[Oeva-list] Battery Balancer

Sun Mar 4 07:00:41 PST 2007

Hi Don,

  Thanks for the excellent feedback you forwarded from Lee Hart regarding the flying capacitor balancer.  Having years of experience can provide valuable information that a theoretical or prototyped approach would not.  I think all of Lee's points are correct.  The charge transfer as the batteries get closer together in voltage becomes very slow.  There are definite losses in the FETs, caps and batteries.  What I liked about the design was its simplicity (no STAMP controller or DC-DC converter).
  My main goal is to make something effective that costs around $10 per battery.  The modification that Paul Compton did sounds very interesting to me.  If we could upgrade the single DC-DC converter to 20-Watts, then we might transfer 1.5 amps instead of 0.3 amps.
  I'd be happy to help you do research on this.  Do we have access to the plot files for the boards?  How do you plan on creating the PC-boards?  Are you planning any modifications from the original design?  When you say "large run," how many are you thinking of?

Regards,
Tim

----- Original Message ----
From: "Theoldcars at aol.com" <Theoldcars at aol.com>
To: oeva-list at oeva.org
Sent: Saturday, March 3, 2007 10:58:21 PM
Subject: [Oeva-list] Battery Balancer

There has been a few posts on this and I wanted to follow up with some 
emails from Lee about the equalizer your working on Tim. 

I am interested in doing a large run of Lee's battery balancer. To get the 
prices down and the quality up. It is going to require a larger quantity then 
has ever been built. I would be interested in some feed back. One if anyone 
would either buy one and at what cost. Also is anyone interested in helping out 
doing this even though the goal is not to make money. 

Below is Lee's response to my email.

Don Blazer

From: Theoldcars at aol.com
> I was wondering if you could take a look 
at this? I value your opinion ? What would the plus or minus be of going 
with 
either?
>
>http://914ev.blogspot.com/2007/02/battery-equalizer-schematics.html
>http://www.smartsparkenergy.com/pdf/batteq1.pdf
>(http://www.smartsparkenergy.com/pdf/batteq1.pdf) 

Be glad to, Don!

This is the old "flying capacitor" circuit. It's 
been around for a very long time. It works, but not very well. There are a 
number of problems:

- The peak current each time the capacitors switch is 
very high. Thus,
   the losses in the circuit resistances (MOSFET 
Rds, capacitor ESR,
   battery internal resistance, etc.) are 
high.

   My Balancer uses essentially pure DC from its DC/DC 
converter to charge
   the batteries, so switching losses are 
negligible.

- If you hold batteries at the same voltage, the current that 
flow between
   them falls almost immediately to a very low level. 
It takes many days
   for them to reach the same state of charge. 
Thus, it balances very slowly.

   My Balancer applies a higher 
voltage to the battery being charged, so it
   charges much faster. 
For example, it can load the pack as a whole to 12.5v
   to charge 
the low battery at 13.5v.

- It takes an unreasonably large capacitance to 
transfer any significant
   amount of charge.

   I 
used a DC/DC so you can charge at any reasonable current (the 
parts
   are sized for up to 30 amps).

- Efficiency is low; 
on the order of 50-75%. It *appears* high if you make
   casual 
measurments with average-reading meters, because they ignore
   the 
high peaks.

   DC/DCs can easily be 80-90% efficient. The 
Batmod I'm using is around
   88% efficient.

- It only 
transfers charge between adjacent batteries. When the "high" and

"low" batteries are far apart in the string, the efficiency losses of 
each
   stage multiply together. For example, if you have a 10% 
loss per stage,
   and you need to move charge across 10 batteries, 
you have essentially
   nothing left at the 
end.

   There is only a single transfer in my Balancer; no 
cascade effect.

From: Tim Kutscha  
<tim_kutscha at yahoo.com>

>Hi Don,
>
> Thanks for 
your  response.  Lee Hart's balancer uses a digital multimeter
> 
to read the voltage on all the batteries in your system, picks the 
lowest
> one and then uses an isolated DC-DC converter to charge just the 
low
> battery for a period of  time.  It then rescans the pack 
and starts again
> with the next lowest  battery.
>
> The 
equalizer described in the whitepaper is far simpler.  It doesn't 
> 
require a controller or multimeter. It just requires a waveform 
generator
> (in this case a 555 timer) and a bunch of isolated FETs and 
capacitors.
> In short, it shares charge between adjacent  batteries, 
thereby taking
> charge off of higher voltage batteries and dumping 
it  into lower voltage
> batteries.  When all batteries are 
equal, no charge sharing happens and
> no power (except the 555) is 
consumed.

His description of my Balancer is correct. However, he's 
comparing a full-featured circuit to a partially developed breadboard. The 
Balancer includes the PC boards, fuses, filters, connectors, and all the other 
incidental parts to make it actually work His circuit leaves all of these 
out.

When he actually builds it, he will find that he needs a lot more 
parts to make it work and be safe. It also consumes power all the time, due to 
switching losses. From his circuit, I would guess that it draws over 10ma from 
every battery even when it isn't doing anything.

It's possible to scale 
the Balancer down. Paul Compton built one of my Balancers, but downsized 
everything to make it cheaper. He eliminated the BASIC Stamp, Batmod, and 30a 
relays; replacing them with a $3 PIC, a $10 5w DC/DC, and 2a relays. It costs 
under $10/battery, but only transfers about 0.3 amps (which is still several 
times more than this capacitive balancer). [In my view, a 4:1 reduction in price 
wasn't worth the 10:1 reduction in performance.]

> I find it  
easily expandable to large strings too.

It is straightforward to expand, 
but the charge transfer between widely separated batteries gets so low as to be 
negligible.

I suspect that once he adds up the cost of all these part, he 
will find that his balancer costs more than the method I'm using.

Hope 
this helps!
--
Lee Hart

In a message dated 2/19/2007 1:16:50 PM Pacific Standard Time, 
oeva-list-request at oeva.org writes:

Message: 
  2
Date: Sun, 18 Feb 2007 22:13:36 -0800 (PST)
From: Tim Kutscha 
  <tim_kutscha at yahoo.com>
Subject: [Oeva-list] Request for feedback on 
  battery equalizer
To: OEVA <oeva-list at oeva.org>
Message-ID: 
  <174824.31402.qm at web36908.mail.mud.yahoo.com>
Content-Type: 
  text/plain; charset="iso-8859-1"

Hello EV folks,

  I'm working with some other engineers on an economical home-brew battery 
  equalizer at the following 
  link:

http://914ev.blogspot.com/2007/02/battery-equalizer-schematics.html

I've 
  prototyped this on my desk and it seems to work fine with 12V batteries.  
  I would appreciate any feedback you might have.  My hope is to get this 
  working and provide the design for free to help EV folks get their battery 
  packs to last longer.

Thanks for your time,
Tim 
  Kutscha
http://914ev.blogspot.com

AOL now offers free email to everyone.  Find out more about what's free from AOL at AOL.com. 
_______________________________________________
Oeva-list mailing list
Oeva-list at oeva.org
http://www.rdrop.com/mailman/listinfo/oeva-list

____________________________________________________________________________________
Finding fabulous fares is fun.  
Let Yahoo! FareChase search your favorite travel sites to find flight and hotel bargains.
http://farechase.yahoo.com/promo-generic-14795097
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/attachments/20070304/1f359914/attachment-0001.html 