Cordless power tool battery tester

Thread Starter

oneminstrel

Joined Dec 8, 2010
4
Greetings all!
Just found this forum. What a cool place! I hope someone can help me with a project I'm working on. I need to build a device to load test cordless tool batteries from 12 volts up to 28 volts, with a 3 AH max. Can anyone help me with a schematic? I would be deeply grateful.
 

Audioguru

Joined Dec 20, 2007
11,248
A battery test results in a bunch of numbers. Who will decode the numbers?
I recently had my car's battery tested and they said I should buy a new one (of course from them). It measured "only" 306CCA instead of its rated 550CCA (when brand spanking new).

Guess what? My car battery is over 5 years old and still works fine. It might fail in a couple of years or more so why should I replace it now??

Why do you want to test batteries for cordless tools? They work or they don't and a test won't tell you anything.
 

wayneh

Joined Sep 9, 2010
17,496
A "test" of a battery is only relevant to the load it will be used for. In use, all we usually care about is hours of service before it needs to be recharged. So one of the most useful tests of a rechargeable is the remaining AH capacity versus its new spec. Darn tough to measure, except by just using it.

There are commercial battery testers available but they're not cheap.
 

Thread Starter

oneminstrel

Joined Dec 8, 2010
4
Why do you want to test batteries for cordless tools? They work or they don't and a test won't tell you anything.
Sorry if I wasn't clear enough. I want to test batteries as a diagnostic tool to help determine a defective battery for my customers. I repair power tools for a living. A load test WILL tell me if I have a bad battery. What happens with cordless power tool batteries is, they can show they have a full charge, but when you put them in a tool and run it, the voltage drops off significantly. Therefor the tool runs slow, or in some tools, not at all. I have measured a few 18 volt fully charged batteries that I put in a tool, and when I ran that tool, the voltage dropped off to about 5 volts. Indicating a defective battery. It was not easy to measure. A good, fully charged battery only drops a couple volts when running. I need a schematic to help me build a "test box" to test batteries.
 

Audioguru

Joined Dec 20, 2007
11,248
A new name-brand battery should be strong.
But a no-name-brand, old or misused (too many deep discharges) battery could be garbage waiting to be replaced.

My new RC model airplanes use Li-Po batteries that work for a long time when new then their discharge life gets shorter and shorter after many flights. The power is still good after a charge but it lasts for only a few seconds when the battery needs to be replaced. It also charges very quickly to its new lower capacity.

You forgot to tell us the chemistry of your batteries.
 

tyblu

Joined Nov 29, 2010
199
You forgot to tell us the chemistry of your batteries.
Cordless tool battery chemistry varies. I'm sure you've seen the "LITHIUM POWER" ads at home improvement stores, and many are lead-acid. Adjustable load circuit shown above will do the job fine, though you'll want to beef it up to handle more power (ie: 2kW).
 

jpanhalt

Joined Jan 18, 2008
11,087
I think the OP wants to measure battery capacity, not tool power. So, for example, if he has a 28V, 3AH battery,he wants to measure the capacity. That is usually done by charging the battery, letting it cool, and then discharging it to a known level. There are lots of those circuits on the Internet and in application notes by chip manufacturers. John
 

Kermit2

Joined Feb 5, 2010
4,162
Battery conditioners, are circuits that discharge and recharge batteries to improve their capacity and operation.

Battery testers do not have to discharge a battery completely, but, merely place a heavy current demand on the battery and show it's voltage output at that current demand. This will give a good indication of the condition and capacity of the battery.
 

wayneh

Joined Sep 9, 2010
17,496
I think that's probably the simplest approach for the OP: Place the battery under load for x seconds, measure voltage. I think it might make sense to have several loads as appropriate for different tools.

The cheapest and simplest loads are light bulbs. Easy to see they're working, cheap and easy to replace, no worries about heat sinking, etc. You could have 1 bulb for small batteries, 2 for medium, 3 for large, whatever. And it would be easy to let it run even an hour or more to see how the battery hangs in.
 

someonesdad

Joined Jul 7, 2009
1,583
There are commercial DC loads that will load the battery for you. I've used two B&K models: the 8500 (about $1k) and the 8540 ($500). These devices are basically some MOSFET transistors on fan-cooled heat sinks with control electronics. You can set them up to maintain a constant voltage, constant current, constant resistance, or constant power (depends on the model) at the terminals. The 8500 can be computer controlled; an example of testing a battery is given in a DC load application note. The 8540 is a small package, but it is manually-controlled only. There are a number of other companies that make DC loads, but since they're aimed at either the engineering development or manufacturing test markets, they're going to cost a decent chunk of money.

