Flash Test Data [RE-wrenches]
Joel Davidson
joel.davidson at sbcglobal.net
Sun Jun 17 13:54:09 PDT 2007
<x-flowed>
Hello Antony,
Manufacturers test and sort cells into 4 to 8 categories or "bins" usually
by current at a fixed voltage and then assemble module cell strings with
matched cells. Some manufacturers squeeze as many $/W DC STC as possible by
rating the same module configuration in 5 or 10 watt increments (for example
165, 170, 175, 180 watts) which is also good for customers because they get
closer to want they pay for.
On the other hand, the first terrestrial solar module manufacturers use to
"pump" the numbers rating, for example, a 90 to 95 watt module at 100 watts
plus or minus 10%. Rarely if ever did customers get as few as 15% of a
shipment above the rated power so "plus" was a myth. Nowadays, most modules
manufacturers use statistical analysis to optimize throughput (volume) and
yield (saleable product) which is why the same size module can be sold as a
165 or 180 watt device.
Minimum power ratings became commonplace for polycrystalline silicon modules
because poly cells lose 3% to 6% of their power shortly after being exposed
to light and then they stabilize. Several years ago, poly cell manufacturers
started using dual ratings (for example, Pmax 175W and warranted minimum
Pmax 166.5W). Amorphous silicon cells and modules also have factory and
light-conditioned ratings.
I like to compare PV manufacturers to bakers. When you buy a dozen donuts,
you like to get 12 nice looking donuts. If you get a baker's dozen (13) good
looking donuts, you'll probably become a repeat customer. If you get 12
wimpy donuts, you won't be happy. And if you only get 11 donuts, you will
complain, demand your money back, take your business elsewhere and tell
everyone about that damn bakery.
PTC is a whole other story. PTC got us a little closer to real-world ratings
by bringing to people's attention the fact that cell and module voltage and
power change dramatically with temperature. However, the CEC eligible
equipment module watts DC PTC times inverter efficiency is a construct used
to standardize system ratings for rebates. Watts DC STC is a measurable and
legal description and so is watts AC PTC. However, PV systems do not always
put out watts AC PTC. Systems with batteries generally put out around 65% of
DC STC and batteryless systems put out around 75% to 85% of DC STC under
ideal conditions.
I give customers 4 ratings:
Watts DC STC (you can take it to the bank or court)
Calculated Watts AC PTC (again, you can take it to the bank or court)
Estimated kWh from PVWatts (just an estimate but from a 3rd party and
customers can play with the numbers on line)
Expected performance from the EPBB (just an estimate but required for the
rebate).
I want to give the customer a baker's dozen so I under-promise and
over-deliver. Customers won't know that they got a good deal from me until
the system has been in operation for a while. The good news is that
customers who bought PV from me years ago, buy PV from me now and refer
their friends to me. The bad news is we occasionally hear about people who
bought PV from someone who did not know PV or pumped the numbers which is
why we have to police our industry and why we need consumer protection laws.
Joel Davidson
----- Original Message -----
From: "Antony Tersol" <atersol at owl.csusm.edu>
To: <RE-wrenches at topica.com>
Sent: Sunday, June 17, 2007 11:32 AM
Subject: Re: Flash Test Data [RE-wrenches]
IF each module IS flash tested what is the exuse for the -/+ 5-10%
warrantys? Isn't that just a way for the manufacturers to overcharge for
what they are delivering? If -0%/+whatever% was required, the
manufacturers could sort their modules into groups of modules that were much
more closely matched, improving overall performance of each installation,
and making it more likely the actual performance would meet expectations.
In California, the new CSI program bases incentives on "Expected
Performance". Yet their EPBB calculator uses STC rather than PTC ratings
(despite PTC being regarded as a better predictor of installed performance),
and uses the manufacturers nominal rating when the warranteed rate is often
much less. [http://www.csi-epbb.com/]
For example, one manufacturer's Limited Power Warranty says "... a power
output less than 90% of the Minimum Peak Power (1) ... "
"(1) “Minimum Peak Power” = Peak power minus the Peak power tolerance..."
So for a -5%/+10% tolerance warranty they are promising to be within 10% of
95% of the nominal rating - for a module sold as 205 watts they're promising
you'll get 175 watts, almost 15% less.
