GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

Would a $5,000 investment in rooftop solar make economic sense?

jamestimmerberg | Posted in General Questions on

I’m building an all-electric home. I am currently paying 7.22 cents per KWH. I’m poor, but could possibly budget $5,000 for solar power, if 30% of that $5,000 is returned to me as a tax credit. Would investing that $5,000 in solar power make sense? (This question is premised on the cost to me, after the tax rebate, being $3,500.) Or, would my $5,000 be eaten up by installation costs, and leave me generating enough electricity to operate an LED light fixture? It doesn’t matter what I could do for $15,000 or $20,000. I don’t have that much money to spend on solar power, unless I want to live in a garden shed.

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.


  1. STEPHEN SHEEHY | | #1

    Five thousand dollars won't buy much pv. I'd call a local pv companyand check, but at 3.50 per watt for a typical system, you'd probably only buy a kw, maybe 1.25 kw. Given the need for an inverter, and the lack of any economy of scale, you might do much worse.
    Depending on where you live, a 1.25 kw array produces around 1,500 kwh/ year. At your rate of 7ยข, you'll get about $100 worth of electricity.

    If i were you, I'd invest the $5,000 in a better building envelope. You might give some thought to providing for a future pv when you build the house, by thinking about how you'd run cables, etc.

  2. charlie_sullivan | | #2

    7.22 cents per kWh is pretty low, so it's unlikely to be a really good investment for you. But how many kWh you get for 1.5 kW of solar depends on how sunny your climate is. If you are in a really good sun location, and you can find an installer who gives you a really good price, it might work out to be a reasonable investment.

  3. GBA Editor
    Martin Holladay | | #3

    If your $3,500 investment yields $105 per year, that's 3% interest. There are worse investments these days.

    Your state might have other incentives -- for example, SRECs (a type of credit for generating electricity from a renewable power source) -- that sweeten the deal. Another option: a no-money down PV lease that offers you electricity for less than the going retail rate. That deal isn't available in every state -- but you never know until you check. Call up a few PV installation companies in your area for more information.

  4. jamestimmerberg | | #4

    Great information. That's exactly what I wanted to know. Thanks guys. I'm in northern Ohio, where the sun behaves like a fugitive on the FBI's 10 most wanted list, and the government believes that climate change is a liberal hoax. Plus, the $5,000 would be tacked on to my mortgage, on which I'll be paying 4% interest, so the return on my $3,500 solar power investment would be negative.

    What an amazing web site. I post a complex question (for me), and 5 minutes later, I have a comprehensive answer. So many people volunteer so much great information, it's astounding.

  5. Expert Member
    Dana Dorsett | | #5

    It really depends on the other financial incentives (if any) are available in your area. The lifecycle cost of power from small scale PV is typically more than 2x the 7.22 cents you're currently paying for retail electricity, even after the 30% tax rebate. But in some low-cost/high-sun areas like Florida (where small scale solar costs about $2.50/watt, installed) it could be close to break-even at 7.22 cents after tax incentives. That may not necessarily be a great investment compared to other things you might spend the $5K on though. Efficiency measures usually have a much better return on investment, but they have to be measured & weighed.

    Here a sunny-location low-cost levelized cost analysis:

    Take a 20 year analysis with a 3% discount rate, with 2%/annum price inflation from the grid, a 22% capacity factor on the PV (more like Arizona than Florida), and $2500/kilowatt-DC installed price the 20 year cost of electricity is comparable to grid power, assuming the power company isn't allowed to nail you with a high fixed monthly fee or demand charges the LCOE of grid power comes out to about 8.7 cents/kwh over the 20 years, the PV comes in at 8.9 cents/kwh, slightly more expensive.

    If you take the 30% tax break the $2500/kw becomes $1750, and the PV comes in at an LCOE of 6.3 cents.

    If you take the tax break and assume a 5% discount rate the PV comes in at an LCOE of 7.5 cents, still cheaper than grid power over the 20 year term.

    With the tax credit , 5% discount rate, and only a 20% capacity factor (Miami/ southern FL, fixed tilt racking) the cost of PV is about 8.2 cents, still slightly less.

    At the US average of about 18% capacity factor (Chicago, Boston) the LCOE is 9.1 cents, which is more expensive than grid-retail at the presumed 2% price inflation for grid power.

