The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: Roger Johansson
Subject: Re: Reducing contact resistance for low volt use?
Date: Fri, 13 Dec 2002 11:16:39 +0100
References: <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org>
NNTP-Posting-Host: dialin-sto-hc-39-52.direct2internet.com (18.104.22.168)
X-Newsreader: Forte Agent 1.92/32.572
jmuchow@SPAMMENOTcamlight.com (John Muchow) wrote:
>Hmm...with the data logger going at 1 voltage measurement per second
>(and knowing the fixed resistance), I could do the math for each 1 sec
>segment of the curve....adding up a bunch of watt-second readings.
It should be enough to have something like 4-10 seconds slices, take
the average of the start and end voltage of each section and use that
voltage value for that section.
>Oops, wait a sec. I don't know if a cell is linear in it's response
>for changing currents, especially where it starts to fall off near the
>end of the curve. The discharge time will be shorter with a CC
We can choose a constant load which gives the same discharge time in
total. We choose the load to give a curve which is a little higher in
the beginning and a little lower at the end, the total measurement
should not give much different results.
>and the cells may not respond the same way as a longer CL
>discharge. The higher cell temperatures for the faster CC discharge
>may influence the resistance of the cell.
There is no reason for CC to be faster in total, we can adjust the
load to give the same total time as a comparable CC test.
The method might need a bit of tweaking, ya know :-)
You might even want to consider changing the constant load in the
middle of the measurement, after 3 minutes, to compensate for
differencies at the end and the beginning.
Use the parallell port of a modern PC and you can have both inputs and
outputs so you can switch in a higher resistance in parallell with the
main load, to adjust it a little.
When the main load is connected all the time you need much simpler
circuits to adjust it during the measurement.
Hey, this could be made into a constant current method again!
If we use a "constant load" copper bar which gives the right current
at one end of the curve, we can switch in other loads in parallell to
make the curve essentially a constant current test.
Use 4 mosfets to add 4 other resistors during the test, that will
adjust the curve enough to be a "constant current" curve.
>It may not be enough of a difference to matter, I don't know.
That is what I think too.
>I would need to work up a couple of easy low-current CC and CL
>circuits and check out the results. Gotta get a data logger going
>first though...or do it manually.
A simple A/D converter for the parallell port for the computer and a
small program in your preferred language.
I still use Borland Turbo Basic so I might not be the authority on
programming around here :-)
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup