SMUD PV systems [RE-wrenches]

Joel Davidson joeldavidson at earthlink.net
Sat Feb 4 14:15:58 PST 2006


<x-flowed>
Hello Matt,

Thank you for sharing your knowledge and experience. The information you 
share is important history from which we can learn.

The Sacramento Municipal Utility District, Don Osborn and Dave Collier were 
real PV Pioneers. They took risks and succeeded in encouraging SMUD 
management and customers to spend money on PV when other utilities and 
government were dragging their feet or pushing back. Don pushed the concept 
of Sustained Orderly Growth, so necessary for PV to become a mainstream 
technology. SOG has finally become a reality thanks to Japan and Germany's 
multi-year PV commitment. Now California has a 10 year 3000 MW plan. A few 
days ago, solar was mentioned for the first time in 25 years in a state of 
the union speech and federal R&D PV funding will return to the level it was 
in the 1970s. (I could go on and remind you all to remain vigilant because 
the anti-solar forces are still lobbying strong, but I'd better get back to 
nuts and volts.)

I stand by my statement that none of the 240 inverters failed due to high 
Voc.

We did have a lot of inverter early failures. Omnion tried to blame the 
installers, but Placer Electric electricians were doing a good job. The 
problem was Omnion's quality control and their short factory burn-in which 
was only overnight (8 - 12 hours) and did not simulate actual field 
conditions. Hans Myers, Omnion founder and PV Pioneer, was the first to 
implement a warranty support policy. He agreed to accept the installers' 
work and pay for all repair and shipping costs if installers signed off on 
my inverter installation procedure (contact me off-line for a copy of the 
procedure). First year inverter failures went from 6% to zero.

I did not do any SMUD systems behind fences. I did 102 customer-sited 
residential systems (SMUD Block 2), several churches and other 
customer-sited "commercial" systems. Typical system configuration was 56 
SQ-80 modules (the world's first 6" round cell modules), an Ananda Power 
Technologies (APT) combiner box designed by John Bernder, another PV 
Pioneer, and an Omnion 4 kW or 6 kW inverter. 14 modules wired in series 
would have totaled 301 Voc if the modules were to 21.5 Voc spec, but the 6" 
round cells were pushing the PV technology envelop and actual Voc was 21.2 
volts so the bi-polar voltage was just under 600 volts at STC. At PVUSA test 
conditions (now called PTC), Voc was over 600 volts. It's hard to say what 
NEC 690.7 Voc is because there are more than one so-called lowest record 
ambient temperature for Sacramento depending on the source and location.

I did not do 42 or 22 SQ-80 modules in series. Perhaps this was done when 
Omnion inverters were replaced with Sunny Boys.

You didn't mention the mounting system that used wood screws into plywood 
instead of lag bolts into framing members. I have not heard of any leaks or 
failures (except for the squirrel's nest under a 7-module panel, but that's 
another story).

Joel Davidson

----- Original Message ----- 
From: "Matt Lafferty" <mlafferty at universalenergies.com>
To: <RE-wrenches at topica.com>
Sent: Saturday, February 04, 2006 8:30 AM
Subject: RE: Lowest DAYTIME temp records [RE-wrenches]


Your free subscription is supported by today's sponsor:
-------------------------------------------------------------------
Sign Up for Packet8 and tell your phone company
to VoIP-Off. First Month Free! Then up to 50% off your
phone bill.  Free Equipment! Packet8 1-877-868-1135
http://click.topica.com/caaer6vbz8Qcsbc0WJAa/ TMONE
-------------------------------------------------------------------

Hi All,

A couple of days ago Joel D referred to 240 systems that were installed in
SMUD territory which calculated at >600 VDC Open Circuit and didn't result
in any inverters failing due to Over Voltage.

I'd like to add some clarification to that statement.

Joel D's experience is only part of the story related to this topic in SMUD
territory.  There are several SMUD-Owned Arrays, large and small, which
exceed 600V when you apply NEC 690.7. Two particular configurations
calculate out to >640V, in fact. To be clear... These are SMUD-Owned,
Utility-Scale, and sited in Sub-Stations, behind fences. SMUD is the one on
the hook for any repairs and they accept that.  Like Allan S said...

Neither of these configurations have resulted in "over-voltage" failures to
my knowledge. Neither of them has ever met their "specs" to my knowledge,
either, but that's a whole different story... Be sure to buy the book one of
these days! (Michael: I haven't written it yet, so I don't think that
comment belongs on the Marketing List yet! Let me know if I'm off-base
here... )

Back to the systems that Joel D was involved in...

