[RE-wrenches] For Big Bank Off Grid System Connoisseurs

Mark Frye markf at berkeleysolar.com
Tue Jun 22 17:15:21 PDT 2010


Thanks Phil,
 
I really appreciate all the input I am getting.
 
As I mentioned, these are not the Solar One batteries, they are forklift
batteries, so I am not even going to talk to NWES.
 
I am working with the regional Enersys service center to have a guy out to
cut and re-weld the interconnects after the cell is replaced. If they want
to help me out because the battery is still within the 5/2 forklift warranty
period, that will be fine.
 
Everything you say makes sense.  The owner is on-board with spending more
time and effort in order to care for the batteries and I am working up a
plan to do that.
 
At this point it looks something like:
 
- Replace the bad cell and continue with the 4 string system size
- Give the bank a full EQ charge. I like your set points of 62-64 V for 3 to
4 hours (just don't get the cells too hot). I would really like to see all
the cells up to 1.280 for once. I can break the bank down to it's individual
strings to do this, in order to give it a bit more punch
- Review and adjust the charge settings 2 to 4 times a year to optimize for
seasonal changes in ambient temperatures and load profiles. This might
include adjusting the volt start settings to insure that the generator is
running more often (every 7-10 days sounds good to me); extending the absorb
time as much as possible (staying above 50% load on the generator).
- Giving an EQ charge as above once every 4 to 8 weeks.
 
I imagine that in this way they will muddle along fine for quite sometime.
When a major capital component needs to be replaced they can make the best
choice at that time.
 
I like a lot of the other ideas that have come up such as more ventilation,
both of the room and the inverters.
 
Please any more thoughts you may have.

Mark Frye 
Berkeley Solar Electric Systems 
303 Redbud Way 
Nevada City,  CA 95959 
(530) 401-8024 
 <http://www.berkeleysolar.com/> www.berkeleysolar.com  
 

  _____  

From: re-wrenches-bounces at lists.re-wrenches.org
[mailto:re-wrenches-bounces at lists.re-wrenches.org] On Behalf Of Phil
Undercuffler
Sent: Tuesday, June 22, 2010 4:29 PM
To: RE-wrenches
Subject: Re: [RE-wrenches] For Big Bank Off Grid System Connoisseurs


I'm still not clear on why everyone thinks that this is a failure which
should be covered under warranty.  If Rob or Jean want to give a free cell,
then that's above and beyond -- but I don't think we should be blaming the
battery in this case.  

Let's explore the situation.  We've got 6760 AH of battery storage, in four
parallel strings.  Four strings in parallel would not be my recommendations,
but let's leave that aside for now.  The solar, if putting out full rated
wattage with no degradation due to heat, off-axis, shading, soiling, or any
other loss will put out a whopping 100 amps, for a mind boggling C/67.  Add
normal derates, and we're down to a C/80 charge rate, and we haven't even
accounted for daytime loads.  The solar is there for show, and to make the
owners feel green.

So how about the generator charging?  We have 10 Outback inverters -- that
should be enough, right?  Well, 10x45A each gives us a C/15 charge rate,
which should be OK.  But is the generator large enough?  That's a lot of
inverters to drive, this load is pretty reactive, and we've not any
information about the background AC loads except that they're probably
pretty large.  Wait, the generator is only 30kVA, which is only good for
24kVA of continuous resistive load (30*.08).  What's the altitude of the
installation -- if it's higher than 1,000 feet the generator should be
further derated.  No inverter is as efficient at charging as it is at
inverting, so of that 24kVA of available generator capacity, only 19kVA is
likely to make it to the batteries -- that's a likely C/21 charge rate if
there were no AC loads, but there are, so it's likely to be lower.

And when do the batteries get charged?  Not every day -- the solar can
barely keep up with the self discharge.  The generator runs every 2 - 3
weeks, so the batteries are in a state of near constant discharge for 14 -
21 days on average.  When they get so low they hit the AGS floor, the
generator kicks in.  However, just as the battery voltage climbs to the
beginning of the absorption state, the generator is shut off.  

These batteries haven't been full in over six years.  They haven't received
a proper charge in over six years, either.  I'm surprised they've lasted as
long as they have.  Let me guess the last time they were equalized -- really
equalized, not just "I hit the button and ran the generator until I got
tired of the noise" equalized -- as in charge until the voltage rises up to
62-64 volts for 3 or 4 hours, and the specific gravity on each and every
cell hits 1.265.  Was the bank commissioned with an EQ upon installation?  

I think that one of the best things that can be done to serve these clients
(other than recommend a huge investment in efficiency) is to decrease the
time between generator charges by 50-75%. You want to get it to the point
where the generator is cycling on at least every 7-10 days.  In addition,
you really want to extend the absorption time -- 12 minutes is just not
enough to do anything but tease.  The customer really needs a bigger
generator to drive this system, too.  Adding vent fans to the power room is
highly recommended -- you're dissipating something like 6,000 watts of heat
when you're running 10 inverters at full charge.  Power the vent fans off
the generator output, so they only come on when the system is charging.

When it's time to replace this bank of batteries, I'd recommend looking at
decreasing the size of the bank by 50%.  And investing more in efficiency,
of course. 

Hope this helps,

Phil Undercuffler
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20100622/2ae237dd/attachment-0003.html>


More information about the RE-wrenches mailing list