GBA Logo horizontal Facebook LinkedIn Email Pinterest Twitter X Instagram YouTube Icon Navigation Search Icon Main Search Icon Video Play Icon Plus Icon Minus Icon Picture icon Hamburger Icon Close Icon Sorted

Community and Q&A

California roof insulation in the High Sierras – what would you do? R38 or R60

zetetic | Posted in General Questions on

TL;DR

We are building in the High Sierras, 6600 ft elevation, 9000 HDD (65F).  Some years we have less than 9000 HDD (65F), other years we have more than 9000 HDD.

We get upwards of 300 inches of snow most years.  Last year (2023) we were gifted with nearly 600 inches of snow.

Our record low is -24 degF in 1997.  During the winter, our average temp is 11F Jan/Feb.  In my lifetime, the lowest I once experienced was -5F at the house.

California Title 24 says R38 for an unvented cathedral ceiling/roof (no attic) for this location (California Zone 16, the most demanding)

IECC says R60 for the exact same roof, for this same location (IECC Zone 6b/7)

Just for fun, IECC says R49 for the same exact roof, located on the California coast (IECC Zone 3, think temperate San Francisco or Santa Barbara)

California tailors its Title 24 zones 1-16 to California without consideration for the other 49 states.  Zone 16 is fairly broad and based on more than just HDD.  So I’m not surprised that there’s a difference between IECC and California Title 24.  But I wasn’t expecting such a large gap!

What would you do?

Am I making a mountain out of a mole hill?  Is IECC overkill and too general, and I should just go with R38 and the wisdom of California, and meet local code and pass inspection?

Go for the IECC R60, required in the other 49 states because the folks at the DOE and Building Science Corp (Lstiburek) walk on water?

Or something else?

Quick References

IECC

IECC requires (Zone 6b/7):

R60 for the roof
R30 for the walls

https://codes.iccsafe.org/s/IECC2021P2/chapter-4-re-residential-energy-efficiency/IECC2021P2-RE-Ch04-SecR402.1.3

DOE and EnergyStar

Both DOE and Energystar align/follow IECC (no surprise, some of the same authors):

https://www.energy.gov/energysaver/insulation
https://www.energystar.gov/saveathome/seal_insulate/identify-problems-you-want-fix/diy-checks-inspections/insulation-r-values

California Title 24

California Title 24 places us in California Zone 16 (most demanding zone), and requires:

R38 for the roof
R21 for the walls

https://energycodeace.com/content/section-1501-performance-and-prescriptive-compliance-approa#ra-chunk–154151

Note: U = 0.048 = R21

IECC/DOE/EnergyStar (Zones 6b/7) vs California Title 24 (Zone 16)

That seems like a big difference – R60 vs R30 (roof), and R30 vs R21 (walls)

GBA Prime

Join the leading community of building science experts

Become a GBA Prime member and get instant access to the latest developments in green building, research, and reports from the field.

Replies

  1. Expert Member
    DCcontrarian | | #1

    I'd go with the thicker insulation.

    1. zetetic | | #4

      Thank you. I actually appreciate any sort of reply, because this is a hard question, maybe even unanswerable?

      My knee-jerk response was similar. "More is better" if it's in your budget and it doesn't create other problems in terms of trade-offs.

      I also respect Joe Lstiburek's work, which is behind some of the IECC standards, especially envelope related. I have his books in a prominent spot in my library and they are well dog-eared.

      On the other hand, might the IECC standard in part be Lstiburek playing the long game, pushing Passivhaus to be more than a boutique program?

      https://www.greenbuildingadvisor.com/article/will-passivhaus-remain-a-boutique-program

      Did California Title 24 get it so wrong that we're under-insulating? Or is R38 actually pretty good / good enough? After all, I'm going from an R4 roof (at best) which eats propane for breakfast, lunch and dinner - so R38 should be amazing, right?

      I've got a copy of "Pretty Good House" on order, arriving in the next day or so.

      R60 seems to be in the range of Passivhaus / PHI, which gives me pause. Overbuilding (if R60 is overbuilding) isn't necessarily green in all circumstances, with hard-to-justify ROIs for some of us.

