The Economic Efficiency of Insulation: How much insulation is enough?
Hi all – I’m stumped, and wonder if anyone can answer the following question:
I got into a heated argument recently, with someone who is building a McMansion with 2×6 walls insulated with fiberglass, and 1″ foam board on the exterior in Michigan. It all started when he told me that his insulation contractor told him that, once you get to R-20 or R-21, the insulation had achieved over 96% efficiency, and therefore, because of the principle of diminishing returns, it made no economic sense to insulate beyond R-21.
In support of his argument, he gave me a paper written by an engineer for a company that specialized in spray foam insulation (see attached).
I did some research on the InterWeb, and found a discussion on either Fine Homebuilding or Green Building Advisor where someone commented that many spray foam installers use that argument because spray foam is so expensive, the only way they can “sell” it to customers is to argue that you really don’t need that much of it. In other words, if it’s R-6 per inch, and R-21 is over 96% efficient, you don’t need more than about three inches worth in your walls.
While that explanation offered by commenter made sense to me, it did bother me that I couldn’t find anything that disputed the insulation efficiency argument, which, again, indicates that you achieve over 96% efficiency at R-21. I understand that the economics of going beyond R-21 depends on many asumptions, including the cost of energy now, the anticipated cost of energy in the future, and the anticipated rate of return on invested cash. But the guy who debated me did seem to have a point – if R-21 is over 96% efficient, then why would anyone go beyond it, regardless of cost?
Are all of these Passive House “superinsulators” wrong?
Replies
The real issue is that even a triple-pane, super-duper window is basically a large hole in the wall, when compared to even a moderately well-insulated 2x6 wall. So beyond a certain point you get to where all the heat is escaping through the windows (and doors, and vent fans, and a few other odd outlets).
If you want to figure it out for your house then sit down and go through the calculations, figuring out the overall heat loss for the house (including windows, ceiling, floor, etc), given a certain amount of insulation, then add insulation and figure again.
Now from an economists perspective you'll have to consider the direct savings, the value of knowing you're helping the environment (existence value), and option value that you have in selling the house for a higher price because it's better insulated than average.
Then you'll want to factor in the time value of money (a dollar saved a year from now is not worth as much as a dollar saved this month), and a person's cost of capital (cost of the money to have the extra insulation installed).
If your cost of capital is high then the overall value is low - if you can afford the insulation it's a great deal. If you live in the moment then your future value of the energy efficiency doesn't mean all that much, but for most it's significant.
These items explain why some people happily pay the high up front cost to make more electricity than they consum even if it doesn't make sense to the rest of us (their existance value is high).
Thanks Dr. Diarge for the tuition on your econ class finally going to good use 15 years later! lol
Maybe start
by explaining what you mean by "96% efficient".
Yes, I too wonder what is meant by 96% efficient. I need a picture of that.
I have not yet read the attachment from the spray foam engineer, but as a rule of thumb, I would be highly skeptical of any analysis offered by vendors of insulation products.
I am curious as to what type of vapor barrier they are using in that new house you describe with the 2 X 6 studs. My guess is that they are not using any warm-side film vapor barrier.
I built one superinsulated house and I am designing another one. I don’t have all the answers, but I have put a lot of thought and research into it. One shining conclusion is this: When it comes to insulation, vapor barriers, and air infiltration control, there is not a topic on earth that has less agreement between the participants.
There are dozens of standard platitudes that people simply repeat as though they were truths. Many of these involve making the case that there is no point in exceeding a certain insulation thickness. But, as been mentioned, this is often a self-serving justification for a practical inability to achieve more than a certain thickness. For a long time, it was a mantra to say that there was no economical benefit in exceeding 2 X 6 stud walls with fiberglass. But this was because of the fundamental difficulty of building with wider studs.
There is tremendous boosterism of spray foam that makes many questionable claims about the benefit of foam and the drawbacks of competing insulation. One of these claims is that fiberglass batts always produce mold, and have no R-value at all because they are permeable to airflow.
There is a new trend of thinking that asserts that vapor migrates outward mostly due to air leaks in the envelop, as opposed to the traditional belief that vapor moves outward in response to seeking regions of lower vapor pressure. This thinking has the effect of minimizing the importance of the conventional warm-side film vapor barrier, which just happens to be one of the most hated and poorly executed construction details. In fact there has been a general shift away from the belief in the importance of the warm-side vapor barrier; and on to an exaggeration of the need to barricade the exterior against air infiltration.
There are proponents of blown-in cellulous insulation that claim that no warm-side vapor barrier is needed because the cellulous is installed so densely that air cannot permeate it. And again, this is based on the new belief that vapor moves by hitchhiking on air, as opposed to moving in response to difference in vapor pressure.
There are very confusing analyses that show the effect that thermal bridging through framing members has on the total R-value of a wall.
Some argue that if a window is R-3, and the rest of the envelope is R-50, all of the heat will escape through the window, and therefore it is a complete waste to insulate any of the envelope above R-3.
The superinsulated house that I am designing now will have R-50 fiberglass batt insulation in the walls and R-90 in the vaulted ceiling.
A real life story
When I take my dog out for a walk in the winter I make an assement of the outdoor conditions and dress accordingly. On "cold" days I start with a wool cap. This covers my head in a comfortable way and lessens the heat loss through my noggin. The old addage, "If your feet are cold, put your hat on" rings true. Then its jacket time: heavy to light, I've got them all. I might select a medium jacket but doing so, I know, won't keep me warm. A windbreaker over the top will add comfort. I know this because my body has told me so through experience.... more jacket (insulation) more heat retention. Note that I'll feel warmer even though I'm not adding anything more to my head which is a major source of heat loss. (And, I've ignored my face entirely... an open window!)
So what you say? Well, one's body is not unlike one's house. Both have internal heat producing mechanism and both suffer heat loss when in a cold environment. We self limit in certain areas: on my body I choose to not wear a terribly cumbersome hat... in our homes, we choose to have windows and doors. Both head and glass are high heat loss areas. These choices are made for better or worse but they are limited to themselves. A bigger jacket or more wall insulation will always keep us warmer.
When does my jacket selection reach 96% efficiency? That depends on the weather which, in part, depends on where I live. Walking my dog in Barrow, Alaska will mean wearing something I don't own (anymore) here in the PNW.
One last point to be made... a rule of thumb should be largly ignored. Except for the one: If your thumb is cold, it's best to put on some gloves. Also: The faster the dog walk, the warmer you'll be. (But that last one gets us into furnace output efficiency. A topic for another day.)
But, nevertheless, there IS a "diminishing return" to insulation -- at some point it's not worth the expense, the hassle, the effect on building design to add more. It's fairly easy to do some rough calculations and figure out about where this point is for a given structure and climate, and an archy, engineer, or HVAC guy with the right software can probably come much closer. At some point adding more insulation is just for bragging rights -- putting mags on your Yugo.
I had this very discussion 30 years ago, long before spray foam days, with a guy who sold styrofoam boards. I was telling him about a house I had seen where they had filled the 2 X 12 roof rafter bays flush with stacked styrofoam. He told me the same thing about the point of dimishing returns. Then he asked me if I had ever seen a walk-in cooler or freezer. I had of course so his next question was if I remembered how thick the walls were. I guessed that they were 4 inches which was the correct answer. He said, "if it's not cost effective to use thicker insulation on a small walk-in freezer why would it be on your house?"