The nice thing about a programmable DC load is that you can use it to simulate an actual load on a device. For example, for a power drill's battery, you could turn it on for a burst of 4 second loads to simulate driving screws, then a longer load for simulating drilling a hole through a chunk of metal. This is nice for R&D stuff, but probably not needed for characterizing batteries to estimate if they're good or bad. I'd imagine the test would be to charge the battery, measure its no-load voltage, load it at a constant current or constant power for some specified time, wait a specified time, then measure the no-load voltage again. The difference between the two no-load voltages would be used to decide whether the battery was good or not. The bad news is that you might have to engineer this information yourself, as the tool companies probably won't release that information from their R&D groups. Here's an article about measuring the internal resistance of a battery; when you see the required stabilization current and the short, (relatively) heavy current pulse, you can see why a programmable DC load is useful to make these measurements.

If the cash isn't available, you can build a suitable load. Here's one thread on the topic. You can also search the web and find numerous designs of a constant-current device that uses an op amp to control a MOSFET (the IRF540 is a popular choice); one input into the op amp comes from a constant voltage source and the other comes from the voltage across a shunt resistor that measures the current; the difference between these two controls the gate voltage of the MOSFET.
 

Thread Starter

oneminstrel

Joined Dec 8, 2010
4
Battery testers do not have to discharge a battery completely, but, merely place a heavy current demand on the battery and show it's voltage output at that current demand. This will give a good indication of the condition and capacity of the battery.
That's it!!!!!
That's what I'm looking For!
:)
 

jpanhalt

Joined Jan 18, 2008
11,087
I don't think it is quite that simple. There is no single, load/voltage relationship that applies to all batteries for predicting residual capacity. That relationship even varies between batteries that are nominally the same (See: slua450). If there is such a simple solution, maybe the secret should be sold to Texas Instruments.

Theory
http://focus.ti.com/lit/an/slua450/slua450.pdf

Cell phone battery
http://focus.ti.com/lit/an/slua455/slua455.pdf

Coin cells
http://nesl.ee.ucla.edu/projects/sensorsim/papers/islped.pdf

John
 

tyblu

Joined Nov 29, 2010
199
If the voltage dives under heavy load after charging that'll tell you that it has a bunk cell that needs replacement or junking. The alkaline ones look like they just have a whackload of C- or D-batteries in them (can't tell the difference). Maybe you could even replace a dead cell with a C-battery?
 

Thread Starter

oneminstrel

Joined Dec 8, 2010
4
Thanks everyone for your assistance!
But as someone mentioned, I forgot the chemical makeup of the Batteries. I'm dealing with Nicad, and Lithium Ion.
 

wayneh

Joined Sep 9, 2010
17,496
I think there is confusion here over just how precise you need the evaluation to be. What decision do you really need to make? I still contend the old fashioned lightbulb test is very useful, simple and cheap. You can easily identify cells that need immediate replacement, and with a little patience you can tell "nearly dead" from "still pretty good".

But it certainly is not a precise tool for telling, for instance, whether a battery still has 60% of its life left versus 40%. For that you need fairly elaborate equipment and time to perform the tests and analyze the data. All in all, maybe not worth it (cheaper to replace the battery than to test it).

In between is a grey area where maybe we can help, maybe not. I've often wished I had a tester that had 1) variable load, so I could simulate whatever the usage was. A programmable (pulsed) load would be fantastic, since "full on" is rarely the real usage pattern, and 2) a timer to tell me how long the battery under test was able to keep going.

Now that I have a data acquisition device, meaning I can use the computer to do all the control and data analysis, such a rig would be much easier to build. Folks here could certainly help you get the circuitry to accomplish the 2 chores I mentioned; pulsing the load and measuring the time to cut-out. I think building a dedicated rig - with it's own micro-controller - like the ones you can buy would be...nuts.
 

John P

Joined Oct 14, 2008
2,025
Responding mostly to Someonesdad, some years ago I worked for a company that did product development, or basically an R&D department for hire. We had a big job to design a system for testing batteries for a household name company. As I recall, they had to be exercised in any programmable way you wanted--like a "radio" with constant low power out, like an "electronic camera" with occasional bursts of high power, like a "flashlight" with occasional longer bursts of medium power, or anything. There was also a requirement for constant-power output (with current rising as voltage dropped) to simulate electronically regulated loads.

It was one hell of a job which wasn't properly predicted (not my fault, I'm happy to say) and our little company almost went bankrupt doing it. But then we realized that over at our client's plant, the boss was on the line to his bosses for getting the thing to work, and he couldn't afford to let us fail. So somehow we made delivery. It actually was successful, but that stuff's not easy.
 

Audioguru

Joined Dec 20, 2007
11,248
The little Li-Po battery cells for my RC model airplanes "wear out". When new a high power flight lasts 15 minutes. When the battery has been used about 100 times then a high power flight lasts only 10 minutes or even only 5 minutes.

I have used some battery cells until a full power flight lasts only a couple of seconds, but at first the power is full. Then a full charge also takes only a couple of seconds instead of half an hour for a new battery cell.

I have 10 of these worn out useless battery cells but they provided thousands of flights the past spring, summer and fall.
 
Top