SunPower Solar Cell Discoloration

[newnmex999]

Hello

I had a SunPower system(E20 327 panels) installed about 6 months ago and just recently noticed that some of the cells are discoloring. It appears to be a few cells in several of the panels.

The only information I was able to find was here: http://guntherportfolio.com/2013/03/...ation-project/

And it appears this gentleman was trying to find out from SunPower if this was just aesthetics or a reliability and power output issue and has not received an answer back yet.

Let me know if anyone on this forum has any knowledge of this issue and can provide me with more information?

Thanks

[Ian S]

Hello

I had a SunPower system(E20 327 panels) installed about 6 months ago and just recently noticed that some of the cells are discoloring. It appears to be a few cells in several of the panels.

The only information I was able to find was here: http://guntherportfolio.com/2013/03/...ation-project/

And it appears this gentleman was trying to find out from SunPower if this was just aesthetics or a reliability and power output issue and has not received an answer back yet.

Let me know if anyone on this forum has any knowledge of this issue and can provide me with more information?

Thanks

I haven't seen anything like those brown cells on my Sunpower 230 watt panels installed in May 2012. I will probably be cleaning them this weekend so will take a closer look.

[newnmex999]

I haven't seen anything like those brown cells on my Sunpower 230 watt panels installed in May 2012. I will probably be cleaning them this weekend so will take a closer look.

Mine are not brown like that but allot lighter then the surrounding dark blue cells

[Mike90250]

uggh. Glad they are not my cells (in the Palo Alto tracker photos)

[Volusiano]

Reminds me of the snail trails problem but apparently this is not exactly the same thing.

[newnmex999]

Mine are not brown like that but allot lighter then the surrounding dark blue cells

These lighter colored cells seem to show up when the sun is directly overhead when you look at the panel from back to front(in other words, from the high end of the panel towards the low end). If I go around to the front of the panel and look at it, the cells in question are a little lighter then the surrounding cells, but not as pronounced.

Here are some photos:
sp4.jpgsp3.jpgsp2.jpgsp1.jpg

[Ian S]

These lighter colored cells seem to show up when the sun is directly overhead when you look at the panel from back to front(in other words, from the high end of the panel towards the low end). If I go around to the front of the panel and look at it, the cells in question are a little lighter then the surrounding cells, but not as pronounced.

Here are some photos:
[ATTACH=CONFIG]3743[/ATTACH][ATTACH=CONFIG]3744[/ATTACH][ATTACH=CONFIG]3745[/ATTACH][ATTACH=CONFIG]3746[/ATTACH]

That's pretty pronounced. Mine are kinda dirty so will check after cleaning. Do you have any indication that you are losing production as a result? If you haven't been keeping track, I'd go back as far as possible and download data from the Sunpower monitoring site for your system. Try to get the 5 minute interval data which should be available for kWh for the entire period. If you go to the advanced section of the monitoring, you can get power data for 5 minute intervals back for 60 days, I believe. Because you have not had the system for a year, you may not be able to get good comparisons but you should be able to identify major losses.

[newnmex999]

That's pretty pronounced. Mine are kinda dirty so will check after cleaning. Do you have any indication that you are losing production as a result? If you haven't been keeping track, I'd go back as far as possible and download data from the Sunpower monitoring site for your system. Try to get the 5 minute interval data which should be available for kWh for the entire period. If you go to the advanced section of the monitoring, you can get power data for 5 minute intervals back for 60 days, I believe. Because you have not had the system for a year, you may not be able to get good comparisons but you should be able to identify major losses.

Comparing September to now I am not really noticing any appreciable loss.

When checking yours, make sure you do it with the sun directly over head because with mine, if you don't look at the panels in the middle of the day and from the correct angle, you can't notice the lighter cells. Position yourself at the high end of the panel and look towards the low end.

The 3rd picture from the left that shows the one lighter cell-- if I go around to the front of the panel I can barely even notice it. Yesterday when I went up on the roof after work, I had a hard time trying to find any lighter cells at all because the sun was not directly overhead. And no matter which angle I looked from it was very hard to see any.

My SunPower E20-327 18 panel 5.88kWh system was turned on the 2nd week of Sep 579kWh, Oct 863kWh, Nov 580kWh, Dec 545kWh, Jan 673kWh. According to PVWatts Version 2 Calculator, the system is outputting more each month then it should based on my location, system size, angle of panels etc.

So the question that's in my mind right now is, were these like this when the were installed 6 months ago- I dunno?

[silversaver]

Would you share your Kiosk model link? I have a similar size system as your in CA 92887. Where is your system location? I just got my system installed Dec 2013, I don't have too much data for comparsion but compare with few SunPower system in OC through Kiosk Mode Sunpower monitoring

Thanks

[newnmex999]

Would you share your Kiosk model link? I have a similar size system as your in CA 92887. Where is your system location? I just got my system installed Dec 2013, I don't have too much data for comparsion but compare with few SunPower system in OC through Kiosk Mode Sunpower monitoring

Thanks

System is located in NM 87122. Not sure what the Kiosk model link is, can you explain? For instance, how would I access yours?

[silversaver]

If you have SunPower solar system, most likely you will also have the SunPower monitoring system. The Kiosk mode let you share your solar system performance. just basic info as outputs. Here is an example:

https://www.sunpowermonitor.com/residential/kiosk.aspx?id=2B0B7671-2507-4702-AB24-DA5805D4FF86&type=address

That is a 6.075kW Sunpower system facing about South 190 deg

[J.P.M.]

Comparing September to now I am not really noticing any appreciable loss.

When checking yours, make sure you do it with the sun directly over head because with mine, if you don't look at the panels in the middle of the day and from the correct angle, you can't notice the lighter cells. Position yourself at the high end of the panel and look towards the low end.

The 3rd picture from the left that shows the one lighter cell-- if I go around to the front of the panel I can barely even notice it. Yesterday when I went up on the roof after work, I had a hard time trying to find any lighter cells at all because the sun was not directly overhead. And no matter which angle I looked from it was very hard to see any.

My SunPower E20-327 18 panel 5.88kWh system was turned on the 2nd week of Sep 579kWh, Oct 863kWh, Nov 580kWh, Dec 545kWh, Jan 673kWh. According to PVWatts Version 2 Calculator, the system is outputting more each month then it should based on my location, system size, angle of panels etc.

So the question that's in my mind right now is, were these like this when the were installed 6 months ago- I dunno?

Re: Using PV Watts as performance predictor. Design tools/estimators like PV Watts are not predictors of performance for any one month. Over many years they may get close to long term averages given inputs that are a reasonable reflection of reality, but any one month is little better than a SWAG. For the most part, each month's output from PV Watts is the result of the input. That input is what NREL has determined as the "most typical" corresponding month (Jan. for Jan, June for June, etc.) month of either 15 or 30 years of months for the location chosen. Not average, not best, not even really, strictly speaking, most probable, with more weight given to solar irradiance data, a lot of which is synthetic, i.e. estimated. These data are known as TMY data, Typical Meteorological Year. For example, Jan. for Miramar, San Diego is from 2004. Oct. is from 1997. Aug. is from 2008, etc. Every month is different. Jan. 2014 will be different than Jan. 2004 for weather and therefore performance. In very sunny climates like NM, numbers for any month will have a higher probability of being similar to the output from the design tool than output for some location like, say, Buffalo which has much more day/day and seasonable variability, but both will still different than actual output far any particular month.
If PV Watts gives an output of say 483 kWhrs. for Jan. for a certain location and the actual output is measured at 625 kWhrs. for example, that probably means one or more, of several things with the first one being quite likely:
1.) The Jan. measured was sunnier than the TMY Jan. that PV Watts used.
2.) The array parameters were higher than the default values used as inputs to PV Watts.
3.) The system is new and/or underated by the manufacturer.
4.) The input parameters PV watts are incorrect.
5.) Something else.

Design tools are not performance predictors. They may have some use as predictors of cumulative performance over long periods (say 10 yrs. or so) with some predictive capacity whose accuracy depends on many predictions about the system and its future behavior and other things like weather. Any such accuracy and usefulness decreases as the period gets shorter. To say any system is outputting more each month than the PV Watts 2 said it would is an incorrect and inappropriate use of the tool and can lead to incorrect conclusions.

[silversaver]

If there's any tool to predict weather, please let us know.

I don't think PVWatt is predictor, I'll take the result as expected result of production. If you want to go into detail of it, which is fine. The weather from last 10 years will never be the same as 10 years from now. The global climate has been change. We just use the result to compare, no big deal.

[newnmex999]

System is located in NM 87122. Not sure what the Kiosk model link is, can you explain? For instance, how would I access yours?

[silversaver]

System is located in NM 87122. Not sure what the Kiosk model link is, can you explain? For instance, how would I access yours?

Just click on the link I provided. I don't have SunPower system. You should search the solar system within your location for closer comparsion. We are here in California and the weather is completely different as your location.

Kiosk is a flash mode to check your production. Just click on the link