Since then I've installed a few walk-in freezers and coolers and they all had 4" thick foam walls. I've worked in a food processing facility where the entire building was a drive-in cooler and had inside that drive in freezers and even a -30 degree blast freezer.. The blast freezer had 6" foam walls, the rest were about 4 inches.
I'd love to see a graph of foam thickness versus percent of savings. My guess is that it would get pretty flat pretty quick.
Diminishing Returns
I am not sure where that the law of diminishing returns draws the line for added insulation. I have never seen an objective, mathematical model that predicts it. And a lot of people tend to talk themselves or someone else into the idea that the line is drawn at a lower R-value than actually may be the case. When you increase the amount of insulation, you have to also increase the framing, so overall there must be an increase in four things:
1) Insulation
2) Lumber and other materials
3) Labor
4) Design
There is thermal diminishing return on each increment of additional insulation. But the other factors of cost do not follow a regular curve. For instance there will be no added labor cost for installing a 5 ½” batt compared to a 3 ½” batt. But the material cost will be higher with the thicker batt. The cost of materials, including the insulation, will also be higher with the thicker insulation.
Once you get to the insulation width of a 2 X 6, it becomes impractical to exceed that width with solid studs. So you can revert to insulation with a higher R-value per inch, but it may cost more to achieve the overall desired R-value for the wall.
To exceed the thickness of a 2 X 6 wall, one may resort to a double-stud wall. However, this requires a large increment of increased design because the double stud system raises complicated issues with corners, headers, firestops, and the detail for windows and doors. But once you go to double studs, the cost increases uniformly as the wall gets thicker from there.
In a climate like Minnesota with major wintertime heating, I do not believe the economic diminishing return threshold can be exceeded or even met with a 2 X 6 wall, even with the highest R-value per inch, attainable with sprayed polyurethane foam. I believe the economic threshold is somewhere higher in the R-value scale.
I believe that spray foam in a 2 X 6 wall gets into a reasonable proximity to the economic threshold, and some argue that it is close enough. But I have enough reservations about the cost, quality, and performance of spray foam to be convinced that it is best to forego it, move beyond the 2 X 6 wall into double stud design, and use fiberglass batts with their benefits of low cost and manufactured quality control.
One thing to keep in mind is that it is easy to talk yourself out of adequate insulation by using the argument that there is a threshold where the added insulation yields no benefit. The threshold is real, but locating it is very complicated and controversial.
Here's an old article I read years ago and saved. It's all anectedotal and written by a spray foam guy to boot but it goes along with what i know about coolers and freezers. And before anyone says it, Yes, I understand that a walk-in freezer doesn't act like a house.
http://www.swbet.net/R-Value-Myth-Foam-Fiberglass.html
I also looked to see what Building Science Corp recommended and found that as far as insulation goes they call for 5.5 inches of foam in Miami and in Aspen.Then that got me to thinking about what Dan said about windows. So, what's the point of insulating a wall to a higher R value than the windows in that wall? Insulating half your attic to R-50 and having no insulation in the other half doesn't give you R-25, it gives you R-0. So, you foam 1500 square feet of walls to R-38 but have 500 square feet of R-5 windows isn't the R value of the whole assembly just R-5?
florida wrote: So, what's the point of insulating a wall to a higher R value than the windows in that wall? Insulating half your attic to R-50 and having no insulation in the other half doesn't give you R-25, it gives you R-0. So, you foam 1500 square feet of walls to R-38 but have 500 square feet of R-5 windows isn't the R value of the whole assembly just R-5?
No, it doesn't work that way. All the heat does not simply leave through the route of least resistance. If you have an envelope with all surfaces insulated to the same R-value, you could say that would be the R-value of the envelope. Then if you put a window in with a much lower R-value, it does lower the average R-value for the whole envelope, but it does not lower the whole envelope down to the R-value of the window.
Anyone
interested in this stuff should make a habit of reading at Green Building Advisor. You can quickly get a clear picture of what folks out there are doing and how it's working.
Alright, so where does the R value end up?
As an example what if we built a 10' X 10' house with no indulation then cut in one 4' x 5' window. We leave the window open on a night when it's 0 degrees outside. Wouldn't the inside temperature in the morning be 0 or close enough to be meaningless?
Now we install R-50 foam in the house and leave the window open again. Would the insulation provide any barrier to the cold then? I think it would still be close to 0 inside but might have taken a little longer to get there.
Now close the R-5 window. What might the morning temp be then?
There is a certain effective R value in the air itself, and in the motion (or lack thereof) of the air. Many homes have leakage equivalent to a medium-sized window left open, but still manage to stay (sorta) warm and the insulation in the walls is (sorta) effective.
But you're right that if the R value were effectively zero (ie, the above-mentioned effects were somehow negated) on even one square inch of wall, all the insulation in the world in the rest of the house would be useless. That one square inch would be a "dead short".
Escaping Heat
florida wrote:
Alright, so where does the R value end up?
As an example what if we built a 10' X 10' house with no indulation then cut in one 4' x 5' window. We leave the window open on a night when it's 0 degrees outside. Wouldn't the inside temperature in the morning be 0 or close enough to be meaningless?
Now we install R-50 foam in the house and leave the window open again. Would the insulation provide any barrier to the cold then? I think it would still be close to 0 inside but might have taken a little longer to get there.
Now close the R-5 window. What might the morning temp be then?
As a practical matter, a building envelope is made up of surfaces of varying R-values. The R-value of the envelope as a whole is an average of the R-values of the areas of differing areas. The R-value of the envelope as a whole does not drop to the R-value of the area of the lowest R-value. If that were true, it would mean that one open keyhole would prevent the entire building from being heated.
So yes, windows matter because they are relatively low R-value. They drag down the overall envelope R-value. If you are trying to save energy by insulating in cold climates, it pays to limit windows. You have to make a choice between the benefit of windows and the cost of energy loss that they represent.
However it is not accurate to conclude that an envelope with one window at R-3 means that it is pointless to exceed R-3 insulation everywhere else in the envelope. It is not true that all of the heat will escape through the portal of least resistance. It ought to be pretty obvious that this is not true.
However, I have heard this contention advanced numerous times as a way to justify limiting the use of insulation.
I understand you point and agree. I suppose the bone of the whole thing is at what thickness of insulation is further insulation pointless.
What's your object to foam?
Objections To Foam
florida wrote:
I understand you point and agree. I suppose the bone of the whole thing is at what thickness of insulation is further insulation pointless.
What's your object to foam?
I object to foam because of its cost, the inability to control or even judge quality. It seems like there are a lot of different variations in foam chemicals, and varying claims of performance, and when it comes to insulation, nobody agrees on anything as it is. It just seems like there would be way too many things that could go wrong. I would be concerned about not ending up with the promised R-value or having foam uncured and remaining in that state indefinitely. I would be concerned about prolonged off gassing from uncured foam. I would also be concerned about the possible separation of the foam from the wood framing and sheathing.
So I prefer to use a manufactured product that I will install and make sure it is done right. I will use un-faced, un-encapsulated fiberglass batts in either normal or high density. I trust the manufactured product to have satisfactory quality control, but I don’t trust custom made and installed spray foam enough to feel comfortable using it.
