BEopt modeling of exterior insulation energy savings
I created a BEopt model of a theoretical house to see what the energy savings would be with continuous insulation.
Size: 2,200 sq ft ranch with 2,200 sqft finished basement
Location: SE Michigan (Zone 5)
HVAC: ASHP
Base Scenario: R21 Fibergalss Batt, 2×6 24 in OC, OSB sheathing
Alternative Scenario: R21 Fibergalss Batt, 2×6 24 in OC, OSB sheathing, R12 polyiso
I ran BEopt in parametric mode without any other alternative parameters selected. The annual difference in site energy use was 437 kWh. Which is an annual savings of ~$83 at my electricity rates ($0.19/kWh). Or the equivalent of leaving a 50 watt incandescent light bulb on 24 hours a day 365 days a year.
Does this seem right?
I understand the benefits of keeping the sheathing warm and dry with continuous insulation, but it seems that those who prefer double stud walls have figured out how to detail walls with cold sheathing via air tightness and vapor retarders.
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Replies
If you are looking at the $83 number, I think you are. The $83.00 is the resalt of a very complex equation. The number of BTU the insulation saves given your local weather get converted to how much fuel is required to produce that number of BTUs. That quantity of fuel times the price you inputted adjuster over 20 years at the inflation rate you selected will give the program a saving number then BEopt take the cost of that insulation you enter in dollars per sqf and adds what it would cost to finance the insulation at the interest rate you selected and subtracts the cost from the saving and gives you the $83 dollars a year savings. There are likely a few more variable I forgot about. You selected a dozen of variables to get your $83 answer.
With BEopt just like any computer program, garbage in equals garbage out.
Do you have a local code you must comply with for wall insulation?
Most modern codes are very close to the breakeven point for insulation.
I agree wall insulation is expensive to install and the fuel savings going beyond code is small.
Walta
Hi Walta, the $83 is not the annual savings based on interest rate, cash flow, cost of polyiso, etc…
It is the annual energy saved, 437 kWh, times my electricity rate, $0.19 per kWh. I thought the raw amount of energy saved seemed low regardless of the more complicated payback calculations.
You do recall that to reduce the flow of heat thru the wall by 25% you need to double the walls current R value. You will run into the point of domination returns pretty quickly given how expensive wall insulation.
The fact is going from an R21 wall to an R33 wall is a relatively small change. So yes, I believe $83 is plausible.
https://www.energyvanguard.com/blog/the-diminishing-returns-of-adding-more-insulation/
Walta
"to reduce the flow of heat thru the wall by 25% you need to double the walls current R value."
Huh? Can you elaborate, that doesn't sound right.
“maine_tyler | Feb 04, 2023 08:59pm | #4
"to reduce the flow of heat thru the wall by 25% you need to double the walls current R value."
Huh? Can you elaborate, that doesn't sound right.”
You are correct my 25% number not accurate.
The number my be closer to 50% my point remains. There is not much heat flowing thru R20 and slowing it with more insulation get expensive in walls.
Walta
Kyle,
I’m almost at the end of the same journey in Canada. After spending weeks agonizing over BEOpt results, then getting similar from HOT2000, I could not figure out how a double wall investment made any sense.
My base plan was 2x6 with R20 plus 1.5” ci (R7.5). When evaluating adding a double wall, the ROI was three generations away. I could only assume I was doing something wrong or overlooking something. I had to make a decision, so asked an energy consultant to do a minimal assessment of my plans. Cost me $500 and I could then sleep with my decision.
My main take always.
1. A lot of the heating/energy savings are going from code resistant heat source, to a super efficient HP. By having the HP as my base case, I had already recognized huge efficiency savings. If you’re assessment has a HP, don’t overlook that the $/BTU is saving you so much.
