GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Audio Play Icon Headphones Icon Plus Icon Minus Icon Check Icon Print Icon Picture icon Single Arrow Icon Double Arrow Icon Hamburger Icon TV Icon Close Icon Sorted Hamburger/Search Icon
Energy Solutions

Sometimes, It’s Cheaper to Install PV Than More Insulation

Before you decide to add more insulation to your attic, compare the cost of the extra insulation with the cost of supplying the necessary heating or cooling with solar electricity

Image 1 of 2
Three photovoltaic (PV) panels that were recently installed on our barn roof.
Image Credit: All photos: Alex Wilson
Three photovoltaic (PV) panels that were recently installed on our barn roof.
Image Credit: All photos: Alex Wilson
We used a combination of cork insulation and spray-applied fiberglass to achieve about R-45 in our walls and about R-60 in our roof.

There’s an age-old question of how much insulation to install in our homes. Conventional wisdom says to add more until the “payback” for the added insulation isn’t worth it — until the energy savings that will result from the insulation doesn’t pay back the cost of that insulation quickly enough.

Energy and environmental consultant Andy Shapiro, of Energy Balance, Inc. in Montpelier, suggests a different approach: basing that decision on the cost of a solar electric system.

Energy conservation and the cost of solar

Andy argues that once we get to very high levels of insulation, it doesn’t make sense to spend more on energy conservation than it would cost to supply that saved heat (or cooling) with a photovoltaic (PV) system used in an air-source heat pump. Air-source heat pumps (often referred to as minisplits) are the heating and cooling system of choice today for many highly efficient homes; they offer two to three times the efficiency of standard baseboard-electric heating systems. Using PV as the benchmark makes sense, because — like conservation — after the up-front investment, there is little to no operating cost.

To illustrate this point, Andy evaluated a 1,000 square-foot roof insulated to either R-60 or R-80. In the 7,700 degree-day climate of Burlington, Vermont, the R-60 roof results in a heat load of 390 kilowatt-hours per year (kWh/yr) or 3.1 million Btus per year (MMBtu/yr), or vs. 290 kWh/yr or 2.3 MMBtu for the R-80 roof. In this analysis I’ll mostly use kilowatt-hours (kWh) as the measure of both thermal and electrical energy, as is common in most of the world; Btus (British Thermal Units) are unique to the U.S.

The savings from providing the extra R-20 in the ceiling is 98 kWh/yr. Andy assumed that the cost of the PV system is $4 per peak-watt ($4,000/kWh-peak) without any tax credits or other incentives, and he assumed that a PV system in Burlington’s relatively cloudy climate will generate 1,100 kWh/yr for every peak kW of rated capacity, while the air-source heat pump has an assumed coefficient of performance (COP) of 2.3.

Given these assumptions — which are certainly up for debate — providing 98 kWh/yr of heat will require 0.089 rated kW of a PV system (98 ÷ 1,100). At $4 per installed peak-watt, the cost of that PV system would be $356, or $0.36/ft2 of roof. With this analysis, adding the extra R-20 to the roof will make sense as long as it costs less than $357. In reality, such a change would cost more like $750, or $0.75/ft2 (assuming loose-fill cellulose and just the cost of the insulation). In other words, it makes better economic sense, in this example with these assumptions, to stick with the lower R-value (R-60) and invest in the PV capacity.

Using investment in PV as a benchmark for conservation investments

I like this approach for figuring out how much we should spend on energy conservation. It could be used not only to evaluate investments in insulation, but also investments in air tightness and some pieces of equipment, such as a heat-recovery ventilators (HRVs). Andy’s calculations assume no tax credits, rebates, or other incentives for PV; with such incentives in place (as is currently the case), the argument is even stronger.

One thing the analysis does not account for is the fact that investments in insulation should continue paying off for a very long time (maybe even a few hundred years if the house is well-built and the insulation protected from damage), while a PV system will need to be repaired and periodically replaced during the life of the insulation. This analysis does not address lifetime costs of PV and insulation; doing so would require an assumption regarding the discount rate and an estimate of future maintenance costs.

My friend Dave Timmons, Ph.D., who is working on models of renewable energy economics and who teaches ecological economics at the University of Massachusetts, notes that for electricity there is a formula for the levelized cost of energy (LCOE), and he suggests that one could develop an analogous calculation for the levelized cost of conservation, so that we’re comparing apples to apples. (But I’ll have to leave that to the economists who are a lot smarter than me.)

Dave also points out that the analysis doesn’t account for the cost of electricity storage. Producers of PV electricity today are able to use the grid as a storage system, but that may change as renewables begin accounting for a larger percentage of electricity production. Viable storage in the grid may increase the assumption we should use for PV cost.

