<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META content="text/html; charset=us-ascii" http-equiv=Content-Type>
<META name=GENERATOR content="MSHTML 8.00.6001.19170"></HEAD>
<BODY>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012>William,</SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012></SPAN></FONT> </DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012>I think we have it all out on the table. Below you cite
2008 690.64 (B) (2). But the requirements of (B) (2) do not come into play if it
is excluded by the provisions of 690.64 (B). And as you already pointed out,
this all revolves around the interpretation of the Code vis-a-vis the word
"capable".</SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012></SPAN></FONT> </DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012>What is the intent of the Code? What reflects reality?
In reality, you cannot find any point in the entire feeder circuit including the
load center itself where the response to a ground fault or short circuit failure
would not overwhelming be controlled by the feeder breaker. On one side of the
feeder breaker you have some inverters, whose ability to produce excess fault
current is limited short term discharge of it's capacitors. On
the other you have an almost unlimited source of excess current and uncontrolled
rise in the rate of energy flowing into a fault. I suggest that in this case, it
will always be the feeder breaker that will open in response to a fault and to
that extent it is perfectly reasonable to size the load center and feeder to the
feeder breaker sufficient to carry the maximum continous ortput of the
inverters.</SPAN></FONT></DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012></SPAN></FONT> </DIV>
<DIV dir=ltr align=left><FONT color=#0000ff size=2 face=Arial><SPAN
class=015365219-22012012>So what was the intent of the Code makers? Why did they
include the language "capable of supplying multiple branch " if their intent was
other than to allow the use of dedicated load centers?</SPAN></FONT></DIV>
<DIV dir=ltr align=left><BR><SPAN lang=en-us><FONT size=2 face=Arial>Mark
Frye</FONT></SPAN> <BR><SPAN lang=en-us><FONT size=2 face=Arial>Berkeley Solar
Electric Systems</FONT></SPAN> <BR><SPAN lang=en-us><FONT size=2 face=Arial>303
Redbud Way</FONT></SPAN> <BR><SPAN lang=en-us><FONT size=2 face=Arial>Nevada
City, CA 95959</FONT></SPAN> <BR><SPAN lang=en-us><FONT size=2
face=Arial>(530) 401-8024</FONT></SPAN> <BR><SPAN lang=en-us></SPAN><A
href="http://www.berkeleysolar.com/"><SPAN lang=en-us><U><FONT color=#0000ff
size=2 face=Arial>www.berkeleysolar.com</FONT></U></SPAN></A><SPAN
lang=en-us><FONT size=2 face=Arial> </FONT></SPAN> </DIV>
<DIV> </DIV><BR>
<DIV dir=ltr lang=en-us class=OutlookMessageHeader align=left>
<HR tabIndex=-1>
<FONT size=2 face=Tahoma><B>From:</B> re-wrenches-bounces@lists.re-wrenches.org
[mailto:re-wrenches-bounces@lists.re-wrenches.org] <B>On Behalf Of </B>William
Miller<BR><B>Sent:</B> Sunday, January 22, 2012 11:28 AM<BR><B>To:</B>
RE-wrenches<BR><B>Subject:</B> Re: [RE-wrenches] Paralleling Multiple
Inverter<BR></FONT><BR></DIV>
<DIV></DIV>Mark:<BR><BR>I would respectfully disagree, based on my
reading. If you have code citations that inform me otherwise, I would be
very grateful to expand my knowledge.<BR><BR>Pending hearing otherwise from you,
here is what I know, based on 2008 code, and assuming load side connection
(2008 citations in italics):<BR><BR>1. <I>690.64(B)(1) Dedicated Overcurrent and
Disconnect. Each source<BR>interconnection shall be made at a dedicated circuit
breaker<BR>or fusible disconnecting means.<BR><BR></I>Each inverter will require
a 40 amp circuit breaker. The value of that breaker is calculated by
adding a 25% continuous duty rating to the maximum AC output, which is 25 amps:
25 * 1.25 = 31.25. The next breaker size up is 40 amps, so you need 40 amp
breakers.<BR><BR><I>2. 690.64(B)(2) Bus or Conductor Rating. The sum of the
ampere<BR>ratings of overcurrent devices in circuits supplying power<BR>to a
busbar or conductor shall not exceed 120 percent of the<BR>rating of the busbar
or conductor.<BR><BR></I>The designer must add the values of the circuit
breakers to determine the back-feed value. We just calculated the circuit
breaker size in step 1, above. Four inverters means four 40 amp circuit
breakers, therefore: 4 * 40 = 160. <BR><BR>3. <I>In systems with
panelboards<BR>connected in series, the rating of the first
overcurrent<BR>device directly connected to the output of a utility
interactive<BR>inverter(s) shall be used in the calculations for<BR>all busbars
and conductors.<BR><BR></I>Therefore the designer must use 160 amps as the total
back-feed value for all panels and feeders in series all the way back to the
service.<BR><BR>Mark, I would really appreciate it if you could reply today with
any information I am missing. I could use my new-found knowledge to modify
the permit application I am submitting tomorrow
morning.<BR><BR>Sincerely,<BR><BR>William Miller<BR><BR>PS: I used to
think, erroneously, that I need only consider the actual maximum AC amperage
from a given inverter. Some time ago I bid on and started a job based on
that fallacy. Mid-way into the job the AHJ informed me that my
calculations were incorrect, that I needed to use the breaker value. This
job used SB6000 inverters and the value required was 40 amps. I researched
this thoroughly and discovered they were right. In order to comply with
690.640(B), I had to downgrade the main breaker at my own expense. The
breaker was not inexpensive, so this is a lesson I learned the hard
way.<BR><BR>This is why I am very interested in any knowledge that might prove
otherwise in this scenario.<BR><BR>Wm </BODY></HTML>