Personally, I'm not sure you need any diodes, assuming fully charged
floodies and AGM's are the same voltage after a recharge (I don't know if
they are). During driving, since the AGM's will always want to give up
their charge faster due to lower internal resistance, the AGM's should
always be at a lower state of charge, so the current (under light, or no
load) should always go from the floodies to the AGM's.
Here's the data I'm thinking we need:
Fully charge a floodie and an AGM (same voltage, 6 or 12).
Measure the voltage across each to make sure they are very similar.
Put an ammeter on each and parallel them, see what current flows into or out
Slap a heavy load (representing brisk acceleration) across the parallel
pair, see how much current is drawn from each.
Reduce the load to around 10% (cruise), see how much current is going into
or out of each
Remove the load completely (coasting or sitting still), see how much current
is going into or out of each
In an ideal world (anybody ever seen one of those?), I'd like to think:
The voltage would be very similar when fully charged.
Paralleling the two would not result in significant current flow when fully
With a heavy load, most of the current would be supplied by the AGM because
of lower internal resistance.
At light load after a heavy load, the higher remaining charge in the floodie
would result in higher voltage so it would provide most of the current or
even provide some charge back to the AGM.
At no load, the floodie provides charge current to the AGM until state of
charge of AGM and floodies are equal.
I'm thinking to do this right, floodies and AGM's should be in buddie pairs
to keep the AGM string from getting too unbalanced (because the floodies
would otherwise just be trying to recharge the AGM's as a string). But the
buddies have to be decoupled during overnight recharge to allow a floodie
string and an AGM string to be charged separately. This implies to me that
a contactor might be required to decouple each buddie pair when recharging.
They wouldn't have to be high voltage since you're only switching the
difference in voltage between the batteries, but they would have to handle
significant current. Could get pricey.
I'm just an ignorant newbie, if the buddie pairs were left connected, and
AGM/floodie buddie pairs were charged like AGM's, what would be the impact?
----- Original Message -----
From: "Timothy Balcer" <firstname.lastname@example.org>
To: "Electric Vehicle Discussion List" <email@example.com>
Sent: Wednesday, August 29, 2007 5:29 PM
Subject: Re: [EVDL] Buddy pairs AGM & Floodi
> Sorry I didn't see this!
> Well, a buddy pair would be the easiest way.. you could just rig up
> one pair, get them juicy and hook em up to various loads, using
> various timing parameters for rest and load to see where the sweet
> spots are and project that onto the driving experience. That would
> actually validate/invalidate the entire model since you would
> technically be connecting two series 6 chain 2V cells together in
> Obviously, you only want current to flow towards the AGM, so you'd
> need to take the load on the AGM.
> The only hitch this may have is the voltage drop using diodes.
>> > That being said, it remains to be seen if that is a -good- thing. It
>> > seems like a neat experiment and I'd think somebody with a few xtra
>> > floodies and AGMs on a bench could try it out with less hazard
>> > know I am going to be once i get mine setup right!
>> I can do this in my Elec-Trak with Exide T-105 equivalents and Hawker
>> Genesis EP-26's.
>> Would you like me to buddy strings or pairs?
>> I'll need to build some zener regs for the Hawkers first, though.
>> Stay Charged!
>> I-5, Blossvale NY
>> For subscription options, see
> For subscription options, see
For subscription options, see