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
Q&A Spotlight

Solar Panels or Exterior Foam?

A homeowner weighs the benefits of two possible energy upgrades to a pre-war Cape

Dark side of the house: The north side of the author's house is showing its age, and the peeling paint and mold don't help. Among the questions he faces: Should a new layer of rigid foam be applied over the sheathing when the house is resided? Or should he skip the foam and invest the money in photovoltaic panels instead?
Image Credit: Apollo S

Apollo S has been making steady energy upgrades to his pre-war Cape Cod style house in Massachusetts. He’s replaced a steam heating system with a heat pump, and with help from the state’s energy efficiency program, he air-sealed and insulated his attic with cellulose.

As a result, his $250-a-month energy bills are one-quarter what they used to be, and Apollo now has his eye on the next round of upgrades.

“In a couple of weeks, I am getting a MassSave contractor … to dense-pack all empty walls with cellulose, so we are looking at, what — R-1 walls turning into R-12 or so?” he asks in in a Q&A post at Green Building Advisor. “Two-thirds of the walls are empty in a 3,000-square-foot house, so this should make a dent in my bills. Most windows are new, too.”

The four or five layers of paint on what is probably the original cedar bevel siding are peeling badly, and Apollo assumes that he will have to re-side the house sooner rather than later.

So here’s how his question shapes up: Will he be better off by adding 3 or 4 inches of rigid polyiso insulation over the sheathing when he re-sides, or using the same $20,000 to buy a 7-kilowatt photovoltaic (PV) system for the house and skipping the foam?

That’s the topic for this Q&A Spotlight.

Doing the energy calculations

Software developed by the National Renewable Energy Laboratory, BEopt (a free download that runs on the Windows operating system), would give Apollo some answers. Dana Dorsett’s guess is that an additional 3 inches of polyiso could even get the house into net-zero energy range, depending on factors such as insulation and windows.

But, he adds, BEopt will not be able to calculate two things: comfort and…

GBA Prime

This article is only available to GBA Prime Members

Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.

Start Free Trial


  1. Reid Baldwin | | #1

    I have a question for people that have either moved from a code-min house to a well-insulated house or done a substantial insulation upgrade on their house. Is it like changing from regular TV to HDTV in the sense of not appreciating the difference until you experience the upgrade? Before HDTV, I didn't see any problem with TV resolution. I couldn't understand why someone should pay extra for better resolution. Now, when I see a regular resolution TV, it looks awful. The 1990s code-min house we are moving from had a couple rooms with comfort problems. However, in most of the house, I never noticed being uncomfortable. When we move into the new house that has superior insulation, will I notice the difference or will I just be more likely to notice being uncomfortable in other houses?

  2. GBA Editor
    Martin Holladay | | #2

    Response to Reid Baldwin
    Many variables affect the answer to your question.

    1. Some people are more sensitive to drafts and radiational cooling than others. Sensitive people will be more likely to notice the differences than oblivious people.

    2. These differences are more noticeable in very hot or very cold climates than in mild climates. The differences are especially noticeable on the coldest day of the year or the hottest day of the year -- but not necessarily in April or October.

    3. These differences are more noticeable if the poorly insulated house was a real wreck -- with uninsulated walls, high levels of air leakage, and single-pane windows -- than if the (old) house really did meet minimum code requirements.

  3. Expert Member
    Dana Dorsett | | #3

    Response to Reid Baldwin
    It depends on where you're starting from (Martin's comment #3) , and the meaning of "...a well-insulated house...." and " substantial insulation upgrade".

    Going from circa 1990 code minimum house to a Net Zero Ready house is something that most people would notice more than simply upgrading to IRC 2015 code minimum. One of the things people notice is that they can no longer tell that it's become extra hot or extra cold outside without consulting a thermometer or going outside. Changes in mean radiant temperature indoors with outdoor temperature become much more subtle.

  4. RD3Sunworks | | #4

    Some Thoughts and Responses
    Apollo, make your own decision, based on your own feelings and analysis, without getting bogged down strictly with ROI. As Scott puts it, "better off," what does that mean to you? It's not all about money, particularly when nobody can really pin down ROI anyway..

    I decided to do both--deep energy retrofit AND PV (see my blog here, under Green Homes).

    Your original question was about ROI comparison between PV and additional wall insulation. That question is impossible to answer, to the satisfaction of everyone. There have been several articles on this website that conclude that deep energy retrofits are simply not cost effective at all. That's certainly one perspective. As Dana noted, it's complicated--there are no simple rules. As Peter said, the inadequacy of any sort of payback analysis is frustrating.

    As for me, I am fully convinced that a person can make the ROI numbers say anything they want. ROI is a moving target, at best, and not subject to strict analysis. I agree with Peter--basically, you would need to predict the future values of many variables to calculate ROI. That is not happening, and perhaps ROI determination is even tougher than that, because economics is probably as much a function of political games as it is science.

