Modeling IAM losses in SAM

Hi Paul,

Do you know how the detailed PV version of SAM models IAM (Incident Angle Modification) losses. My colleagues who use PVSyst note that IAM losses are significant in the waterfall loss chart (~2.5%). Does SAM treat this internally behind the scenes? If so, what are the assumptions that go into it?

Best,

Nathan

Hi Nathan,

The incidence angle modifier or IAM is a way to account for optical losses that occur when the solar angle of incidence on the array surface is greater than zero. In other words, when the array surface is not perfectly normal to the rays from the sun, some energy is lost to reflection from the module surface materials.

How SAM accounts for these losses depends on the module model that you choose. For a detailed description, please see the PV reference manual. (You can download it and other technical documents from the Performance Model Documentation page.)
Here's a summary:

The Sandia Module Model uses a set of empirically derived coefficients (b0...b5) and a fifth-order polynomial equation to account for the losses. These are based on measured data from modules in the field (see Equation 9.3 in the PV reference manual).

The CEC Module Model characterizes the losses based on the optical qualities of the module cover using physical equations for transmittance of surface normal, beam, sky diffuse and ground diffuse irradiance and an IAM equation (see Equations 9.17 and 9.18 in the PV reference manual).

The Simple Efficiency Module model does not account for these losses.

SAM reports these losses on the loss diagram as part of the "Module loss." Because of the way the models work, the IAM losses and other losses are not calculated as the model runs. There are ways to estimate the portion of the total module loss that can be attributed to specific factors like incidence angle, spectral losses, temperature effects, etc., but we chose not to try to estimate those contributions and instead just report the total module loss because it more accurately shows actual results from the model.
Best regards,

Paul.

Paul,

Thanks for the thorough explanation.

One more question.

If a client provides a PAN file for a particular module that can only be opened and inspected in PVSyst what model from SAM (Sandia, CEC, etc.) would you recommend to use to translate those PAN file parameters into SAM?

Best,

Nathan

Hi Nathan,
You can use the CEC module model. If you are modeling a module that is not in the library, then you can enter the parameters yourself in the "CEC Performance Model with User Entered Specifications" version of the model.
SAM's CEC module mode is an implementation of the De Soto "Five-parameter Module Model," which is a single-diode model like the PVsyst model. You can see a description of both of those models on the Sandia PVPerformance Modeling Collaborative website. As I mentioned above, a full description of SAM's implementation of the De Soto model is in the PV reference manual.
Thanks,

Paul.

Thanks again Paul.

In the SAM reference manual is there supposed to be some sign between the "20" and "1" for eq. 9.36?

Hi Nathan,
There are a few parenteses missing from that equation.
Please see the errata sheet for the PV reference manual for a correction to that and a few other equations: sam.nrel.gov/performance .
Thank you for pointing out the error.
Thanks,

Paul.