      You'd think California would be nudging the state toward Passivhaus if they could. I wasn't quite expecting to see R60 coming out of the IECC and not California.

      It's tempting to ask, "Is R60 over-building, or R38 under-building?" Usually a question like this is settled by the building code, created by experts with lots of real-world input and years of education, experience, and experiments. But what do you do when two codes conflict created by their respective experts, and the difference is significant? Things can get spicy (!), philosophical and idealogical even.

      *** So the better question would be: ***

      "Does anyone have a home with an R38 roof in IECC zone 6? If so, is it comfortable and energy efficient enough, or do you wish you'd gone with more insulation in hindsight?"

      The success metric ("good enough"): you're not kicking yourself too hard, or complaining on a daily basis about that R38 roof.

      1. Expert Member
        DCcontrarian | | #5

        There is no stock answer, it's a cost-benefit analysis. To do that analysis you need to know the costs and the benefits. What's the cost difference between R-38 and R-60? Are there other factors, like having to redesign the structure?

        On the benefit side, how much are going to save on your heating bill? Have you done any energy modeling?

        Since the costs are today and the benefits are in the future, you have to somewhat discount future benefits and decide what your time horizon is. Energy efficiency improvements tend not to increase the sales price of a home, so for most people the time horizon is how long they intend to own it. For a spec builder that can be zero, and even for an average owner that might be seven years. This generally leads us to under-value efficiency improvements because they will generally last the life of the building.

  2. Expert Member
    BILL WICHERS | | #2

    R49 is usually the point of diminishing returns for roof/attic insulation, meaning more than that and you usually aren't saving enough in energy over time to justify the expense of the additional insulation. This can vary a bit with the complexity of the layout and the type of insulation used though. If you can fit R60 of cellulose, for example, that's not going to cost much more than R49, so it's something you should consider. R38 is a bit light regardless. I'd try for R49 as a minimum unless you have something that makes that difficult (such as insufficient rafter space, etc.).

    Bill

  3. Expert Member
    Michael Maines | | #3

    Insulation return on investment can be plotted as a curve, which drops off above R-38 even in cold regions, but here in IECC CZ6, with about 7,000 HDD, I have found it cost-effective to use R-60 in roofs for over ten years now.

    But that's when using an insulation with low embodied carbon emissions, such as cellulose or wood fiber. When using insulation that has a large, negative environmental impact, which includes all spray foam, I go with the least allowed. Or I use a hybrid system called flash-and-batt or flash-and-fill, which involved using just enough foam to prevent condensation, and filling the remainder of the space (or R-value requirements) with a more benign insulation.

  4. walta100 | | #6

    What set of rules is your local authority going to use to set the require minimum R value?

    For me choosing to go beyond that legal requirement would be about return on investment. I can see other people have different goals, motivations and budgets.

    If the goal is to maximize return on investment. Delete the cathedral ceiling this will make it easy to cover the 50% smaller flat ceiling area with much lower cost insulation to R 60.

    If you are stuck on the cathedral ceiling take the time to review this article and consider one of its 5 options. Do the math that it cost much more to insulate to a lower R value with twice the surface area losing more than twice the energy. This assumes the pitch about 12-12and not a hipped.

    https://www.greenbuildingadvisor.com/article/five-cathedral-ceilings-that-work

    The only real way to understand the return on investment for your home is to build the computer model and input your costs. BEopt is free to use.

    https://www.nrel.gov/buildings/beopt.html

    Walta

  5. zetetic | | #7

    This is all amazing feedback, with plenty of nuts-and-bolts details.

    @DCcontrarian, you are right on the money, so to speak. It does come down to budget - but also a long time horizon. We've had our home for 25 years, and hope to have it for another 25 if we're lucky. We did energy modeling for the HVAC assuming Title 24 - so we know what the R38 numbers look like. There's still plenty to be done on the existing home (maybe $s better spent) like trying to tighten the existing envelope. It's built to 1966 standards and detailing - lots of gaps to daylight in difficult to reach locations. When the power goes out, it has all the romance of glamping. Replacing the single pane windows in the existing building would be good. We're keeping our vintage 1990s furnace for the moment but improving the ducts (sealing, better returns, better duct insulation) in preparation for heat pump in the future, but the new addition will be entirely heated by heat pumps.

