musingsheader image
0 Helpful?

The High Cost of Deep-Energy Retrofits

A pilot project generates cost data on deep-energy retrofits of four buildings in Utica, New York

Posted on Mar 2 2012 by Martin Holladay

How much does it cost to perform a deep-energy retrofit at a 100-year-old single-family home? Thanks to a recent study in Utica, New York, we now know the answer: about $100,000.

The research was sponsored by New York State Energy Research and Development Authority (NYSERDA), an agency that administers programs funded by public benefit charges tacked onto electric utility bills. The program paid for deep-energy retrofits at four wood-framed buildings in Utica, New York.

The project manager for the study was NYSERDA engineer Greg Pedrick. At the recent Better Buildings by Design conference in Burlington, Vermont, Pedrick gave a presentation, “Research Findings and Momentum for Deep Energy Retrofits,” explaining the scope of work and summarizing the costs of the retrofits.

A big fan of deep-energy retrofits, Pedrick explained, “I want to see a fatter house with a smaller mechanical system.”


An ambitious goal

Pedrick’s team selected four wood-framed buildings; brick buildings were deliberately excluded. All are owned by low-income families who had applied for weatherization assistance. Three of the buildings are single-family homes; the fourth is a duplex. All of the buildings are about 100 years old.

The work was paid for by NYSERDA; there were no out-of-pocket expenses for the building owners.

The researchers’ goal was to reduce energy use by 75%. To achieve this goal, the plan was to retrofit slab floors to R-10, below-grade walls to R-20, above-grade walls and roofs to at least R-40. The windows would be upgraded — either with low-eLow-emissivity coating. Very thin metallic coating on glass or plastic window glazing that permits most of the sun’s short-wave (light) radiation to enter, while blocking up to 90% of the long-wave (heat) radiation. Low-e coatings boost a window’s R-value and reduce its U-factor. storm windows or new windows — to achieve a maximum U-factorMeasure of the heat conducted through a given product or material—the number of British thermal units (Btus) of heat that move through a square foot of the material in one hour for every 1 degree Fahrenheit difference in temperature across the material (Btu/ft2°F hr). U-factor is the inverse of R-value. of 0.25. The airtightness goal for the homes was 0.15 cfm @ 50 pascals per square foot of surface area.


The basement insulation was installed on the interior

To insulate the basement floors, the contractors first installed a layer of Platon dimple mat on top of the existing concrete slabs, followed by R-10 rigid insulation and a layer of Durock cement board. The Durock was not fastened down; it just floats over the XPSExtruded polystyrene. Highly insulating, water-resistant rigid foam insulation that is widely used above and below grade, such as on exterior walls and underneath concrete floor slabs. In North America, XPS is made with ozone-depleting HCFC-142b. XPS has higher density and R-value and lower vapor permeability than EPS rigid insulation., held in place by gravity.

The basement walls were insulated with two different types of rigid foam. Next to the foundation walls, the workers installed 2-inch-thick Dow Perimate (XPS with vertical drainage grooves). The second layer was 2-inch-thick Thermax, a type of rigid foam that can be installed without a thermal barrier (that is, without gypsum drywall protection). The Thermax was held in place with cap screws.


Above-grade walls were insulated on the exterior

All of the existing siding was removed from the above-grade walls. Once the walls were stripped to the sheathingMaterial, usually plywood or oriented strand board (OSB), but sometimes wooden boards, installed on the exterior of wall studs, rafters, or roof trusses; siding or roofing installed on the sheathing—sometimes over strapping to create a rainscreen. boards, contractors installed a layer of Thermoply with taped seams as the air barrierBuilding assembly components that work as a system to restrict air flow through the building envelope. Air barriers may or may not act as a vapor barrier. The air barrier can be on the exterior, the interior of the assembly, or both.. Then they installed two layers of 2-inch-thick polyisocyanurate with staggered seams, followed by vertical rainscreenConstruction detail appropriate for all but the driest climates to prevent moisture entry and to extend the life of siding and sheathing materials; most commonly produced by installing thin strapping to hold the siding away from the sheathing by a quarter-inch to three-quarters of an inch. strapping and new siding.

