DIY Electric Car Forums banner

1 - 3 of 3 Posts

70 Posts
Discussion Starter #1
It's time for me to upgrade my EV transportation. I've been running VRSLA for
the last five years or so. In my current EV, a 2003 Gizmo, I'm running a
comfortable 25 mile range in ordinary temperatures. Unfortunately, it's been
rather cold here in the frozen south and I'm also wanting to have an increased
range overall.

Clearly the answer to this is LiFePO4, at least in my alleged mind. David
Nelson, also a Gizmo owner, has a rack of TS cells and his experience is good.
I'm not so sure about TS cells though and would like to ensure my investment
remains as healthy as possible over time. I envision five thousand miles or more
a year, almost exclusively flat ground.

My envisioning also includes the ability to record each cell performance during
operation and charging, similar to the information that the PakTrakr provides
now in my pack. I'm thinking I should also have some low voltage and high
voltage protection during discharge and recharge, respectively. Bluetooth output
to an alleged smart phone, android-based would be a bonus, although ordinary
RS232c via netbook is acceptable.

My current pack is eight Deka 8G31 batteries, wedged pretty tightly in the
battery box, tightly enough that the top flange on each battery prevents the
pack from sitting flat in the box. I've solved that by putting floor-foam-tiles
under alternating batteries, those things that you can put down for play areas
or just something comfortable on which to stand. The factory battery box is the
only location in which to place batteries and I do have a bit of leeway in terms
of height, but that's about it. I'd like to have as close to 200ah as possible.
The pack in place now provides nearly that figure, ignoring the Peukert effect,
of course and the fifty-percent discharge limit.

As David Nelson has discovered, our controller and DC/DC converter can both
handle 60v, so I'm aiming for that figure, more or less. I'll have to confirm
the numbers to be more certain, but it means that it's likely that 20 TS type
cells should work. David worked within the physical constraints of his battery
box by using 2x100ah cells in a 2p20s configuration. I'm not sure of the
terminology, but it's pairs of cells in parallel, strung in a series

My pack is currently four batteries in serial, duplicated. I'm thinking that
David's is two batteries in parallel, twenty times in serial. Is there a reason
to not string 20 cells in serial, then parallel another pack of 20 with it?

As part of my objective, I'd like to have some form of equalizer or balancer or
dare I use that term BMS? I'm running PowerCheq equalizers on my pack now and
have had it on the long-gone Xebra and attribute the long life of the packs to
those devices, along with my strict limitation to fifty percent DoD. I'd like to
keep that capability in the next chemistry.

This is more-or-less a request for quote, in a less-formal manner. I'd like to
hear from vendors especially, to have them make me an offer I can't refuse, as
long as the general consensus agrees with the conclusions provided by the
vendors. I would also want a charger "tuned" to the pack, even though I
understand that mine is capable. I don't want to have the vehicle out of service
by the additional time required to ship the Zivan out for reprogramming.

Comments, suggestions and proposals are welcome.

Onward into the fog...

-------------- next part --------------
An HTML attachment was scrubbed...
| REPLYING: address your message to [email protected] only.
| Multiple-address or CCed messages may be rejected.

70 Posts
Discussion Starter #2
Hi Fred,

If I were to do my pack over I would probably go with CALB cells since
they are a little stiffer than the TS cells but they do cost a little
more. If I could have found 200Ah cells equivalent to two 100Ah cells
in physical size I would have gone that route. There is plenty of
height to the battery box so every thing clears while sliding the box
into the Gizmo after installing the batteries. Even with the cold you
are getting now I wouldn't bother with insulating the box like I did.

As for doing a 2p20S or 20s2p arrangement note that these prismatic
cells are really a stack of paralleled cells inside the package. By
paralleling two I'm merely doubling the number of paralleled stacks.
Because of the characteristics of LiFePO4 cells' voltage curve on
charge/discharge where the low end drops sharply and the top end rises
sharply I think you can see that you could even get by with
paralleling two cells with significantly different capacity, however
the mechanical connection might be a pain. If you were to parallel two
20 cell strings you would lose this benefit and would have twice as
many cells to monitor. Some have argued that by paralleling two cells
if one shorts it will kill the other one. I agree. In my case I would
lose a 200Ah equivalent cell. If I had single 200Ah cells and one
shorted I'd lose the same thing so I don't see the argument as a valid
reason not to parallel sets of cells like I have done.

As for your charger programming. After running 18 cells with the lead
acid profile and having to adjust the voltage to 4.00Vpc so that I
could get a full charge before the charger timed out and then going to
a LFP curve set to 3.65vpd and 19 cells and running it on 20 cells it
is well worth the wait time to send the charger in to get it
programmed. I wish I hadn't waited so long for the reprogram.

Yesterday I went 68+ miles with lots of hill climbing, rain, cold, and
strong headwind. I used 175Ah for the run (in nice weather this same
run would have only taken ~140Ah). When I got home I plugged in the
charger and then just before going to bed I unplugged it and then
plugged it in again. This is because the NG1 will time out before
fully charging. It runs 9-10 hours and shuts off with an error. Keep
this in mind if you go with a 200Ah pack. Remember you can safely use
70-80% of your pack and the voltage doesn't drop like it does with
lead acid. My point: you might consider something less than 200Ah to
get a form factor that will fit the battery box with only 20 cells.
Remember that 200Ah at 48V is only 9.6kWh where as 200Ah at 64V (20
cells at 3.2v nominal) is 12.8kWh.

I recommend you look around at the different cell sizes available and
do a mock-up to see what will fit and then get quotes from someone on
the batteries you want.

Also, view this video if you haven't already:

David D. Nelson

| REPLYING: address your message to [email protected] only.
| Multiple-address or CCed messages may be rejected.

70 Posts
Discussion Starter #3
Hi Fred,

In your quest for LiFePO4 battery info you might throw in to the ring
the GBS batteries as available at
The terminals are at opposite ends of the cells and I think they can
be placed in any orientation. They might fit the Gizmo battery box but
I haven't really looked into that.

David D. Nelson

| REPLYING: address your message to [email protected] only.
| Multiple-address or CCed messages may be rejected.
1 - 3 of 3 Posts