Analyzing batteries on a budget

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
Some of us may want to estimate the capacity of our batteries, and either we don't own a CBA III (or I or II), or even if we happen to own an MH-C9000 we might want to test other cell chemistries than NiMH or NiCd.

Well, there is an answer. If you can obtain a digital multimeter with a computer logging interface, then with a few simple components you can get some results. The nice thing about the DMM is that even when you are not using it for logging it still has a full set of DMM functions, so the purchase price does not have to be justified as a single use tool. It can sit on your tool bench and be used for other things too.

Here is how to go about it. The theory is to set up a discharge circuit with voltage logging like this:

dischargecircuitdiagram.png


You discharge the cell or battery under test through a resistor of known value. The voltage across the cell terminals then tells you everything you need to know. Given the voltage and the resistance value you can calculate the current using:
I = V/R
When you know the current you can calculate the capacity. For instance if a current of 0.500 amps flows for 1 minute, the accumulated capacity is:
∆Q = 0.500 x 1/60 x 1000 = 8.33 mAh
Using the recorded table of voltage measurements and time intervals, you can therefore use a spreadsheet to sum up the total capacity during the discharge.

This is not quite the same as constant current discharge, but for many practical purposes this does not matter. For one thing many cells have a somewhat flat discharge profile, and for another many cells have a capacity that is relatively unaffected by the discharge current.

One thing you have to do manually is watch the progress near the end and stop the discharge when the voltage reaches the appropriate threshold. You can't go to sleep and forget about it.

Here is what my test configuration looks like:

img1400w.jpg


I used a couple of 1 Ω 10 W resistors from RadioShack connected in series. I measured the total circuit resistance by simultaneously measuring the voltage across the measurement points and the current through the resistor. In this case I obtained a value of 2.17 Ω.

The circuit is attached to the cell under test using two strips of polished copper foil held against the cell terminals by strong magnets.

Finally, here is a graph of a test result. I discharged an Eneloop that was previously charged on the Duracell Power Gauge charger. This is an older Eneloop of 2006 vintage that has not received very much TLC. It is feeling a bit tired nowadays with a capacity measured at about 1830 mAh.

eneloopdischargecurve.png
 
Last edited:

MarioJP

Enlightened
Joined
Sep 2, 2009
Messages
933
how do you log something like that and come up with a graph like that. That's pretty cool.
 

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
For comparison purposes I ran a discharge test of the same Eneloop on the C9000 at 500 mA and obtained a capacity of 1824 mAh.

Here are the last few entries in the discharge table from the test in post #1:

Code:
Time        Voltage  Current  Capacity
(hh:mm:ss)  (V)      (mA)     (mAh)
03:09:30    1.048    483      1808
03:10:00    1.031    475      1812
03:10:30    1.012    466      1816
03:11:00    0.989    456      1820
03:11:30    0.963    444      1823
03:12:00    0.933    430      1827
03:12:30    0.898    414      1831
The cell capacity is therefore measured at about 1830 mAh at the 0.9 V discharge cut-off. This value and the C9000 value are in good agreement, giving some validation to the test methodology.
 
Last edited:

Light Sabre

Enlightened
Joined
Sep 8, 2008
Messages
404
Location
Tucson, Arizona
I used a couple of 1 Ω 10 W resistors from RadioShack connected in series. I measured the total circuit resistance by simultaneously measuring the voltage across the measurement points and the current through the resistor. In this case I obtained a value of 2.17 Ω.

The circuit is attached to the cell under test using two strips of polished copper foil held against the cell terminals by strong magnets.

Finally, here is a graph of a test result. I discharged an Eneloop that was previously charged on the Duracell Power Gauge charger. This is an older Eneloop of 2006 vintage that has not received very much TLC. It is feeling a bit tired nowadays with a capacity measured at about 1830 mAh.

eneloopdischargecurve.png


I don't see how you can come up with mAh using your set up. The current is going to go down as the as the voltage drops. You would have to add a 2nd DMM measuring current and import both into Excel and calculate the mAh per second (or set interval) and add the results to come up with a decent approximation of the mAh. The current will not stay the same the whole time of your test. You would need a constant discharge circuit to drain the battery then all you would need would be a time graph.

I have several of the exact same DMM's. I made my own version of the milk cartion fixture from Flashlight-Reviews using CDS cells and sensors and measure how long a flashlight lasts before it falls out of regulation, whether it has a gradually decreacing output after that or just quits all together, or whether it is not regulated at all. BTW: Battery vampires last a whole lot longer with alkine batteries than with lithium or NMH's. This test would analyze how long a particular flashlight lasts with a particular type of battery. Yesterday I tested at PT Tec 40 with a Terralux TLE-1F dropin and Duraloops and it lasted 9 hours regulated and then gradually dimmed for another 2 hours of useful light. That's the type of information I want to know. The 1W and 3W versions of the Rayovac sportsman flashlights give 0 warning when end of regulation is reached. The results would be a constant power reading not mAh's tho.

Thats MHO. YMMV.


The above was written before reading Message #3 which Mr. Happy wrote and submitted while I was writing this one. His way seems to be accurate more than I was thinking with an Eneloop anyway. A test with an alkaline battery would not be as accurate.
 
Last edited:

Mr Happy

Flashlight Enthusiast
Joined
Nov 21, 2007
Messages
5,390
Location
Southern California
I don't see how you can come up with mAh using your set up. The current is going to go down as the as the voltage drops. You would have to add a 2nd DMM measuring current and import both into Excel and calculate the mAh per second (or set interval) and add the results to come up with a decent approximation of the mAh. The current will not stay the same the whole time of your test. You would need a constant discharge circuit to drain the battery then all you would need would be a time graph.
There is no need to measure the current. That's the beauty of this setup, it will tell you everything you need to know with just the voltage measurement. If you measure the voltage you can calculate the current by using the known value of the resistor (which is what a DMM on the current scale does internally).

As you said, you then put all the results into a spreadsheet add up the mA over time to get the mAh. A sample of that calculation is shown in post #3 above.

Shown graphically, the capacity is the area under the current curve, like this:

capacitymeasurement.png


You can see from the graph that even though the current changes, it does not change all that much.
 
Last edited:

Light Sabre

Enlightened
Joined
Sep 8, 2008
Messages
404
Location
Tucson, Arizona
Duhh, dumb me. Never thought of the constant resistance part and calculating the changing current that way. :poke: That's what I like about CPF. If someone forgets something as simple as this, someone else points it out.

I like the idea of my flashlight tests. Lets me know the lights behavior at the end of regulation. The 1W & 3W Rayovac Sportsman lights shouldn't be counted on for a camping trip, search & rescue, etc type of situation. Or that you need a backup light and a 3rd light as a 2nd backup. Also how fast the diminishing light output is good to know. Is it 5, 10 or15 minutes, 2 hours or a lot more I have 1 flashlight that runs 24 hours in regulation with 4 days of useable diminishing output after that. 1 LED, 2 AA's, decent on brightness. Good for an extended power outage.
 
Top