Thinking About Buying Solar
Buying solar hasn't been on my parents' mind for a while. They recently bought a Tesla, which I will talk about in another post. I wouldn't describe them as environmentalists, though they are certainly concerned about the environment, and are logical, intelligent people - which really should be equal with being concerned about the environment.
However, I recently pushed them to re-examine solar power, something that they had briefly checked out several years ago, before discounting it due to the high up-front cost and brutal (think 35 years) payback period.
But, as occasionally happens, sometimes I come up with good ideas, and regardless if they're good or not, I'm usually good at constructing good arguments, and this time the facts were in my favor.
The Background on Solar Energy
If you want Elon Musk's opinion on buying solar and solar energy, here's a great article from Wait but Why.
Later, we will examine down to the dollar the cost of implementing and buying solar, but first, let's step back and take a look at some interesting maps and statistics. The world receives enough solar energy in an hour to power it for an entire year. An entire year!
Alternatively, for those who are American, PV panels installed on just 0.6% of the area of the United States would power the entire country.
Or, if you prefer it this way, here's a map of the area needed to be covered by solar panels to power the whole world. In total, it's about the size of Spain - not very big at all.
And, in just the last 3 years, the cost of manufacturing solar panel modules have dropped in price by 60 percent, and are predicted to decrease a further 40 percent in the next two years. In the last 35 years, solar has become over 100 times cheaper.
So how cheap is buying solar, then, exactly? Well, based on data from the California Solar Initiative, the average cost per watt of installed solar over the last 12 months (rolling) was:
- for installations <10kW: $5.35 USD
- for installations >10kW: $4.38 USD
Now, in current Canadian dollars that seems a bit high. But, based on the recent quotes we received in Nova Scotia, the cost of an installation is really around $3.67/W CAD, installed. This is a non-tracking, fixed roof mount system using microinverters, which is common for current installations. And this cost varies little on most residential installation sizes, which we will talk about later.
Is Buying Solar Viable?
But is solar viable for those of us who live in less-than-ideal solar climates?
My parents live in Nova Scotia, Canada, which isn't quite Southern California in terms of sunshine. So it's not ideal for solar. But it matters much less than you'd think.
Here's a map of the solar irradiation of the world, North America, and finally, photovoltaic potential in Nova Scotia (here's the Canada map). Individual values of photovoltaic potential given per month (and annual totals), by province and municipality, can be found here in table form.
Not sure what solar irradiation or photovoltaic (PV) potential are? Simply put, solar irradiation is the expected amount of solar energy from the sun that hits the Earth's surface at a particular spot. Photovoltaic potential is the amount of AC energy expected to be generated by a solar panel installed at a given location and orientation (essentially, how much output could be expected from a solar panel installed there).
Still confused? Check the links here and here.
So really, at my parent's location in Nova Scotia, they could only expect just over 1000 kWh/kW annually. Southern California could expect close to double, and even Boston, which is barely south in terms of latitude, could expect about 15% better.
So how could it be worth it?
In most places, it makes the most sense (if your utility allows it), to feed the power you generate back into the energy grid, rather than store it and use it to power your home, or store it and feed it back into the grid at a later time, the most notable exception being if you have time-of-use pricing (I'm looking at you Ontario).
So, the equipment required for a roof installation is really just the panels themselves, micro-inverters for each panel, and the mounting hardware. Micro-inverters are currently one-per-panel, which has the added benefit of letting you view your generation by each individual panel, via a web-based interface (check out some public installations here).
My parents have a sizable roof, half of which faces almost due south, which is definitely lucky. The slope of the roof doesn't actually matter as much as you might think at their latitude, as you can see by the small variation in radiation received based on tilt in Nova Scotia.
The Payback Period
So down to the cost. The upfront cost still isn't small. Buying solar is basically like buying a new car. But instead of having to sink money into this one every week, it will start paying you back. And the new payback period? For an average house: ~15 years. If you consume more than average, quite quickly it can be 10-12 years. After that you'll be making money, and most panels are guaranteed for 20-25 years, with current life expectancy up to 35 or even 40 years.
Keep in mind the price of power varies, and this largely affects the payback period. Don't forget to change the values to match your electricity provider if you use the spreadsheet.
So where does that payback period come from? First of all, there are limited subsidies available for residential solar power in Nova Scotia (check here for Nova Scotia subsidies) (Update May 2016: this link leads to something "coming soon"; the previous link I used was removed). Those elsewhere in Canada can find information about their province here (if you live in Ontario, you need to check these out - you can actually make money), and in the US you can find them here.