2. Windows suck. We have 13% window to wall ratio. The energy analysis flagged that our best investment (over R20+R7.5ci) was triple pane windows, because double pane were R3 vs R6.5. That’s “double” the insulation for 13% of the walls. I was shocked that going double to triple pane was only ~10% addition $ for us, so ROI was easy decision.
3. The amount of additional wall insulation I had to add to make any significant ‘annual’ savings was crazy.
4. Air infiltration seemed to be the largest piece of the heat loss ‘pie chart’, so my hope is to get that as low as possible.
I’d price triple pane window upgrade, get R-value for them, and run numbers with improvements to see if that could be better investment. Windows are such a huge heat loss compared to walls. Cutting that huge heat loss in half can be more than ci or double wall savings, or close and more cost effective.
What I didn’t spend on a double wall is now set aside for PV.
Here’s a thread where I outlined my similar confusion.
https://www.greenbuildingadvisor.com/question/r24-rigid-foam-wall-vs-r28-double-wall
Thanks Mark, triple pane windows are definitely on my radar. I have triple panes in my current house and am sold on the increased comfort.
Just keep in mind it's not all about the money. I have high R-value walls and very air-tight. It's such a comfortable house. I find I keep the thermostat much lower than I normally would due to lack of cold walls and drafts. I have one room in a remodeled area that became a nuisance to get ductwork to, so I decided I'd do it later. I may never get to it, as it's comfortable with a little return air passing through.
I checked the results of a project I modeled in climate zone 6 a few years ago, about 1500 sq ft with R-49 in the roof, plus an insulated walk-out basement and a modest amount of high-performance windows with good orientation. Going from walls insulated with R-20 cavity insulation to R-20 cavity plus R-12 continuous exterior insulation saved 390 kWh/yr, or about $75 worth of energy. Going from R-20 cavity insulation to R-40 double stud walls saved about 650 kWh/yr, or $125.
While improving the wall insulation package may not have the best ROI, it's at least as good as putting your money into a bank CD or similarly safe investment, and it has other benefits: often the heating and cooling systems can be downsized; the loads for my Pretty Good House designs are 1/2 to 2/3 what is needed in a code-minimum house. With exterior insulation, the walls are kept warm and dry; I've been involved in hundreds of renovations and I can't recall one that has not had some level of moisture-related damage. Comfort is increased by warming the interior surfaces, which lets you keep your thermostat set a bit lower, which is hard to model but something we see consistently in practice.
My client for the house I modeled had this to say, which I put on my website last summer when we were having a stretch of very hot weather: "Our house stays below 70 on these hot days without any a/c. Even the second floor is cool. Thank you for designing this excellent house."
Thanks Michael for helping validate that I didn't do something wrong in BEopt.
I guess what I'm trying to separate is how much above code wall insulation really helps if you incorporate the other aspects of a pretty good house. Assume you have a basement slab with R10, basement walls with R15, R60+ cellulose in the attic, triple pane windows, correctly sized and engineered fully ducted heat pump, <1 ACH50 as a starting point. Will you really have a noticeable comfort difference in a home built with 2x6 R21 batt walls vs 2X6 R21 batts + 2" of foam?
BEopt is telling me that if I add a fridge to my garage, I've basically wiped out the energy savings from the 2" of foam. So is the comfort difference worth it? I assume air tightness and U value of windows will dominate the comfort side of the equation. I know these questions are impossible to answer, but it's helpful to hear others thoughts.
At least where I live the other aspects of a pretty good house aren't that hard to incorporate in a new build. But once you start talking about exterior foam or double stud walls, builders aren't interested or there are extreme markups.
Thanks everyone for helping me put this in perspective.
The PGH prescribed numbers are simply what Joe Lstiburek and company recommended as being reasonable targets for high-performance homes in 2010 or 2011, and a good target if you aren't using energy modeling, as few of us were ten years ago. When using energy models, as I now do on most projects, depending on the construction system and contractor, I often find that I can reduce some of those values a bit, such as going with R-30 walls, without making enough difference to matter.
