[RE-wrenches] Cell temperature (back of module temp) effects on yearly kWh predicted generation

[Jim MacDonald]

Hello wrenches

We have a customer that wants us to promise kWh generation on a system, by relying on just an irradiance sensor.
I think that a cell temperature sensor should be used as well to get a more accurate picture of the actual weather conditions that the array will experience.  I'm trying to figure out what percent that hot cell temperature will effect an arrays output over the course of a year.

Each Celsius degree over 25 drops a standard C-Si module's power output by about a half a percent.
Really hot solar noons in the NE (37 celsius ambient, 57 celsius cell temp) mean losing 16% right off the bat.

Im wondering how much that effects kWh over the course of a year.

If we have 2 meteorological data sets to judge an arrays performance:
#1: only insolation data
#2: Insolation data and cell temp data

How much will the expected yearly PV AC kWh # diverge from those 2 scenarios?

For scenario #1:
Say we got 1500 kWh/ m2 per year for 2010, recorded by your pyranometer.
For a 1 kWdc C-Si system, we would expect 1500 kWh (without derate.)
Derate the system by .85, and we're at 1275 kWh.

For scenario #2:
Now, factor in cell temp (of course this varies based on mounting system, geography etc.)
Consider a system flush mounted on a standing seam metal roof in NY.
How much further could that 1275 kWh/yr be dragged down by factoring in cell temp?  5%? 10%?

PV Watts shows 1203 kWh for a 1kWdc system @ .85 derate, (1500 kWh/m2/yr) (4.11 kWh/m2/day)
Therefore 5.6% less than 1275 kWh/yr.
I'm hoping this logic makes sense.

How bout out in the desert?  10%+ losses?

Thanks for your thoughts on this
PS- I know PVSyst will tell me this, havent figured out how to ask it properly yet.. ;)

Jim MacDonald
Solar Energy Systems, LLC
Brooklyn, NY
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20110112/aedf790b/attachment-0002.html>