New fiberglass-framed windows from Serious Energy (R480 series) were installed in most of the buildings. The windows had Heat Mirror glazingWhen referring to windows or doors, the transparent or translucent layer that transmits light. High-performance glazing may include multiple layers of glass or plastic, low-e coatings, and low-conductivity gas fill. consisting of two layers of glass with a plastic film suspended between the panes (a type of glazing that performs like triple glazing). The windows were installed as “outies” in new window bucks that projected 4 3/4 inches out from the old wall sheathing. Many of the existing exterior doors were replaced with new insulated doors.

Two of the buildings got new metal roofing installed over 4 inches of new polyisocyanurate. The roof strapping was extended to increase the width of the roof overhangs. Two of the buildings had roofing in very good shape, so those buildings didn't get new roofing. At these buildings, it made sense to install insulation on the attic floor. After all of the existing insulation was removed and discarded, the exposed lath-and-plaster ceilings were sealed from above with a thin coat of closed-cell spray polyurethane foam to seal air leaks. Then a deep layer of cellulose was installed.


Using a tankless gas water heater to provide space heat

The design heating loadRate at which heat must be added to a space to maintain a desired temperature. See cooling load. of these renovated buildings is less than 40,000 Btu/h. The existing forced-air furnaces in these buildings were all removed, and new hydro-air heating systems were installed.

The hydro-air systems use a natural-gas-fired Rinnai tankless water heater to supply heat; hot water is circulated through heat-exchange coils in an air handler, and the space heat is distributed through the existing ductwork. The same Rinnai heater also supplies domestic hot water. To be sure that the Rinnai’s limited output of hot water is adequate, each unit is connected to a 12-gallon electric water heater (with the electric resistance element removed) that acts as a buffer and storage tank.


Unforeseen conditions

According to Pedrick, construction crews encountered “a lot of unforeseen conditions,” including undersized electrical service, damp basements with improper drainage, failing black-iron sewer pipes, and lead water-supply pipes. (They also encountered lead paint, but that wasn't unforeseen.)

Of course, the project’s goal was to create safe, energy-efficient, code-compliant homes. “Anything we found that needed to be fixed — we fixed it,” said Pedrick. Correcting unforeseen conditions at the four buildings cost $81,680 — an average of $20,420 per building.


Deep energy savings

The energy retrofit work greatly reduced the air leakage rate at all four buildings; final results ranged from 2.2 to 5.0 ach50. The homes had impressive levels of energy reduction; however, the energy-reduction goal of 75% was not met. Overall energy use (including space heating, domestic hot water, and electricity) was reduced by 60% to 65%. Electricity use in the four buildings actually went up. (Among the new appliances that added to the electricity load were the homes’ mechanical ventilation systems.)

The bottom line

This was a very valuable research project. The retrofits resulted in very significant energy savings, and the gathered cost data are extremely useful. Before the retrofit work, the homes were drafty, uncomfortable, and out of compliance with local building codes. After the retrofit work, all of the homes have new siding, and some of the homes have new roofing and windows. All of the homes are safer, more comfortable, and less expensive to operate.

However, the energy savings alone can't possibly justify the very high costs of this type of retrofit. The average cost for the work was $112,000 per building, or $89,783 per housing unit. The average annual energy savings was 393 therms of natural gas (11,486 kWh) per housing unit. Since the cost of natural gas in Utica is $1.65 per therm, the average annual energy savings are $647 per housing unit.

In other words, the simple payback period for these retrofits was 139 years.

