Cleaning my panels - 14% gain in electricity

[John_Dumke]

I live in SoCal, where we go the whole summer with no rain, and significant dust that builds up. I installed my panels in Dec 2018 and haven't cleaned them since. A week ago I got up on the roof and used water and car shampoo. My power is up 14%. I think the improvement was pretty accurate since I have had several days with no sun and consistent daily generation numbers.

Then several days later, I added 2 more 385 watt panels to the existing 24 panels. Theoretically this should raise my power by about 8% annually. Needless to say cleaning the panels was a better cost per watt improvement.

Maybe next time I won't wait so long to clean, maybe every 60 days or sooner might make sense.

[BoloMKXXVIII]

I wonder is a product like RainX would help keep them clean.

[Mike90250]

I wonder is a product like RainX would help keep them clean.

A query was made specifically about rain-x, and it can void the panel Mfgs warranty. It could damage the AR coating on the glass, or the edge seals. It's purpose is not a cleaner, but a water repellent.
I use plain water out of the hose on a dewey morning before the sun hits the panels. You don't want to thermal shock hot panels with cold water.

[BoloMKXXVIII]

Makes sense. I just wonder if a (factory) coating of some kind could make the panels less susceptible to collecting dust/debris. Of course this would add cost to the panels and I doubt the manufacturers care how clean their panels stay once sold.

[J.P.M.]

I wonder is a product like RainX would help keep them clean.

This question comes up about 1X/yr. or so. Several years ago, one of the moderators (Russ), contacted RainX and was advised to not use their product on PV panels.

The questions becomes what will happen to the antireflective characteristics of the PV glazing (still unknown at this time in any definitive sense), and how will such a treatment affect any warranty claim(s).

As a comment to the OP's post, given that array fouling is highly variable and also very site dependent, 14 % fouling penalty for a almost a year's built up with no cleaning seems to mostly agree with what I've found.

In my eccentricity, I've been keeping a pretty close eye on my array's performance for close to 6 years. One big reason among several goals for the effort is to try to get some idea of how my array fouls (or doesn't) as f(weather, atmospheric junk, time, rain/lack of, etc.). FWIW, I think I've got my array's performance and input/output parameters dialed in pretty good.

I'm in a semi rural environment (zip 92026). Precis of how my array behaves with respect to performance as impacted by array cleanliness in general:

1.) For periods of no rain my array fouls at a rate such that array output falls at a rate of, very approximately, 0.75 to 1.0 % per week. With no rain, that rate seems to be fairly steady for somewhere between 10-12 weeks. After that, without rain or cleaning, the rate of performance decrease appears to tend to become somewhat asymptotic. I'm still working on figuring that one out. Without cleaning or rain, the worst fouling penalty I've measured with my array is ~ 9% or so.
2.) I often have a nocturnal marine layer that leaves puddles on my mostly flat and mostly horizontal deck over my garage and, if heavy, will show up as precipitation of 0.01"- 0.02" in my rain gage/weather station located ~ 4ft. north of my array. That condensation can help or hurt array performance. A heavy dew will run off the array a bit and tend to improve performance a bit. Light fog/dew tends to dry in place and seems to cake the dust/dirt on the array, particularly at the bottom 8" - 12" or so of the panels (the bathtub ring as can commonly be seen).
3.) Rain events in general seem to help keep the array clean. However, beyond the general qualitative statement that more rain means a cleaner array, things are harder to quantify. Very roughly, precip. rates of 0.25" or so over several hrs. seems to restore ~ half or so of the performance lost due to fouling. A rainy day with about that much precip. will probably remove a bit more dirt. A heavy rain (heavy for around here anyway) of an inch or more will restore ~ 3/4 or more of the output lost to array fouling. A period of several days of such weather will tend to restore most performance lost due to fouling. This is not an exact science - or any science at all really.
4.) For my situation/array, a simple hosing of my array from the north with no soap/wiping/scrubbing/squeegeeing using potable water at a rate of ~ 3 - 4 gal./min. and a quantity of ~ 3/4 gal. per panel seems to restore ~ 1/2 to 2/3 of the array's performance that was lost due to array fouling. By measurement more times, and with more detail than I care to recount, I'm pretty sure that soap and another rinse gets the rest of the fouling off the array.
5.) As for performance penalty for leaving hard water spots on the array, after 5+ yrs. of doing all of the above and some careful measurements comparing performance after cleaning with soap/water/rinse and performance measurement, and then carefully removing the left behind hard water spots with vinegar, some elbow grease and a soft cloth and remeasuring, I have been unable to measure a performance difference. My conclusion is a confirmation of observations over the years that the human eye is a poor indicator of array performance and that any special treatment of the cleaning fluid or extra effort beyond simple hosing is not worth the effort, at least not with respect to keeping/restoring my array's performance.

Array tilt, 18.75 deg. Array azimuth, 195.75 deg.



Some day, after I stop all this fooling around, and as I suggest to other's in my area/neighborhood, I'll probably hose the array ~ every 4 weeks or so if it doesn't rain and accept an average/overall fouling penalty of ~ 2% to 3 %.

[John_Dumke]

In my eccentricity, I've been keeping a pretty close eye on my array's performance for close to 6 years. One big reason among several goals for the effort is to try to get some idea of how my array fouls (or doesn't) as f(weather, atmospheric junk, time, rain/lack of, etc.). FWIW, I think I've got my array's performance and input/output parameters dialed in pretty good..

J.P.M. - Great Post!

[J.P.M.]

J.P.M. - Great Post!

Thank you. Opinions vary. I'm only here to help the team.

Take what you want of my stuff. Scrap the rest.

[w00dy]

This is my first year for various reasons that I haven't gone up on the roof to clean our panels. So far for this year we have produced less than in 2018. However, due to the seasonal changes that can happen it isn't that far out of line with the averages we have had over the full 4 years of ownership. I do feel like I could have had higher numbers this summer if I had gotten myself up on my roof to clean things...

Solar 2019.png

[J.P.M.]

This is my first year for various reasons that I haven't gone up on the roof to clean our panels. So far for this year we have produced less than in 2018. However, due to the seasonal changes that can happen it isn't that far out of line with the averages we have had over the full 4 years of ownership. I do feel like I could have had higher numbers this summer if I had gotten myself up on my roof to clean things...

Solar 2019.png

What's your zip and array orientation ?

Thanx in advance.

J.P.M.

[w00dy]

95060 - 350 ft elevation (somewhat above the heavier fog zone) - our array is S-SW facing...around 210°

[J.P.M.]

95060 - 350 ft elevation (somewhat above the heavier fog zone) - our array is S-SW facing...around 210°

Thank you. I looked back this P.M.and got some of that from old posts . Apologies for not saying so and causing you extra work in responding.

BTW: How's the pool cover doing as a pool water heater ?

[w00dy]

Thank you. I looked back this P.M.and got some of that from old posts . Apologies for not saying so and causing you extra work in responding.

BTW: How's the pool cover doing as a pool water heater ?

No worries - I have no plans for solar heating for the pool - we get really nice warm water starting in late spring to around now. We have just started to get cooler water (low 70's) in the last week or two. During the warmest periods I have to open the cover to keep it from getting hotter than 90...

[J.P.M.]

No worries - I have no plans for solar heating for the pool - we get really nice warm water starting in late spring to around now. We have just started to get cooler water (low 70's) in the last week or two. During the warmest periods I have to open the cover to keep it from getting hotter than 90...

If anyone asked me, after reading your posts from back when, I'd say your dark cover does double duty as a cover and a pool water heater.

Regards,

[voxelman]

I came to this thread as the result of a search on "Rain-X". Thank you J.P.M. for your thoughts on this subject as well as the discussion regarding dirt accumulation.

Just to add a data point to the dirt accumulation discussion. I have a 44 panel 11 kW array in rural Southeast Iowa that was installed in September of 2013. There is a fair amount of wind blown dust in this area due to agricultural activity. The panels are oriented east/west with the roof slope of about 26 degrees. None of the panels had been cleaned since installation until I cleaned 4 of them in August of 2019. I noticed an immediate 8% average improvement in output between the cleaned panels and 4 identical panels mounted in the next row up on the roof. Needless to say panel cleaning will be a priority in the spring.

On another note; in June of 2019 I installed a 12 - Silfab Model SLG -M 345 panel addition to the array on the one remaining open roof surface that I had. In Iowa we have the potential for a fair amount of snow that can keep the panels covered for days or even weeks. Recently I noticed that the new panels shed the snow much faster than the old panels. I think to some degree this may be due to the dirt that coats the older panels but I'm also wondering if the panels incorporate coatings that inhibit dirt accumulation and facilitate snow shedding.

[J.P.M.]

I came to this thread as the result of a search on "Rain-X". Thank you J.P.M. for your thoughts on this subject as well as the discussion regarding dirt accumulation.

Just to add a data point to the dirt accumulation discussion. I have a 44 panel 11 kW array in rural Southeast Iowa that was installed in September of 2013. There is a fair amount of wind blown dust in this area due to agricultural activity. The panels are oriented east/west with the roof slope of about 26 degrees. None of the panels had been cleaned since installation until I cleaned 4 of them in August of 2019. I noticed an immediate 8% average improvement in output between the cleaned panels and 4 identical panels mounted in the next row up on the roof. Needless to say panel cleaning will be a priority in the spring.

On another note; in June of 2019 I installed a 12 - Silfab Model SLG -M 345 panel addition to the array on the one remaining open roof surface that I had. In Iowa we have the potential for a fair amount of snow that can keep the panels covered for days or even weeks. Recently I noticed that the new panels shed the snow much faster than the old panels. I think to some degree this may be due to the dirt that coats the older panels but I'm also wondering if the panels incorporate coatings that inhibit dirt accumulation and facilitate snow shedding.

For anything you got from my input, you're most welcome.

Thank you for your input.

I preface what follows with the caution that none of this is rocket science, highly variable and site dependent. Take it as little better than (maybe) informed guesswork based on observation and measurement over 6+yrs.

Given the widely varying nature of the mechanisms of panel fouling due to, among other things, location, what seems to be highly local dependence on geography, climate and (to restress) local weather and site conditions, I found your noted ~ 8% improvement in output for cleaned vs. uncleaned panels at your site to be about/sort of /maybe/something like in line with my SWAG that my array's fouled performance, if it doesn't rain, seems to level off at ~ 8% - 10% or so performance deterioration due to fouling. Other arrays in my HOA that I somewhat casually monitor seem to have about the same characteristics with respect to array fouling as f(time).

After that, the performance deterioration may "tend" to become asymptotic and not increase all that much. That's somewhat in line with the anecdotal observation/guess that skylites, for example, probably don't have 0% transmission after, say, 100 weeks of not being cleaned.

Some other measurements I've done on irradiance transmission through daily cleaned and never cleaned portions of the same lite of plain glass seem to agree with my preliminary guesses that fouling may be somewhat asymptotic in some way(s), at least at my site.

If that is a fair representation of reality in some fashion, it may indicate that if my panels are not cleaned (hosed off basically) on a regular basis, it may be just as well to not bother beyond major events like dust storms of big owl skrocks, accept some relatively high fouling level of dirt at the outset like 10 % or more and add that to the sought annual electrical load offset.

Then, there's also the opinion I've formed through observation and measurement that moderate to decent rain will restore something like 1/3 to 2/3 of the lost performance due to accumulated fouling.

Up to that ~ 8% fouled level, at this site anyway, array output seems to roll off at a rate of something like 0.75% - 1.0 % per week in continued rainless conditions which most likely seems to happen around here from ~ May through Oct. or so. However, the longest absolutely rainless period since I've had the array and a weather station to measure precip. and other weather variables has been 51 days, including and accounting for days with heavy morning dew/marine layer that registers as precip. of 0.01" or more.

Some day, if I ever stop measuring array output and the rates of fouling on my array, I expect I'll hose off my array ~ every 4 weeks or so if it doesn't rain and live with an ~ 2% - 3 % performance penalty due to fouling.