<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">Mick,<br>
<br>
The SunExtender technical manual says to charge at 2.385 ± 0.015
volts per cell at 25°C until the charge current drops to 0.5% of
the battery capacity. That's 28.6 ±0.2 volts for a 24 volt system
and 28.8 volts is the top end of that range. If you are seeing
signs that the batteries have vented, the voltage probably has
been too high. Since venting causes permanent water loss, with
these batteries it's probably better to have the absorption
voltage a little on the low side rather than on the high side of
the recommended range. It's really important to have a temperature
sensor and a temperature compensated charge controller too. <br>
<br>
It takes a lot more than one hour of absorption time for the
current to drop down to 0.5% of capacity; three or maybe four
hours might do it.<br>
<br>
If the end amps is set to 45, then the battery capacity should be
9000 amphours, otherwise it'll be terminating the charge too soon.
Realistically, you may as well disable the end amps feature.<br>
<pre class="moz-signature" cols="72">Kent Osterberg
Blue Mountain Solar, Inc.
<a class="moz-txt-link-abbreviated" href="http://www.bluemountainsolar.com">www.bluemountainsolar.com</a>
</pre>
On 7/7/2012 2:53 PM, Mick Abraham wrote:<br>
</div>
<blockquote
cite="mid:CABOJ=XQpVaoxuMAnvXuhxMAUAxN60=R0W7ETNtFs1GkXNjYT1g@mail.gmail.com"
type="cite">Greetings, All~<br>
<br>
My client's string of SunXtender two volt AGM's won't hold a
charge...after only two years! The prior AGM batteries also turned
to toast prematurely. There must be a better way, so could we
review the Outback charge setpoints? <br>
<br>
Side note: Delta-v from the highest 2v cell to the lowest shows a
gap of only .148v during a charge cycle so cell imbalance is not
the likely problem. I think they're all working together equally
bad. <br>
<br>
A small amount of dry powder--light green color resembling
verdegris finish on copper--can be seen around 30% of the +/-
posts, in an amount smaller than 1/8 teaspoon. From this sign, I
suspect that the cells have vented and dried out. The property is
vacant 90% of the time so I am mostly focused on the MX60 solar
charge controller as the culprit.<br>
<br>
This is an early vintage Outback controller, firmware version
5.11. Temperature compensation is in place; automatic EQ is
defeated. Bulk & absorption setpoints exactly match those
requested by Concorde/SunXtender: 28.8 volts bulk/absorption &
26.4 volts Float. The controller does regulate at these voltage
points so it's technically "working"--just grinding up batteries
on a regular basis.<br>
<br>
The Concorde people do not furnish suggestions regarding
absorption time or "rebulking" so that may be the source of
trouble. Controller setpoints have the Absorption Time Limit set
for "0 minutes minimum & 1 hour maximum". The controller is
also set to "rebulk" the battery at a 23 volt trigger and there's
an inscrutable setpoint called "End" that's set for 45 amps. I
have a theory about the meaning of that but I'll keep the noise
level down...<br>
<br>
If I recall correctly, the Outback end of charge routine in early
days was to start a clock & see how long it takes to move the
battery to the bulk/absorption voltage then apply an equal amount
of time for an absorption charge before shifting to float. Steve
Higgins may confirm if that is indeed the protocol for a version
5.11 controller but: has that method now been changed in the newer
version Flexmax controllers? It seems those now have an absorption
duration value that can be changed so maybe that's what we need--a
replacement controller along with good information on the
absorption duration and/or float setpoints, etc. <br>
<br>
I see that some controllers can even be set to not Float at all.
Maybe that's better: run the charge then stop it all for the rest
of the day. OR: thinking outside the box here...what about a
controller that won't initiate charge cycle at all unless the
battery voltage crests down to a trigger point. All I know for
sure is that two sets of valuable batteries have now been ruined
so "normal" charge setpoints are not working well in this
situation.<br>
<br>
If some better electronics (such as blending in the Outback
capacity monitor gizmo) might help, I'm all ears. Next time maybe
I'll suggest flooded cells to the client instead of AGM's, but the
cells are in a crawlspace that gets freezing cold at times. <br>
<br>
Thanks in advance for advice & suggestions. A second off grid
client has a similar setup so I need to hotfoot it out there and
tweak that (more recent) Outback controller to match the suggested
setpoint revisions...before bad things happen again. <br>
<br>
Jolliness,<br>
<br clear="all">
Mick Abraham, Proprietor<br>
<a moz-do-not-send="true" href="http://www.abrahamsolar.com"
target="_blank">www.abrahamsolar.com</a><br>
<br>
Voice: <a moz-do-not-send="true" href="tel:970-731-4675"
value="+19707314675" target="_blank">970-731-4675</a><br>
<br>
</blockquote>
<br>
<br>
</body>
</html>