Read back a a few days. Lee Hart explained the process entirely in an
earlier post on this . You just had to read his post on the subject.
You charge them, let them sit for x number of days, then discharge them
to see how many amp-hours remain, then you charge them up again to see
what their total remaining capacity is. The difference between the
discharge amp-hrs (after x days) and the subsequent recharge amp-hrs is
the self-discharge. He also explains that a simple voltage test won't
tell you state-of-charge.
To run such a scientific test, takes care, patience, and time. You will
notice that none of the "anti-BMS" crowd seem to want to bother to
actually do these tedious long-term self-discharge tests.
I have found over the years than when you argue with any of Lee Hart's
experimental tests, you will find that you are typically mistaken. I
have been doing this EV thing for a long time, (20 years?) but Lee Hart
has been doing it _much_ longer than I have. Once in a long while I
manage to find some error in Lee's logic, but that is extremely rare.
Here is Lee Hart's recent post that describes how he does the
self-discharge test. Read it, and if you have any questions, I'm sure
that Lee will answer them.
>>>>>>>
Cor van de Water via EV wrote:
I have Lithium cells (CALB 180Ah) sitting in my garage since 1.5 years
and have measured their self-discharge and plotted it in an Excel
spreadsheet.
I can *see* their self-discharge.
I can *see* effect from ambient temperature on the change in
self-discharge rate...
I can *see* differences up to a factor 2 in self-discharge between
cells...
I've done the same thing with Thundersky. The results were the same,
except that the self-discharge rates differed by more than 4:1.
Additionally, the self-discharge rate is dramatically faster at higher
states of charge. The closer it gets to 0 SOC, the slower the
self-discharge rate.
Since the voltage-vs-state of charge curve is so flat from about 20-80%
SOC, voltage alone is a poor indicator of state of charge. This will
fool you into thinking there is "no self-discharge", because the voltage
won't change enough to measure between these SOCs.
To determine how much charge was lost over time, I fully charge the
cells. Then let them sit in parallel for a day. Then remove the
connections, and wait X days. Then measure the amphour capacity of each
cell. Recharge the cells, and repeat the process, but with successively
larger values of X. So for example, I might find that a 60ah cell yields:
- 60ah for X = 1 day after charging
- 58ah for X = 30 days after charging
- 56ah for X = 6 months after charging
- 54ah for X = 1 year after charging
Now, 10% a year is a pretty low self-discharge rate. 6ah per year
corresponds to an 8ma self-discharge current. The problem is that it
varies so much between cells. It means the cells can drift 10% apart per
year. This won't matter in the short term, but it adds up over time.
I'm also testing a set of A123 cells. The self-discharge rate is
similar, but the differences between cells is much smaller; less than
2:1. I'm coming up on the 2-year point for them next month.
>>>>>>>>>>>>>
On 6/28/2015 1:43 PM, David Nelson wrote:
On Sun, Jun 28, 2015 at 9:27 AM, Bill Dube via EV <[email protected]> wrote:
There were folks right here on this list that performed a _scientific_ test.
They cycled cells to find their capacity, charged the cells fully, then
placed them in storage. After extended time on the shelf, they tested the
state of charge by repeating the cycle once more to ambivalently determine
the state of charge and capacity once again. They found, scientifically,
there was indeed self discharge.
Bill, Isn't repeated cycling testing capacity not self-discharge?
Lee, is your data available somewhere? How did you determine if you
were measuring self-discharge and not capacity fade?
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.evdl.org/private.cgi/ev-evdl.org/attachments/20150628/82912e7a/attachment.htm>
_______________________________________________
UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub
http://lists.evdl.org/listinfo.cgi/ev-evdl.org
For EV drag racing discussion, please use NEDRA
(http://groups.yahoo.com/group/NEDRA)