Yes you will really have a noticeable difference if you go from code-minimum to PGH-type values for insulation. Whether it's "worth it" is subjective. I have heard a lot of people say that we should sell the added comfort, but not everyone is super sensitive to thermal comfort, or desire for their entire house to be the same temperature. So I also show what range of financial ROI to expect, how the extra insulation promotes better indoor air quality, and how it provides a significant boost to resilience. Depending on the HVAC system, the improved envelope can make a real difference in sizing, cost and performance.
If none of those elements are important to you, or important enough to seek out a contractor willing to do something slightly different from what your neighbors are doing, than you don't have to do it. Most people don't. But from what I've seen, once people understand the benefits and relatively small percentage of cost increase compared to conventional construction, most people want the higher-performance home.
Per Michael's post - exactly. ROI is not a good metric for all things. Comfort is a pretty good one.
Per Walta's first post, the GIGO principle is likely at play here. Much software behaves in a black box manner. It has it's own set of assumptions and (often wrongly) assumes that you know what they are or don't care that you know what they are. Not a good starting point.
Per other posters, this may seem a day late. So, in my usual over explained manner, off to the races.
Adding R12 exterior insulation to your standard 2x6 wall with R21 batts will almost halve your losses.
I have had no contact with BEopt and have no idea what assumptions they have built into the program. Maybe someone here has done a deep dive into the background assumptions of the program and can help define the critical inputs to pay most attention to. Instead, I think looking at two basic elements, the wall and window loss rates, when developing a preliminary house design is a good place to start getting your head around things.
Simple Wall Assumptions:
2200 sq foot print. 25% framing factor. CZ6, design point 0F. Interior 70F
2x6 (R1.5 per inch) = R-8.25 plus 1/2" ply and 5/8 drywall, so round to R-9.
Thru wall value is R-21 batt plus ply and drywall, so R 21.75.
Square root of 2200 is 46.9 ft each side of square, round to 47 ft x four sides 188 linear ft of 9' wall exposure yields 1692 sf. The framing factor of 25% means 423 sf of low R-9 and 1269 sf of high R-21.75.
At a delta T of 70 the low R area loses 3290 btu/hr
Same delta T of 70 the high R area loses 4042 btu/hr
Effective R value for 2x6 stud wall with just R-21 in bays is 1692 sf x 70 dT divided by 7332 btu/hr = R16.15 whole wall
Add the R-12 exo-insulation and the 423 sf of now R-21 would lose 1410 btu/hr
the now R-33.75 thru wall area of 1269 sf will lose 2632 btu/hr.
Effective R value becomes 1692 sf x 70 dT divided by 4042 btu/hr = R-29.3 whole wall.
The difference of 3290 btu/hr is quite close to 3412 btu/h or 1 kwh. 24 kwh a day into the 437 kwh difference you are seeing on the BEopt model 18-19 days of at design point weather. More realistically, a few months of actual weather.
As markmac has discovered, the windows and door losses frequently exceed the wall losses. Attempting to make up for the windows losses by increasing wall insulation is a often a bad bet economically. Doubling the the windows R value is much more useful simply because they start with such poor numbers that halving 10,000+ is both dramatic and cheaper as he found out.
In his case, 13% of wall area was 540 sf of R-3 windows, so the loss for a lower design dT of 65 would be 11,700 btu/hr. Bumping the windows to R-6.5 reduces the loss to 5,400 btu/hr. Not sure why the program told him 27k btu/hr. Coastal Canada should be CZ5 I think.
The proposed wall of 2x6 with R-7.5 exo-insulation does come in at R-24 whole wall, so 3690sf x dT65 divided by 24 = 9642 btu/hr. The 13% of R-3 window area is almost 2,000 btu/hr over the 87% of remaining wall. markmac noted that the upcharge on going to triple pane windows was 10%. I will guess about $4,000 usd.