Your point about the overall envelope R-value is supported by the REScheck program offered by DOE as a means of determining compliance with the Energy Code.
As an alternate to using the cookbook, or prescriptive path, in the Code, you can use REScheck to enter R-values and U-factors of the various elements of the envelope (windows, doors, walls, ceilings and floors), and the program spits out a value for the total U-value of the building. As long as this U-value total is less than the total that the prescriptive path would provide, the design will meet the Energy Code.
In practical tems, this means that you could use better wall and ceiling insulation to offset the loss from a greater window area than would be allowed by the prescriptive path. Or, better windows could allow lesser amounts of ceiling or wall insulation.
All of this is not meant to suggest that compliance with the Energy Code is the best approach to insulating a home, it just helps clarify the concept of how differing insulation values affect the overall efficiency.
Effect Of Windows
rdesigns wrote:
Your point about the overall envelope R-value is supported by the REScheck program offered by DOE as a means of determining compliance with the Energy Code.
As an alternate to using the cookbook, or prescriptive path, in the Code, you can use REScheck to enter R-values and U-factors of the various elements of the envelope (windows, doors, walls, ceilings and floors), and the program spits out a value for the total U-value of the building. As long as this U-value total is less than the total that the prescriptive path would provide, the design will meet the Energy Code.
In practical tems, this means that you could use better wall and ceiling insulation to offset the loss from a greater window area than would be allowed by the prescriptive path. Or, better windows could allow lesser amounts of ceiling or wall insulation.
Thanks for that information. It is nice to see that clarification. Generally, I have observed that there is considerable resistance to the idea of superinsulation. I interpret this to mean that the builder and owner simply do not want to do any more that what is necessary. It is complicated and costly enough just to put up a building.
So this leads to the need for a technical argument against insulating above just a minimal requirement. And the most convenient argument is that it is not cost effective to insulate beyond what has always been done traditionally.
The most focused detail of this insulation-limiting argument is this:
Building insulation R-value has no effect above the R-value of the lowest R-value of any window in the building.
So what that means is that if you have one little window at R-3 in a house, then there is no point in insulating the whole house above R-3.
Building insulation R-value has no effect above the R-value of the lowest R-value of any window in the building.
I don't recall anyone here saying that. But, with any structure there does come a point where additional insulation is pointless -- the benefit per added inch or whatever becomes vanishingly small. And this point is reached much sooner with a large number of windows or other major leakage paths.
where does the r value end up
Here is a real world scenario.
I have a bedroom with R14. During the daytime the temp reads 19C During night the furnace is off alltogether. At 10pm I open the window (size 72"x36") 6" to the outside temp of 2C. At 8am the room temp reads 17C
Ceiling insulation is R40
Ok you energy engineers go to work
As I said earlier, a lot has to do with the movement of the air. If there is no strong air current there will be relatively little heat exchange. In particular, by opening the window "just a crack" as you do, you assure that any airflow will be only one direction, and unless there's another opening in the room (or elsewhere in the house, with the door open) then little air will flow. (Or, in some cases, the net airflow may be outward, drawing in warm air from the rest of the house.)
Heat exchange is not instantaneous -- a lot depends on the airflow within the structure.
As an EE, I tend to think of heat loss problems using an electric circuit analogy:
Let's say you have a house with a total of 100 square feet of surface area (it's a very small house). Initially it's not insulated, and that represents an effective R value of, let's say, 2.0. So my electric circuit analogy is 100 2 ohm resistors in parallel.
Let's say that it's 70F inside and 20F outside -- 50 degrees difference. We'll call that 50 volts. 50 volts across a 2 ohm resistor is 25 amps, and multiply that times the 100 parallel resistors and you have a whopping 2500 amps.
Now we add R-11 fiberglass to our walls, bringing the total R to 13. 50 volts across 13 ohms is 3.85 amps, or 385 amps for the entire house.
Next let's assume 3.5 inches of polyurethane at R-24 plus the original 2 for R-28. 50 volts across 28 ohms is 1.79 amps, or 179 amps for the entire house.
But let's give our house a window. Let's say a low-E argon-filled window (U-0.33) that's 4 square feet. U-0.33 is R-3, so our house is now consuming 96 x 1.79 amps plus 4 x (50/3) amps (ie, 4 x 16.7 amps) for a total of 171.8 + 66.8 == 238.6. So the window that occupies only 4% of the surface area of the house is accounting for 28% of the heat loss.
OK, so double the insulation to 7 inches of poly -- R-48 plus the original 2 to R-50. One amp per square foot, so we have a total of 96 amps for the non-window portion of the house, and still 66.8 amps for the window, for a total of 162.8, with 41% of the heat exiting through that window that still only occupies 4% of the total surface area.
Double the insulation again to 14 inches. Now we have 49 amps + 66.8 amps = 115.8, with 58% of the heat escaping through our window.
What did we gain through these steps? Going from the original R-2 wall to the R-24 poly wall (figuring the window in both) would have saved us 2228.2 amps, or 637 amps per added inch. Going from 3.5 to 7 inches saved us 75.8 amps, about 22 amps per added inch. Going from 7 to 14 saved us 47 amps, about 6.7 amps per added inch.
That sure looks like a "diminishing return" to me.
What you said!
as an EE
And all this would be moot if you had a heat source that could supply endless energy to keep you warm - Sun
But I also read that there is a diminishing factor adding more insulation after a "certain" point.
Maybe we should blow 5 foot high cellulose into the attic.:=)
Design has in impact. I look at these high-end custom homes on the Vancouver waterfront . High R values specified.
Complete endwalls consisting of glass (for the view). Been in them. They felt cold and uncomfortable in the wintertime. And it is not even cold here
Two feet of cells in the attic isn't that uncommon. But it's cheap and easy to do without having to distort the structure. Plus insulation in the ceiling is somewhat more effective than insulation in the walls, especially in the summer.
Expansive Windows
People who want to bring the outdoors into their house by the use of glass, and still want a highly insulated building envelope, are in a pickle. I wonder if there is heat loss by radiation through glass to cold objects outside. Normally, one thinks of glass as being a potentially relatively cold surface because of its low R-value.
But say you somehow heated the glass to a temperature higher than the indoor ambient temperature, and yet it was very cold outside. Would heat transfer through the glass to that colder area by radiation even though there was no potential to transfer to the warmer glass surface itself?
I'm sure you know the various ways heat moves, not just conduction. Low e is common, with several configurations.
No pickle, "highly insulated" is relative. I'm normally involved with houses with large glass, several choices available. Our place has 28% of square footage in glazing area. This accounts for the majority of our heat loss. Uncomplicated to quantify. My choices are to live with it or use moveable insulation of some sort. Substantially better glazing is available, but very expensive. Amory Lovins demonstrated a north facing window with net gain, not inexpensively.
We are your "people who want to bring the outdoors into their house". We prefer no window treatments and it costs us 3º inside on an annual basis. An easy choice when our heating/cooling is so cheap (passive heat storage).
