Changin' The Rules

So, as I mentioned previously, there are some pretty big changes in the works for the PV industry. In addition to the recent approval of a BIPV acceptance criteria, there are two other major shakeups in progress.

First, the adoption of a new incentive program in California is barrelling ahead.

On the large commercial side (systems over 100 kW), a Performance Based Incentive will be put into place starting January 1st. This is basically the European feed-in tariff model, with solar generation paid $0.39 / kWh. This was legislated in the California Solar Initiative (CSI) managed by the California Public Utilities Commission. It contains triggers that reduce the incentive amount based on the total rebate reservations in the system, so that as the industry grows the rebate shrinks. This seems like a good model, and hopefully will be as successful as the European and Japanese programs that it is based on.

For small commercial and residential retrofit, things don't change much. There is a declining incentive schedule, but it is still a capacity incentive, starting at $2.50 / Watt.

Where things get interesting is the New Solar Homes Partnership (NSHP). Unlike the other rebates, this is managed by the California Energy Commission. While still not fully defined, there is a draft guidebook. This program is for residential new construction, and it is still a "predicted" performance based incentive program with an up-front rebate computed based on system performance modeling.

Some of the issues that I identified earlier have been dealt with. Specifically, there is a provision for "flexible installation" where a blanket rebate can be taken for homes that fall within certain parameters. This reduces the problem with timing deals versus site planning.

One very interesting and positive aspect is that the PV incentives are linked to energy efficiency, allowing builders who implement energy efficient features (better than 30% lower consumption than already stringent Title 24 energy efficiency code requirements) to gain bigger rebates. Intriguingly, it also writes PV systems into Title 24 for the first time, treating them like any other energy-reducing measure.

Overall, however, the program is starting to look quite onerous, especially for BIPV. The way that data is gathered and BIPV is modeled, this technology is hit with a series of small performance penalties that don't exist in the real world. The resulting decrease in incentive is in effect an increase in cost. The program requires a series of inspections, both by the installer and a 3rd party - adding more cost. It will also eventually require testing to international (IEC) standards that are not currently part of the US certification process, something that will be unique to this market segment. Finally, there is fierce debate over a proposal that will require a unique module power rating system for this market segment.

The "dings" that BIPV takes are particularly problematic because many builders are unwilling to install solar, period, unless it is BIPV. By making BIPV more expensive, the simple and obvious end result is that less PV will be installed and the program goals are less likely to be met.

The intention is to protect the consumer, and ensure accurate system ratings, to which I say bravo. However, it seems to me that this is a poor choice of venue to tackle some of these issues.

The problem boils down to two things.

First, what is the acceptable deviation of an individual module from it's "nameplate rating"? As it currently stands, if you go out and buy a 200W module from company X, UL says that the module must have power within +/- 10% of that rating (or the tolerance on the nameplate - typically +/- 5% - whichever is less). It should be noted that PV manufacturers can readily hold tolerances within +/- 5%. So, according to UL, a 200W module that puts out 180W is perfectly OK.

The problem historically is that if you tested a whole bunch of modules, you would probably find that the "average" 200W +/- 10% module was more like a 190W +/- 5% module. You weren't likely to find any 220 W modules!

As buyers have wised up, this problem has diminished because they've insisted that the average power over some (large) sample meet nameplate specifications, while allowing for a reasonable tolerance.

The second issue, once you've defined how much power a module is expected to make under specific test conditions, is how much energy it will produce in real life. That's what people really care about. So, the question becomes how does one properly model PV system performance? This is quite complex, and as always, garbage in garbage out. The model may be properly designed but if the input parameters aren't collected in a well-defined, repeatable, and very well thought-out way there will be problems. Unfortunately, a lot of the testing protocols are effectively being defined in a committee process at the last minute; some of them simply have large inherent uncertainties; and many are fairly expensive and are not currently required. The CEC software model itself has not been publicly released, so no one has a had a chance to "test drive" it and see how well it predicts the performance of existing systems.

Attempting to implement back-door regulatory controls onto a relatively small subset of the global market, in order to steer the PV industry in a certain direction, is not a good idea in my opinion. There are other venues for this that, while slower, will make changes more uniform.

It should also be noted that in really competive markets, like Germany, manufacturers are guaranteeing very tight tolerances (such as -0% / +2.5%) on modules. This is not because of any government regulations but because consumers are educated and the PV manufacturers jockey to position their products in this manner. It seems to me that the issue will take care of itself as the California market heats up.

Nonetheless, changes are afoot at the national level as well. Currently, work is ongoing to "harmonize" the venerable UL 1703 PV Module standard with IEC 61215 and IEC 61646 . These latter standards are not just for safety, as is the case with the UL standard. They are also concerned with qualification and performance. That is, there is testing included which is designed to ensure that the modules perform as advertised, and will continue to perform for 20+ years. Performance will probably not be part of the scope ruled into 1703, but qualification probably will be.

In any event it makes sense to me that rather than requiring early adoption of IEC 61215 in a subset of California's market, that the CEC simply allow the changes of 1703 to take their course and then fill in gaps if required at that time. They will probably have to require a subset of the testing in 61215 just to gather performance data. This shouldn't be a problem for anyone. However, the full qualification sequence is expensive and time consuming.

The bottom line is that things are changing, and on the whole, in the right direction. However, the devil is in the details. Getting solar on new homes is far easier than retrofitting it later, and it would be a shame to see this powerful section of the market gummed up in red tape. Unfortunately, it seems as if that's a likely outcome under the current plan.