    @Bill and @Michael, I believe the builder wants to spray foam. We'd love to run continuous insulation on the exterior, but that's out of budget. Thank you for the nudge to look more closely at flash-and-batt or cellulose as an alternative. Maybe we can get to R49 without a lot of changes and additional cost.

    @walta100, the cathedral ceiling/roof is over most of the existing building. Classic 1960s with nothing more than 4" thick planks on the ceiling between the interior and the great outdoors. It's staying, but we hope to up that R-value. BEopt looks like winner! Thank you for turning me on to this. I used to write Finite Element Analysis software for structural dynamics - this is right up my alley.

    1. Expert Member
      DCcontrarian | | #8

      If you get the house modeled in BEopt, you can try out different options and see how they pencil out.

    2. Expert Member
      BILL WICHERS | | #9

      Spray foam would air seal a leaky ceiling like that pretty well. That's an advantage for that material in this case, but note that there are other ways that can work too. If you can provide some more information on what you're working with, we can give you ideas of other ways to try to achieve your goals, be those goals least cost, best performance, avoiding spray foam, etc.

      When you're only talking about code requirements, the discussion ends up being more of a theoretical exercise. When you provide details on your structure, then things can be more practical, because there is something to actually build! :-)

      Bill

      1. zetetic | | #15

        Bill, don't laugh too hard... The roof really is R4.

        From the inside-out:

        Living space
        Two sistered 8x20 (40 ft length, no splices) beams holding up the roof, 6 beam pairs
        Each beam pair on 8x20 posts.
        Beam pairs are 8 ft O.C.
        Roof is 6.5:12 slope
        4x6 T&G planks, butts are T&G as well
        Tar paper
        Asphalt shingles
        Snow in winter

        No air barrier
        Finished floor to peak of cathedral ceiling is 22 feet
        One wall is entirely single pane glass

  6. FrankD | | #10

    The wall requirements are actually pretty close. The Title 24 wall U-factor is most likely for the whole assembly, whereas the IRC R-value is for insulation only. A 2x8 wall at 24" o.c. with R-30 batts has a U-factor of about U-0.049 (assuming R-9 for 2x8 framing, a 20% framing factor, and ignoring the other layers in the assembly). The IRC also gives an assembly U-factor option of U-0.045, only a bit better than California's U-0.048.

    When it comes to the roof, the 2024 IRC reduced the requirement to R-49, still a significant difference but much less so. Since California's zone 16 seems to cover a large area with a wide temperature range and you are in the coldest part of it, I'd aim for R-49 as a minimum.

    1. Expert Member
      BILL WICHERS | | #11

      Are you sure on your numbes there? Your full wall U value would give that wall around R20 or so, my quick check shows closer to R26 or so. A 30% hit to effective R value from framing seems a bit steep just from a logical standpoint here too.

      Bill

      1. freyr_design | | #12

        According to Ekotrope it is r22.68 or u .044

        https://www.ekotrope.com/r-value-calculator

        1. FrankD | | #37

          That sounds about right. GWB, sheathing, and air films add about R-2, plus whatever cladding you used.

      2. FrankD | | #16

        Bill, I used R-1.25/in for the lumber and a 20% framing factor, as a typical value to account for studs, plates, corners and opening framing:

        R-1.25 x 7.25 ~= R-9
        (0.20/9) + (0.8/30) = U-0.0488 ~= R-20.5

        I don't know if California has any specific requirements for how to do the calculation, but the IRC commentary suggests using 78% cavity, 18% framing lumber and 4% headers.

    2. zetetic | | #14

      I missed that update to the 2024 IRC - thanks for pointing that out. R-49 does seem achievable for more homeowners than R-60, and seems to make more sense (i.e. not Passivhaus).

      Did a little hunt on GBA and found the reference to the 2024 reversion to R49:

      https://www.greenbuildingadvisor.com/question/will-irc2024-be-reducing-roof-insulation-values

      Informative post. Thoughtful comments (#11 by Bill).

