Most optimistic Annual Yield figures - PVGIS or PVSOL?

[cookiemonster]

I guess the question is in the title! Would anyone care to comment on how these compare. I currently use PVGIS and Sunny Design to estimate annual yield and for the most part for a given slope, azimuth etc their estimated yields seem to be pretty similar. i have seen some quotes from competitors which appear pretty optimistic e. g. an extra 20% higher for a virtually identical kWh system. Is this just deliberate falsified info or are they getting their figures from PVSOL I'm wondering?

 

[MorganPVI]

Some southern companies assume the insolation is 1200 w/m3. This leads to esitmates of over 4000 on some systems, creative accounting Enron style.

 

[Morgan keith]

Are you sure their figure isn't just the SAP calc? I've kind of given up on giving customers the figures from PV Sol (unless there are shading concerns) as they're almost always lower than the SAP and figured it probably puts you on the back foot more than anything because the customers think it's more to do with your equipment etc. Don't get me wrong though, I do think the SAP calc is a joke!

 

[tlsambo]

SAP is a very crude calculation. You will have noticed this if you've used it. It starts off by assuming you're in the middle of the UK so if you're further north then the PVSOL/GIS figures will be lower. It's next failure is it only allows for 15 degree increments on the roof pitch and 45 degree increments on orientation. Then the shading, do you know what significant shading looks like? I don't, it's so subjective it's insulting. Then the crucial floor is that there is no allowance for the difference in quality of panels.

As for 20% more from a system of the same kWp, this is possible as it highlights one of the floors with SAP. Good quality panels like Conergy or BP will give much more output than Sharps, Yinglis etc.

It is possible for a good set up with a good roof to generate more than 4000kWh annually in Cornwall (I design systems for Cornwall Solar Panels).

 

[whlphil]

@Tlsambo I take it you are using PVSOL what weather data and irradiance are you using? TMY2 data? Which location? I know in v4.5 there wasn't much data past Devon although I'm using v5 now with Meteonorm and I haven't looked at Cornwall yet, I'm guessing there may be some more files from Perinporth or other RAF bases

 

[tlsambo]

yes, we use PV Sol and have some weather data files taken from PV GIS for Bideford, Launceston, St Austell, Newquay, Truro, Penzance and the Scilly Isles. These don't have enough wind data to operate the dynamic wind calculation so it relies on the linear. This should still demonstrate the difference in quality of panels. However when you use the dynamic calculation it does show Sanyo up a bit. Here's some comparisons using Plymouth weather data

Dynamic Linear
Sanyo 17 x 235 3.995kWp 4031 3998 +33
BP 21 x 190 3.99kWp 4127 4083 +44
Suntech 21 x 190 3.99kWp 4042 4002 +40

So as you can see, due to Sanyo's extra thickness they hold on to heat more which makes them less effecient!

Hope all that helps.

 

[tlsambo]

Sorry, should have pointed out that the results are in kWh/annum generated from the inverter.

 

[tlsambo]

and, as the forum software has removed all my carefully positioned spaces the first result figure is the dynamic result and the second is the linear.

 

[whlphil]

yes, we use PV Sol and have some weather data files taken from PV GIS for Bideford, Launceston, St Austell, Newquay, Truro, Penzance and the Scilly Isles. These don't have enough wind data to operate the dynamic wind calculation so it relies on the linear. This should still demonstrate the difference in quality of panels. However when you use the dynamic calculation it does show Sanyo up a bit. Here's some comparisons using Plymouth weather data

Dynamic Linear
Sanyo 17 x 235 3.995kWp 4031 3998 +33
BP 21 x 190 3.99kWp 4127 4083 +44
Suntech 21 x 190 3.99kWp 4042 4002 +40

So as you can see, due to Sanyo's extra thickness they hold on to heat more which makes them less effecient!

Hope all that helps.