    But that's a VERY sunny place, and a Florida-low type cost. The average cost of rooftop solar in the US is about $3.50/watt, and the capacity factors (basically a measure of how sunny your location is), lower than 20%.

    Play around with this calculator a bit, see where it might come in:

    A few random locations for approximate PV capacity factors:

    But if your state offers compensation via solar production credits or other incentives to sweeten the pot it can tip the balance pretty quickly.

  6. Expert Member
    Dana Dorsett | | #6

    At $3.5/watt installed cost (the US average) less the 30% tax break, using a 4% discount rate (your cost of money) and an 18% capacity factor (about right for northern OH) you're looking at 11.7 cents/kwh LCOE on a 20 year analysis.

    The cost of PV is still falling fast, but it doesn't sound like it's the best place to park your $5K this week. Stay tuned. Two-fifty PV is happening in Florida right now, it's under $2/watt in mature markets like Australia and Germany. Buck-fifty PV is a realistic world price average possibility before 2020, at which point it would be price-competitive with grid power even without the tax incentives, at about 7.2 cents a 4% discount and 18% capacity factor. That's your current retail price. The tax incentives don't step back until after 2020, and with that incentive the buck-fifty installed cost boundary can happen sooner.

  7. jamestimmerberg | | #7

    Given Dana's premise that it might make more sense to invest in solar power 3 years from now, if I am building a new house, what simple, inexpensive steps should I take to prepare for a possible future solar installation, if any?

  8. STEPHEN SHEEHY | | #8

    James- I'd take my drawings to a solar installer and explain that you plan on pv in the future and ask for what can be done to make the installation easier. Talk to your electrician as well. A lot will depend on the layout of the house. You'll need space on the roof, a place for the inverter, maybe a chase for wiring.

    When we were planning my pv installation, the installer put in the box that takes the wiring from the panels on the roof into the building and ran the wiring internally over to where the inverter would go. Then, after the roof was on and the electrical panel was wired, they came back and installed the panels.

    One benefit to waiting is you'll have an idea of how much electricity you use and thus how much pv you need if you aim for net zero.

  9. Expert Member
    Dana Dorsett | | #9

    Design & orient the house ridge line that runs east-west, with a simple roofline gabled roof, or at least no dormers or other interruptions on the south facing pitch to optimize the available real estate suitable for solar is a good start. The angle of the southerly pitch hardly matters from a solar-geometry point of view, anything from 2:12 to 12:12 would have pretty good annual yield. But angles of 4:12 or lower it may hang on to snow for weeks, whereas at 6:12 or higher it will usually glide off on the first above freezing day, improving the mid-winter uptake a bit. In colder climates than yours it can take an even higher pitch to shed snow well, but in northern OH a 6:12 roof should be pretty good. The closer you are to heavy lake effect snows the more important this factor becomes- Ashtabula is more susceptible than Akron.

    If it's a lightweight trussed roof design, make sure that the truss spacing and top chords have sufficient capacity for the additional 3-4lbs per square foot dead load weight + wind loading of a PV array. Beefing up load capacity later is more expensive than doing it now. Most raftered roof designs would already have sufficient margin but when in doubt, reduce the spacing to 16" o.c. &/or go one lumber size larger than a low-load code min roof, eg: If a generic roof design specifies 2x6 rafters 24" o.c. go with 2x6 16" o.c. or 2x8 24" o.c. , or have the architect/ engineer just do the math and verify that the original design still works with up to ~4lbs per square foot of PV panel & racking (usually does.)

  10. user-2453173 | | #10

    Dana has it right.

    Also, if you get the roof trusses pre-engineered, ask them to include 3 psf additional dead load for a future solar PV array. But the most important part of that is to ask that they indicate that the additional capacity was added FOR SOLAR on the actual truss drawings that have their engineering seal. Having that with an existing seal can save you an additional $500-1,000 when the time comes for the array. Otherwise you may need to have the truss company create a new signed/sealed engineering letter for your permitting department, and they love to charge a few hundred to run it through their software a second time.

