On 1/28/2013 5:17 PM, Bruce EVangel Parmenter wrote:
[ref Elcon PFC-2500 thread]
My S10 Blazer EV (with a 132VDC pack of 6V T145s) had metal battery
racks from its creation which were grounded to the chassis. In the 15+
years I enjoyed the heck out of my EV conversion, charging and driving
anywhere and everywhere, I never had a GFI trip from leakage current
through the metal battery racks, whether I was charging using a:
K&W BC-20 with a boost transformer for up to 144V
Zivan K2 or NG5
Manzanita Micro PFC 20, 40, or 50
+more
I swear I have a memory of looking at your engine bay some 10-15 years
ago, and it had polypro (or similiar) cases for the front pack(s). If
you had metal racks actually touching the tops of your batteries, I
would certainly like to know how you managed to avoid GFI trips through
unisolated chargers.
I had some occasional GFI trips through the K&W BC-20 with my first
battery pack (1994-2001). However, with the liberal use of elbow
grease, I was able to bring the voltage between the most + post and
chassis ground close to zero. During my second pack, I lost being able
to do that - it would never go down. By that time I had a Zivan K2
(isolated), so GFI trips were a thing of the past. Since then, even
with re-powder coated racks done in 2009 at the beginning of this 4th
pack I'm on now, I've never been able to drive the voltage down. I even
installed something similiar to what ElectroAutomotive was selling in an
isolation kit for the hold-down racks - it didn't make much difference,
although I found it rather difficult to ascertain if I didn't have metal
touching through the hold-downs.
Well, it could be through motor carbon tracking, as Roland has often
mentioned. One of these times I may crawl under the car in my apt.
carport and disconnect the motor. I don't have a contactor on the
negative leg of my pack - the VoltsRabbit didn't come with one.
With visions of Lee watching over me (since I think he has explained
this a few times in the past), a few months ago I decided to find out
just what the magnitude of the chassis leak is that I have. I figured
measuring through a high-impedance digital voltmeter can pick up a lot
of phantom readings, since it is so sensitive. So I got out my "light
table", which contains eight screw-in light bulb bases rigged up in
parallel. I hooked up this light table through 6-ga cables to the most
+ post and to chassis ground. The first bulb I tried, a 300W
incandescent, I figured wasn't going to do anything, since the chassis
leak resistance would be too high to light up this bulb, but it was a
starting point. Wound my way down through 60W and so on to try and find
the "sweet spot". Well, a 7W/120V incandescent got me to the sweet
spot: it was something like a 30V drop across the bulb at some 300mA as
measured with a clamp-on ammeter. Bulb was barely glowing. But that's
about the magnitude of my chassis leak. So I wonder can I just multiply
that voltage by four and say at 120V I have a 75mA leak (which is way
over the typical 5mA GFI limit) as concerns an non-isolated charger?
[snip]
_______________________________________________
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)