Does anyone have any actual results to compare this against, I'd be very surprised if these results actually stacked up as correct. Obviously PVSOL et al dont allow for better diffuse pick up from the amorphous layer so I highly suspect that all the results quoted for Sanyo will be very definitely on the low side, this is borne out by my German friends findings. I have a had a word with Sanyo about producing a revised file for PVSOL to show a more accurate result but whilst they produce good panels asking them to sort technical queries is like banging your head against the wall..

 

[tlsambo]

The low light performance of the panels is taken in to consideration in PV Sol as their database is kept up to date by the manufacturers. I don't like to over estimate the impact of the amorphous layer as the majority of the power is still from the crystal. Also, as I pointed out earlier although Sanyo temperature co-efficient is quite good their ability to dissipate heat is not as they're so much thicker than other panels due to their extra layers. PV Sol calculates the cooling effect of the wind and temperature on the panels which is why Sanyo panels don't generate as much as many people think.

Sorry to shatter any illusions but it would appear that Sanyo panels are overrated.

 

[hruda]

If you really care about global solar data with the best overall precision, use solar radiation data from SolarGIS database (there is a nice PV simulator, too). Especially when I work in the areas outside the well known climate conditions (like e.g. most of Germany).
There are a few comparisons of available solar databases on the market, e.g. study of University of Geneva, they recommend their data. PVsyst recommends their TMY.
I gave up to use PVGIS data, especially that "classic database" - this is interpolated from ground stations. And if you have not enough ground measurements and more complicated mesoclimate regions, you are in trouble and only input data can cause me 20-30% mistake.
As far as I know, PVSOL is similar, Meteonorm almost similar (they still use interpolation methods, even they declare they do some enhancements from a satellite models).

 

[confounded]

I have a had a word with Sanyo about producing a revised file for PVSOL to show a more accurate result but whilst they produce good panels asking them to sort technical queries is like banging your head against the wall..

We have installed many Sanyo systems, we are still lacking any data to justify a superior performance relative to many other respected manufacturers. They offer a greater energy density but higher yields are not being born out by our data.

To cover some of the quires regarding PVSOL yield predications we have found our systems have outperformed the PVSOL yield by 15% or more in some cases. We also have some local competitors predicting 4400kWh's from a 4kWP Sanyo system. It is possible but to provide a customer these figures as a given for 25 years is pushing it a bit....

 

[SolarCity]

We also have some local competitors predicting 4400kWh's from a 4kWP Sanyo system.

How one Earth do they come to that figure?

 

[whlphil]

How one Earth do they come to that figure?

If you have a well designed system in Southampton or Cornwall you should be seeing that sort of figure with a good panel!

 

[SolarCity]

You may well see it - however, you certainly shouldn't be predicting it.

I haven't installed a system yet that has reached that in a year. 4,200 or so is the highest I have seen.

 

[confounded]

You may well see it - however, you certainly shouldn't be predicting it.

Absolutely, we have systems that could achieve the 4400 in a year but to base the 25 year payback on it is crazy.

 

[SolarCity]

Absolutely, we have systems that could achieve the 4400 in a year but to base the 25 year payback on it is crazy.

Crazy and against REAL rules. Not that that seems to matter.

 

[liquidity]

There's poster on navitron who is reporting just over 4900 kWh in 2011 from a system in Littlehampton - 17 Sharp 235s with a 4000TL. Wonder what he would have got with Sanyo panels?

 

[SolarCity]

4,900?! I don't have any annual figure yet for my systems with TL inverters so we'll see how they come in.

 

[tlsambo]

I've run Winaico Panels on Penzance weather with a Fronius IG4.0 and got 4173. This is the best combination I know of (south facing, 37 degree roof) for PV Sol. Unless there is more irradiation on the Channel coast? This is the best combination for Cornwall that I know of. We keep a league table of panels which we run on an optimised roof on Plymouth weather data so we can compare them on a level playing field. We run each panel with Power One, SMA and Fronius inverters.

It is interesting to see the gaps between panel manufacturers close or widen depending on shading and orientation variables too!

If anyone else runs some better figures I'd be keen to know what combinations you've used.