But in NS, the current cost is $0.154/kWh (for details about the difference between a kWh and a kW, check here). We will assume that the rate increases at 5% per year, which is consistent with historical rates.
Since the solar panels are sold with micro-inverters for each panel, the size of installation doesn't matter too much, with the only economies of scale coming with installation costs, and as mentioned previously, based on recent quotes, the installed cost per watt (W), including tax, is about $3.67 CAD (or $3.19 CAD pre-tax).
So, if we took a 10kW installation for example, then used the PV potential values we looked at earlier, we get can summarize the savings you might expect in the following spreadsheet (if using yourself, change values in green as needed):
As you can see, the savings are pretty easy to justify if you can afford the up-front cost; payback period using this model is just over 16 years.
If you're a higher-than-average electricity user (like if you have an electric car, or use electric heat, etc.), then the payback period quickly becomes quicker than 15 years.
But, we can still expand the analysis. There are environmental benefits to solar panels too, as the energy you're feeding back into the grid is more energy that isn't being generated using dirtier methods like coal. We can try and translate this environmental effect into dollars and incorporate that into our payback spreadsheet.
Nova Scotia currently uses about 60% coal energy, and have published the historic emission of their systems. Total life-cycle emissions for solar panels are about 25-50 g CO2 equivalent per kWh. Assuming the decrease in carbon dioxide equivalent for NS Power averages the same as the last 9 years (-2.98%), we can estimate the carbon dioxide equivalent output for the coming years. Based on our example, we would be saving about 6,978 kg CO2 equivalent in our first year. The average passenger vehicle emits 411 grams CO2 per mile, so that's the equivalent of driving 27,323 km (16,978 miles) extra per year!
Now, here's where things get a little complicated. There are different models which predict the "social cost of carbon", and based on these values from the US EPA. Canada has used estimates based on these in the past, but the values for different models vary widely.
You can play around with different values in the spreadsheet provided, but for the sake of this example I'll go with Environment Canada's $112.37 CAD value, which in today's dollars is $117.16.
Here's the revised spreadsheet. Now the payback period is only 13 years!
Or, if you prefer, the total carbon savings over the 25 year period examined are the equivalent of driving 478,734 km (297,472 miles)!
Solar Payback Period Including CO2
Now, this analysis isn't perfect. We haven't quantified the savings in particulate and other emissions from fossil-fuel generation. And, conservatively, we assumed that the cost of CO2 equivalent is going to stay the same, when it will almost certainly increase (if only for inflation).
Disaggregated Solar Power
The final argument I'll make for solar is the end-goal of disaggregated power generation.
To understand this, let's first look at the typical power consumption curve for almost any region, and certainly any mildly urban area:
Source: U.S. Energy Information Administration based on data from Independent System Operator New England
As you can see, there's a spike in the morning, and another one in the evening. The goal of time-of-use pricing is to try and flatten this curve; energy is more costly during peak periods, to discourage use, while it's cheaper during off-peak hours to encourage use.
The reasoning behind this is that any utility must be able to generate enough power to supply the peak demand during the most demanding time of year. And rarely do the peak-demand or low-demand times correspond with the greatest generation efficiency. So for a large percentage of the time, the power grid is operating inefficiently.
If, magically, we consumed exactly the same amount of energy, but it was spread evenly over the day, so the power consumption curve was instead a flat line, we could save a huge amount of energy by just cutting out generating plants so that we had a smaller number of plants operating at optimal efficiency.
But there's another way: disaggregated generation. In this scenario, everyone in the neighborhood has solar panels on their roof (or they've all gotten together and invested in a large array - community solar). Not enough to completely power their homes, but enough to be a significant percentage. And now, since they're so forward-thinking, they've installed power storage solutions like Tesla's Powerwall. They probably have a Tesla too, so they can charge their car using solar power.
So all day, their solar panels charge the batteries in their home. Then, at night, when they get home, turn the lights on, start getting supper ready, and generally start consuming more power, the battery in their home starts feeding power back into the grid. And so does their neighbor's. And their whole neighborhood. And their whole city. And all of a sudden, the power consumption curve as seen from the utility's perspective looks like that flat line. And since their utility invested in green energy too, they can match this much lower demand curve with only renewable energy generation.
And all of a sudden, all the power being used in that city is renewable. That's the goal of disaggregated generation.
This is a simplification - your utility will need to have planned well and implemented technology to make disaggregation possible, which is why you should lobby them to invest in smart grid technology. But that's the eventual goal.
The first step is getting your solar panels. And if you can afford them up-front, or can afford to get a loan, buying solar is worth it. Not only that, but your neighbors will be jealous that you're such an environmentally-upstanding citizen.
