Old + new batteries? (Comment from Trojan Tech Support)

 Here's what Trojan tech support had to say on the issue of mixing old
and new batteries. (Posted with permission.) Not exactly the
definitive treatise on the subject, but such a thing might not exist.

-=s

<<<<<<<<<

Hello Scott:

Unfortunately, it's one of those knowledge that everyone knows and
talks about but you can never find any study or research to back it
up.  What I can tell you in general (and I'm speaking only in term of
lead acid batteries) is that  you should not connect old batteries
with new batteries together in series or in parallel.  An imbalance
does occur and will affect each string and also the entire system.  

The key is what is considered as old; one year, two years?  It all
depends on usage.  A five year old battery that is hardly used could
be considered as new and a one year old battery that is cycled
everyday is definitely old. We found that in systems that are
regularly used, about a year and a half to two years is old.  

Of course, technically you can connect old and new batteries together
and nothing is going to blow up but the performance of your system
performance and the life of your batteries will suffer.

If you need further assistance, please contact me.

Best regards,

Jim Le
Technical Support Engineer

<<<<<<<<<

  Seems like a reasonable answer to me. It is similar to what I have heard
here as the "common wisdom".

Now, if we were a group of medical doctors, we would have a number of
conflicting  statistical studies to reference.

 

     It is easy to see why you would not want unmatched batteries in series,
that goes well beyond "common wisdom" into plain electrical theory.  The
only reason that I can give why you would not parallel batteries that are
the same except for their age is if their fully-charged voltage varies over
their lifetime, as has been suggested here.   Does anyone know if that is
true?

Vaughn

 I agree completely.

Being part of the "common wisdom" does not exclude it from being backed by
theory. It more likely to become common knowledge when it can be directly
observed through direct experience. Most people understand through
experience that batteries degrade with usage and age. If you take that as a
given, you can apply well known theory to deduce the rest.

I think that when the words "age" is used in reference to a battery, what is
really being stated is it's electrical effectiveness, and not necessarily
chronological age although there is definately  some correlation between the
two in lead acid. Even sitting on a shelf, a lead acid battery looses
charge, and is therefore electrically aging.

An "old" battery might indicate a similar "fully charged" voltage, but its
voltage will drop off faster than a new battery under load. Therefore, an
old battery in parallel will degrade a bank even if it is in parallel. Fully
charged voltage is not a good indicator of a batteries effectiveness. A
better measure would be a voltage and current measured over the full
discharge cycle.

When two batteries in parrallel are "out of phase" with respect to their
voltage/current curve, some portion of current will flow from the higher
potential to the lower.  That means that you have current flowing from the
good battery to the bad, and less into the output circuit of the bank.

At some point, the old battery acts more like a load than a contributor,
even if it is in parallel. An "old" battery will get to the end of its
discharge cycle faster than a "new" battery.

The degree to which this hurts a bank dependes on the relative differences
between discharge curves. A highly degraded battery can become a big drain
on a bank, where a slightly degraded battery will have a lesser negative
effect.

that's my $.02


series,

 

From what I've heard, "old batteries" have higher self-discharge rates
(because of small shorts between plates?) and more series internal
resistance (because of reduced plate area?)


With its higher self-discharge rate, it might discharge a newer battery
in parallel faster than the newer battery would self-discharge...


Would putting a battery with a 5%/week self-discharge rate in parallel with
one with a 2%/week rate "lower the current" into the output circuit?
I don't think so, but the pair might discharge more quickly.

Nick