    The solar company I work for services your area. So here is what i'd tell you to do if you want to get "solar ready".
    - Install a 3/4" EMT (it needs to be metallic inside the house it its going to be DC, and since you don't know what the right type of system for you is yet, it's better to be ready for AC or DC in the house) conduit from your attic to an area right next to your breaker box. Make sure the run is pull-able (less than 270 degree of bend between pull boxes, and leave a pull rope in the conduit for the installer)
    - Leave a solid 3' wide open area on the wall next to your breaker box.
    - If you are installing shingles, have the roofer install a few permanent roof anchors at the top of the roof, they are not expensive, and will allow for the installer to work on your roof without any extra holes for temporary anchors, it will also lower their labor number if they know you made their life easier.
    - Try to keep the roof pitch between 4/12 and 7/12. lower than 4/12 with shingles and your roof is more prone to ice dams and other roof leaks (totally separate from the solar install, shingle roofs leak more when they are low slope) and 8/12 and higher become harder to work on and the solar labor number goes up after you get to 8/12. The ideal year round roof pitch for norther Ohio is usually 8/12, but the difference in production between 4/12 and 8/12 is less than 2% in most cases, the labor increase for 8/12 will overpower the extra production given today's prices. The snow sticking to the array is really not a significant factor in most of the state, because the shallower the tile, the less energy you get in the winter due to the solar angle alone. Adding the losses from more snow coverage really doesn't move the needle. Unless you live in a snow belt, then it's a good idea to go on the steep side, but that's the exception rather than the rule.

  11. user-2453173 | | #11


    Latitude +15 only makes sense for off grid sites in most reigons. When grid-tied, the best solar production is from a lower tilt angle, around 30-35 degrees holds true from Yuma, AZ to Cleveland, OH. Bias to the summer. Since winters in that area are very cloudy, you need to ignore winter time production. If you run the PVwatts simulations, you'll see what i mean. The latitude +15 is an old rule of thumb and is misleading for grid-tie systems.

    Off-griders need the extra power in the winter, and they have waste energy in the summer, so it makes sense that you really push for winter production in those cases if it's practical. Hopefully Martin can confirm that. I just design these systems, Martin lives with one.

  12. RD3Sunworks | | #12


    I think you are close to being correct, for PV, if maximizing annual yield in Ohio is the goal. I mixed it up with my old study of solar collectors,for which collecting maximum heat in the winter is often important.

    Regardless of the ideal tilt, most important to me is that if I would have had a workable ground site, there is no way I would have put PV on my roof at all--it would be ground mounted. It sounds like it is even less desirable to put PV on a roof in your/James' area (northern Ohio) than it is in mine. You didn't actually state it specifically, but you implied that a roof loading analysis is part of the permitting process in that area. For other people possibly reading all this, roof loading analysis is not required in all local jurisdictions and PV is so light weight that there is usually no need for it. It is not required in my area (other ridiculous things are).

  13. STEPHEN SHEEHY | | #13

    Ground v. roof mount:
    We're in Maine. I looked into a ground mounted system, but the numbers just didn't work. It was much cheaper to add more PV to the roof than install a tracking system on the ground to end up with equivalent production. Even a fixed ground mount was more expensive than the roof option. We had plenty of room on the ground, but didn't really want a big array in our front yard, although we'd have gone with it if it weren't so expensive. The roof mounted system is fine, although the snow does stick for a while on both the 8/12 and 4/12 roofs. A ground mount may be the only viable option for some houses without a decent amount of south facing roof.

  14. STEPHEN SHEEHY | | #14

    Rick-I think you misunderstood my comment. Perhaps I wasn't clear.

    For my new house, I looked at three options for PV. I needed about 8000 kwh/year.

    1. PV on roof
    2. PV on a fixed ground mount
    3. PV on a ground mount that would track the sun

    Option 1 was the cheapest system that would get us the 8000 kwh/year. The tracking system would produce more kwh per installed wattage, but at such a premium price that it was cheaper to simply use more panels on the roof to match the output of the tracking system.

    The fixed ground mount was cheaper than the tracking system, but still more costly than the roof system. The benefit to the ground mount being set at a more beneficial angle than on our roofs was surprisingly small.

    My understanding is that the hardware for ground mounts is more expensive than for roof mounted systems. I don't know why that is the case. I suppose one needs a crane for erection, maybe a wide load permit for transportation, or?? The motors and hardware for the tracking system are clearly more complicated and costly.

    The only drawback I can see from roof mount is the snow issue and it's pretty minor.