If the same amount of money ($89,783 per housing unit) were invested in a photovoltaic(PV) Generation of electricity directly from sunlight. A photovoltaic cell has no moving parts; electrons are energized by sunlight and result in current flow. (PVPhotovoltaics. Generation of electricity directly from sunlight. A photovoltaic (PV) cell has no moving parts; electrons are energized by sunlight and result in current flow.) array instead of a deep-energy retrofit, you could buy a 20-kW PV system with an annual electrical production of 22,401 kWh (worth $3,248 at the local electricity rate of 14.5¢ per kWh). The value of the PV electricity would be 5 times the savings achieved by the deep-energy retrofit.

[Postscript: The September 2012 issue of the Journal of Light Construction includes an article, “Tightening Up a Two-Family Home,” that describes a deep-energy retrofit project at a century-old wood-framed duplex in Massachusetts. The cost of the work was $275,000, or $137,500 per unit.]


Last week’s blog: “Carl and Abe Write a Textbook.”

Tags: , , , , , , , ,

Image Credits:


Mar 8, 2012 3:06 PM ET

Conclusion or Alternative?
by Marcus de la fleur

I quote from the blog post:

"However, the energy savings alone can't possibly justify the very high costs of this type of retrofit."

Sure, but what would be the alternative? Leave the house (and the all other several thousands of homes) alone. Screw comfort, safety, energy savings (even though they are small) and global warming?

I don't want to be cynical - just looking for an alternative to the DER that could take its place.

Mar 8, 2012 3:45 PM ET

Response to Marcus de la Fleur
by Martin Holladay

Q. "What would be the alternative?"

A. My own proposed alternative is to perform retrofit measures that yield energy savings per invested dollar equal to or greater than an investment in a PV system.

The Utica retrofits went well beyond such common-sense measures.

Mar 8, 2012 8:05 PM ET

High Cost of DERs
by mike keesee

At the risk of getting lost at the end of 50+ comments, I'll add my two cents. I think Martin's latest posting was an excellent follow up to the "Pretty Good House" discussion of the past couple of weeks. I worked on 6 DERs with a goal of 50% annual source energy savings as part of SMUD's R&D efforts(you can read the case studies @ NREL assisted us in this effort providing energy analysis,. monitoring and evaluation support. I'll also be presenting a paper of the projects at the upcoming ACEEE summer study session. which provides more details about costs and savings, including monitored results to date.

I mention all this because it's critical that we get real data and monitored results from these efforts, especailly given the high costs associated with these projects - costs that are often subsidized by utility rate payers.

In brief, the lessons learned include many of the points made by Martin and the commentators:

- DERS are expensive but provide unquantifiable benefits including increased comfort, rehabilitating abandoned, foreclosed homes, providing low income families a decent place to live, etc. therefore, DERS should be undertaken as part of bigger efforts to rehabilitate housing and not necessarily be undertaken for their energy savings alone.

Other lessons learned from the SMUD DER demonstration experience include:

- intelligently packaged efficiency packages geared to specific climate zones and housing vintages can lead to large energy savings. The rule of thumb being is to bring the home up to new construction levels.

- peak demand savings are under appreciated by utillities and represent far greater value than simple therm or kWh savings. In particular, DER programs should be geared to reducing AC (and heating) equipment sizing and utilities should seriously consider paying for right sized equipment rather than kWh or heating therms. This will require a major re-think on the part of state utility regulators and utilities.

- PV can make a big difference, expecially for peak demand savings

- the home performance and accompanying utility and state energy office efficiency programs are missing (ignorning) the biggest, most lucrative DER market - the existing home re-sale market - where the energy efficiency mortgage and 203 k loans could easily underwrite the cost of most energy upgrades at net postive rates of return and case flow (are we ever going to stop talking about payback!?)

- we need to be very careful about claims about overall electric savings as the explosion of plug loads can easily overwhelm thermal electric (AC) savings (all electric homes are the exception where huge savings are possible).