What about with lower insulation levels?

I’ve used this argument for deciding between really high levels of insulation: R-60 vs. R-80. How does it work when applied to insulation levels most builders are using?

If we are considering boosting attic R-values from R-19 to R-38 (also an increase of about R-20)—the economic argument for investing in conservation is far different. In this case, the savings in heat would be 620 kWh/year to go with the additional insulation and the cost of PV needed to deliver this heat would be $2,250, or $2.25/ft2. Clearly, the extra insulation, at $750 ($0.75/ft2), is a better investment.

This second example illustrates the argument I’ve long made that it makes sense to invest in energy conservation first and only after that put in the PV system. But if you go far enough with conservation, as Andy argues, you eventually reach a point where it doesn’t make economic sense to invest in he additional insulation.

Does this reasoning make sense? I’d be interested in your thoughts.

Alex is founder of BuildingGreen, Inc. and executive editor of Environmental Building News. In 2012 he founded the Resilient Design Institute. To keep up with Alex’s latest articles and musings, you can sign up for his Twitter feed.


  1. Skip Harris | | #1

    But of course!
    One might also consider that, given two investments that are each guaranteed to pay off, one would first fully fund the one that pays best, then continue with lower value investments until one runs out of capital: as long as you have enough money and you get a decent payback one should invest in both PV and insulation....and stop when the payback becomes unattractive.

    On the other hand, I can not agree strongly enough with your economist friend that we need to consider the levelized costs and benefits. I've seen too many examples of folks buying very expensive refrigerators instead of PV because of the savings in capital costs, ignoring the far lower probable useful lifespan of the refrigerator. Same goes for heat pumps: one must look at the likely total costs rather than just capital costs. Would you buy an apartment building for the rents and ignore your upkeep and other expenses?

  2. Jin Kazama | | #2

    still relative to payback definition...
    I am sure we all agree that it is a priority to calculate investment and paybacks on every move concerning buildings as they are all economically tied to owners.
    This is all still economy, and then some energy/green .

    It is a very different thing to consider economic payback for the current ( or near future ) owner(s)
    to working toward green building or a conjunction of both.

    PV at current price are very attractive, even more if you diy some or if you have access to incentives,
    but they are still electrical devices with a limited year if you compare them to insulation materials.
    Insulation materials will continue to pay in 25, 50 maybe even 100 years ...
    This is only good though when one considers "green" in his calculations.

    And as you pointed out and was discussed in prior threads on the subject,
    with current HP , we are seeing COP of 2-3 ..will probably go at 3-4 in the near future.
    HVAC is not to be compared in longevity, as it is a must , the same cannot be said about PV.

    It is fairly easy to understand how insulation can be added up until the returns are ridiculously low.
    We all need to take into account that adding insulation doesn't have the same cost on different situation ( construction type ), different cost VS depth or R value required ( probably is much more expensive to add R20 to an existing R40 wall than the first 2 R20 because of the required thickness of the wall assembly with current technology ( this could change with new materials soon though )
    so the cost of adding insulation is not linear, and this has proved to be hard to input into calculations on design stage without prior installation experience.

    I personally like to use between 10 and 20 years for economical payback ( insulation should be more in the 20-30 years ) but it is very hard to convince the one that is paying that it will only payback within 20 years or more ..people tend to think in 5-10 years ...further than that is pushing their time clocks a little bit.

  3. Edward Mallett | | #3

    It makes perfect sense
    In econo-speak its called 'diminishing marginal returns' and it applies to all kinds of investment decisions. Your analysis would be helped by one or two more simple calculations to determine the shape of the curve and where precisely the PV benefit will intersect with a beginning R-value. Eyeballing it, I suspect near R-45 is the point at which an extra R-20 insulation benefit cancels out the cost of supplemental PV.
    While I agree with the idea of levelized costs, it would be tough in practice to take into account all the finicky variables, including technological change, regional product and installation cost differences, installation quality, discount rate preferences, future costs of energy etc. I have been grappling though these on my own Meaford ON build. My site is ideal for passive solar design and there is no NG service. I have already mentally migrated from ground-source radiant to minisplit, but PV is creeping into my thought process.

  4. Jin Kazama | | #4

    Edward : if the current fails
    Edward : if the current fails rate of the PV industry gets "fixed" in the few coming years,
    and prices continue to work toward what they should be ( 2-4$ installed per watt )
    you will have no other choice to move into PV on existing buildings,
    as adding insulation on a finished building is much more expensive than at the design/building stage.

  5. Mike Eliason | | #5

    a couple of thoughts...

    a couple of thoughts...

