, and I have never found
a need for anything more fancy. For small gel cells, (as are used in gliders) I
use a digital voltmeter and a 50-watt sealed beam light. For larger batteries,
I use a commercial load tester that you can buy at NAPA or any similar auto
parts store for about $50.00 US.

Vaughn

Eric Greenwell
January 23rd 06, 02:03 AM
Vaughn wrote:
> "COLIN LAMB" > wrote in message
> ink.net...
>
>> In fact, you can often hang a auto lamp across the battery with a digital
>>voltmeter and watch the voltage drop. The less drop, the better the battery.
>
>
> That is exactly how I have been doing it for years, and I have never found
> a need for anything more fancy. For small gel cells, (as are used in gliders) I
> use a digital voltmeter and a 50-watt sealed beam light. For larger batteries,
> I use a commercial load tester that you can buy at NAPA or any similar auto
> parts store for about $50.00 US.

It sounds like you are effectively doing an internal resistance
measurement. Do you have chart of volt drop, load, and good/ok/bad
decision for various size batteries? I've not seen that sort of thing in
battery data sheets, so I assume you have to develop the numbers
empirically.


--
Change "netto" to "net" to email me directly

Eric Greenwell
Washington State
USA

Georg Holderied
January 23rd 06, 02:57 PM
wrote:

> I'm wondering if there's a simple or straightforward way to test an SLA
> battery for general condition. Or a device I can buy that sucks a
> charge out of a battery over a few hours and reports its actual
> amp-hour capacity?

I've built Peter Nyffelers amperehour tester:

http://www.ppc.ethz.ch/~peny/sf/ah_test.html

The simple circuit is easy to build out of standard components.

It discharges the batteries with a fixed constant current until the voltage
is down to 10.5 V then it switches the battery back to the charger.

With a discharge current of 1 A the timer shows the capacity in ampere
hours.(at 1/2 A multiply by two).

This tester is convenient when you take care of a lot of batteries:
- Charge battery (I do this with a second charger)
- Connect the battery
- Reset the hour meter
- press the discharge button
- The next day the battery is fully charged and the hour meter shows you its
capacity.

Sometimes it is possible to revive tired batteries with a few cycles of
discharge and recarge (use a smart 3 phase charger).
If 5 cycles don't help you can give up.

George

January 23rd 06, 04:10 PM
Georg, what is the "M1 Elapsed Time Meter"? This looks like a simple
enough project but I'm not sure I can identify enough of those
components (ask me about software!). I do know how to handle a
soldering iron though...

Vaughn
January 23rd 06, 11:38 PM
"Eric Greenwell" > wrote in message
...
> It sounds like you are effectively doing an internal resistance measurement.
Exactly. I find that internal resistance is a great measure of the health
of a battery.

>Do you have chart of volt drop, load, and good/ok/bad decision for various size
>batteries?
No, but I usually consider any drop over a volt or so to be suspect
(assuming an appropriate load for the size of battery). In my work, I often
have the advantage of having identical batteries to compare against each other.

>I've not seen that sort of thing in battery data sheets, so I assume you have
>to develop the numbers empirically.

True, but for auto-sized batteries the tester I mention (a very common
type) comes with markings that make a great starting point.

Vaughn

Google