To conclude, then, I'll echo what I said about pretty good homes, which is that the pretty good home approach is what we should be aiming at with a special focus on right sized equipment in DERs, say 30-50% savings. Quite simply, the scope of the challenge - dramatic energy savings in a very short period time and re-starting our construction industyr -demands that we come up with simple, easy to replicate models on a mass scale - read millions - if we hope to make any kind of impact.

Martin deserves kudos for furthering this discussion. Keep up the good work

Mar 9, 2012 7:22 AM ET

Edited Mar 9, 2012 7:30 AM ET.

Response to Mike Keesee
by Martin Holladay

Thanks for you pertinent comments on the need for a climate-specific approach.

I think the focus on efficient air conditioners, PV systems, and peak electricity reductions are very important in California, but obviously less important in Utica. Appropriate energy-efficiency measures will certainly vary by climate.

Readers who click on the case studies at the link provided by Mike will note the following project costs for Mike's deep-energy retrofits:

  • Mascot: $25,000
  • 9309 Quintanna Court: $66,500
  • 3893 32nd Avenue: $77,000
  • 1110 Jean Ave: $120,000
  • 2380 North Avenue: $184,000

Of course, the project costs listed above (except for the $25,000 project) include a variety of measures, not all of which were directly related to energy improvements.

Apr 8, 2012 2:31 PM ET

Edited Apr 8, 2012 2:35 PM ET.

Sometimes it works well...
by Skip Harris

Some years ago my neighbor asked us to look at her manufactured home to try to make it look more like a normal home. It had rotting siding (pseudo-Swiss-chalet), no eves, a 1/12 roof pitch, enormous windows on the west end, and dark interior paneling. And it turned out that the builder had used the entire uninsulated attic as an air return to the furnace/AC.

The architect and I designed a $30k retrofit: a nicely pitched roof with 2' eves, extending a full 10' to the west. Low-E windows and glass doors, R13 fiberglass in place of the rat-infested R7 in the walls, and an uninterrupted layer of R30 over the old roof deck created a home in which the dog could no longer hear cars enter the drive, the propane tank needed filling less than half as often....and the AC ran for a few days a year rather than weeks. They raved over the comfort and saved at least $2k per year while the efficiency improvement costs probably took under $5000 of the budget.

I realize this was only possible because we were piggybacking the energy improvements onto already needed/wanted improvements and because the original manufacturer was Bozos-R-Us, but is it possible in some circumstances...

Apr 16, 2012 3:18 PM ET

Edited Apr 17, 2012 11:33 AM ET.

Bringing down costs of DER and assessing DER markets
by David Guenette

I've come late to this discussion, but if nothing else, I'd like to thank Martin for one of the first and most interesting pieces--and valuable comments from others--concerning the issue of DER cost. Nice to see some numbers getting posted, even though we have a long way to go, still.

As you point out, and also, especially, as seen in many of the reply comments, there are many variables in DER costs beyond building envelope insulation and air sealing improvement, such as new energy efficient mechanicals and appliances. One set of variables is, as pointed out by many replies above, the condition of the roof, windows and doors, insulation, and/or siding, to name the major components, right along with the state of air sealing. If a house needs a new roof and re-siding, I think that we'll pretty much all agree that the DER cost will be relatively modest compared to a DER of a home with good roof and siding, if for no other reason than the need to do roof and siding work on the first house regardless of DER (and so the cost here is not attributable to DER, mostly), while the second house must carry the costs new roof and residing as a direct DER cost. While the same amount of money, more or less, will be spent in the first case as in the second, there is clearer understanding that the first house's relative poor condition requires that much of the money still needs to be spent to renovate the building, and so the additional spending on DER is less. A lot of what the DER market development requires are clear guidelines and market sizing of existing homes that represent the first case instance. Here's a rough list of the sorts of building characteristics to be investigated as cost-accessible DER prospects:

Insulation level of house
air-sealing of house
roof condition
siding condition
style of house (e.g., Greek Revival tend to have good roof overhangs, mitigating the need to extend roof lines over built-up exterior walls)
climate zone location
window efficiency

A matrix of features and conditions will go a long way to highlight likely house candidates for DER, and then, presumably, market sizing can be researched; for instance, there are figures showing percentages and gross numbers of housing getting re-sided each year.