    1. i'm horrible at maths, but the IEA calculator says that 3.1 MBTU = 909 kWh/year and 2.3 MBTU = 674 kWh/a - which would reflect the heat load for the roof, no? (e.g. PRIOR to selecting heating equipment). dividing that number by 2.3 (ASHP's COP) does give the estimated consumption of heating equipment.

    2. increased insulation could reduce heating/cooling equipment costs, so just looking at problem this way can be too simplistic.

    3. speaking of cooling, while burlington has a relatively low ~500 CDDs, other places have significantly higher and should also be evaluated.

    4. lifecycles costs not taken into account (what about NPV, IRR calcs?)

    5. tax incentives for insulation changes results.

    6. carbon tax changes results.

  6. Expert Member
    Dana Dorsett | | #6

    Many moving targets...
    The spot price of energy, the LCOE of PV and efficiency of mechanical systems are all moving targets, that all get skewed even further by subsidies, and calculations of pre-tax dollar investment for post-taxed energy cost savings, etc. There's no simple model that works in advance- it's all moving, making the decisions more difficult at purchase time.

    Estimates of future efficiency & costs of the replacement mechanicals or PV at the end of a 25-40 year lifecycle are also going to be a WAG at best. Yes there are trend curves one might try to project, but there are also quantum-shifts on both costs and efficiency that don't necessarily fit the curve. And the future compensation for power uploaded/down loaded from the grid for small distributed power generation in 25 years aren't very likely to follow the simple rough-justice "run the meter backward" on a fixed per-kwh price. (Witness the more nuanced calculations in motion currently in Austin TX: ) Electricity markets and local grids are looking at a tsunami of distributed power generation coming within the lifespan of a PV array, or even the grid-tied inverter- this isn't 1990 anymore!

    So, we pays our money, takes our chances- the best calculated financial model this afternoon could be obsolete before the insulation or PV gets installed, and will certainly be far from reality in a 20 year time-frame. It's all a best guess- better than a WAG, but not something that's likely to follow the financial spreadsheet for decades, no matter how carefully crafted. That's not to say it isn't worth making the calculation, just keep in mind that the error bars are quite large in a lifecycle time frame.

  7. Jamie Wolf | | #7

    Favor envelope over machinery
    This dictum was stated by Marc Rosenbaum two decades ago. PV's have proven to be reliable machinery, but even they have a very different range of service life and service quality (two important conditions) than the envelope does (and Alex acknowledges this). Going to R-80 costs essentially twice the cost of PV's, but that investment should be working, and working at full service quality, more than twice as long as the PV's.

    I think that Alex is right to call our attention to the importance of determining when to stop investing in the envelope and where to consider to choose (to quote Marc again) where the next dollar is best spent.

    One other aspect of this decision tree is which investments are one time opportunities (below slab insulation, the thermal envelope and its four boundaries) and which we will have the opportunity to revisit over the life of the building (glazing, mechanical equipment, renewables, etc.).

    Thanks for exploring this Alex!

  8. Curt Kinder | | #8

    Relative Merits of Alternative Conservation Investments
    There are probably a whole lot of energy conservation investments that make better sense than increasing insulation from R60 to R80 anywhere other than a lunar colony.

    In a somewhat similar vein consider the economics of a costly complex solar thermal water heater compared to 4-6 PV panels and a heat pump water heater.

    My bread and butter is HVAC contracting, but the cost delta between a SEER 16 and SEER 19 system will often pay for either of a HPWH or variable speed pool pump, both of which nearly always provide greater conservation (faster payback) in my AO, and I'm honest with clients about that.

    A responsible building performance contractor will discuss unquantifiable comfort aspects of alternative improvements such as:

    Variable speed pool pump is quieter
    Heat pump water heater has longer recovery (bad), but cools and dries basement or garage (mostly good)
    Spray foam improves humidity control and evens out temperatures.

  9. Andrew Henry | | #9

    Need to account for Resilient Design
    Hi Alex,

    I posted a comment to a GBA guest post of Mike Eliason's related to this topic which I'll partially re-post here, with some edits and additions, because the super(duper)-insulation vs PV discussion always seems to gloss over the issue of Resilient Design (which I know you well appreciate). The discussion should account for resiliency at the building system level and at the grid (or micro-grid) system level and how this relates to the energy constraints that need to be managed with climate change.




    In terms of global warming if we are going to "prevent what we can't adapt to" we have to reduce CO2 emissions from fossil fuels to zero very quickly (see Kevin Anderson's Cabot Institute presentation to grasp the urgency). Coal, oil and gas in the U.S. accounts for greater than 80% of energy supply. So what must be done, and it must be done, is no small task.