    Appearance wasn't the determining factor. But on our house, we'd have needed to put a ground mount in a highly visible location. Our roof that faces south is on the back of the house. The land slopes down steeply on the south side, so the panels are almost invisible for anyone not in an airplane.

  15. RD3Sunworks | | #15

    Response to Stephen:
    Your post is a little flawed and misleading. You make a cost comparison based on adding PV to an existing site with roof PV AND including tracking for the ground mount. It is misleading because James was asking about preparations for PV at a new home site. Also, a fixed ground-mounted array will produce as much as a fixed roof-mounted array, all other considerations being equal. Your post just confuses things in that you are talking about overall production for a mixed system, where ground mounted PV is added to an existing roof system. No, I would not usually recommend that. From scratch, depending on the specific circumstances, a ground-mount could certainly be a little more
    expensive than the roof mount, but I think most people would end up saying that it was a small price to pay to avoid the many negative aspects of roof mounting.

    I stand by my position, for the reasons given and more, that a ground mounted array is better over the long term than a roof-mounted array, if the owner has the site to accommodate the ground mount. Now, certainly, very many people do not have a good site for ground mount, and are not starting from scratch with no home, but James asked for advice for a new site, There are untold thousands of us out here that retrofitted PV, without any say over several important siting variables. If someone is starting from scratch, then optimize every variable, including using a ground-mount, if possible.

  16. RD3Sunworks | | #16

    First, there's really no predicting whether or not it makes more sense to wait 3 years for PV. The PV world is changing in a number of ways---but if you don't have the money now, anyway, it's academic.

    The orientation of a simple, preferably single roof is important, like already mentioned. Also (if you must put PV on the roof rather than a ground mount) a standing seam metal roof would be nice, rather than shingles.

    Speaking from experience, here is some other advice:
    ---if you have the land, keep it off the roof. Advantages are: A lot of roof penetrations are needed for
    roof installation--sooner or later, things could leak. It might be easier to orient the array to solar south with a ground mount. Ground mount can easily be set for the optimum tilt, whereas roof pitches are often a compromise. Easier to clean snow with a ground mount--trust me. Easier to clean and inspect any time of year. Probably cheaper to insure than roof mount--not sure about that. Easier conduit runs, typically.

    I have 6-12 roof and the recent blizzard snow in the east did NOT slide off in the first few warmer days.

    --Plan to keep shadows, like trees etc., well away from the array. It is easy to model your house and shadows using Sketchup, with the shadows being very accurate throughout the year, as long as you have everything drawn to scale and oriented with respect to solar directions.
    --Ideally, you would want also to perform an energy analysis of your proposed house, and coordinate the size of the array--similar to what Stephen said.
    ---If you put PV on the roof, check local fire code requirements. They can vary a lot. You may not have as much roof space as you initially think--some localities require as much as 3 feet offsets on the sides and top of an array, which the fire code calls "ventilation opportunities."

  17. RD3Sunworks | | #17

    Response to Stephen:

    Yes, as I said, the ground mount hardware is, typically, more expensive than for roof mount. There is more of it, usually, and a little more erection labor. But, that labor does not have as many possible associated safety concerns. There are also footers and some site preparation involved.

    But there are more drawbacks to roof mounting than just snow clearing. In addition to the other drawbacks I have already mentioned earlier, with a ground mount there is less chance of getting hurt or damaging panels (or the roof, obviously) when cleaning them or inspecting them any time of year. Another concern, unless perhaps you have a metal roof, is that panels may well last longer than shingled roofs. One installer that I know about will gladly charge a heavy fee for taking your array back off the roof, in case you end up needing a new one.

  18. Chaubenee | | #18

    I intend to carefully design a south-facing grilling pavilion/tool shed with a shed roof at the optimal pitch My solar guy/gal instructs me to build, in conjunction with my architect. It will do double duty as utility/party/storage space and a place to mount my solar panels when I am set to go. I am even running the conduit out to it that connects to my service area where there is mucho room left for the needed controls. Not only is there room in this potential design for picnic tables, grills on the north side, but plenty of space for a luxurious tool shed/potting shed and lawn tractor storage. If you have land, think about that. And God forbid if the penetrations cause roof leakage, we aren't talking end of the world.

Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |