Thursday, December 24, 2009

Our solar installation is nearly complete, and with that in mind, I have been looking into some of the more subtle, yet quite important, issues
of photovoltaic installations. Foremost among these is the problem of shade.
Shade can have an impact that is surprisingly disproportional to its apparent coverage. This is because of the construction of a sol
ar panel. A given panel generally consists of dozens of cells connected in series. If one of those cells is shaded, it may not only not contribute to the output of the panel, it may detract from it by becoming a resistive element. The impact of one non-contributing cell of the dozens in a panel can be as much as 50% of the panel output (see these links:,,, and there are lots of others).
There are a number of ways solar panel manufacturers try to limit the effects of shade. Bypass diodes are one of the most common, but it is surprisingly difficult to find much information on how these are employed. (Here is a link to a simplistic explanation.) For example, our CS6p-200p panels have bypass diodes, but none of the documentation states how many are in the panel. These function by passing current when a cell becomes resistant, but when the cell is operating the diode is reverse biased, and therefore resistant. There are also higher-tech solutions, such as SolarMagic®, from
National Semiconductor, which acts as a voltage regulator. It has two benefits: first, to regulate the panel voltage level so that it continues to contribute to the overall output of the installation, and second, to report on the panel performance, which helps to identify when a panel is under-performing.
Shade is a problem for us in the morning, which is when Puerto Rico receives the most reliable sun (afternoons tend to get convection-generated clouds). There is a ridgeline just east of us that has many tall trees. We have trimmed some of these and will have to trim more. By mid-morning, the sun is above the trees.
Finally, we just found out that the law will not allow us to turn on the system until we complete two final steps. These are certifications, one by
the design engineer and the other by the power company. The first is simple and should be done soon (with a slight Christmas delay). The second is hard to predict, as is everything having to do with government agencies. However, if the power company does not respond within 10 days of notification, we have the right to turn on the system ourselves.
This website, in Australia, has some useful tools if you are interested in the benefits of solar installation.
The following are a selection of photographs of our experience.
The pictures are:
1. The SMA Sunny Boy 4000US Inverter
2. The panels from below, showing the mounting
3. The view from the ridge east of our house.

Friday, November 20, 2009

hardware is here!

We have just received 10 200-W solar panels and the 4 kW inverter. I will post photos soon. We are aiming for installation the first week of December.

Thursday, November 12, 2009

tax credits for Puerto Rico solar power installations

Prior to the passage of law 248, the Puerto Rico legislature created the Administración de Asuntos Energéticos (AAE) to monitor and regulate solar electricity installations in Puerto Rico.

I read law 248 of 2008 (in Spanish) last night. There are a few useful points to highlight.
1. The 75% credit was during Puerto Rico fiscal years 2007-2008 and 2008-2009.
2. The 50% credit applies for FYs 2009-2010 and 2010-2011.
3. After FY2010-2011, the tax credit is 25%, and this seems to extend for an indefinite period.
4. The credit applies to both the equipment and installation, although it also applies only to equipment that carries 5 years or more warranty.
5. The equipment must be installed and certified by solar-licensed electricians and engineers.
6. The tax credit can be carried over for up to 10 years, so if for some reason you cannot receive some or all of the credit in one year, you can carry it over to the next, for up to 10 maximum. This may be important, as the total available in tax credits for any given FY is limited.
7. The tax credit can be transferred once.
8. Apparently, the documents that are required for the tax credits must be submitted with one's tax returns for 6 years following the installation.
9. The sales tax (the famous IVU) does not apply to the acquisition of solar electric equipment.

Later, I will detail the process of applying for the tax credit.

finalizing specifications

Choosing a grid-tied system has its costs. The main one is that when the utility power goes down, so does the solar electricity. This is for several reasons. First, "grid tied" means what it says, tied to the grid, not to the house. It serves to mitigate some or all of the use of grid power, but the house still runs off the grid. Second, the house cannot run off power from the panels alone (inverted, of course). This is because of the fact of brownouts caused by clouds passing in front of the sun, or surges in demand. So a storage system is required as a buffer. This consists of a battery bank, which I decided against for reasons detailed below.

Our contractor visited with the solar energy engineer on Tuesday. They measured the estimated average daily production based on the hours of sun that can be expected per day. This measurement consists of first, a device that records the sky exposure of the site. Based on the annually integrated solar track, shadowing from trees, adjacent buildings, even our TV aerial, cloud-cover data for our location, and other parameters, it computes a number that they claim is quite reliable. In our case, we will get an average of 5.4 h/day of production. This comes to 300 kW-h/month, which is just over half of our energy usage.

We will purchase 2 kW in panels, but with a 4 kW inverter, so if we choose to add capacity, it means only additional panels. The panels are 200 W models from Canadian Solar. The inverter is a Sunny Boy 4000US. The reason to forego the batteries that would have given us a standalone system is mostly cost. The batteries are expensive and short-lived. In fact, I calculated that whatever we save in electricity costs, we will spend in replacing battery cells, which have roughly 5-year lifetimes.

Our contractor is PR Green Tech Corp. Julio Correa, the owner, is friendly and attentive, and appears to be quite devoted to his avocation. I hope this will serve us into the future as technologies evolve.

Saturday, October 3, 2009


After most of a year of delay, I have restarted my solar home project.
Yesterday (October 2), I gave a presentation on home solar power at a local parochial school, "Nuestra Señora de Carmen", in Hatillo, PR. I was very impressed by the abilities and interests of this group of 11th graders. They asked a lot of interesting questions, and in this post, I will begin to address them. We start with anticipated power cost savings.

The simple calculation is to estimate how many kilowatt-hours per month my system will produce and how much electrical power will cost over the coming years. We are installing a 2kW system. We can reasonably expect to have full sun on the panels for 6 hours/day, 25 days per month, which adds up to 150 h/month. We currently pay 20¢ per kWh, so this will be worth $30/month. At this rate, we will reduce our electric bill by $360 a year, or $3600 over 10 years, and it will take about 20 years to pay off our ultimate construction cost, which is close to the 25-year 80% efficiency lifetime of the panels. (A total cost of $7200 includes the tax credit we expect to receive from the PR government.) In reality, it may take 25–30 years to pay off the system.

The above calculation makes a big assumption: that electricity costs will not go up. However, only a little over a year ago we were paying close to 30¢ per kWh, and this cost will increase in the coming years. The question is, by how much? This is a big problem to estimate, as electric power costs do not change with any predictable pattern. We can assume that these costs have bottomed out, given the state of the economy, the environment, and of known reserves of petroleum (see this link for estimates of future petroleum costs) (in PR there are currently no other sources of electricity). If electrical power averages 30¢ per kWh over the next 10 years, we will save $5400 rather than $3600.

I will do some more research on electricity costs. In future posts I will touch on changes in our plans, solar-thermal power, and wind power.