  7. walta100 | | #13

    You are likely find the 16 BEopt training videos worth your time to watch.
    https://www.youtube.com/watch?v=GdMYCuwp0AY&list=PLHC0xDtkdjgec8QhVt7exJY3tpSLEFk-d

    Walta

  8. jollygreenshortguy | | #17

    R-49
    Individual cases require individual analyses, but as others have pointed out, in most cases, you're into "diminishing returns" beyond R-49 and the cost of additional insulation would probably be better spent on other things.
    I think this is one reason why the IRC 2024 is reducing back to R-49.

    1. begreener | | #22

      Interestingly, R49 is exactly in the middle of R38 & R60 ...

      I was going to suggest - just split the difference & call it a day!

  9. jollygreenshortguy | | #18

    Correct me if I'm wrong but isn't the R-38 only allowed for a relatively small area of the house? I recall it being something like no more than 20% or 500 sq.ft., whichever is less.

  10. LukeInClimateZone7 | | #19

    If you have the rafter depth, go R60. More is better unless its really expensive, in which case do the modeling that people are advising.
    The title 24 weather file for cz16 is based on blue canyon. Not as extreme. Also cz16 doesn't get a lot of attention since not huge developments.

    Minor point-- there is no prescriptive pathway for unvented cathedral ceilings. The prescriptive approach is R38 at ceiling and R19 at the rafters with a vented attic . That's whats in table 150.1

    Given the rest of your building, this is quibbling over second order decisions when you have single pane windows with orders of magnitude greater importance in the efficiency of your enclosure.

    1. Expert Member
      DCcontrarian | | #20

      So they want two layers of insulation, one at the ceiling (R38) and one at the rafters (R19)? That sounds really odd. I mean R38 in CZ7 sounds a little off but two layers like that sounds cracked.

      1. LukeInClimateZone7 | | #26

        R38 at ceiling and R19 at rafters. Yes, this is the code requirement. The rafter insulation is to temper the attic in the cooling season. There was another thread about this sometime back. And one last wrinkle on this, the prescriptive path also requires an air gap between the roof deck and the roofing material, so spanish tile roofing would qualify, but asphalt shingle applied directly to deck wouldn't.

    2. Expert Member
      DCcontrarian | | #21

      "Given the rest of your building, this is quibbling over second order decisions when you have single pane windows with orders of magnitude greater importance in the efficiency of your enclosure."

      I have to quibble with this, it's a common line of thinking and it should be discouraged.

      Going from an R-38 roof to an R-60 roof saves the same number of BTU's per year whether the underlying house is a passivhaus or is sub code minimum. The cost benefit analysis of an individual improvement is irrelevant of the condition of the rest of the house.

      The only time that the rest of the house would come into consideration is if you have a limited budget for improvements, and would want to make the ones that have the highest payback first.

      Now, an improvement to a leaky house may end up in a small percentage-wise improvement in energy usage. But that is a completely meaningless measurement.

      I say this because this line of thinking is often used to dismiss improvements that in fact make economic sense.

      1. Expert Member
        BILL WICHERS | | #23

        I completely agree. There is no reason not to try for energy efficiency in one area just because somewhere else isn't doing so well. This is especially important during renovations that may be done in stages: bring each area up as you go, with each stage adding some improvement. Eventually you have a much better performing house, you just get there over a number of smaller steps.

        I think some people mistakenly believe that heat has "pressure", so if they plug up one "leak" with more insulation, it will just make more heat "leak out" of some other "more leaky" (less well insulated) place, thus negating whatever improvements they did. This isn't how it works though -- heat doesn't have "pressure", so if you add insulation in one area, you reduce losses there (as an absolute number of BTUs), while the other, less well insulated area, continues to lose the same amount of energy as it did before. You still end up doing better overall, and you can always upgrade the less well insulated area later in a future project.

        Bill

        1. LukeInClimateZone7 | | #25

          Ironically, I do think of heat as pressure...
          It's just that pressure isn't always that intuitive.
          I'm still working on that intuition about pressure...

      2. LukeInClimateZone7 | | #24

        DC--
        I take your point and appreciate it and agree with it.
        And you steel man my take here with the following:
        "The only time that the rest of the house would come into consideration is if you have a limited budget for improvements, and would want to make the ones that have the highest payback first."