    Buildings will have to reduce energy demand to the greatest extent possible before even starting to meet that demand with PV. This needs to be done for a number of reasons.

    First, there is this vast legacy of poorly built and extremely inefficient homes that likely can't be retrofitted to Net Zero, even if building site and orientation weren't an issue. Given the layout of suburban developments many legacy homes are poor opportunities for PV.

    Second, peak winter electricity demand is an issue at the building level in any climate, particularly cold climates. We will largely be using electricity to heat our homes and the sun isn't shining when it's coldest outside. PV can't meet winter night time peak demand in a cold climate. PassivHaus on the other hand can do what PV can't, which is reduce a buildings demand for electricity. If all new homes were built to the most efficient standard the amplitude of winter night time peak demand will be much lower than if we just built "good enough" houses and then added PV. We have to think of how the grid is going to meet electricity demand in a Post Carbon future. It is much easier for a distributed and diverse renewable energy supply to meet a low amplitude peak demand. So our built environment must be designed to reduce demand as much as possible. In a way I see PassivHaus in effect providing "Thermal Storage and Load Balancing" indirectly. My sense is that the cost to store PV generated electricity at either the building or grid level for night time has to cost more than the cost of the extra insulation in the example.

    Third, meeting summertime peak electricity demand in a warming world will be difficult. Night time lows may not drop below temperatures that make passive cooling (opening windows) effective. Also, many existing coal and more importantly nuclear electricity generating stations are dependent on surface water sources for cooling. Warmer water sources and also reduced river flows will disrupt electricity generation at coal and nuclear power stations. So again, collectively it will be important to reduce the amplitude of summertime peak demand. PV won't be able to help cool buildings on the hot nights, (and we should anticipate that they'll get hotter).

    Fourth, for a variety of reasons one should expect electricity to cost a great deal more than it presently does. We can't replace 80% percent of our energy supply for free. That said, you pay nothing for the energy you don't use.

    Buildings built to PassivHaus insulation levels reduce energy/electricity demand. A lot of buildings built to PassivHaus cumulatively reduce a lot of electricity demand. And this demand destruction will make it easier for the available PV, wind, and hydro (which have their supply limitations) to get us through the peaks.

    Given how the impacts of global warming are out-pacing forecasts I don't see any other option but to get behind the most energy efficient building standard and demand that all new buildings be mandated to meet it. From my perspective PassivHaus is that standard.

  10. GBA Editor
    Martin Holladay | | #10

    Response to Andrew Henry
    Two points in reaction to your comments:

    1. You wrote, "There is this vast legacy of poorly built and extremely inefficient homes that likely can't be retrofitted to Net Zero, even if building site and orientation weren't an issue. Given the layout of suburban developments many legacy homes are poor opportunities for PV."

    You didn't mention that these legacy buildings are also poor candidates for Passivhaus retrofits. Of course it can be difficult to install PV on existing buildings, especially if there is no unshaded south-facing roof. But it's also very expensive to retrofit the walls of older buildings to R-40. In many cases, PV beats superinsulation for older buildings.

    2. You wrote, "A lot of buildings built to Passivhaus cumulatively reduce a lot of electricity demand." The key problem with suggesting that building to the Passivhaus standard is a solution to global climate change is the "a lot" part. To make a significant dent in U.S. fossil fuel use, the Passivhaus solution is unworkable in the necessary time frame. As you pointed out, "we have to reduce CO2 emissions from fossil fuels to zero very quickly" -- in 30 years or perhaps only 20 years. If you do the math on the required construction schedule, neither new Passivhaus construction nor a Passivhaus retrofit program can be implemented on the necessary scale during that time frame.

  11. Alex Wilson | | #11

    Retrofitting costs
    The need to insulate (and perhaps solarize) millions of houses very quickly does indeed seem overwhelming and impossible, but it's partly about priorities. I just did a quick Google search on what the Iran and Afghanistan wars are costing American taxpayers and see an estimate that the total cost will be $4-6 trillion ( If we made a similar $4 trillion national investment over 30 years in a national deep-energy retrofit program, that could work out to $80,000 per house for 50 million houses (if I did my math right with all those zeros).

    I'm not suggesting that that could happen (or even should happen), but these wars have given us an interesting benchmark for massive government spending.

  12. Andrew Henry | | #12

    Missing my point

    You miss my point!

    It is because we have this vast legacy of poorly insulated homes that going forward we need to be building to PassivHaus levels of energy efficiency. It's the only way to make the transition to a post carbon energy infrastructure possible.