    Response to Apollo and Peter: The BEopt software is really not difficult to use, at all, and once you define variables, it is quick and easy to change them to see the results on energy usage. I must admit, I totally ignored the financial aspects. As I said, I don't believe them. The program does require some initial time expenditure, to download weather files, to download an energy simulation engine (like Energy Plus) and to set up wall configurations, etc. But, I assure you, it is easy after that and well worth it. I have been able to accurately model my electricity usage before my vast retrofit AND afterward,easily down to a few hundred kWh per year.

    Peter: I would like to hear some detail as to why BEopt was not gratifying for you. As for the difficulty you mentioned, more than one wall configuration, that is pretty simple to simulate, at least to the degree necessary to get good results. In BEopt, you can use a custom wall configuration, including specifying the R value. Although you can only end up with one R value for all the walls, you can estimate individual wall contribution to that R value as in this example:
    Wall configuration "A" is R40 and represents 75% of my total wall area.
    Wall configuration "B" is R20 and represents the other 25% of the total area.
    Then, a usable estimate of my wall R value is: 40(0.75) + 20(0.25) = R35.
    I actually used just such a simple calculation, since my brick veneer front (insulation added inside) is a little different from the rest of my walls.

    One thing I know about ROI is that the sun does not send a bill. More than 50% of my heat is provided by the sun, using passive solar energy. If it is sunny and warmer that about 32F outside during the day, and no colder than the low 20s at night, no extra heat is needed. I am talking about right now, in my area.

    Another thing to consider is: If you are interested in getting to net-zero, can you even fit enough PV on your roof (or yard) to do it without a substantial energy retrofit? The answer is no for very many houses, including mine.

    Dana mentions a really neat point about tight, highly insulated houses. You can really get confused about what it may be like outside, until you go out. I have walked out and could not believe how cold it was, with neighboring heat pumps running all over the place, and my heat not even on. Somehow, the inside heat is also more evenly distributed, naturally. Before my retrofit, it was always much warmer upstairs in the winter than downstairs--I believe that I had been feeling all my heat rushing upstairs on its way out of the badly insulated/air sealed attic. Before, I would always know when it was raining. Now, unless it is pouring, I am often surprised to step out and find that it's raining.

  5. Reid Baldwin | | #5

    Composite wall R-value
    Rick, the formula you used for wall R value is not quite right. If 75% is R40 and 25% is R20, the effective R value is: 1/(0.75/40+0.25/20) = 32. This doesn't change your central point in that paragraph.

  6. Expert Member
    Dana Dorsett | | #6

    Solar PV ROI is really all over the place in MA
    If you build today and qualify for solar renewable energy credit ( SREC) incentives AND net metering AND the 30% federal income tax credits the ROI is extremely good (arguably TOO good), with an insanely high IRR if you get in before some of the incentive doors are slammed.

    But the MA SREC subsidy is already on it's second phase which will only be available up until 1600 Mw of distributed solar is subscribed (and it's already more than halfway there). The first SREC program launched in 2008 was to incentivize getting 400 Mw of solar installed in the state by 2016, a number that was reached in 2013. My best guess is that SREC-II will be done before the end of 2017, and it's not clear if there will ever be an SREC-III (or if it would even be needed), especially if net metering caps get raised significantly.

    The net metering caps have already been reached in many MA locations, and the legislature failed to come up with a plan going forward in the last session of 2015, leaving many projects on hold pending clarification. A haphazard small raise on the net metering going forward to kick the can down the road doesn't appear likely, but the final form of the new incentives plan isn't even in the eyes of competing lobbyists and utility industry stake holders- there's some legislative sausage to be made here.

    It takes a better crystal ball than mine to predict what the ROI of solar will be, since it depends so highly on how incentives and utility regulations will evolve over the next 5-15 years. Figuring out error-bars on the ROI of a high-R wall assembly are quite simple by comparison, since the effects of policy changes & subsidy aren't nearly as large. But that's not to say it's actually simple.

  7. agurkas | | #7

    This last comment isn't

    This last comment isn't helping the PV cause, especially when per watt installed price of panels in 128 belt is ~40 cents per higher than say outside of 495 belt.
    Big driver is to get this house less dependent on the fossil fuels (and in MA we get our electricity from fossils). We have had electric rate double on us from season to season. Same for natural gas.
    Dead siding isn't helping the cause either. It will have to go to the dump one way or another.

  8. Dana1 | | #8

    All things come to those who wait.
    The cost of PV drops ~25% every time the production rates double, and the doubling period in the US is less than 2 years.

