Destructive testing of LiPo batteries?

Regular readers know of my side interest in RC helicopters and quadcopters. All of these models use LiPo batteries. Prior to my interest in RC, I used LiPo batteries during many QRP excursions.

Early on, I bought a West Mountain Radio Computerized Battery Analyzer- a great little device that allows battery capacity to be determined. And over time, trends become apparent with various batteries.

I’ve always been meticulous in my care of LiPo’s – they’re not cheap and I want them to last – and I have them in sizes ranging from 1S-250 mAh to 3S-4000 mAh.

Now about the battery analyzer:

Software is used to set up the analyzer’s parameters: chemistry, initial voltage, desired rate of discharge, etc. Each battery chemistry has a default value to which it will be discharged and at that point, the battery analyzer is finished and the graph (see below) can be saved.

For 3-cell LiPo batteries, the battery analyzer ends the discharge (at default values) when the battery under test’s voltage drops to 8.4 volts (2.8V per cell). I believe this is too low and has shortened the life of my three most recently purchased batteries.

Wikipedia claims that LiPo’s should not be discharged any lower than 2.7 volts per cell, other sites mention 3 volts as the minimum.

Looking at the chart below, it can be seen that each of these minimums are reached within minutes (or seconds at higher discharge rates) of each other due to the steep nature LiPo chemistry’s end of life “knee”.

In other words, 2.7 volts pretty much equals 3.0 volts in terms of how quickly the voltage drops when in this range.

The default LiPo cut-off of West Mountain Radio’s battery analyzer is, I believe, too low to allow testing of a LiPo battery without destroying it.

I noticed all this when all three of my 3S-2200 mAh batteries started giving drastically reduced flight times. Putting them on the tester, I found that their capacity had dropped to 1/3 of what it had been only a few weeks earlier.

Once I realized what may have been the culprit, I recalled past examples of other batteries whose “brothers” purchased with them enjoyed longer lifes than those that were tested half a dozen times or so.

I now define end of life for a LiPo battery to be 3.4 volts per cell. This is the point at which the curve first starts to go south and I now use this value for the cut-off point when testing them.

.

3sLiPo

Green horizontal line at 8.4 volts indicates cut-off for 3-cell LiPo tests – too low!

.

.

  6 comments for “Destructive testing of LiPo batteries?

  1. Keith Hamilton N8CEP
    February 21, 2014 at 6:30 pm

    Very useful information! I have seen different cut-off points listed at various websites and I have used 3.0 volts in the past. I will now begin to use 3.4 volts! Thank you for the discussion! N8CEP

    • February 21, 2014 at 6:58 pm

      I should have realized when viewing the graph after my first LiPo battery test that nothing is gained by having the battery analyzer cut off so far below the obvious point of a battery’s end of life – and that there is something to be lost (battery life). That’s what I get for blindly trusting default values…

  2. Roderick Young
    February 22, 2014 at 12:11 pm

    Thanks for posting your graph. I’ve been considering Lithium Phosphate as a replacement for the old Lead-Acid batteries in my electric bicycle. Weight is not as critical a matter as for your model planes, but it still matters. Have you done any experiments with charging? The manufacturers have recommended curves of a constant current and constant voltage period of charging, but I’m hoping someone can validate those.

    • February 22, 2014 at 3:57 pm

      Hi Roderick,

      I have never used Li-Phosphate (LiFePO4) batteries but I am interested in doing so and recently bought a new battery charger that is capable of charging them (which I love BTW – charges everything I have and displays a lot of info):
      http://www.amazon.com/gp/product/B00466L0BW/ref=oh_details_o04_s00_i00?ie=UTF8&psc=1

      They come in a wide array of form factors including those of some of the common battery types such as CR123, AA and others (although the AA cells are 3.2 volts, not 1.5!). I have no idea of the charge or discharge characteristics but if when I buy one I will post the discharge curve as well as the charge characteristics that the Tenergy charger applies to it.

      • Roderick Young
        February 24, 2014 at 8:01 pm

        If the Tenergy charger is as good as their batteries, it will be a treat. I did some (not terribly rigorous) testing on some small batteries here http://pididu.com/wordpress/blog/more-battery-life-measurements/ .

        On the Li-Fe-PO4 batteries, what I would like to use is these huge ones to replace 12 AH sealed lead acid, but the price for finding out whether they are any good, or how easily they are damaged, is steep. I actually would like to build my own charger as part of another project, but am trying to gather all the info I can before starting to spend.

        I see that you’re a radio amateur. My father was into that when I was growing up, KH6GMX (Hawaii), but I never had the persistence to learn code. But he did start me off in electronics, which eventually became a career.

  3. February 25, 2014 at 10:58 am

    I think most commercial applications of what you want to do are still using LiPo batteries such as in these electric motorcycles:
    http://www.zeromotorcycles.com/technology/ and paramotors which also require large amounts of current and enough charge/recharge cycles to make them economical. I haven’t seen much in the way of LiFePO4 in commercial products and there could be a number of reasons for that…

    You should again look into getting your ham ticket – Morse proficiency is no longer required and there are a large number of new digital modes that may appeal to you. Google ‘WSPR’ ‘JT65′ and ‘PSK31′ for an idea of what many are doing now.

Leave a Reply

Your email address will not be published.