To save 6300 btu/hr by upping the R value of the wall would mean taking the wall value from R-24 to R-65. Roughly 11" of EPS or 8" of polyiso added to the outside 3690 of a 2x6 wall with R-21 batts. Neither option is cheap or practical. Certainly more than $4k
It is my personal opinion that similar loss reduction avenues abound in foundation and slab design. Won't go there just now.
All the same, it is important to remember that all these btu numbers are the result of some design point driven by local conditions and/or code set values. Hot or cold, they are the maximum values. Life is spent in between dealing with much lower delta T's. The numerous local choices that need to be made when finalizing how one builds a home is obviously more complex than the simple prior analysis.
The thermal juggling event called home envelope design may best be handled by complex programs like BEopt, but I remain unconvinced.
“markmac noted that the upcharge on going to triple pane windows was 10%. I will guess about $4,000 usd.”
Yep, that’s about correct. It was a given to invest that vs chasing more wall.
As I noted, the other realization was that the HP efficiency was making the investment side of the decision tougher. I had to cut BTU/hr so massively to make up any significant savings at the efficiency of the HP.
As you noted, to get 6,300 btu/hr reduction would be a massive undertaking in wall design and investment. With local utility at $0.13/kw, the prices of PV reasonable, and the local utility buying back excess at 1:1 it’s hard to chase more R value once I know I have a comfortable amount.
Here's another reason why a good envelope is... good: here in Maine we just had a rare -20°F night with near-zero temps before and after. Stephen Sheehey, who often posts here, lives nearby in a Pretty Good House with R-40-ish walls. His heat pumps kept his house warm and cozy.
Meanwhile, good friends of ours nearby have a large house (~4000 sq.ft.) with excellent solar orientation and 2x6 walls filled with fiberglass. They recently had their electric baseboards taken out and mini-split heat pumps intalled as the sole heat source. They did not consult me first or I would have warned them that the heat pumps might have to work too hard in very cold temperatures to keep their interior up to temperature, and unfortunately I was right--the heat pumps couldn't keep up and they got down to 35°F INSIDE the house. Then they lost power, their generator wouldn't start and their on-call support service was too busy from taking care of all of the other poorly-performing houses to even return their call. They were able to get through the night but it was hairy.
During another cold snap about ten years ago, a friend of mine was out of state when they lost power for ten days during an ice storm. He had recently retrofitted his house to Passive House-level performance, and even with poor solar orientation, a wooded sight and mostly cloudy days, his house never got below 45°F.
A good building envelope is an investment in security and comfort. And can eventually pay for itself, though returns will usually be in the single digits or lower.
I’m just north of Maine and it was pretty chilly! Got down to -47F with windchill and settled in at -12F all night (w/o windchill). The HP worked down to -4F, albeit not as efficient. The resistant heating had to kick in for a few hours to ride through.
Code here won’t allow HP only, unless it has electric resistant backup. Friday night and Saturday illustrated why.
Kyle, I'm not sure how close to theory vs reality your schedule truly is, but I believe Michigan is currently evaluating an update to the 2021 IRC insulation minimums which would be R20+5 for climate zone 5 (or R30 cavity option) and R60 attic. I'm also in SE Michigan, so I've been keeping an eye on that. Once it's code minimum to hit R30 or do that +5, I imagine the options will get cheaper, as right now it's probably uncommon for a residential contractor to do exterior insulation around here.
Thanks Tim, that’s good to know. Timing is probably in the next few years, assuming some normalcy returns to supply chains and labor markets.
I would be building in a rural area, so I imagine enforcement would take a while.
That's true, it may take a while to roll out. We are planning for Washtenaw county, sort of rural, but not far from the action. Most of the townships seem to have plenty of inspections from what I've seen for Washtenaw. I've done a lot of reading on ordinances as we shop for property.
To answer your other question, the double stud walls deal with cold sheathing by allowing the sheathing to dry to the outside. By using polyiso on the outside, you are blocking that ability to dry.