These are simple decisions, not unlike return on investment for additional insulation (which can easily be never). Thermal modeling makes those choices clear though we don't all choose for the same reasons. For instance, a Georgia house nearing completion will have 1/2" thick glass with argon between the 2 panes. That thick glass has significant transmission reduction due to reflectance. Less light entering, both darker and colder inside, not to mention rather expensive. Nothing I would want, but the owner decided that the thicker glass was justified, for him. He also chose large columns outside his glazing, giving 13% shading. It will still be a bright house inside. To get the building permit we were required to prove the house would maintain 68º in January. No problem.
After I built our place I did heat calculations and discovered that my wall insulation was excessive. The next house, with similar glazing, got less. Less than 1º of difference inside from my place. Not huge savings in insulation cost, but no reason to waste it.
Comfort inside is greatly affected by structure temperature. We use extreme high mass. My aged mother is comfortable here so long as I don't tell her what the air temperature is. She keeps her low mass house 7º warmer in winter than she finds perfectly comfortable here.
Comfort being the primary goal. And not everybody wants their neighbors peeking in windows. Not a problem for us.
The reason I use the word, "believe" is because in the final analysis the computation is too complex to yield a firm conclusion. From practical experience with what I have built so far, I would say that the cut-off point of diminishing return is further out than I expected. So I would go further the next time. I realize that heat loss can be calculated, but I don't see how a person comes to a specific conclusion as to what amount of insulation is enough. It depends on the future cost of money and the future cost of fuel among other things. It depends on how long a person plans to live in the house. It depends on the cost of materials and labor beyond just the cost of insulation and labor to install it.
I don’t want to use spray foam, and I have determined that I want higher R-values than what I can get with a 2 X 6 walls and solid rafters in a vaulted ceiling. So I will make the jump to double studs and scissors trusses. Once I spend the money to go to that stage, I don’t want to skimp on insulation. At that point, there may be diminishing returns on the insulation, but adding insulation cost is nearly irrelevant compared to the extra cost of getting the framing into the double stud and scissors truss design. So I will err on the side of too much insulation rather than too little. With this reasoning, I land at R-50 walls and R-90 in the ceiling.
I could crunch a bunch of numbers and it might say that R-80 in the ceiling is enough. If it said that, I probably would go to R-90 just to be sure. There would be so many variable inputs to the computation that it probably would suggest a range of possible thresholds. There would also be a fairly large margin for error.
When I say there is a threshold where adding more insulation yields no benefit, I am referring to the very diminishing return threshold we are talking about where one decides that the added benefit is not justified by the added cost. But I agree that the insulation return or benefit does not actually stop.
Looking For The Threshold Of ROI For Insulation
The reason I use the word, "believe" is because in the final analysis the computation is too complex to yield a firm conclusion. From practical experience with what I have built so far, I would say that the cut-off point of diminishing return is further out than I expected. So I would go further the next time. I realize that heat loss can be calculated, but I don't see how a person comes to a specific conclusion as to what amount of insulation is enough. It depends on the future cost of money and the future cost of fuel among other things. It depends on how long a person plans to live in the house. It depends on the cost of materials and labor beyond just the cost of insulation and labor to install it.
I don’t want to use spray foam, and I have determined that I want higher R-values than what I can get with a 2 X 6 walls and solid rafters in a vaulted ceiling. So I will make the jump to double studs and scissors trusses. Once I spend the money to go to that stage, I don’t want to skimp on insulation. At that point, there may be diminishing returns on the insulation, but adding insulation cost is nearly irrelevant compared to the extra cost of getting the framing into the double stud and scissors truss design. So I will err on the side of too much insulation rather than too little. With this reasoning, I land at R-50 walls and R-90 in the ceiling.
I could crunch a bunch of numbers and it might say that R-80 in the ceiling is enough. If it said that, I probably would go to R-90 just to be sure. There would be so many variable inputs to the computation that it probably would suggest a range of possible thresholds. There would also be a fairly large margin for error.
When I said there is a threshold where adding more insulation yields no benefit, I was referring to the very diminishing return threshold we are talking about where one decides that the added benefit is not justified by the added cost. But I agree that the insulation return or benefit does not actually stop.
I like your r values!
It just dawned on me while working today why commerical freezers aren't better insulated - while it would seems businesses should look to long-term paybacks, they really don't function that way. A manager paying twice as much for thicker freezer walls doesn't look as good at the end of the year bonus time as the manager that saved that up front money but will pay higher bills year after yaer.
Yoour knowledge of how business managers are typically reviewd and how capital investments are accounted for is lacking. Investing in an efficient freezer would not impact the profit of the manager's business unit as you suggest because it is a capital investment not an expense. In the long run his operating expenses could very well be reduced by making the investment and as a result he would look better. Business are not stupid and do not ignore returns on investments.
I know plenty of managers who make decisions like this every year and they go cheap on things that would save them money in the long term so their balance sheets look better at the end of the year. If you don't think this actually exists you haven't been around many managers. Capital investments are not blank checks - many business owners I know would laugh themselves silly if one of their managers wanted to start spending money on improvements with 10 year paybacks.
I'll eat a dog turd if someone can show that businesses are operating only in the most long term efficent way possible! lol How many business are holding off spending money on more efficent ways of doing business because they don't even know if they'll be around 10 years from now? A bunch of them!
you can believe all the mythology you wish. And you can fabricate a fantasylands in your mind to vilify business or call them stupid if you like. Of course it is possible you surround yourself with stupid people. I assure you Corporations do look at returns on investments. I have work for many of them for over twenty-five years.
It depends...
"How much is enough" depends on what you are trying to achieve. If you;re designing to a lower deltaT, then less might be more. Way up in the Great White North, more might be more. Heat flow is for the most part a proportional equation. Double the R-value of the insulation and your BTU requirements drop by half.
If you use Fourier's equation, it's fairly simple:
Q = (A x deltaT)/R
Q is the heat loss in BTU/hr
A is the area of the assemby in sqft
deltaT the temp differential from inside to outsde in F
R is the R-value of the wall
A basic non-insulated wall, between wood thickness and air films, comes in a a little less than R-1. If we have 1000sqft of wall (a simple example, no foundation or roof calculations) and a temp differential of 40F:
Q = (1000 x 40)/1 = 40000 BTU/hr
Add 1" of foam at R6, costing $1/bdft, and Q = 6700 BTU/hr. For $1000 you reduced your leat loss by about 83%.
Add another inch of foam, and Q = 3300 BTU/hr. For another $1000 you reduced your heat loss by 92% when you compare it to no insulation, but it only gained you another 9% over that first inch of insulation.
Add a third inch of foam, your new Q = 2200 BTU/hr. Now you're down to 94.5% of your original loss, but that third $1000 only gained you an extra 2.5% overall.
Add a fourth inch of foam and Q = 1700 BTU/hr. You're at 96% less than your original value, but only 1.5% better than your 3" value.
Doubling your insulation thickness halves the heat loss. I rounded off the numbers above, but you can see going from 1" to 2" halves your loss from 6666 BTU/hr to 3333BTU/hr.
Going from 2" to 4" again halves it, from 3333 BTU/hr to 1667 BTU/hr.
