My First Solar Install - Notes and Experiences (Part 2)

During our serious investigations and talks with potential contractors, we were introduced to PVWatts (as mentioned in my last post).  With the rough estimate, I worked out probably yearly and monthly production numbers based on the PVWatts simulation.  Now, I have a more complete and near-final layout in hand.  Time to plug in the numbers!

The Layout

Certain building codes require set-backs.  I didn't know about those when I was first playing with the idea.  You evidently also cannot saw the vent pipes shorter (not that I really expected that to be allowable).

We had to sign off on the splits in the arrays, otherwise all the panels (at least on the east face) would be together.  The split was caused by a vent pipe in an inconvenient spot.  I don't care much about aesthetics, personally - I'd rather have MORE PANELS!

Simulation Configuration

The orientations of each roof is given in red, in degrees.  The 4/12 pitch comes to about 18.4 degrees.  Surprisingly, an extra panel was able to be added to the south-facing roof.

The configuration I used for each roof in PVWatts was:

• DC System Size: 320W * number of panels / 1000 (kW)
• Module Type: Premium
• Array Type: Fixed (roof mount)
• System Losses: 14.08% (default)
• Tilt: 18.4
• Azimuth: azimuth of each roof

Numerical Results

I didn't bother with PVWatt's power cost estimates, since my utility is...a little complicated.  That said, here are the results:

This should provide about 2,600 kWh per month (average) production, 200 kWh more than my original target.  Good thing, since certain people in my household have recently taken to extra-long, extra-hot showers.

Column details:

• AC System Output (kWh): The amount of power the solar system should generate for each month.
• Expected kWh Usage: Based on my 2020 usage, with the last three months extrapolated.
• Production Over/Under: Positive numbers indicate we produced more power than we consumed.
• Credits: If we produced more than we consumed, the provider will roll our overage.
• Energy after Solar: The amount of energy we didn't cover with solar (which is the negative of our Production Over/Under, not ironically).
• Credits Accumulated: Since our provider rolls credits month-to-month, this is the "current balance" for the given month.  Cannot go below zero.
• Energy after Credits: The amount of power not covered by both solar and credits.
• Energy Costs: Cost of our energy usage after everything has been accounted for.
• Costs without Solar: What it would cost us for the power used, without solar input.

Note that the costs are at my provider's current rate schedule, and are not extrapolated beyond the anticipated inputs.  Also, for simplicity this considers a constant rate schedule, whereas my provider tends to change their rates several times during the year.

Whereas most companies will estimate how many hundreds of thousands of dollars you'll save in the future, this table tries to estimate what we might have saved or spent for just the power we used in 2020, with the system as designed.  With energy rates expected to increase at around 4% per year for our provider (YMMV), the "savings" as it were should only go up.

This table does not show how much the system will cost on a monthly basis (we're anticipating around $250/month), so the final Savings number is slightly misleading.  But, if we saved $3,806 on actual energy costs for the year, and spent $3,000 on the loan, we'd be $806 ahead.  That's over three months of payments we could make to pay down the loan quicker (or about 100 sub sandwiches a year from our favorite sandwich shop - my preferred way to calculate money saved and spent).

Heat Losses and Discussion

I had to do a little digging, but PVWatts actually considers performance degradation due to heat.  This is encoded in the Module Type parameter.  Their loss number was a little nicer than my actual panel specs:

• PVWatts power loss due to heat: -0.35% per C
• My panel's power loss due to heat: -0.37% per C

But, with such a small percent, we are not talking a considerable amount.  After all, at +30C operating temperature, we're talking a difference between 10.5% losses and 11.1% (0.6% off).  Just roughly, that amounts to 1 kWh less production for the year, if I did the math right.  Bear in mind that a lot of marketing for panels, inverters, etc, will tout efficiency improvements around 1-2%.  At the end of the day, what matters is how much power it produces, and whether or not that power is worth the cost of those percentage points.

So, let's call that "margin of error," especially since this tool is also using historical weather data from a location at least an hour away to drive its simulation.

If things go according to plan, we should see an insufficient production for the first two months, but then zero energy costs (aside from the monthly fee) through to the end of the year.  If we're able to run the house at a lower consumption rate, that will just mean a better buyback at the end of the year.  If panel performance isn't as good as we expect, we'll (at worse, hopefully) be paying upwards of $300 for energy for the year.  That amount is based on double the system losses due to heat (which I added in before realizing PVWatts already accounted for it).

Guess we'll see how it goes!