DIY Electric Car Forums banner

1 - 4 of 4 Posts

70 Posts
Discussion Starter #1
I certainly enjoyed the ZEVA study, and appreciate a comparison of the
LiFePO cells (which are wonderful). First of all, let me say that I have
been driving 166V (and 166 lbs!) of A123 cells since July 3 (made by
repackaging 23 DeWalt DC9360's into a 36X3X9 box, 5 boxes for now). I
have a Agilent 34970a data acqusition system in the car (for now) to
measure voltages from all 50 cell groups during driving and charging.
(I'll try to get some of that data published--I'm way behind updating my
website). They run so cool that I'm not even cooling them (not even in
summer). However, I need to do a better job of measuring the temperature
inside the cells (all channels are taken measuring voltage). Bottom
line--I'm about 6X lighter than lead AGM batteries, partly due to the
relatively high power of the A123's (less Peukert effect). I ordered an
eval kit from A123, and found their data to be very accurate. I used a
Sorenson DCS8-125 power supply, an ACDC EL750BR-2-1 electronic load in
CC mode, and an Agilent 34970a data acquisition system. I used 4 gauge
wire for the connections. So, back to the study. The discharge part
seems ok--I'd have to run constant resistance mode to recreate it.
Normally I use constant current, but a better measure of actual EV
performance is constant power. But A123's at least are so flat that it
doesn't make a lot of difference. The charging part probably needs a bit
of work. Notice that the battery voltage does not reach 3.6V until after
the power supply has gone into CV mode. This is especially apparent on
the surprising 2C charging on the A123's. I suspect that there is a bit
of resistance in the wiring between the power supply and the cell under
test that is not terribly significant until you reach higher currents.
Note that the A123 cells also have the highest capacity, and 2C is
higher on them compared to the other cells. This can be remedied by
using larger wire between the power supply and cell, and by using a
Kelvin connection (a separate pair of wires to measure the voltage at
the cell independent of the wires used to deliver the power). Some lab
supplies will use Kelvin connections to regulate the voltage, and will
thus deliver 3.6V at the cell rather than their output terminals. I
usually don't bother to do this, but it does make a difference at the
higher currents. The LiFePO4 cells are great, and it's good to see more
vendors getting into them. It's also great to see real data (despite any
shortcomings) from independent testers. Thanks for going to the trouble
of testing and sharing the results! Gary (hopefully
I'll get going on updating it). ============================= Date: Sat,
13 Oct 2007 12:51:04 +0800 From: Ian Hooper <[email protected]> Subject:
Re: [EVDL] Price Winner - PHET PE-1150 To: Electric Vehicle Discussion
List <[email protected]> Message-ID:
<[email protected]> Content-Type:
text/plain; charset=US-ASCII; delsp=yes; format=flowed The page has been
accessed almost 4000 times now - makes it all worthwhile :) My test rig
was a little "grass roots" I admit :) but hopefully people will realise
this and won't take the results for gospel. I like to think that at the
very least testing a variety of cells on the same rig gives useful
qualitative/comparative results. -Ian On 13/10/2007, at 12:53 AM, Travis
Gintz wrote:

> > I thought it was a good read anyway :)
> >
> > Some guy on another list (Doug Korthof, Electric vehicles for sale)
> > thought the data was innacurate, and when I asked him to point out
> > "WHY" he didn't respond. I'm still curious why he thought the data was
> > full of error. I'm not saying its 100% dead nuts on, but its very
> > close. I worked at GE testing tons of stuff (power supplies, IGBT's,
> > controllers) and It looked like it was set up pretty well.
> >
> > On 10/12/07, Dan Frederiksen <[email protected]> wrote:
>> >> it was posted here a few days ago and Ian (zeva) is a member of this
>> >> list : )
>> >>
>> >> Arak Leatham wrote:
>> >>
>>> >>> Here's a very interesting article putting 5 different LiFePo
>>> >>> batteriesagainst eachother in charge and discharge tests. Good
>>> >>> data, lots ofgraphs, cost effectiveness, cost per Ah, cost per
>>> >>> cell..... etc. Goodread.
>>> >>>
>>> >>> Found on
>>> >>> regards,Travis Gintz1986 Honda VFR AC conversionHttp://
>>> >>>>>>
>>> >>>
>>> >>>

For subscription options, see
1 - 4 of 4 Posts