    The net zero sleight of hand ignores the temporal disconnect between when PV produces electricity and when peak winter time heating demand occurs. The energy to meet that peak heating demand has to be delivered as electricity in a post carbon world. It's the peak demand that will matter most.

    It's a "bird in the hand or two in the bush" issue. Added insulation is the bird in the hand. PV on the roof is the "two birds in the bush". The added insulation is certainly going to reduce peak winter night time heating demand for electricity. PV on the roof will most certainly not reduce peak winter night time electricity demand.

    The energy transition will have enough challenges as it is. We shouldn't be adding to these challenges by missing out on the easiest and cheapest way to reduce peak electricity demand... added insulation.

    It will be far more expensive for a jurisdiction (state, province, grid-level) to build the storage solutions necessary to supply peak demand than it would be if we didn't have to meet that demand in the first place.


  13. GBA Editor
    Martin Holladay | | #13

    Response to Alex Wilson
    Fortunately, U.S. taxpayers haven't yet had to pay for a war in Iran -- at least since 1953, when the CIA orchestrated the overthrow of Iran's democratically elected Prime Minister, Mohammad Mosaddegh. I think you meant to write "Iraq."

    You have pointed to one possible problem to implementing a massive government-subsidized program of deep-energy retrofits: funding. There is another: the availability of skilled labor.

    Here are some points I made during a presentation at this year's Greenprints conference (these notes come from PowerPoint slides, so the arguments are not fully fleshed out):

    New home construction in the U.S.:
    There are 132 million homes in the U.S.
    Annual U.S. population growth rate is 0.9%.
    We need 1.19 million new homes a year just to match the population growth rate.
    In 2050, the U.S. will probably have 185 million homes.
    In recent years, we’ve built 587,000 new homes per year. This number rises to about 1.85 million homes in boom years.

    What if the building code mandated adherence to Passivhaus? By 2050 (in 37 years), we would have built between 22 million and 68 million new Passivhaus homes. That would represent between 12% and 36% of all homes in the U.S. in 2050. (The high end of the range is very unlikely.)

    Even if the U.S. government required all new homes to meet the Passivhaus standard, beginning this year, it’s likely that 80% of all homes in the U.S. in 2050 will be old homes that don’t meet the Passivhaus standard.

    There are cheaper ways to reduce C02 emissions.

    Investing in other measures (for example, utility-scale wind projects) yields greater CO2 reductions per dollar invested.

    It’s hard to make the math work.
    To address climate change, we need to make huge reductions in CO2 emissions.
    CO2 emissions are still rising.
    Residential electricity use is still rising.
    Even Passivhaus homes use more energy than Passivhaus advocates declare.
    We can’t build homes fast enough to make a significant dent in CO2 emissions.

    Deep energy retrofits: Energy savings of 45% to 65% are possible if you are willing to invest $74,000 to $90,000 per housing unit.

    Can we perform a DER on every house in the U.S.?
    $50,000 per home x 132 million homes = $6.6 trillion

    For comparison, this is:
    6 times the U.S. federal deficit in 2012 ($1.1 trillion)
    9 times Obama’s stimulus package ($700 billion)
    41% of the entire U.S. national debt ($16 trillion)

    Can we afford it? Probably, but just barely. The challenge is on the order of fighting World War 2.

    Is such a program politically feasible? Probably not, unless massive new storms, heat waves, and droughts cause hundreds of thousands of deaths and huge property losses.

    Even if we could afford it, and even if politicians endorsed it, there are at least two other problems:
    There is a nationwide shortage of qualified home performance contractors.
    If the work is rushed to meet the timetable of our climate change emergency, we’re likely to ruin a lot of homes.

    Does your local siding contractor know:
    How to attach 4 in. of rigid foam to an exterior wall?
    How to flash an innie window?

    What are the risks?
    Adding exterior rigid foam complicates window flashing details, raising the chance of water entry.
    Insulated walls have a slower drying rate than uninsulated walls.

    To sum up: there are several hurdles and risks to instituting a massive program of deep energy retrofits:
    There is a financial hurdle.
    There is a political hurdle.
    There is a practical hurdle (the shortage of qualified contractors).
    There is inherent risk (because rushed work can damage buildings).

  14. Expert Member
    Dana Dorsett | | #14

    The reality of peak power (response to Andrew Henry)
    Andrew writes:

    " The added insulation is certainly going to reduce peak winter night time heating demand for electricity. PV on the roof will most certainly not reduce peak winter night time electricity demand.

    The energy transition will have enough challenges as it is. We shouldn't be adding to these challenges by missing out on the easiest and cheapest way to reduce peak electricity demand... added insulation."