Heat loss calcs for a structure are cumulative, in it's most basic form you simply break the house down into all of it's varying R-value components, and based upon the square footage of each of those components, you come up with a basic heat loss number for the entire structure. In the above house, we have no windows. Using the 2" of insulation number, the house has a loss of 3300BTU/hr. If we replaced 25% of the wall structures with R-3 windows, then we'd have the heat loss for the house equaling the sum of the wall loss plus the window loss:
Q = (750 x 40)/12 + (250 x 40)/3 = 2500 + 3300 = 5800BTU/hr
If you left a window partly open, and for a basic example I'll use R=0.25 for the open window, then Q = (750 x 40)/12 + (249 x 40)/3 + (1 x 40)/.25 = 2500 +3320 + 160 = 5980 BTU/hr. But that's all theoretical. That open window won't drain your house of heat unless you have no walls or doors within the house. Plus the mass of your house retains heat, so while you are losing BTUs through the window, the room is being recharged by the mass of the house. So even in that one room, the temp may drop, but not neccessarily to the outside temp. Again, I'm taking liberties with these basic examples.
So yes, when you compare dollars spent versus BTUs saved per dollar, your first dollars have the most bang for the proverbial buck.
The first $1000 saved you 33000 BTU/hr, or 33 BTU/hr/$
The second $1000 saved you an additional 3300 BTU/hr, or 3.3 BTU/hr/$
The third $1000 saved you an additional 1100 BTU/hr, or 1.1 BTU/hr/$
The fourth $1000 saved you an additional 500 BTU/hr, or 0.5 BTU/hr/$
And for grins, going to 5"? The fifth $1000 spent would save you another 330 BTU/hr, or .33 BTU/hr/$
And the sixth inch, the sixth $1000 would save you an additional 220 BTU/hr, or .22 BTU/hr/$
Diminishing gains? Yes. But when to stop depends on what you are designing for. You might be trying to save a little money over the life of the house. You might be shooting for a passive house. You might be trying to not have to upsize your heating or cooling system.
This is a very good explanation - Thanks!
Here is another resource:
http://www.energystar.gov/index.cfm?c=home_sealing.hm_improvement_insulation_table
Michigan is Zone 5 or 6.
More To Think About
While there are diminishing returns on each added inch of insulation, is it fair to assume that the price increase for each added inch is linear? I would say that each additional inch of insulation costs less that the previous inch. So there will be diminishing cost per inch of insulation that will help offset the diminishing performance return per inch of insulation. So perhaps the performance return per dollar of cost is not really diminishing at all. That part needs to be in the calculation too.
Knowing where to stop is the key, but how do you decide? The law of diminishing returns does not answer that question. It only offers discouragement like somebody bidding against you at an auction. Because it make it seem like you got such a good deal on the first inch of insulation, so you are getting a bad deal on the last inch. Another way to look at it is that since you paid all those dues on getting the R-value so high, why stop before the job is done?
Adding additional inches of insulation might cost less per inch than the first inch, then again, it may not. As in all construction, it depends on how the structure is/was built. Filling stud cavities with insulation requires more material but little additional labor. But what happens when the cavity is filled and one still want more insulation? Then the price per inch takes a big spike upwards.
it depends yet again...
"While there are diminishing returns on each added inch of insulation, is it fair to assume that the price increase for each added inch is linear?"
Again it depends. There are foam installers who will shoot 5" of foam in a single pass. More foam with less labor, you might get a discount per inch after the first few inches. Some do, some don't. There are others who won't shoot that depth in a single pass due to too much heat being given off in that deep of a spray. So those crews might make two complete passes over the cavities and shoot 2-1/2" each time. There you have more labor than tthe first crew. So yes, when you go for more depth of foam, there might be one set-up and one spray, or there might be one set-up and two sprays. You'll have to ask your local crews what they charge for varying depths. But a buck a board foot is a fair generality in my area. You can easily substuture different costs in the above calculations. Again, everything I wrote above was one huge general example.
But big-picture, remember, foam cost versus prevented heat loss will never be linear. You're halving your loss of heat each time you double the thckness of spray. But your foam cost won't be halved. So dollars spent per BTU saved will still increase the deeper you spray.
"Knowing where to stop is the key, but how do you decide?"
Again, that has to be determined by what yo are designing for. One the one hand if you're looking to minimize construction costs but you really want a flash coat of foam for air sealing, then there you go. If your goal is a net-zero type of house and you're trying to walk a fine line between winter heat loss versus passive solar gain, then more foam may be justified to get you over that thermal BTU loss-versus-gain hump. Then there is everything in between.
In conventional stick construction, my personal preference would be for foam to be on the exterior of the sheathing. Say two 1" thicknesses of rigid polyiso board with the seams offset and taped. That takes care of most of your thermal bridging, you'll simply have a run of screws through the foam for furring strips for the siding. It gives you the benefit of better overall insulation, plus a rain screen.
2" of polyiso in New England is thick enough to keep the dew point within the foam thickness. I'd fill the framing bays on the inside with dense-pack cells, or as a last choice, FG batts.
In an earlier post I think it was you who expressed concern over proper foam sprays to achieve proper curing. The use of rigid foam board eliminates that concern, and it assures you that when you spec out 2" of foam, you get 2" of foam. Not 1-1/2" here and 2-1/4" there.
Exterior Foam Board Question
Mongo wrote:
In conventional stick construction, my personal preference would be for foam to be on the exterior of the sheathing. Say two 1" thicknesses of rigid polyiso board with the seams offset and taped. That takes care of most of your thermal bridging, you'll simply have a run of screws through the foam for furring strips for the siding. It gives you the benefit of better overall insulation, plus a rain screen.
2" of polyiso in New England is thick enough to keep the dew point within the foam thickness. I'd fill the framing bays on the inside with dense-pack cells, or as a last choice, FG batts.
In an earlier post I think it was you who expressed concern over proper foam sprays to achieve proper curing. The use of rigid foam board eliminates that concern, and it assures you that when you spec out 2" of foam, you get 2" of foam. Not 1-1/2" here and 2-1/4" there.
I understand the objective of exterior foam reducing thermal bridging, and eliminating the need for a warm-side film vapor barrier by stopping the outward flow of vapor before it can reach the dew point. Wouldn't the vapor pass through the foam board joints until it reaches the tape on the exterior side? And wouldn't condensation therefore occur in each of the foam board joints?
I have wondered about this effect. It seems like the joints should be taped on the warm side of the foam for the proper performance. You mentioned using two layers of 1" foam and staggering the joints. I can see how that would help because it would stop the vapor half way through the foam layer. But isn't it common practice to use just one layer of foam board with taped joints?
This is really a good thread. I'm learning a lot about insulation. Still searching and have found some good graphs at last.
http://www.insulright.com/PDF%20FORMS/economical%20thickness.PDF
And a good, but very long thread along the same lines as this one on Green Energy Advisor.
http://www.greenbuildingadvisor.com/blogs/dept/musings/it-s-ok-skimp-insulation-icynene-says
And I'll betcha if you run the numbers on that first one the differences in the three models ("good", "better", "our stuff") is more due to the differences in ACH than due to the insulation R value.
(In fact, on re-reading, I see that's the point of discussion in both articles.)
Smoke and Mirrors
There is an exaggerated emphasis on air sealing, along with a corresponding de-emphasis on R-value of insulation. If follows a common perception that heat is warm air, and heat is lost through air leaks. This is reinforced by the fact that much of what is perceived to be cold drafts in house is actually radiant loss to cold spots on the walls or ceiling. It feels like cold air coming at you, but it is actually heat being drawn away from you by radiant loss.