    In mature markets like Germany & Australia (not exactly cheap-labor markets) the average installed cost of PV is under $2/watt, using the same racking, panels, and inverters used in the US. The biggest cost difference is in the customer acquisition (advertizing, bidding, explaining, hand-holding) , and the paper shuffle with utilities & regulators adding to calendar time.

    With policy & regulatory support such as streamlining the permitting & inspection etc. the cost of PV inside of 128 will eventually be there too, probably before the federal tax subsidies go away completely in 2022. So if you go ahead and build out to something close to Net Zero Ready now, the PV can be added later.

    In MA it's possible to select different power generators- it's a "decoupled" market. In anticipation of last year's wintertime rate hike I took a 3 year contract for 100% wind power at a fixed 12.9 cents/kwh from a 3rd party broker. The utility's distribution & grid charges remained the same, but it turns out to be about 21-22 cents/kwh year-round, cheaper than the standard "green up" offering for 100% renewables through the utility. The prices and contract terms vary, but if you catch it at the right time you can lock in at a favorable rate, or just let it float with the market. Another possibility is to buy into a community solar virtual net metering, or lease panels from a Yeloha community solar project at a guaranteed minimum net metering:

  9. user-1052275 | | #9

    Whole house R-value calcultions
    Not sure either of the computations presented above are correct. The article linked below is relevant but does not account for the fact that different walls of a house see different delta-t's or that distance makes the assumptions about the fungibility of heat-flows more complicated than the model presented.

  10. Expert Member
    Dana Dorsett | | #10

    A simple 2D parallel path method is good enough.
    Yes, the details are a lot more complex than that. But there's more error factor in the occupant use than the difference between a full-on 3-D simulation w/ solar surface heating and windwashing factored in visa simple 2D parallel path analysis. The actual vs. labeled R-value vary by quite a bit with both temperature and delta-T, and there are phase-change & thermal mass aspects to the materials as well, but taking it to a high level of precision doesn't really matter.

    Focusing on the moss on the trees rather than figuring out the forest isn't very useful. The simple model is good enough to be predictive of both peak and average energy use within 5% under any number of typical conditions.

    Not included in the Reid Baldwin's parallel path analysis is the R-value of the continuous sheathing & siding & interior wallboard, and the interior & exterior air films, all of which are normally included in a wall's U-factor or whole-wall R. Here's a fairly concise 3- pager on the topic:

  11. RD3Sunworks | | #11

    R Value Calculations and Cost of PV
    I standby the R value calculation approach I gave for use in BEopt. There is no need for the approach that Reid suggested (treating the walls something like resistors in parallel in a DC circuit).
    Of course there are also assumptions of equivalent delta T among walls, and many other assumptions. They don't matter in the effort to predict energy usage in a home.

    As far as the cost of residential PV in the US, and why the average cost here is around $3.48, while in Australia its much less, I believe there are other, more likely, reasons than "customer acquisition" or permitting:
    --The US is full of opportunists--get rich fast --take every opportunity to make a buck.
    --The residential population is on a bandwagon--get PV now, regardless of whether or not I have an efficient house to put it on.

    In front of the bandwagon are incentives, most notably the 30% tax credit, making PV extra enticing, and seemingly cheap, a real bargain.

    Enter the US PV installers (at least in some cases and markets), the opportunists. Much of that extra cost to get PV in the US is going straight to profits for the installation company owner. Argue against that if you want, but I have seen the cost of panels, rails, other equipment, labor, permitting, the Australian cost, etc. Take a look in System Advisor Model and get it to add the costs up for you.
    When you add up all those costs, there is one thing left that hasn't been talked about here, and that is
    the profit margin/overhead.

    Bottom line for my PV installation: It was something more than the US average of$3.48/watt. I know what the fixed costs were. There were certainly some costs associated with inefficiencies in the operation of my installer, but at least $1 per watt went straight to the profit of the installation company owner.

  12. Calloused_Thumb | | #12

    Suppose Apollo adds an inch
    Suppose Apollo adds an inch or two (or more) of foam on the outside of the wall in the pic...... won't he be nudged into at least some roofing work as well? There isn't any overhang on the rake to play with. Also, Apollo... What is the original sheathing material? If its pine boards instead of 4x8 sheets, would that let you keep the claps, assuming they're in good condition and you invest the labor? Now that you've read up on lead and already have the right kind of vac, have you tried the

  13. Expert Member
    Dana Dorsett | | #13

    The soft costs of PV (response to Rick Milller)
    There's been a lot of analysis & tracking by RMI and GTM on the installed cost structures of PV in the US compared to other countries. As of 2012 the soft costs broke down something like this:

    In the mean time the inverter and module costs have dropped considerably, and installation costs have been cut nearly in half, making up the majority of the cost reductions down to ~$3.50/watt in the US, but the contractor margin, marketing, and paper shuffle that comprised more than half the cost then haven't fallen as much, and according to SolarCity's analysis, their customer acquisition costs have been flat, or may even gone up slightly since then, even as labor & financing costs have fallen.

