Measuring voltage on a LiPOFe4 cell has *_no bearing_* on the
state-of-charge. Only at near 100% SOC or near 0% SOC does the voltage
change significantly.
You have to actually fully cycle a LiFePOFe4 cell to determine its SOC.
If you don't do a full, 100% - 0% - 100% cycle, you can't determine the
SOC. Time consuming and tedious, especially for many cells in series.
The patience and care which is required to measure self-discharge in
LiPOFe4 cells is probably the root of many misconceptions about
self-discharge, balance, and the need for a BMS.
The self discharge for LiPOFe4 cells is indeed small, but it is real and
it is a strong function of temperature. It also varies from cell to
cell. Just because it is small, does not mean it can be neglected. We
have all hear the familiar saying "rust never sleeps". Well, "self
discharge never sleeps" either. Given time, it will ruin a series pack,
or in worse case cause a fire.
If self-discharge were equal, and remained equal, there would be no need
for a BMS, but it is unequal. As the number of cells in series
increases, the problem becomes more pronounced, and more deadly.
Here is a link showing the temperature dependence of self-discharge:
http://www.batteryspace.com/prod-specs/9444.pdf
You can see that the self discharge gets _much_ greater as temperature
increases. (There is a cornucopia of self-discharge information
available. All you have to do is look.)
What most people don't realize is that the cells in a battery pack do
not stay at uniform temperature. The end-most cells, for example, are
connected to the outside environment by large cables. These cables make
the end cells swing in temperature with the outside temperature, while
the middle cells tend to stay at the average pack temperature. The
middle most cells, are somewhat insulated from the outside, and tend to
heat up when the pack is cycled. When the temperature rises, the self
discharge goes wild. When it gets cold, the self-discharge returns to a
tiny rate. Each temperature rise gets the end cells more out of balance.
Kind of like a ratchet. (The cold swings don't "correct" for the warm
swings, unfortunately. The exponential nature of the Arrhenius curve is
to blame.)
If you log the individual cell temperatures, this disparity becomes
obvious. Without any BMS or any sort of battery monitoring, you don't
realize there is any problem at all, until your garage burns down. :-)
Bill D.
On 5/14/2018 7:18 AM, paul dove via EV wrote:
Hey thanks for your input. I’m sure a lot of people have varied experiences. I
must’ve met there probably differences between manufacturers. However that was
not my experience. I bought Bestgo sales hundred amp hour. I did extensive
testing before installing them in a vehicle. I put 40 for 100 amp our cells in
a 1986 Toyota Celica. I charge them 3.65 V per cell or 160.6 V. After sitting
for a while the voltage dropped 148.7 v or 3.38 volts per cell. I drove the
vehicle every day for two years. There was never a variance at the end of
charge. All the cells measured 3.38 V several hours after charging. I took the
cells out of the vehicle And they said on the shelf for year and a half or so.
I measure the voltage and they were all 3.38 V. However they do experience
reversible capacity fade. I discharged all the cells and Got 45amp hours the
first time. The second cycle I got around 65 amp hours and it continued to
increase for five cycles. On the last cycle all the cells measured Close to 100
amp hours.
Sent from my iPhone
On May 12, 2018, at 12:23 PM, Lee Hart via EV <[email protected]> wrote:
From: Cor van de Water via EV <[email protected]>
I also did tests on LiFePO4 cells and while self-discharge was low, I was able
to prove from my measurements over many weeks, that there is was about a factor
2 difference in self-discharge current between the best and worst cell. Sample
size was over 40 cells.
This is what I have found as well. Brand new cells, all bought at the same time
from a quality source are very similar. Their amphour capacity, internal
resistance, and self-discharge rates are very close.
But cells from cheap or low-quality sources have a much broader spread in
characteristics. Cells also get worse as they get old, or get cycled, or as the
temperature changes. Differences between cells accumulate over time, getting
worse and worse. You may get by without a BMS initially, but it gets needed
more as the differences between cells grows Not having a BMS means shorter life.
This is exactly what the author of the quoted article found, that the capacity
of the cells had not degraded, but the cells had gotten out of balance.
Yes. The BMS was too primitive to do its job of balancing the cells. Only
having upper and lower voltage limits prevents against catastrophic failures;
but does not compensate for differences between cells.
--
Excellence does not require perfection. -- Henry James
--
Lee A. Hart http://www.sunrise-ev.com
_______________________________________________
UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub
http://lists.evdl.org/listinfo.cgi/ev-evdl.org
Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
_______________________________________________
UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub
http://lists.evdl.org/listinfo.cgi/ev-evdl.org
Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.evdl.org/private.cgi/ev-evdl.org/attachments/20180514/89cbf559/attachment.html>
_______________________________________________
UNSUBSCRIBE: http://www.evdl.org/help/index.html#usub
http://lists.evdl.org/listinfo.cgi/ev-evdl.org
Please discuss EV drag racing at NEDRA (http://groups.yahoo.com/group/NEDRA)