Another explanation for this new focus on air leaks instead of R-value is that foam proponents have to compete with lower cost alternative insulation. So they de-emphasize the need for R-value, and emphasize leak sealing because foam is very effective at leak sealing.
I read the two links above, and do not believe the conclusions of the Icynene article with its charts and graphs. The second link to the Green Building Advisor blog is very informative, and I see that many there also distrust the Icynene article.
There is one issue that I have never understood about the doubling of insulation halving the heat loss. That is the issue of the starting point. If you start with zero insulation, you have 100% heat loss. How do you double zero? If you don’t start with zero, where do you start? The value of the starting point makes a big difference in the full analysis.
If this were only a scientific problem, it would be confusing enough. But there is plenty of spin introduced by competing insulation manufacturers. There is also plenty of misunderstanding and miscommunication of the details of insulation, vapor control, and air infiltration. And what is just beginning is a massive body of new bureaucratic regulations to tell us all how it must be done. There is nothing like a committee to come up with the right answer to a technical problem.
The starting point is never zero.
You wrote: "There is one issue that I have never understood about the doubling of insulation halving the heat loss. That is the issue of the starting point. If you start with zero insulation, you have 100% heat loss. How do you double zero? If you don’t start with zero, where do you start? The value of the starting point makes a big difference in the full analysis. "
Since we are considering insulating a structure, one must assume there is something to start with. In his example, Mongo started with sheating, studs, drywall, etc that has an assumed R value of 1. He could have assigned it any other R value and the example would still be sound. Even if the structure is simply a nylon tent, that layer of nylon will have an R value. It will be small, but it won't be zero.
Diminishing Returns Starting Point
I understand, but just for clarification, say we just consider thickness of insulation and heat loss. If you start with one inch of insulation, there is some degree of heat loss. Now double the insulation to 2" and the heat loss is cut in half. Now run that out further and look at the result of diminishing returns.
Okay, now start over with a new starting point. Let's say start with 1/16-inch of insulation. I don't know what percentage of heat loss that would be preventing, but it must be a very small portion. Let just guess that it is stopping 5% of the heat loss, so there is 95% heat loss occurring. Now double that insulation to1/8-inch. Adding that 1/16-inch to the first 1/16-inch supposedly cuts the heat loss in half. Therefore, adding 1/16-inch of insulation reduced heat loss from 95% to 47.5%. Is that correct?
If you run this out further and look at the result of diminishing returns, it presents a much different picture than the scenario of starting with 1” of insulation. It diminishes sooner than starting with 1/16-inch.
to clarify...I hope!
KDESIGN wrote:
I understand, but just for clarification, say we just consider thickness of insulation and heat loss. If you start with one inch of insulation, there is some degree of heat loss. Now double the insulation to 2" and the heat loss is cut in half. Now run that out further and look at the result of diminishing returns.
Okay, now start over with a new starting point. Let's say start with 1/16-inch of insulation. I don't know what percentage of heat loss that would be preventing, but it must be a very small portion. Let just guess that it is stopping 5% of the heat loss, so there is 95% heat loss occurring. Now double that insulation to1/8-inch. Adding that 1/16-inch to the first 1/16-inch supposedly cuts the heat loss in half. Therefore, adding 1/16-inch of insulation reduced heat loss from 95% to 47.5%. Is that correct?
If you run this out further and look at the result of diminishing returns, it presents a much different picture than the scenario of starting with 1” of insulation. It diminishes sooner than starting with 1/16-inch.
I might be misunderstanding you, so let me write this out:
The two plots (one starting at 1/16" and doubling and the other starting at 1" and doubling) are actually part of the same heat loss curve. However, if you want to compare the slope of the first four plotting points when you start doubling from 1/16" (1/16, 1/8, 1/4, 1/2) to the slope of the last four plots when you start doubling from 1" (1", 2", 4", 8"), then yes, the slope of the 1/16" to 1/2" plot will be steeper than the slope of the 1" to 8" plot. Your BTU savings per doubling is greater per double with your initial doublings than it is in your later doublings. Your savings per doubling is higher initially because there are so many BTUs being lost, so there is more to save.
The savings curve flattens out as your foam thickness increases and that's where you get your diminishing returns, that's where the curve flattens. There are fewer BTUs to be saved later, so you're getting less bang for your buck the thicker your foam is.
Again, 10000sqft of wall with deltaT of 40.
This time I'll add the R1 of the uninsulated wall into the insulated numbers and I'll double the thickness of the insulation each time, starting with 1/16" of insulation like you mentioned:
Uninsulated R1 Q = 40000 BTU/hr
1/16" = R1.375, Q = 29090 BTU/hr, a 27.3% reduction from R1
1/8" = R1.75, Q = 22857 BTU/hr, a 42.9% reduction from R1
1/4" = R2.5, Q = 16000 BTU/hr, a 60% reduction from R1
1/2" = R4, Q = 10000BTU/Hr, a 75% reduction from R1
1" = R7, Q= 5714 BTU/hr, an 85.7% reduction from R1
2" = R13, Q= 3077 BTU/Hr, a 92.3% reduction from R1
4" = R25, Q= 1600 BTU/Hr, a 96% reduction from R1
8" = R49, Q = 816 BTU/hr, a 98% redcution from R1
You can see that going from an uninsulated R1 wall to adding 1/16th" of foam, that first 1/16th" of foam saved you 11000 BTUs. The second 1/16th" that took you to a total thickness of 1/8" of foam saved you another 6250 BTUs.
However, going from 4" to 8", that last 4" of foam saved you less than 800 BTUs.
So initially your savings curve is steep. You're losing a lot of BTUs in the beginning, so your initial increases in R-value give you substantial savings. Later on when the foam gets thick, you've already stopped most of your heat loss, so there are fewer BTUs to be saved. That's where your curve flattens.
So the last part of your post that I quoted where you wrote : "If you run this out further and look at the result of diminishing returns, it presents a much different picture than the scenario of starting with 1” of insulation. It diminishes sooner than starting with 1/16-inch. "
If I'm correctly interpreting what you wrote, yes, you are correct. Starting with 1" of foam and doubling from there, your "diminishing returns" shows up faster, because so much of your BTU loss has already been reduced with your first few doublings from 1/16" to 1/2". It reinforces what folks have been writing, that your initial thickness of foam gives you the biggest bang for the buck.
Using my numbers above, going from zero foam to 4" of foam reduces your heat loss by 96%. Adding four more inches for a total of eight inches will only save you another 2%. Diminishing returns.
KDESIGN,
You say "If you
KDESIGN,
You say "If you run this out further and look at the result of diminishing returns, it presents a much different picture than the scenario of starting with 1” of insulation. It diminishes sooner than starting with 1/16-inch"
No it does not. Lets assume at 1" you loss 90% of the heat. AT 2" you loss 45%. But what does that mean? Let's assume that a 90% loss is a loss of 10,000 units of heat (what ever a unit is does not matter), then we are saying we loss 5000 units at 2 inches.
Ok, now if we start the discussion at 1/16 of an inch we loss 160,000 units, at 1/8 - 80,000, at 1/4 - 40,000, at 1/2 we loss 20,000, and again at 1 inch we loss 10000 units.