    Peak power draws in the US NEVER occur during the nighttime hours, even for most houses heated with electricity the peaks occur during the post-dawn AM. The REAL peak grid loads in all of the lower-48 US states are highly correlated with cooling load, which have a large overlap with peak PV output. It's not a perfect correlation- the peak AC load lags the peak PV output by 2-3 hours, largely due to the traditional habits of commuters turning off the oversized AC during the day then turning it on when they arrive at home, but this can be mitigated in many ways. Even without demand-response control strategies by the utilities, the raw un-controlled overlap is still quite substantial, and the marginal cost of that PV output is FAR cheaper than the spot market price from low capacity-factor gas or #2 jet peakers, and is a stabilizing factor for the local grids during the cooling peaks.

    Yes, insulation brings down the peak cooling load somewhat too, but from a peak-power bang/buck point of view PV is by far the better deal.

    The AM peaks occur as everyone is waking up, the 500-1000W of air handler on the fossil fired furnaces kick on when the overnight setback steps up, folks turn on 100-500watts of lights, 200 W of TV, 500-1000w of coffee maker, 1000W of toaster oven, then takes a shower, turning on the 4.5kw of electric water heater. In the aggregate this is a much more dramatic & sharper load peak load than a modulating heat pump or even the average cycling loads of resistance baseboard heaters. That's why off-peak rates are almost universally in effect between midnight and 6AM, which are highly correlated with the hours of peak heating load.

    There is even an issue of excess nighttime wind power that becomes common in areas where wind exceeds ~25% of all grid kwh, as well as the necessary power dumping of nukes during those off peak hours searching for loads. As renewables increase, these resources have to PAY to dump power onto the lightly loaded grids. This isn't an argument for less insulation, just pointing out that insulation is far from being "... the easiest and cheapest way to reduce peak electricity demand...", since it's has almost no effect on the heating related peak loads which occur during periods of excess grid capactiy, and beyond code-min insulation has nearly zero value for reducing for peak cooling loads. If reducing peak power loading is your priority, heat rejecting WINDOWS or exterior SHADES/AWNINGS are orders of magnitude ahead of more insulation, and in most ( but not all) instances ahead of PV. In places with high summertime cooling peaks, were PV owners paid spot-market wholesale rather than flat net metering at the fixed residential rate they'd be better of with SW or W oriented panels- the annual total production would be lower, but the payback in increased revenue during the ACTUAL peaks would be very rich compared to the loss of the output during lower-demand hours.

    Peak load & demand response strategies are a more complex issue than can be laid out in a web forum. I encourage anyone interested in green-grid issues to regularly check the blog bits at , which cover the evolving power grid issues and utility business models/regulation in some detail. This stuff is evolving far more rapidly than most people think, and faster than many utility operators can handle under the current regulatory structures in some market. Small scale PV is now cheaper per kwh than fixed-rate retail in many markets already, and with the mushrooming of third-party ownership models of rooftop PV over the past 3-8 years, where the owner can sell fixed rate to the homeowner at below grid-retail, collecting any subsidies or power purchasing agreements for their shareholders, the landscape of distributed PV is changing at an accelerating speed. In CA last year PV installation by private parties exceeded the amount that was actually subsidized by the state, simply because it was a good investment, and when PV drops south of $3/W this market ripple will rapidly reach tsunami proportions. Some utilities and state regulators get it (but don't always respond in the same way- business models and gored oxen seem to matter) and are able to handle it, one way or another. Others are clueless, and are bound to become road-kill. But the net-metering deals are bound to evolve- the "run your meter backward" approach won't cut it going forward. But the PV explosion is already taking a chunk out of the true peak loads in many areas.

  15. Andrew Henry | | #15

    Reducing the peaks

    Thanks for pointing us to greentech media as it is helpful when thinking of building energy efficiency to also be thinking at the bigger system level, that being the grid from which the building draws its power from (and supplies power too). David Roberts over at Grist has also been writing quite a bit on this topic and of course Chris Nelder at his site and at greentech media.

    This is where I am coming from and where I think we have to go... we have to transition to a post carbon energy system in pretty short order. This will be done with renewables and the energy they produce will be delivered as electricity on the grid. Our renewable options are hydro, PV solar, thermal solar plants, and wind (and energy efficiency). We know solar doesn't deliver at night, wind may or may not deliver on a cold winter night. Good hydro resources pretty much already have a dam on them. So as it stands we have to install a huge amount of wind and solar. And though I am hopeful that a large number of broadly distributed points of renewable generation will flatten out the intermittency of renewables, it's still pretty clear solar won't be generating electricity at night. So some energy storage will have to be constructed. What is that cost? Might it not be cheaper and more resilient to reduce the need for storage?