Unfortunately, even when a house's roof , siding, and windows need replacing, DER still contributes significant expense. The other thing needed is a method for bringing down the cost of DER, and I'm glad to see Greg Pedrick, of NYSERDA, funding some further field trials related to DER cost reduction. DER, today, remains a one-off process with few efficiencies available; one study suggests that the current practice of exterior DER requires 8-10 trips around the structure (siding demo, various foam layers, taping, sheathing, siding, etc.).

RetroSheath (, patent-pending, is an effort to systematize exterior DER, leveraging both digital technologies (measurement capture, CAD and CAM software) and existing materials and processes (R-Control's Nailbase and Premier Panel's ci Panel, to name two, and closed-cell soft foam sheets) to reduce labor costs, both from less time spent on site and the use of general construction labor, as opposed to the very highly skilled DER contractor crews currently required. There are one or two other potential cost reduction processes, such as encapsulating old siding instead of removal and remediation, and window/door build-out or extension methods. There's plenty of information available on the RetroSheath site.

I've been searching for partners, either as fellow field trial participants and/or to help move the provisional patent into a full (utility) patent application. DER may not be, in the near term, a huge market, but it is likely a big enough market, especially in the Northeast, where housing stock is old and amenable to DER, at least where the infrastructure of natural gas isn't available! The next step for RetroSheath is to determine the real level of cost reduction available through its process, since every point of cost reduction expands the potential market for DER. I need to move from back of the envelope figures to some real cost analysis. Comments welcomed!

Aug 15, 2012 1:10 PM ET

Still Cost Effective
by Joshua Nelson

I just did some quick math, and if you are looking for a green-built house, a DER still makes more sense. A weighted average of the costs (not including the unforeseen complications) comes out to around $46.16/sqft. Including the unforeseen (which is much more variable, depending on the condition, age, location of the house) and that bumps up to $75.80/sqft.

That is still reasonable compared to the costs of a super energy efficient new home, I would think - not to mention better on the environment from a life-cycle analysis standpoint. I think it's important to note that the unforeseen complications can become more foreseen with experience and are incredibly variable (as seen in those four examples above), so it would probably be better to assume $50-60/sqft for a DER on an older home (100-ish years), less for a newer one.

Great article!

Aug 15, 2012 1:26 PM ET

Response to Jeff Wilson
by Martin Holladay

I'm sorry to hear that you paid $8 a watt for your PV system. If you bought it today, there's no reason you couldn't get the same system for $4.50 to $5.00 a watt.

Of course, you're right that the value of a deep-energy retrofit is equal to whatever the client is willing to pay. You seem to be happy with your choices -- so it's worth it.

Those looking for policy solutions to our climate-change crisis face a different type of arithmetic, however, and it's hard to imagine that our country will be able to afford to spend $80,000 to $120,000 per house to reduce energy use. While every deep-energy retrofit is a valuable exercise, no one has yet come up with a way to achieve deep cuts in energy use in a way that our society can afford.

The rapid drop in PV prices is changing the economics of deep-energy retrofits month by month. In many areas of the country, a PV system -- as big a system as you can squeeze onto your roof or south lawn -- has an immediate financial payback. Surprisingly, you can't say the same for 4 inches of foam or triple-glazed windows.