You can not approach the analysis on a percentage basis. Percentages are a nice way to conceptualize the dimishing return aspect, but not actual heat losses.
I am a believe in air sealing because of this: When we resided about 25 years ago, replacing old hardboard siding with new hardboard siding (that was actually slightly thinner and therefore probably lower R value), the one insulation-related thing we did was to install Tyvek, taking great (obsessive) care to tape and seal every joint, tape around every window, etc. We did nothing else to seal things other than to caulk the new siding. This was a house built in 1976, with brown-board sheathing tightly stapled, so reasonably "tight" for the pre-Tyvek era.
The net result was a house that was dramatically warmer and more comfortable in a Minesota winter -- the dining room, which previously (being on the NW corner) was virtually uninhabitable, became our favorite place to hang out, and overall "drafts" essentially disappeared. We didn't compare heating bills (hard to do with the variability of Minnesota winters anyway), but I would guess that our heating bills dropped 20-30%, since the house could be comfortable at a lower temperature, in addition to actually cutting heat loss.
So, while I don't know what the numbers would be for a new house, obstensibly properly sealed and wrapped, I do know that it made a dramatic difference in our case, and I would not be surprised that air sealing could make a major difference in even an "well sealed" "modern" house, where the builder put forth the usual "minimum to pass code" effort.
Fiberglass
NRTRob wrote:
- Being airtight is very important. "Seal it tight, vent it right". Dump the fiberglass.
Why dump the fiberglass?
I'm actually rather shocked that many people are still slow to accept the importance of tight construction - it's like they've never heard of blower door testing and how it's relatively easy to put concrete numbers to the value of sealing everything well.
Along the same lines, you can't have a super tight house without covering the subject of air quality including venting, replacement air (air heat exchangers or not), and whatnot.
Airtight
IdahoDon wrote:
I'm actually rather shocked that many people are still slow to accept the importance of tight construction - it's like they've never heard of blower door testing and how it's relatively easy to put concrete numbers to the value of sealing everything well.
Along the same lines, you can't have a super tight house without covering the subject of air quality including venting, replacement air (air heat exchangers or not), and whatnot.
Do you run into a lot of people who do not accept the importance of tight construction? I know for a while, when I told people that I wanted airtight construction, they would always say no, you must let the house breathe. My objective is to get as close to 100% airtight as possible and then ventilate with an air-to-air heat exchanger.
But I have also run into a line of thinking that airtight construction is a substitution for insulation R-value as seems to be the premise being put forth by Icynene and their spray foam.
I'd be suprised if 1 in 20 houses is even close to being well sealed - it's like extra insulation and people are talked out of the extra upfront cost to their own detriment.
I do wonder what the climate is where makeup air heat exchangers are no longer reducing the overall cost of ownership over the life of the building. That depends somewhat on usage and other factors (windy areas especially), but I can't recall ever hearing a discussion that had anything that addresses it.
But I have also run into a line of thinking that airtight construction is a substitution for insulation R-value as seems to be the premise being put forth by Icynene and their spray foam.
Look at it this way. If you can, for the sake of argument, tightly seal a house with R10 and get the same heat loss as R20 without sealing, about half of that R20 is being wasted.
Ignoring Icynene's parochial interests, when sealing makes that much difference, additional insulation (for the unsealed structure) isn't really going to be very effective.
The proper statement is that "More insulation is not a substitute for airtight construction."
Airtight + Insulation = Happiness
DanH wrote:
But I have also run into a line of thinking that airtight construction is a substitution for insulation R-value as seems to be the premise being put forth by Icynene and their spray foam.
Look at it this way. If you can, for the sake of argument, tightly seal a house with R10 and get the same heat loss as R20 without sealing, about half of that R20 is being wasted.
Ignoring Icynene's parochial interests, when sealing makes that much difference, additional insulation (for the unsealed structure) isn't really going to be very effective.
The proper statement is that "More insulation is not a substitute for airtight construction."
Yes, I understand those points. You seem to believe that I am advocating insulation, but not air sealing. But on the contrary, I certainly do believe that more insulation is not a substiture for airtight construction. And I would also say that more airtight construction is not a substitute for insulation. Neither airtight construction or insulation is a substitute for the other. They complement each other like salt and pepper. You need them both.
However, the earlier point I was making was that I believe many are indeed putting forth the idea that air sealing is a substitute for insulation. If one brings up the idea of going beyond just average R-values, for instance, half the world is right there trying to talk you out of it. And yet they are so concerned about air sealing that they think you need to use foam because a film vapor barrier just cannot be made perfect enough.
Yeah, I'm no big fan of foam -- it has its place, but it's no cure-all. I just feel that often "balance" is lost when people go overboard with one of the three:
Sealing
Insulation
Windows and/or the lack thereof
to the exclusion of the others and to a point where one is "polishing a turd" -- trying to make up for shortcomings in other areas by being excessive in one.
Not if you do what I do...
Edit: I musy have mis-clicked, this reply is to Kdesign:
"Wouldn't the vapor pass through the foam board joints until it reaches the tape on the exterior side? And wouldn't condensation therefore occur in each of the foam board joints?"
I'll install the sheets with a slight gap. Say 1/4"-3/8" or so. Then I'll seal the gaps with foam. Then after the foam is cured, I'll shave it flush, and tape with Al-faced tape.
Ever since Carter took his peanut wisdom to Washington, I've been listening to drivel about 'energy' and 'insulation.' I want to remind all that insulation is only part of the picture.
Someone mentioned a walk-in cooler. Anyone ever do some serious work in such a place? Not very nice ... in no time at all the walls have a sheen of moisture, the air is muggy, and it's no fun to breathe. Why? In a word: ventilation. You're breathing your own bad breath.
Again, the 'engineers' here should have stayed awake during their core curriculum. The amount of insulation you need is directly related to the temperature difference you want to maintain. You'll want a lot more in +120 degree Arizona and -40 degree North Dakota than you'll want in relatively mild 60 degree San Francisco. Double the temperature difference you want to maintain, and you'll want four times the insulation.
Insulation does a lot more than just insulate. It can, according to the materials and the design details, deaden sound, stop air and moisture movement, stiffen the walls, and affect the fire resistance of the structure. It can also have a direct effect on any mold, odor, argon, and vermin issues.
Let me put this in 'real world' terms:
I'm doing a complete-gut remodel of a 1957 house. After a rather drafty winter, I wasn't all that surprised to find significant gaps in the insulation of the 2x4 walls. There was extreme air leakage from the original windows. Various construction details let quite a bit of heat to flow directly into the attic by way of the wall cavities. Some sections of wall were such heat losers that you could feel the draft as a breeze near them. Other areas were constantly wet or severely rusted, as the result of water vapor condensing on the cooler surfaces.
That's not to say everything needs to be 'air tight.' Making that effort results in trapped moisture and mold. If we learned nothing else from the 70's, it was that fact. No, we have to CONTROL the ventilation. Fresh air in where needed, old air out where needed.
I would suggest that the biggest effect foam insulation has (as compared to other types) is to create an extremely effective air and moisture seal at very modest thicknesses. Nor does it take very much thickness, even in Arctic conditions, to place the 'dew point' well within the foam thickness - meaning there will not be moisture condensing on the face of the foam. That's why so many have 'discovered' the virtues of using a thin layer of foam, covered by additional fiberglass. Such a method is seen as taking advantage of the best virtues of each material.