First of all, these systems were installed using the old Omnion 2200 & 2400
series inverters. Many of these inverters required repair throughout the
years. Repairs were done by SMUD's Electrical Technicians and records were,
and still are, NOT kept regarding what type of failures and/or causes
ocurred. Repairs were made on an "as-required" basis to get the systems back
online. Control boards, capacitors, relays, cooling systems, whatever... I
do not personally know of any "over-voltage" failures, but that is only
based on my general conversations with the Techs over the years. With all
due respect, Joel, I don't see how anybody can make a statement that "None
of the 240 inverters failed due to high Voc."

For those of you unfamiliar with the Omnions, they are a Bi-Polar DC Input
inverter platform. What this means is that the input circuits are
"center-tapped" which results in the inverter "seeing" two different
voltages... A "+" and a "-" voltage as measured between each pole and the
center-tap. (The entire string is split into two source-circuits) The
inverters were rated at "+400 / -400" nominal.  In those days, the +/-400
referred to "Nominal" which is the operating voltage instead of the Open
Circuit voltage.

The vast majority of the systems used the Solec SQ-80 modules, which have a
21.3VOC rating.  String configurations were as follows:  42 modules in
series with a center-tap between module #21 & #22.  This calculates out at a
total string VOC of 894.6 if measured between + & -.  The actual source
circuit voltages were +447.3 & -447.3 VOC, which results in a VMP of +359.1
& -359.1 respectively.  As you can see, this is well below the "Nominal"
rating of the inverters.

Some notes about this type of configuration:
The NEC used to consider each side of a Bi-Polar circuit as a separate
circuit for voltage ratings.  This is no longer the case. A Bi-Polar
cirucuit is now required to be rated at the total + to - voltage and all
conductors & components in that circuit are required to be rated at no less
than that figure.  In this case, the original configuration would not be
allowed on a residential application for two reasons... First of all it
would be >600V. Secondly, the wires are only rated at 600V, as are the
modules.

The Solec modules never reliably produced their rated voltages. Still don't.
We have been refurbishing these systems for customers who elect to purchase
them for the past two years. This process is called "converting" due to the
ownership being transferred from a SMUD-Owned system to a Homeowner-Owned
system.

During this process, the entire system is removed from the site, including
the inverters, conduit, etc. Roofs are repaired &/or replaced.  Inverters
and conduit are disposed of. Modules (Panels of 7 each) are taken to our
shop, dis-assembled, thoroughly cleaned & inspected, individully tested for
VOC & ISC and sorted according to the test and inspection results. We
experience an average of approximately 20% failure rate due to one of three
primary factors. (Physical failure of or in the J-Box; Low-VOC; Low-ISC.)
Modules which fail at this stage are transported to SMUD's Salvage facility
for sale at a later date in a Salvage Auction. (Caveat emptor! Remember,
they look nice and shiny by now!) Failed modules are replaced using "good"
modules from a supply of modules which come from "retired" systems. "Good"
Modules are then re-panelized on UniRac Solarmount rails. After roofing work
is completed, the system is re-installed with a new Sunny-Boy. Systems are
re-configured to meet current codes and go through the entire permitting
process. String configurations are 21 Modules in series (447.3VOC rated).
Complete commissioning tests are performed. Results are recorded. Under
favorable conditions (good sunlight & low temps) it is common to see VOC's
come in at 85-90% of the rating.  ISC's come in around the 70-80% range.
These values are "as compared to corrected for irradiance and temperature",
FYI.

For those of you concerened about the Customer... SMUD uses a very
aggressive formula for "de-rating" the entire system which comes out very
much in favor of the Customer. (The Customer ends up getting more system
than they pay for and they pay a very low $/Watt for the systems). SMUD also
provides a complete-system warranty of 5 years. In addition, before a
Customer is allowed to purchase their system, a complete site evaluation is
performed, including Orientation, Shading Analysis, and Structural
Evaluation. Due to historical siting standards, not all systems are deemed
to be suitable for conversion and get retired.  Which is where the "makeup
stock" of modules comes from...

Hopefully some of you find all this to be at least somewhat interesting.

>From the trenches of history, old and new...

Pray for Sun and please don't forget to donate to the Hammock Relief Fund!
(Yes, rum is considered an acceptable donation...)  :-)

Matt Lafferty
mlafferty at universalenergies.com

Your free subscription is supported by today's sponsor:
-------------------------------------------------------------------
Shopping for a new car?
Pay 2-3% below invoice on a new car of your choice.
Save thousands with our free car discount finder service!
http://click.topica.com/caaerMmbz8Qcsbc0WJAf/Dealix
-------------------------------------------------------------------

- - - -
To send a message: RE-wrenches at topica.com

Archive of previous messages: http://lists.topica.com/lists/RE-wrenches/read

List rules & how to change your email address: 
www.mrsharkey.com/wrenches/etiquette.php

Check out participant bios: www.mrsharkey.com/wrenches/

Hosted by Home Power magazine

Moderator: michael.welch at homepower.com





</x-flowed>



More information about the RE-wrenches mailing list