André Mermoud

This is a difficult question. For Africa and Europe, the PVGIS data are rather good for recent climate. For the USA, there are several sources, I don't know which is the more reliable. For the rest of the world, if you have doubts, you can also buy paid data.

When creating a synthetic hourly data file (from external monthly data), PVsyst uses the algorithm which is included in the Meteonorm program (DLL used in PVsyst). During this process, the program Meteonorm establishes the Diffuse component (from global) using the Perez-Ineichen model (also named DirInt). Now if you specify Monthly values for the diffuse, PVsyst will renormalize the Diffuse computed from the model, in order to get the exact specified Monthly sum.

Your project file (and probably the associated calculation verions) is corrupted. This usually arises when you try to read a file which has been elaborated from a more recent version of PVsyst. In this case you have to remove the corrupted files from your workspace.

The site of the projects should contain Valid Monthly meteo data. Thes data will not be used in the simulation, but are involved in several sizing tools. If your imported data are not complete (as it is your case), you should define valid and complete monthly data, what you can import for example from Meteonorm.

The shading factor is the ratio of the shaded area, with respect to the total sensitive area. For a same collector tilt, if the pitch increases, the mutual shadings (shading factor) will decrease of course. I don't understand the rest of the question. In the 3D construction, you can have a look on the shading factor using the "Animation" (button on bottom left).

The number of MPPT's is a physical feature of each inverter. In PVsyst, you normally use the MPPT features of the inverter. But in a pre-study, you have the opportunity to get rid of this and consider the inverter "as a whole", i.e. one MPPT input (option "Uses MPPT" unchecked). Now in one sub-array, you have to specify the number of strings, and the number of MPPT inputs used for this sub-array. For example if you have 8 strings, you can distribute them on 4 MPPT inputs. If the inverters have 1 MPPT, this will mean 4 inverters; with 2 MPPT per inverter: 2 full inverters; with 3 MPPT per inverter, 1 inverter and a third. You can use other MPPT inputs in other sub-arrays.

I had a look on your file: it is not an .OND file, but a copy of the EXCEL line representing an inverter. You could import it in PVsyst by "copying" the file contents, and importing it in the Inverter's definition (button "Paste from Table"). But you file is not valid: the Voltage definitions for the 3-voltage efficiency are missing.

For an update of your license, you have to contact our administration admin@pvsyst.com.

Yes you can easily import SolarGIS data into PVsyst. In "Databases", please choose "Import Meteo Data" and then "SolarGIS". The help gives a summary of all meteo data sources which you can easily import into PVsyst. Please see in the help "Meteo Database > Import meteo data".

This is a problem with the domes manipulation. We have corrected it for the next version 6.43, to be released on March 16th, 2016.

The Rule of thumb remains valid: if your system size is less than around 0.2 miles, you can use the Far shadings option. Now you can also use both options, and analyze the discrepancies.

The Solar geometry calculations are fully described in the help "Physical models used > Solar Geometry"

Please send us the concerned controller file (to support@pvsyst.com).

Everything is fully explained in the Help: "Project design > Grid-connected system definition > Power optimizers > SolarEdge Architecture".

The Hay model is fully described in the help: "Physical models used > Incident irradiation models > The Hay transposition model". For the Perez model, it is more complicated, you have to follow the original article: R. Perez, P.Ineichen, R. Seals, J. Michalsky, R. Stewart: Modeling Daylight Availability and Irradiance Component from Direct and Global Irradiance. Solar Energy 44, no 5, pp 271-289, 1990.

Please see our FAQ When starting PVsyst I get: "The application can not start because it is not signed by PVsyst SA."

This not yet possible. We are working on this. However if it is for example a factory rooftop, you can record a plane image from your Sketchup construction, and use it as a background image in the 3D construction tool of PVsyst.

Sorry, it is a "youth" bug in the version 6.42. You can use the tool "Degradation" only with a new calculation version, not with a calculation version that has been read on a file. This will be corrected in the next version 6.43.

In the Orientation part, you choose "Tracking, tilted or horizontal tracking axis". If your trackers are horizontal, you set "Axis tilt" = 0. You can check "Backtracking" if you want to use this strategy. You define your system (array with PV moudles and inverter(s). Then in the "Near shadings" 3D editor, you create your array by choosing "Object > New > Tracking PV planes", and you define the same parameters as in the "Orientation part. You can follow the tutorial present in the software (just below the "Help" in the menu), especially the part dealing with the shadings scene construction. This doesn't concern a tracking system, but the principles are the same.

Make sure that this file is issued from the program Meteonorm, with the correct mention "Output for PVsyst" chosen at the export time. We didn't have any complaints about the import of hourly files correctly created by Meteonorm since more than 15 years.

PVsyst supposes that you use an image which is a plane view. You cannot use an image seen from a not vertical view. I don't see the problem for getting such an image from GoogleEarth.

No, we have not scheduled such a training.

With a Helios3D scene, the orientation taken into account by the simulation is always the average of all "real" orientations. However we don't have any way for quantifying the error due to the orientation distribution in the present time.

Are you sure that your noon corresponds exactly to the "solar noon" (i.e. sun's Azimuth = 0)? This cannot be the case along a full year. Therefore you always have a slight tilt of your trackers, which may explain this difference.

Have you checked the head for these operating conditions ? There may be many parameters for explaining the behavior of the pump during the simulation.