    In Quebec, which is a good example of a jurisdiction that largely uses electricity for residential heating, winter night time peak loads often exceed the ability of it's huge hydro resource's generation capacity. The peak is handled by light oil and NG generating stations.

    This all gets me to your point...

    Peak power draws in the US NEVER occur during the nighttime hours, even for most houses heated with electricity the peaks occur during the post-dawn AM.

    I am not sure that you are dealing with the aggregate residential demand in that statement, that said what concerns me is how best to meet winter heating demand entirely with electricity . If we look at residential NG demand in the US it's pretty clear the peak demand is in the winter, and it's a safe bet that it's at night. Transitioning that demand from NG to electricity is going to really spike winter night time electricity demand.

    Managing grid stability is an important part of the discussion. We can't discuss building energy efficiency and PV in isolation, buildings are part of the grid, now even more so as more and more PV gets thrown on roofs.


  16. Curt Kinder | | #16

    Nifty thing about a PV / Hydro combination
    is that daytime PV can allow hydro to be banked for use during night. If we get enough PV on the grid, we can start looking for wider deployment of storage technologies ranging from Ice Storage for cooling, pumped hydro, etc.

  17. Jin Kazama | | #17

    Andrew : about your Quebec reference ...
    2 factors mainly drive this hard winter time peak ...

    1- poor past building code and insulation on existing houses for our extreme climate
    ( last week was pretty much +30c and 70% humidity during day, last winter we had 3 weeks under -20c with nightime near -30c and very humid mornings ) also super leaky houses because lack of supervision

    2- almost everybody here is using electrical resistance heating
    ( heat pumps are only very very slowly getting in houses, and most made the jump from wood/oil to electrical heating during the 80-90's + late 2000's ..also on all new houses )

    that said, i believe that we have been using a lot less our oil plants in the last 10 years
    ( i have no number, but i used to live 1km down the river of one of the main plant which i don't recall being on as often as it used to ) because of better grid design and hydro deployment.

    on the topic,

    there are alot of great posts on this thread from many of you ...
    very interesting to read this and see how "deep" some of you are interested and involved in this big picture !!

  18. Expert Member
    Dana Dorsett | | #18

    Too many conflations (response to Andrew Henry )
    Conflating seasonal (or even daily) natural gas-grid peaks with daily electric grid peaks is completely illegitimate, since the costs of grid-storage (either CNG or LNG) for gas is dirt-cheap and massive compared to what it practical and cost effective with electricity.

    And once you're at 1.5-2x code min, the net benefit to the gas-grid peak of more insulation is also negligible too. Here too, improved window performance may outstrip the bang/buck value of more insulation or PV for reducing the gas-grid peaks, but not the net financial benefit to the homeowner. Of course, PV (or any solar technology.) isn't going to have a significant positive effect on the heating season daily or seasonal gas grid peaks.

    As of this year just over half of all kwh pumped on the ISO-New England power grid uses natural gas as the source fuel, and regional gas-grid storage & distribution resource are now hitting pretty close to the limits during the heating season, given the recent rates of both home heating and power operation changeover to natural gas, but homeowners pay a fixed rate per MMBTU (at or near historical lows) just as most are paying a fixed rate per kwh (not at historical lows), at the current residential retail power rates, PV cost, and subsidy, the point where PV becomes the better financial measure has moved significantly toward PV in the past 3-5 years.

    Yes, the gas grid will be needing to improve the grid infrastructure (mostly storage, but also pipeline), if those giddy folks drunk on the frack water kool-aid prevail, but the daily-peak issue is solved with more distributed storage far more cheaply than adding another R10 whole-wall to an R30 whole-wall house. And it's pretty orthoganal to the power-grid peaks that are very much indeed reduced by PV. IIRC even in Quebec the summertime grid load peaks exceed the winter peaks- despite the large number of people heating with electricity, the off-peak hours for power in winter are still overnight-hours. If that changes, there might be an argument for spending the money on mini-split heat pumps, which still run a COP of ~1.6-1.8 at Quebec style 99% outside design temps. At ISO-New England power rates it's often a tossup between heating with natural gas vs. mini-splits in higher-R houses on a raw operating cost point of view, but whichever heating source is chosen the financial crossover between ever-more-R and PV is coming at a ,much lower R than anyone would have predicted 10 years ago, and should not be ignored. The net benefit of PV for stabilizing annual peaks (which are always cooling peaks) is large, but not yet compensated under the various net-metering deals available from most of the regional utilities.

