[RE-wrenches] Test procedure for Concorde batteries

[Matthew Sirum]

Greetings Gary,

I like what John and Tump provided for advice on this.  They both have
a great deal of experience with batteries.  I agree some of what John
described can produce misleading numbers.  The Concorde
(manufacturer's) recommended test procedure will take a fair amount of
time, but observing actual amp hours "in" (charge) and amp hours "out"
(discharge) while observing voltage will give the most accurate
results.  These are AGM VRLA batteries, so measuring specific gravity
with a refractometer as part of testing is not an option.  If these
batteries are about 9 years old it may be time for their owner to
invest in new replacements depending on how well they have been cared
for and how much exercise they have, or have not had.  Labor time
testing and trying to recover these batteries can be viewed as
expensive, especially if they end up needing to be replaced anyways.
If they were still under warranty the manufacturer's recommended test
may have to be performed in order to obtain a warranty replacement.  I
doubt this is the case with this ~9 year old battery system.

I can look at a case like this in a couple of different ways...  Do we
want to see if we can find one or two batteries that are causing the
issue and try to recover them, or do we need to determine the actual
battery system Ah/kWh capacity?  Sometimes we need to do both.

With a battery bank like this the technician should be able to at
least determine if any one or two batteries stand out as being the
issue(s) ("weakest link in the chain").  Internal resistance testing
is one helpful way to help determine this.  Since these are 12V
batteries an automotive battery internal resistance tester may be
used.  If I was troubleshooting this battery system I would take and
record voltage and current measurements as the system is.  Then I
would un-wire the batteries from each other; removing the interconnect
cables, fuses, and inverter cables.  I would take and record VOC
measurements of each battery, ideally after they were at rest for at
least a few hours.  Next I would use a 12VDC internal resistance
battery tester to test each battery.  A cold cranking amp (CCA) value
may need to be entered in to the tester before testing.  Since these
are not starting batteries, I don't think Concorde publishes a CCA
rating.  A good way to come up with a value to enter is to see if
Concord does publish a CCA/MCCA rating for one of their somewhat
similar batteries sold to the marine or heavy equipment industry.  Or
look at CCA/MCCA ratings for other 12V Group 4D batteries.  If the
tester used is designed for passenger car and light truck staring
batteries you may just end up using the highest CCA setting of the
tester.  It is NOT critical the CCA value used is exactly correct for
the battery being tested here, just that it is in the general
ballpark.  For these Concorde batteries this may be approximately
850CCA-1250CCA.  What is critical to the testing is that the same CCA
value be used for testing each of the batteries.  If this testing is
performed carefully it may allow the technician to identify one or
more batteries which stand out as having a higher internal resistance
than the rest of the batteries in the system.  Ideally this same
testing would be performed when the batteries are new, first placed in
service.  This would provide a benchmark to compare them to on
preventative maintenance service visits and troubleshooting etc.  In
the case of this existing battery system it would be great if the test
could be performed on a new Concorde of the same model for comparison
(if available).  After individual internal resistance testing I would
probably give each 12V battery an equalization charge based on
Concorde's specifications.  Then I would let them rest, disconnected,
for 24 hours and then perform the testing again.  If a battery charger
capable of charging each battery individually was not available I
would consider pairing batteries with the most similar test results
into series strings and charging the strings separately, one at a
time.  And then disconnect them, let them rest for 24 hours, and
retest.

A couple other pieces of advice with these Concorde AGM batteries:
Care must be taken not to charge or equalize them at voltages higher
than specified in the manufacturer's literature.
Also, care needs to be taken when attaching/detaching battery cables.
Over torquing or leveraging the hardware connecting cables to battery
terminals/posts can torque the battery terminals/posts and disrupt the
seal between the terminal/post and the case among other potential
issues.  If this occurs the battery may be capable of offgassing
around the terminal which will likely cause sulfation on the
connection over time.
I'm assuming these batteries are configured as a 48V system (2 strings
of 4 batteries each).  If you end up with at least 4 good batteries
perhaps you may consider rewiring the best 4 as a single series
string; so the system owner has some backup capacity while they
consider total battery system replacement.

I hope these thoughts may be helpful.

Best regards,

----MATT

Matthew Sirum
P.O. Box 1227
Greenfield, MA 01302-1227  USA
phone: +1.413.773.0611
email: MatthewSirum at gmail.com


Date: Thu, 1 Sep 2016 12:51:46 +0000
From: Gary Bassett <Gary at hudsonsolar.com>
To: "RE-wrenches (re-wrenches at lists.re-wrenches.org)"
        <re-wrenches at lists.re-wrenches.org>
Subject: [RE-wrenches] Test procedure for Concorde batteries
Message-ID:
        <4c9022358cd54ad38ff888fd1101fad5 at Hosted-MS-3.elcomp.local>
Content-Type: text/plain; charset="us-ascii"

We have a grid tied battery backup system that uses 8 Sun Xtender
PVX-2120L batteries, about 9 years old. The grid has been going out
frequently - about 4 times in the past 3 weeks. When the grid goes
out, the battery voltage gets too low and shuts the system down pretty
quickly. One of the times, this happened within 4 hours. We want to
test the capacity of the batteries and we have a testing procedure
from Concorde that seems like it would take a lot of time. Is there a
quick way to test the battery capacity?

Gary

Date: Thu, 1 Sep 2016 09:31:58 -0400
From: John Blittersdorf <john.blittersdorf at gmail.com>
To: RE-wrenches <re-wrenches at lists.re-wrenches.org>
Subject: Re: [RE-wrenches] Test procedure for Concorde batteries

Gary,
   I have just two methods I use to test all batteries.  I do a voltage
check of each battery under load with solar input shut off.  You will
quickly find a battery with a bad cell or two. It is best to do this
test when the batteries are already somewhat discharged to see clearer
results.  When under charge, and even floating or with a light load, a
battery with bad cells can look OK in the voltage check.  Bad cells
can show up as down about 1.5  to 2 volts for each bad cell from
normal.  I have also seen batteries that not only dropped in voltage
but actually changed polarity. Had a 6volt L16 read  -2 VDC.
Thatdropped the string by 8 volts!!!   Typically bad cell problems
will show up after some discharge and if a single string, the string
voltage suddenly drops down.  In parallel strings the numbers get
really random but other batteries in same string will rise high to
compensate for the low battery. I have seen a 6 volt go to +9 volts
and was boiling violently.  (hence why multiple strings are not the
best design). If the above checks are OK, batteries still appearing to
be weak have reached the twilight of their lives and a plan is needed
to retire them.  $$$$  9 years seems about right from my experience to
see serious decline in output.

   The other quick test for parallel string batteries is to use the DC
clamp meter to check each string as being roughly equal in current
under load or under full charge rate.  This will find bad cells,
different aged batteries, and /or bad connections.  When you lose a
string, that will definitely reduce the coasting time.

    I don't check specific gravity for two reasons.  It is time consuming
and moving acid around is not good in my book.  I have found individual
batteries with low specific gravity readings that went on to work OK for
many years,  If the batteries pass my voltage and current checks, I call
them good to go.

John Blittersdorf
Solar Guru
Rob Stubbins Solar (division of Rob Stubbins Electrical and General
Contractors)
280 Quality Lane
Rutland, VT 05701

802-775-1484 ext 512

Date: Thu, 1 Sep 2016 09:43:53 -0400
From: Tump <tump at swnl.net>
To: RE-wrenches <re-wrenches at lists.re-wrenches.org>
Subject: Re: [RE-wrenches] Test procedure for Concorde batteries

Gary I would suggest that the battery manufacture?s recommended method
is followed. Yeah it IS time consuming BUT?
Checking w/ a clamp on during a charge or discharge depending on the
# of strings &/or the way the strings are configured will be
misleading.
Your customer is paying by the hour and a the choice is theirs replace
OR pay your tech to watch the voltage/current over the manu?s
discharge time.
These batteries can be run thru an equalization cycle, following the
guide lines recommended by Concord, then do your load test. Full
charge to discharge over time IS the only way.