Sunday, April 1, 2012

What's in a Solar Production Estimate?

A potential customer contacted me a couple of weeks ago. She lives in a town with a municipal utility. She asked me about a solar array on her house. The first thing I did was pull up the satellite image.

From the satellite image, you can see that the roof is nearly east/west facing. I know, from using PVWatts, that east/west facing roofs are low solar energy capture roofs.

Based on PVWatts, east/west roofs with a 6 pitch (26.6 degrees) or lower produce between 80 and 84% of an ideal roof. Roofs of 7 pitch (30.3 degrees) and above produce at below 80% of an ideal roof. It's important to note, that these PVWatts production readings DO NOT factor in the production loss due to shade. (The partial table below was built using PVWatts and shows the easterly solar production relative to the ideal roof in the Worcester MA area.)

To understand the impact of shade requires a site visit and use of a shade evaluation tool such as Solar Pathfinder (~$300 + camera), Wiley ASSET (~$500, includes camera), or Solmetric SunEye (~$2,000).

Each of the devices listed above works slightly differently but all will give you roughly the same answer.  (The main difference between the SunEye and the other tools is speed of result. The Pathfinder and ASSET require you to analyze your digital photos on a computer. The SunEye does the analysis on board.)

In all cases the devices start with roof pitch (angle) and heading (azimuth) information. Why? Because they will all apply production adjustments from PVWatts then subtract out the shade impact.

Ultimately we are looking for the site's relative production when compared to an ideal site. An ideal Massachusetts site faces true south (azimuth 180) and has a tilt of around 40 degrees (42.5 if you ignore weather, 37 if you don't). In Massachusetts, the CEC (solar rebate organization for customers of National Grid, NStar, Unitil, and Western Mass Electric) will not approve a rebate unless the site performs at 80% or better of an ideal site. This makes sense because they don't want to fund a site that will perform poorly. (Unfortunately, the municipal utilities do not have the same requirements or oversight.)

Now, back to my customer above.... After I told her I thought the roof would perform poorly, she forward me an email from another solar installer (I don't know which one).

Here is what that installer said (un-edited by me except removal of some potential identifying information)
First to start with your shading numbers I got from the Solmetric Suneye.  The day I was there, the shading readings I got were between 90-94% on either side of the roof.  In the quote I sent you, I used the lower reading of 90% just to be conservative.  Now to understand how we came up with the estimated output I'm providing what factors go into it and the industry standard for calculating it.

1.  We input the Azimuth of each roof as well as the pitch angle into the quote model
2.  We take a conservative shading percentage based on Solmetric suneye (in your case 90%)
3.  We plug in the modules and inverters being used in estimate
4.  The quote model the uses PVWatts to estimate your output.  PVWatts is an industry wide tool used to estimate solar output.  It takes 30 years of weather history to make the calculations.  
While the process described above appears sound, by my standards, something is missing. The solar installer never told the customer how well the roof would actually perform. They say that they put the angle and azimuth into their quoting model along with the 90% shade and from that information, they calculate an annual production (which they then use to estimate electricity savings and economic return). But they never say how good or bad the roof is. In fact, they leave the impression that the roof is a 90% roof.

So how good is her roof? Since the back of the house has a heading of about 97.5 degrees and the roof pitch is 35 degrees (I visited the site and measured the roof), we can expect the site, AT BEST, to perform at 80 to 81% of an ideal roof. If the shade reading was 94% (the installer's best estimate), then the production is 76% of an ideal roof. If the shade is 90%, than the production is 73% of an ideal roof. In either case, the site is well below the standard set by the Mass CEC.

While the customer is free to do what they want, solar is a big investment. I think they should be told that their roof is far less than ideal. They should also know that the low production level will dramatically extend the break-even time for the array.