Mar 21, 2016 11:58 AM ET

DER real cost or MER as better option? (Moderate Energy Retrofit
by Thomas Kacandes

Good article, but bad headline: I'd suggest that "DER costs from $44,000 to $70,000" would be more accurate. Why? Obviously you take out the "unforseen conditions" because replacing the lead water main is not an energy issue, even if it's a really good move! Also, my numbers come from abandoning the basement retrofit because I'm advocating for making the basement ceiling the boundary layer for the energy envelope of 100-year old houses. Yes, you CAN insulate their basements and it's a good thing, but the basement did not start as conditioned living space so adding cost to even get it in that direction is a good way to burn up the budget. Again, nice and def. can add to energy overall savings if you do it, but you can fix and insulate plumbing and build a small insulated room around mechanicals for far less $$.
I'd submit that many of these comments get at the difference between a "pretty good" or Moderate Energy Retrofit vs. NYSERDA research version DER. Starting with buildings that are in bad condition, some of which were never designed to be comfortable, just skew the "DER" numbers more, even if the exercise is instructive! Thanks for the article Martin and GBA!

Mar 21, 2016 12:45 PM ET

PV cost, $ and environmental; "comfort" is many forms of value
by Thomas Kacandes

Construction and building science are my avocations, but I get paid to design and install solar PV. The use of solar economics as a foil to the cost of a DER are a good exercise, mostly because solar PV is so easy to measure in kWh output. Meanwhile Btu's moving unseen in and out of a building are hard to measure, especially over time. Of course, the Btu's you needed to be comfortable but were not there are never measured, so we toss them an analysis black hole called "comfort". For me, comfort included quieter house to work and sleep in, elimination of annual yellow jacket invasions by closing up all holes in walls and properly venting the roof in a wasp-proof way, eliminating the need for AC except on high humidity days, and not worrying about running out of oil when I'm out of town. Forgot that new trim, sidng and windows make me comfortable to point at the house and say "it's mine"!
Back to PV:
1) the environmental cost of PV is definable and as someone who has actually manufactured solar panels, just cut it out and stop kidding me will ya? Here's the short answer: there is some water use and some very, very controlled hazardous chemical use in making solar cells. There is less and less energy use to make a watt of solar, mostly due to the same crystal being cut into more (thinner) cells and less waste in every feature of every stage of production.
BUT EVERY OTHER FORM OF ENERGY (except wind) uses LOTS of water! Yes, the answer is water makes all the difference because getting water, treating water, cleaning up the water, even discharging the water - it all uses lots of energy. This is now known as the "water-energy nexus" because the two are interlinked so thoroughly. Unless you use PV or wind which use NO WATER post manufacture. They also use about no fossil fuels post installation where as even nuclear and hydropower (much less) use fossil fuels for operations and maintenance and getting rid of waste is kind of a big thing, too.
The other big joke about even bringing up the idea of "factoring in PV's environmental impact" is that PV reduces the centralized transmission of electricity because it is generated at distributed points on the distribution grid (aka "distributed generation"). The largest, most complicated "machine" in any developed country is the electricity grid and it costs trillions of dollars to maintain and millions of tons of CO2 to operate, but has a final efficiency of less than 40% by most estimates. PV on your house supplies electricity about 40 feet from the source over a wire that requires NO maintenance and some significant % of the time, does it at the exact same time that your house would otherwise demand peak load from the most complicated and subsidized machine in the country. I call that a pretty good substitution in any environmental terms you choose - and that's not counting the complete avoidance of water use.
3) PV costs for residential installs as of 2016 are $4.00/W dc or less mostly. Any custom system installed at your house (incl. all PV) can vary a lot, so your mileage WILL vary. However, solar PV is really an after-tax deal which is my final comment on beware of comparing PV and DER: it is harder to figure as a true comparison due to this, all the POSITIVE environmental benefits of each that are "externalized", and the fact that PV generally is forgotten about once installed. Outside of taking some heat out of your sunny areas of roof (40 deg F on back of roof sheathing on a hot day), PV will not add to your Btu comfort like a DER will, it doesn't make your house quieter or healthier, and it will not keep insects out of your walls.

Register for a free account and join the conversation

Get a free account and join the conversation!
Become a GBA PRO!