I was finally able to get the time lapse video posted to Youtube.com
This was the full roof installation. Before this was done the guys from Select Solar and Generator hung the inverter and set the wiring from the panels to the inverter and from the inverter to the DTE generation meter, and from the generation meter to the breaker panel.
In case you didn’t see it before this is 42 panels, each with a 230 Watt capacity for a total DC power rating of 9,660 Watts.
Mail this post
Sizing the system will depend on your current usage, the amount of roof space you have, your budget, and how much time you want to spend optimizing the system.
Using the basic numbers from the PVWatts site and from our bill it looked like we would be good asking DTE to approve a 7,000 Watt DC rated installation. A couple of installers confirmed that range should be approved so I knew my baseline at that point.
However, I know the marginal cost of installing a system goes down as the system gets larger. This is a critical point to understand. A number of the quotes I received were derived and justified based on a certain cost per watt. But the actual cost per watt gets lower as the size of the system gets larger. (Think cheaper by the dozen.)
My desire was to utilize the maximum amount of the DTE renewable energy credit which paid $2.40 per DC watt, and also size my system so I wouldn’t need to come back later to add additional capacity. (Which would probably cost a lot more money per watt.)
I also knew that we currently use wood as our primary heat source and that as we get older we might want to use it less. And, we are planning on installing some super efficient mini-split heat pump systems to supply heat and air conditioning in the future.
I did some basic heat energy calculations and determined that our electric energy use could easily increase by 25 to 30% over the coming years as we used less wood and more electricity for heating.
So, I targeted a larger system for my proposal to DTE. I spoke to a couple installers who agreed it made sense to plan this way. The target number I ended up with was a 10,000 watt DC system.
I also have a 3,000 square foot roof so I know that about 1,500 square feet is facing South West. That gave me 1,500 square feet to play with which would turn out to be plenty to install our system.
Note that there are at least two different ways to describe the power capability of a solar system:
1. The maximum possible output of the solar panels themselves. (This is referred to as the DC Watts output or the “Nameplate Watts.”)
2. The maximum possible output of the system. (This is referred to as the AC Watts output.)
So I targeted a 10,000 DC Watt system, knowing that it would rarely if ever be possible to actually get a full 10,000 watts of output from that set of panels because it would take perfect conditions as far as the sun’s location in the sky, the temperature on the roof, shading, and the weather.
I ended up with what Detroit Edison called a “9.66 DC/ 8 AC kW Solar System” on my roof. And, today as I write this there have been a couple minutes where conditions have been good enough to hit the 8 kW level of system output, so I can confirm it is actually working as designed.
Since DTE is currently paying $2.40 per DC watt after installation, this means that when the installation was complete and had met all the approvals, DTE would write me a check for $23,184 to help pay for the system. (And after a few nervous weeks I actually did receive it!)
Mail this post
So we’ve decided to install a grid tied system to take advantage of the utility and government subsidies and not have to deal with batteries.
The next choice is microinverters or strings.
String systems have a number of solar panels connected in series much like a flashlight might have two or three 1.5V batteries connected in series. In series connections the voltages add. So a flashlight with 3 1.5V batteries has a total of 4.5V to the bulb. In a solar installation a set of 13 solar panels might be connected in series. If they each put out about 30 Volts then the sum would be 390V. Many solar panels have about 200W of power output under full sun so that might mean that the panels produce a current of about 7 Amps. (Assuming 210 Watt panels.)
If you need more power than 390 V at 7 Amps, (2,730 Watts), then you need to add additional strings. You can’t practically mix string voltages so this makes it a little difficult to design and build. In our example you have a choice of a 2,730 Watt system or a 5,460 Watt system or a 8,190 Watt system. Trust me, it starts to get a little complicated.
Also with a string system the whole string can be negatively impacted if one panel has a problem. For example if a shadow from a tree is over one panel that will limit the output from the whole string, even if the other panels aren’t shaded.
The alternative that has just been practical for the last few years is having an inverter connected directly to each panel. One inverter per panel. The output of the inverter is the 240VAC that can feed back into your breaker panel. These inverters that mount on the back of the solar panels are called microinverters. The most popular one is from Enphase.
Up to 15 of these can be grouped together on one circuit, and there is basically no limit to how many groups you can have.
This is a very flexible alternative because with microinverters you can start with just one panel and one inverter and build from there.
Again since I have a background in power supply design I found these systems fascinating. There were just two problems:
1. I am concerned about the long term reliability of that many power converters in that extreme environment. Roof temperatures in the summer can get quite high and obviously during the winters they can get quite low. And that temperature swing will be happening every 24 hours for the next twenty/thirty/forty years. (Solar panels can easily last 40 years, however their output will drop some over time.)
2. I wanted the option of doing something else with the power from the panels at some point in the future. (This is a geeky engineer thing most people probably won’t care about.)
The microinverters do cost a bit more on a per Watt basis, but the installation is easier so they end up fairly close in total price.
In the end we decided on a string tie system with three strings of 14 panels.
Mail this post