Other construction details have a huge influence on the insulation you need. For example, a well vented attic has the effect of stopping all the sun's heat at the roof deck, effectively keeping your house 'in the shade' all day. A 'rain screen' wall siding detail has the effect of taking the 'wind' out of the wind-chill effect. Sceptical comments about windows aside, there's no denying the difference a good-sealing window makes. Even an unconditioned two-stage entry (think: enclosed porch) greatly reduced the losses from opening the door; that's why 'mud rooms' came about.
In short, look at the big picture, rather than seeking a magic bullet.
Your anger is misplaced and ... ah .... just wrong. If you have a moderately insulated house which would cost you $2000 to heat and you "super" insulate it to the point that it only costs you$500 to heat it, then how much would you or should you pay to drive that figure to $400. The fact is that you will never get teh figure to zero and each incremental step (say $100 savings) will cost you more and more than the last incremental step. Just a fact. So there is a limit beyond which further expenditures are just stupid.
Great thread, although a bit cumbersome
I believe Fine Homebuilding published an article about 2 years ago regarding the diminishing returns of insulation, specifically SPF, and charted results. I also highly recommend the BSC podcast regarding insulation. In it, people with more letters after their names than I have, make a point that, realistically, if the entire envelope of your home were R-5, you'd probably be pretty comfortable.
I am disappointed that the R-value arguments are still going on, and we haven't moved towards a more progressive means of looking at efficiency as it relates to insulation. Diminishing returns is a HUGE issue when it comes to SPF, and I see it on a daily basis. I own a spray foam truck, I buy multiple sets of drums per week, and I also own a house....what's in my attic? (NJ) .....5" Closed cell. Why?...because any more than 5" and even though it costs me half the cost per installed BF of SPF, the ROI would be so long, it doesnt make any economic sense. This is not because SPF is expensive, but because like many building strategies, alternative energy sources, etc....the graph trails off pretty quickly after a certain point.
I find it to be much more important to look at proper moisture management and air movement in buildings, than nitpicking R-values. When all elements of the building are considered; windows and doors, thermal bridging, extra framing factor, uninsulated headers, solid corners, etc....all the things found on various jobsites....yes, there sure are diminsihing returns. How long will it take to actually pay back that 4th inch of closed cell foam in a wall cavity? A long time!
When I walk into an ICF home with R-50 walls, for an attic, ceiling, and soundproofing quote, I may suggest R-values that get into the "superinsulated" range, but a well sealed stick framed home, with SPF in the walls, rims, rafters, (and sealed crawlspaces), at current code R-values, performs quite well.
My point is, instead of splitting hairs on R-values, from the SPF contractor's perspective, yes there are severely diminishing returns, R-values are only part of the issue, and a comon sense ROI view of insulation works.
I'll put in 2 cents here. The concept of diminishing returns is what the guy is talking about. If you graph heat loss vs. R-value, you will quickly see the curved line where at R-1, the loss is relatively high and it is a steep line that then curves and adding e.g. R-6 after you've already had R-19 adds very little reduction in heat loss over the first R-6 over the R-1.
A simple concept. However ... that's where it gets a little complicated. We have heat loss ... then we have economics. Energy codes have generally evolved over what is GENERALLY economical and practical to install. Attics require high R-values because they are easy (i.e. more economical ... less expensive) to insulate to a higher R-value w/out impacting const. cost.
During new construction, it makes sense to put the insulation in NOW ... for to pay to increase it later will be [economically] a much harder 'sell'. So I tend to tell people to spend as much as you can afford and w/in reason of being practical. Some people want to go the extra mile and super insulate. Some might want the extra floor area from 2x4 walls and use foam (which is more expensive).
Dan H has a good point about the windows. We insulate ... even super insulate the walls/ceiling but don't do much about the windows. So as you increase a few R-values above e.g. R-21 in the walls ... you still have a hand full of window sqft that leaks maybe 50% of the heat in the house ... I've done the calcs hundreds of times. The trouble is ... windows are expensive (relatively speaking) and options can be more limiting ... e.g. Heat Mirror glass (by Southwall Technologies) is a very nice product ... but it's not as readily available as common low-E glass. I've seen windows w/ R-values up near R-12 (you can go higher) ... But at a cost.
So you have a sieve ... pour water in it. It leaks out. Insulate your walls more ... and the leaks slows ... but it still has significant holes in it (your windows). Even if you plug all the wall holes, you still have the window holes. If you had a dam full of small leaks and large leaks ... which ones do you plug? ... given also that the large leaks are more 'difficult' (expensive) to plug.
These days high density fiberglass batts are inexpensive ... why would you install the conventional R-11 or R-19? Applying 1/2" or 1" foam on the exterior is fairly inexpensive, too and adds the thermal break. Common strategies that were unheard of e.g. in the 80's. Products and technologies change ... as do the economics of 'what's best'.
Pay your money, take your choice ... your economics is what you have to live with day to day. You can do all the theoretical analysis you want ... but YOUR economics is where it's at (and maybe that balances your sense of doing something better/right with the paycheck you bring home).
I promote taking a few wasted sqft out of the new house and using that $100+/sqft of savings to buy more insulation or better windows. Too many times I've see overly large houses being built where the owner 'can't afford' more insulation or better windows ... what BS!!
Sorry Idaho Don ... I must respectfully disagree and say what you say is BS!!!
I respect your point of view. However not everyone owns their house for the life of the house and a new buyer will likely be oblivious to 'super insulation' and the relative value of it.
Dan H had a good point .... You insulate e.g. the roof to R-60 and yet you use cheap low-E windows ... which you have to admit are the industry minimum at this point.
There is lots of ways to look at situations ... energy savings ... economics ... my personal finances ....
You super insulation freaks will go to the ends of the earth to add R gazzilion to the walls/roof and then put in cheap windows w/ a U-value of maybe 0.30 ... I could say that would be total BS!!! But I don't. You spend an extra 5 grand insulating walls to the extreme while you don't do the same to the more leaky part of the construction ... windows.
So BACK OFF!! Do you have windows in your house that are U-0.10? If you do then you live up to your 'super insulation' claim ... if you don't ... you maybe are BS.
If you plan to build with double studs for the sole purpose of getting more insulation value, then the cost of the additional studs and labor IS PART of the COST of your insulation. I find it hard to beleive that this added expense will ever yield a reasonalbe ROI.
I think there is a hesteria around insulation. If it costs you $2000 to heat a "base" insulated home and you can increase the insulation to get that figure to $1000, it is probabaly low hanging fruit. And to save that $1000 you could spend $10,000 and get a 10% return on the investment. ANd you likely would not ahve to spend $10,000 to get the savings. Now, you KNOW it is impossible to save another $1000 off that heating bill no matter how much you spend (or nearly so). SO now what does it cost to save teh next $250 off that bill? Then you have to look at whether you can save that $250 by spending less than say $2500 or $3500 depending on your taste for required rate of return. How much do double stud walls cost you. AND remember you loss floor space withthat approach, and therefer you have a bigger footprint which means more foundation expenses, more floor construction, a biggger roof, etc. WOW!!!! For what $250 a year?