Lithium battery capacity derating: C-rates
December 2, 2009 9:39 PM Subscribe
Asking for someone else: is it possible to calculate or estimate the capacity derating of a lithium polymer battery when discharged at high C-rates?
Or can this information only be obtained empirically or from manufacturer data? Does the capacity derating follow some sort of known relationship with respect to the C-rate?
Or can this information only be obtained empirically or from manufacturer data? Does the capacity derating follow some sort of known relationship with respect to the C-rate?
Manufacturers will often publish the battery's capacity at a handful (3 or 4) of different discharge rates, from which you can roughly interpolate.
AIUI, the main factors affecting this are internal resistances and the kinetics of the actual electrochemical reaction, both of which are things that the manufacturer can and will tweak to make their battery better or to adapt it to specific uses. So you won't be able to get more than a rough estimate without either testing the battery itself or looking at manufacturer data.
posted by hattifattener at 10:00 PM on December 2, 2009
AIUI, the main factors affecting this are internal resistances and the kinetics of the actual electrochemical reaction, both of which are things that the manufacturer can and will tweak to make their battery better or to adapt it to specific uses. So you won't be able to get more than a rough estimate without either testing the battery itself or looking at manufacturer data.
posted by hattifattener at 10:00 PM on December 2, 2009
The Ubuntu software polls the battery on a regular basis, and builds a profile curve for the battery life over time. If I remember or find the name of the user daemon that does this, I will be sure to follow up with that info. It is an open source program, of course.
posted by idiopath at 10:37 PM on December 2, 2009
posted by idiopath at 10:37 PM on December 2, 2009
I am not sure if this is what Ubuntu uses, but if you want to poll battery power in order to create a profile and predict battery life under Linux, I have used Ibam to much success.
posted by idiopath at 10:41 PM on December 2, 2009
posted by idiopath at 10:41 PM on December 2, 2009
For example, here's a pretty typical manufacturer's info page, here [PDF] is another one, and here [PDF] is another one.
posted by hattifattener at 10:43 PM on December 2, 2009
posted by hattifattener at 10:43 PM on December 2, 2009
The manufacturers generally include a number of datapoints or, at least in some cases, an actual plotted curve, giving Wh as a function of discharge rates. You might need to find out what cells are actually inside the pack you're using, and then go to the manufacturer for the datasheets.
The first link provided by hattifattener is about what I recall seeing. You'll need to basically combine two charts — the discharge curve at various C-rates and discharge curve at various temperatures — if you're going to be using a high C-rate at nonstandard ambient temperature.
If you can't find the actual manufacturer's spec sheet then you'll probably just want to find the closest similar part that you can find a sheet for, build in a safety margin, and then test it. When I was playing with lipos (for RC helis) there didn't seem to be that much variation between manufacturers of similar cells, but that may have changed as things have been refined.
posted by Kadin2048 at 1:04 AM on December 3, 2009
The first link provided by hattifattener is about what I recall seeing. You'll need to basically combine two charts — the discharge curve at various C-rates and discharge curve at various temperatures — if you're going to be using a high C-rate at nonstandard ambient temperature.
If you can't find the actual manufacturer's spec sheet then you'll probably just want to find the closest similar part that you can find a sheet for, build in a safety margin, and then test it. When I was playing with lipos (for RC helis) there didn't seem to be that much variation between manufacturers of similar cells, but that may have changed as things have been refined.
posted by Kadin2048 at 1:04 AM on December 3, 2009
This thread is closed to new comments.
posted by tehloki at 9:57 PM on December 2, 2009