Saturday, February 7, 2015

Q&A: Question About Over-discharged LiPo--How do I quantify the amount of damage done to the battery pack?

Subscribe by Email!
By Gabriel Staples
Written: 7 Feb 2015
Last Updated: 7 Feb 2015

Related Articles:
High-current bench-top battery test setup by FliteTest (source: 4:08)

Question I Just Received:
-from "Lakshman

"Hi Gabriel.
I am referring to your post

I accidentally discharged my battery to 2.8v/cell today due to a faulty lipo alarm. It did not go off at 3.6V while flying my tricopter and within a minute or so, the voltage dropped to 2.8v/cell. I recharged it back to 3.8v storage at 0.2A without any issues. This was before I read that article.

How do I quantify the amount of damage done to the battery pack? I usually stop flying at or before ~80% discharge. Will it be a lousier battery after this incident?

Thank you."

Here's what I had to say...

My Answer:


As an engineer, I'm trained to be thorough, so I'm assuming you want the long answer. Here goes:

When you say 2.8V/cell, do you mean to say that your 3S LiPo (I'm guessing it was 3S) was down to 8.4V (avg of 2.8V/cell), or are you saying you checked the battery with a cell meter and the lowest cell was 2.8V? This is an important distinction. A 3S Lipo below ~10V or so, for instance, oftentimes can have vastly different cell voltages at this point because once a cell gets below ~3.6V, its voltage will very rapidly and steeply decrease for very small differences in remaining capacity (see chart below, from my Parallel Charging article):

In other words, a reasonably-well balanced 3S LiPo that is at 8.4V might actually have a cell at 2.1V, a cell at 3.1V, and a cell at 3.2V....rather than each cell at 2.8V.

If your lowest cell did in fact only drop to 2.8V, I think the damage would probably be pretty minor. Keep using the battery and you may not even notice a difference in flight. If it dropped to 2.1V, for example, I think the damage would be more significant. You'd probably notice some difference in flight, and that cell might be the one to drain first in it's capacity might be diminished, and its internal resistance increased.