And if it's a -10%/+10% it's 19% less. Both of these drops are bigger then
incurred by installing the modules flat instead of at optimum tilt to the
south!
Since priced by the rated nominal watt, the customer can end up paying for
more then they get. Since priced for 205 watts, if the module only delivers
175 watts, the real cost is (205 - 175) /175 = 17% more per delivered watt.
And as installers, you have customers getting 15% less production then
expected.
Ahh, but you say, "how about the +10%? The customer might get more then
they expected!" Well yes, they might, but do you expect that manufacturers
rate their modules according to the average rate, or the best they get?
Which leads to more revenue? And even if some of the modules have a higher
rating, production in a string is determined by the module with the lowest
rating.
"Won't market forces reward the best and punish the worst?". If quality
isn't brought to the fore prior to the purchase, how can it influence
behavior?
A large part of our task is educating potential clients. Try explaining to
a client why the official program's "Expected Performance Based Buydown
Calculator" is mistaken in predicting production using STC rating. And try
to convince the client that the system with a higher PTC rating, but lower
EPBB production estimate is better than a competitors system with a lower
PTC rating, but higher EPBB production estimate. Wouldn't the client tend
to think the "official" rating is the one to believe?
By the time the customer is experiencing lower production then promised it
is too late. The competitor is gone, out-of-business, or merged into one of
the new PV marketing conglomerates. Shades of the '80s?
If the programs reward overselling and underdelivering, don't be surprised
if that is what results.
The CSI program has introduced an order of magnitude more work for the
installer for compliance, over differences in ratings of a few percentage
points. All to "ensure the best possible performance".
Why introduce a huge amount of bureaucracy to chase after a few percent of
expected production differences, while letting 15-19% slip through the
STC/PTC/tolerances crack (more like a gaping hole)?
[Compared to the old program, CSI greatly increases the amount of paperwork,
site-surveying, and inspection, while making the rebates sensitive to the
"Expected Performance" due to geographical factors (location and weather)
and design factors (tilt, azimuth, shading). They've chosen an arbitrary
reference location, such that other locales are penalized if they have worse
expected production, but locales with better expected production are not
rewarded. And they've created more work for the administrators - more forms
to enter, more things to check, more inspections - so much so that it has
been reported that not one single system has been approved and connected
under the new program.]
Best way to improve performance: tighten up the quality requirements. The
relationship between STC and PTC is established; PTC is regarded as a better
predictor of installed performance. Make sure the customer is getting the
STC for which they're paying, and within tighter tolerances, and the
performance of the systems will be improved.
Require that each module be tested, and mandate a -0% delivery warranty!
There are _some_ manufacturers who so warrant. But if it isn't required,
then those that use "smoke and mirrors" guarantees are rewarded.
Best way to get industry-wide warranties that better protect the consumer?
Have the California program require it. Have you noticed the 10 year
warrantys the inverter manufacturers are offering?
Antony Tersol
Applied Solar Energy
jay peltz wrote
"Date: Sat, 16 Jun 2007 17:46:11 -0700
From: jay peltz <jay at asis.com>
Subject: Re: Flash Test Data [RE-wrenches]
HI Joel,
Given that each module IS flash tested and referenced to the serial
number I can't believe that it would cost much to produce that via
email or paper copy on the module or some other form.
And given that ratings are -/+ 5-10%, well wouldn't you want to know
if you are getting what you are paying for?
Thanks,
jay
peltz power"
- - - -
Hosted by Home Power magazine
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/
Moderator: michael.welch at homepower.com
- - - -
Hosted by Home Power magazine
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/
Moderator: michael.welch at homepower.com
--^----------------------------------------------------------------
This email was sent to: michael_welch at sbcglobal.net
EASY UNSUBSCRIBE click here: http://topica.com/u/?bz8Qcs.bz9JC9.bWljaGFl
Or send an email to: RE-wrenches-unsubscribe at topica.com
For Topica's complete suite of email marketing solutions visit:
http://www.topica.com/?p=TEXFOOTER
--^----------------------------------------------------------------
</x-flowed>
More information about the RE-wrenches
mailing list