    The average Australian customer today can get 5kw of solar installed for less than just the no-labor soft costs (excluding the contractor margin and financing costs) of 5kw in the US. It's not merely feeding contractor margins with subsidies. (And contractors still need to eat to, y'know?) It's a moving target, but there's a lot that goes into the soft cost.

    As of 2015 NREL's benchmarked 5kw system runs $1.40/watt in raw hardware cost (which is more than the average installed cost in Australia as of Q4 2015), 33 cents/watt in labor, and $1.36 to cover all the rest, including contractor margin. I doubt that margin is a buck a watt. See figure ES-1:

    Conservatively the sales & marketing and all other customer acquisition costs are running ~50-75 cents/watt based on the reported cost of the two largest US installers of small scale solar:

    Higher volumes and incremental learning curve improvement will continue to erode the hardware costs, but the soft costs are really still the slower moving elephant in the room.

  14. RD3Sunworks | | #14

    Response to Mark and Dana
    Mark: You asked, won't Apollo be nudged into some roof work to add foam siding? The answer is, it sure looks like it. See my recent blog, Making and Old Tract House Sunnier and More Efficient, under Green Homes on this website. I extended two roofs and its a lot of work, but to me it was worth it. I didn't see any other way to do it, if I wanted 4 more inches of foam. I wouldn't say I was "nudged" into it but, rather, bludgeoned into it, since it is not a fun thing to do.

    Dana: PV contractors need to eat, and they deserve to make a profit, but that profit is a little high,
    right now, I think. I cannot speak for the entire industry, but in the case of my contractor (owner of the PV installation company), I am pretty sure he planned to make $1/watt or more in profit. What I didn't really mention before is that the inefficiencies AND mistakes his people made during my installation did cut into that $1 more than he wanted.

    Think about it, rather right or wrong, profits will be padded whenever possible. With PV so much en vogue, along with the perceived bargain due to incentives,the contractors know they can "get in on some of that," and charge more than they could without the popularity or incentives. It's somewhat similar to the new housing market. When interest rates are high, prices are low. When interest rates are low, prices are higher. Costs haven't changed when prices go up but the sellers want to "get in on some of that."

  15. Expert Member
    Dana Dorsett | | #15

    In vogue? PV? It's hardly trending on most of the US... yet
    The PV party is just getting started! Except for Hawaii the penetration is miniscule, but as prices fall the installation rates will accelerate. There is a huge unserved market that would be cost-effective if net metered and priced at the world market price for PV.

    Sure, contractors too busy to even respond to RFQs will be padding quotes to protect the potential downside and pay for expanding their business. Once the US industry matures to the level of Australia or Germany it will be a more competitive leaner-meaner type of market, but simplifying the paper shuffle still goes a long way toward making a living while offering a low price to the customers, without a fat gross margin.

    For instance, in Germany the deal with utility and government is the same nationwide, using identical forms, and the contractor licencing allows them to self-certify and self-inspect (and screwing that up would risk their license.) A homeowner in Germany can call a few solar contractors for quotes & financing options, pick one, the contractor shows up, installs the system, turns it on and tests it, and mails in the forms to the local utility & necessary government agencies using the standardized forms. DONE! Even the most streamlined (or most loosey-goosey) locations in the US take up a lot more overhead related to permitting & inspection, dealing with multiple governments and utilities to get 'er done.

    That said, I understand the cost of PV in Florida is averaging under the $3/watt mark as of Q4 2015, in a state where the utilities are actively fighting the entry of third party ownership models (which means some of the bigger lower-cost players aren't even doing much there- yet), but that may be including larger scale PV, not 3-10kw residential stuff.

  16. agurkas | | #16

    Some answers
    Going to answer couple questions here:
    1. Clapboard is pine. Had siding specialist look at it. It isn't slavagable, because over year, as the house was expanded, previous owners just took the cheapest route possible with siding.
    2. Sheathing is T&G pine with tar paper over it. Tar paper is so old, you touch it, it crumbles.
    3. Roof extension - I have no choice, it would have to be done, since moment gutters are removed, I have only 3-4 inch overhang.
    4. I am in MA. Our economy here has always been good. Even in downturns people are buying homes, renavating, etc. Trades here are extremely busy and it shows in their pricing and responsiveness. The getting has been good for a while and it will not change. Can't fault them, because if same is in my business, I am not going to lose sleep over being more profitable.

    How about doing insulation and smaller (maybe 5kW, instead of proposed 8kW) PV setup with higher efficiency panels. If I add electric car down the road or see demand is still high, I can just add on to the system later

Log in or become a member to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |