Comments on 7.2kw system in San Diego

[J.P.M.]

[QUOTE=sensij;n303756]
The lower consumption is sort of interesting. Conventional wisdom is that people increase their consumption with an array, because, free power. I think there is a segment of the population that actually becomes more efficient, because their electrical awareness is substantially higher if time was taken during the PV planning to understand more about how residential energy works. /QUOTE]

Just a thought.: I do believe most folks will increase their consumption when solar is added on the idea that the perceived cost of supplying power will be less. But I wonder if some segment of users will decrease absolute consumption - those being the folks who begin to monitor daily consumption. That thought being based on my believe that if you want to improve something, keep an eye on it and measure it regularly - what's measured will probably improve, in this case, go down.

BTW: Sensij, did you see my post about your irradiance readings ?

[jflorey2]

Appreciate any comments or recommendations!

In Santee you are in a pretty good location for solar. You'll see higher temperatures than the coastal area (=slightly lower power) but you'll avoid most of the June gloom that plagues the coastal areas. If you see power interruptions in your area you might want to consider one of the SMA inverters that provides a backup power supply.

[sensij]

BTW: Sensij, did you see my post about your irradiance readings ?

Yes... I want to get on the roof this weekend and re-verify level now that it has been up a couple of weeks and had a chance to settle, get some pictures of the installation in general. I'll get a shot of the south-facing view as well, perhaps there is a source of reflection to explain some of the difference in readings.

[ericf1]

Like some of you posted, part of the "exceeding predictions" is due to SE's metering. Also, PVWatts default losses seem excessive, and I believe the LG panels probably fall somewhere between Premium and Standard. By the time I have 1 full year (March 5th) I should be at around 13.3 mwh or so, x .97 to correct metering error still brings me back to about 12.9. Selecting Premium panels and reducing losses to 9.15% gets PVWatts to 12.7 mwh. Pretty close considering all the variables, including weather.

The main cause of my over production is the large decrease in my consumption. I'm certainly not going to try to use more energy, but won't be putting a lot of effort into reducing.

As for price, I was initially at $3.55/w, but decided to upgrade the service panel, increasing the price to $3.76/w before tax credit. I did get a Sunpower bid from the same company that installed my LG panels, and it added an almost 25% premium.

[J.P.M.]

Like some of you posted, part of the "exceeding predictions" is due to SE's metering. Also, PVWatts default losses seem excessive, and I believe the LG panels probably fall somewhere between Premium and Standard. By the time I have 1 full year (March 5th) I should be at around 13.3 mwh or so, x .97 to correct metering error still brings me back to about 12.9. Selecting Premium panels and reducing losses to 9.15% gets PVWatts to 12.7 mwh. Pretty close considering all the variables, including weather.

The main cause of my over production is the large decrease in my consumption. I'm certainly not going to try to use more energy, but won't be putting a lot of effort into reducing.

As for price, I was initially at $3.55/w, but decided to upgrade the service panel, increasing the price to $3.76/w before tax credit. I did get a Sunpower bid from the same company that installed my LG panels, and it added an almost 25% premium.

Thank you. Running PVWatts w/a smaller derate, ~ 6-10% or so, seems to get closer to actual vs. PVWatts estimates for clear days and new(er) arrays. FWIW, mine's about 8% or so.
For me, SAM runs LG 300's about 0.5% or so better than S.P. 327's on a kWh/yr. per nameplate kW basis for clear days. That agrees pretty well w/ my clear day model which matches SAM quite well using Miramar data.

All this looks to me like more data showing S.P. ain't worth the premium based on greater output/$ spent up front.

[J.P.M.]

Yes... I want to get on the roof this weekend and re-verify level now that it has been up a couple of weeks and had a chance to settle, get some pictures of the installation in general. I'll get a shot of the south-facing view as well, perhaps there is a source of reflection to explain some of the difference in readings.

That the data is consistently ~ 5.5 to 6.5% higher at your end compared to mine sort of makes me think the difference is not due to albedo augmentation which I'd think would vary over the course of a day. I'm not saying which data is more/less accurate, but there is a difference in the GHI of about 6% or so, and like Inetdog or Mike suggested, this is similar to the situation of having 2 watches and never knowing what time it is.

You have another site near you with similar GHI to yours, and I have one (Boulder Knolls on the Weather Underground) close to me that agrees pretty much with my data. That would suggest to me the diff. is location, but 6+ % or so is a whopper of a consistent albedo caused GHI boost. On the other hand, if your data is more accurate, that would make your array instantaneous efficiencies lower, perhaps lower than spec, which would not be my first suspect.

But, your daily clear day output pretty much constantly runs about 57-59 % of my output for an array that's (3.12/5.232)= .596 as large with my orientation perhaps being a tiny bit more favorable, and my roof temps generally being bit cooler from your data.

My GHI model and my sensor data seem to agree fairly well with the NREL Bird model, and seem to give reasonable and consistent fouling #'s in my searches in that area, keeping in mind that if anyone gets within 1% of repeatability, that's about as good as one can expect with the best equipment available, and even that may be a real stretch under the best of conditions. If my GHI was 6% low, my efficiencies, which are about what I'd expect, would be unreasonably high with cell STC's >21.2%.

On albedo effects at either your or my end, reflections could be from any direction or from any source that's above the plane of the sensor, including forward scatter from objects to the north of the array, particularly if "uphill".

Are we having fun yet ?