How do you check the health of a cell?
When a cell is damaged it will perform different in one or both of these two key areas (they are actually coupled [related] too, though not the same thing):
  1. its useful capacity will be decreased. ie: it will have a reduced run-time.
  2. its internal resistance, or "Equivalent Series Resistance," ESR, will be increased. This means the cell cannot output as much current or power, it will have a higher voltage drop under any given load, and it will produce more heat for any given current. 
Here are some techniques to check #1:
  • You can simply check this by hovering your tricopter (in no-wind conditions, as wind increases lift and I suspect also increases hovering time) with a damaged battery vs an identical healthy battery. See how much the hovering time has decreased. This test assumes you use the exact same test vehicle at the exact same flying weight in both conditions, and that you are comparing identical batteries, and that the identical batteries performed equally well when new, and that you are performing identical flights. Ex: try doing a steady, no-wind, no acceleration hover for the whole flight. A decrease in hovering time is indicative of damage. 
    • If you want to calculate the capacity lost you will need to perform the flights with a current meter on the vehicle so that you can get an avg. current reading for the flight. You can then calculate the useful capacity in each battery as follows: Capacity[Ah] = time[hrs] x Current[A]. Ex: if your current meter says you pulled an avg of 13A, and your flight time was 15 min, then your capacity is 13A x 15min/60min/hr = 3.25Ahr = 3250mAh. A battery is generally considered "bad" once its useful capacity has decreased to 80% of its original capacity. Nevertheless, don't let this make you stop using the battery. Use the battery until it no longer flies your vehicle as you desire. I have some batteries down to ~60% useful capacity that still fly one of my (extremely overpowered) planes, so I still use them since it flies just fine with them.
  • You can also check the capacity by using a charger to discharge and charge the battery. Check the mAh used during a full discharge. Nevertheless, unless your charger is discharging at a realistic flight-similar discharge rate (ex: 13A in the above scenario) the value given by your charger could be a gross overestimate of the battery's useful capacity. The more damaged a battery is (the higher its ESR), the more of an *over*estimate this method will give you, since a realistic in-flight voltage drop will not be produced during the test. Instead, try a bench-top test like Josh Bixler does here, so that you can produce a realistic in-flight-equivalent discharge current during the test (picture shown at top of this article):
Here are some techniques to check #2:
  • Perform the above bench-top rapid-discharge test while monitoring the battery *temperature*. The higher the ESR of a battery, the more its temp will rise. Don't let it rise over 140 deg C during discharge. 
  • Perform an experiment to calculate the ESR. I'll let you google for this one. Make sure to use a realistic in-flight test current or else the ESR calc will not be realistic. Many modern chargers can test ESR, but they may do it at such a low test current that the reading may not be very useful. 
  • Buy and use an ESR meter, such as this one (shown below-right): 
  • Notes:
    • Remember, the higher the ESR of the battery, the worse the battery is. I have not, however, quantified what might be considered a "bad" ESR for any given-capacity battery. I'll let you decide what is "bad" by comparing to similar batteries you have.
    • Be careful about comparing ESR values between batteries of different labeled capacities. ESR will *decrease* as C-rating OR capacity *increases* for any given battery. This means that it is feasible that a an absolutely horrible high-capacity (high-mAh) LiPo could have a lower (better) ESR than a really good low-capacity (low-mAh) LiPo. This is because ESR is a function of both battery capacity (in mAh or Ah) and battery C-rating.
You can also perform the tests above on a *single* cell at a time:
  • If you want to check an individual cell, you can connect up to the balance lead. Cell 1's ground is the negative-most wire, and positive is wire 2. Cell 2's ground is wire 2, and positive is wire 3. Cell 3's ground is wire 3, and positive is wire 4, etc. Warning though: just because you now know what the ground is for each cell, do NOT attempt to connect the grounds of each cell together, as of course this would be short-circuiting the battery. The cells are wires in series, so the grounds are NOT common. Also keep in mind that the balance lead is not rated for high currents, so don't pull more than a few amps max through a balance lead, and monitor the lead itself for temperature (just feel it during discharge). If you pull too high a current and blow your lead, that's on you.
Here is a technique to check #1 and #2 simultaneously (this is the best and most useful test, but also the hardest to do):
  • Acquire and compare discharge curves of the batteries and/or individual cells. You can purchase equipment to do this, or you can custom-make something with a microcontroller platform such as Arduino. Plot and compare the results. You will be able to see total capacity *and* voltage drop/sag during discharge (indicative of ESR), of each battery or cell, throughout the whole discharge process. This is very useful. See the figure below for example. As discharge current increases, voltage sag increases too, as indicated by a *lower* discharge curve. For any given constant discharge current, and if comparing multiple batteries of the same labeled capacity, the worse the battery is (ie: the higher the ESR/the lower the actual C-rating is) the lower the discharge curve will be on a plot such as the one below.
  • Discharge curves, showing capacity and voltage sag at various discharge rates (for more info, read here: source)
  • Some cheap chargers (see my article here for example) or high-end chargers can even do datalogging during charge and discharge, but again you run into the problem of often-times not being able to do this for high (realistic flight condition) test currents, since chargers generally have very low maximum discharge rates.
  • Some power meters can do on-board datalogging too, and you can then download the data after a bench-top test, in order to obtain discharge curves similar to those shown above.
I hope you weren't looking for a super simple answer. Rarely with anything electronic or engineering-related is the answer super simple. If you ever get a super simple answer it simply means that a lot of assumptions were made. For the general public, this may be good enough, but the application and exact use would have to be known or assumed before-hand.  The quality of a shorter answer would depend upon the accuracy of the assumptions made.


Gabriel Staples

***Subscribe by Email!***


Thanks for your comment or question! If it is a question, I will try to get back to you quickly. Notice to spammers: I personally remove all spam promptly and report spammers to Google, so please don't do it.

Note: some HTML tags are allowed in your comments. To learn how to add bold (<b>...</b>), italics (<i>...</i>), or hyperlinks (<a href="URL">NAME</a>) to your comments, read here.

~Sincerely, Gabriel Staples.

P.S. Yo hablo español también. Je parle français aussi. (I speak Spanish & French too).