    [edited to add]

    While not as good as a peak-day minutely or hourly profile, the time of day use rates for Quebec hydro seem to bear out that the peak grid loads are NOT during the night when peak heating loads are occurring. Whether winter or summer, the cheapest power rates occur between 10PM and 6AM., though they are modestly more expensive during the waking-hours in winter than in summer, they are in fact cheaper during the winter overnight hours than in summer:

    Note for those taking advantage of TOU+ metering, the "critical peak hours" are 7 a.m. to 11 a.m and from 5 p.m. to 9 p.m. in winter, and come with advance notice of when those very-expensive rates are in effect. but note neither 7-11AM nor 5-9PM are peak heating load periods- it's the AM commuter & business wake-up spikes, and the evening cooking/TV/gaming loads, not heating loads.

    So, even in Quebec with it's high penetration of electric heating, PV would always be delivering it's output during the more expensive peak-hours, though taking far more off the summer peaks than the winter peaks.

  19. Robert Rinehuls | | #19

    Reflective roofs
    As long as you're splitting hairs don't forget that reflective roofs supposedly reduce energy consumption and pv panels could help fill that bill.

  20. Mike Nelson Pedde | | #20

    There are many kinds of 'costs', monetary costs being but one of them. Generating power (no matter the source) that would not otherwise be needed seems like false logic to me. Insulate, upgrade windows, etc. and THEN look at alternative power sources.


  21. Expert Member
    Dana Dorsett | | #21

    Bob: Reflective roofs add to the heat load... cold climates like VT. Cool roofs that reflect solar gains lower the average temperature of the roof, increasing the average heating load, and increasing the mold hazard at the roof deck by running cooler (== slower drying.)

    Mike- Alex Wilson's article was talking about the advantage of going from an R60 roof (already 20% beyond code), to an R80 roof (closing in on 2x code), and he did the rough-estimate math. I suspect the rest of the house was a LEAST 20% beyond code for R- your precriptive " Insulate, upgrade windows, etc. and THEN look at alternative power sources." was already being taken.

    Alex was probing for the crossover point on costs/benefits of going ever-higher R, given the current low (and falling) cost of PV, not arguing for PV & code-min. A code-min wall in VT has a whole-wall R value of about R15, but few performance builders stop there- most shoot for between R40-R50. There's a false logic to going with R200 walls (even using lowest cost methods) instead of R50 walls when the savings add up to 200kwh/year, and consume substantially more materials with a 100 year carbon footprint exceeding that of the additional energy savings, using any realistic guesstimate of the energy sources over the next century. At some point the argument for PV becomes a no-brainer, whether the costs are about atmospheric carbon, raw materials extraction, or strictly financial.

    The value of solar for offsetting the current high-carb peak grid power sources is very real- typical gas-fired or #2 jet peakers run ~30% thermal efficiency from source-fuel-to-load, and all PV output is during higher-demand periods. The amount of this high-carb peaker-power that another R20 of roof insulation on an already code-min roof offsets is at least an order of magnitude lower than the peak-power output of PV. That may or may not be true for exterior shades over west facing windows, but you'd have to run a fairly careful simulation to prove it. Offsetting peak power use has much higher value (in carbon or dollars) than the simple BTU or kwh would imply when applied to some fixed-rate model, but the value of that varies with the local grid sources, and it's not just about the carbon or dollars:

  22. Bob Irving | | #22

    "I don't see any other option
    "I don't see any other option but to get behind the ...PassivHaus ...building standard and demand that all new buildings be mandated to meet it."
    I agree with Andrew to a point, but we need to scale up to that standard for this to happen on a mass scale. We need the materials, the dispersed knowledge and the experience to build these homes economically enough for wide appeal. That is all coming but not fast enough; for now we can use the Building Science 5/10/20/40/60 guidelines to build, in large numbers, economical low energy homes.

    Same thing with older homes - until the gov'mt decides to subsidize this (and don't hold your breath) we need to find ways to upgrade them inexpensively. The more we do collectively, the faster we'll learn. For instance, designing an inexpensive 6" I joist which could be hung off existing sheathing & studs, filled with cellulose and covered with a good air barrier would give us a similar benefit to foam at less cost.

    I understand the insulation vs PV argument but didn't we hear the same argument when oil was $.19/gallon? Lets build really good houses that hold the heat, then the heat source can be an academic discussion.

  23. Bob Irving | | #23

    I didn't mean to offend
    I didn't mean to offend anyone with my oil/pv conflagration, but rather those who want to build standard or slightly above average performance houses since they are planning on using solar. Build a high performance house first; there is no downside.

Log in or create an account to post a comment.



Recent Questions and Replies

  • |
  • |
  • |
  • |