        I work on projects with clients that effectively have unlimited budgets (e.g. well endowed universities as well as individuals). But are they budget constrained on the enclosure? Absolutely. Agressively so in most cases. My personal experience is that all projects are budget constrained. I can flesh out the negotiation here, and it involves a lot of anchoring and so on and so forth.

        To the OP-- would I advise them to use closed cell in that cavity to achieve the R-60 at 3x the cost, relative to R-38 and invest the rest into air-sealing and other improvements? I don't know. As has already been recommended-- I'd recommend to run the numbers...

        1. Expert Member
          DCcontrarian | | #27

          Point taken. But sometimes the reason an existing building has never been upgraded is that it isn't cost-effective. The low-hanging fruit may actually be in the new construction.

          Ultimately houses are like snowflakes, every one is unique.

          1. zetetic | | #29

            Exactly right. The new construction requires us to integrate the new roof with the existing roof.

        2. zetetic | | #30

          We let mother nature run the numbers, and nature says prioritize the roof over single pane windows. See the ice dam photos in one of my replies. We did lose a few shingles that time.

          It's like having your own personal glacier and avalanche, all in one.

      3. Malcolm_Taylor | | #38

        DC,

        You are of course entirely right that each part should be considered in isolation as independent exercises in whether the improvement make economic sense. But at the same time it's still worth analyzing whether the improvement will make an appreciable difference to the overall energy consumption.

        We had this discussion a while ago about insulating three walls of a basement and not the fourth. Each insulated wall may in isolation make sense to do, but if you still end up with the all the basement walls averaging out at R-4, you need to know you will get that result to decide whether your efforts are best spent elsewhere, or in some other way.

        1. Expert Member
          DCcontrarian | | #39

          I disagree. Let's say you have two houses, one with an energy bill of $200 per month and the other with an energy bill of $1000 per month. You have the opportunity to make an improvement that will save $100 per month.

          The fact that this improvement leads to a 50% saving in one case and only a 10% saving in the other is completely irrelevant. Or rather, it is a purely psychological concern. In both houses, a hundred bucks is a hundred bucks.

          "you need to know you will get that result to decide whether your efforts are best spent elsewhere, or in some other way."

          But that's always true, with every improvement of any kind.

          The reason this is a hobby-horse of mine is that I care about questions of policy and macro effects. The psychological effect is real, you hear people all the time say things like "I don't think it's worth insulating my basement until I deal with my windows." We need to push back on that kind of thinking, in the aggregate the impact can be substantial.

    3. zetetic | | #28

      No rafters. Literally 4" thick T&G between the interior and about 10 ft of snow. Just a layer of tar paper and composite shingles. No air barrier.

      If you ever wondered what sort of ice dams an R4 roof produces, see attached photos of our home from Spring 2023, a 701" snow season.

      Certainly debatable, but the ice dams are why we are choosing to go after the R4 roof ahead of the single pane windows.

      1. Expert Member
        DCcontrarian | | #31

        I have to admit that looks really cool.

        1. zetetic | | #32

          Thanks.

          You could hear the house creak and grown the day before, but no jammed windows or doors. On the day of the slide, I decided to dig out the steps while waiting for the ice dam to come down. It was sunny, I knew it would trigger.

          When it slid, it felt like a solid California earthquake when it impacted. I was standing 40 feet away, shoveling snow, of course. Lost a few shingles.

          We'd been planning to do the addition since 2018 - we have no garage or driveway. This particular slide, and having to dig out the propane tank from under 15 feet of snow that winter to fill it, helped make the decision for us to finally go forward with the addition.

          I think the neighbors will miss our ice dams more than we will.

          My favorite feature of the project is the new and bigger 1000 gallon propane tank under the new driveway *with a manhole cover in the driveway* over the filler. Driveways get plowed by the HOA. Amazing it was approved. One less thing to dig out.

          Need the propane tank (definitely not green) because power goes out for days at a time (wildfire, or an avalanche takes out a power pole and transformer), so we need a whole house generator. Nowhere to put solar when we need it most during the winter. No power and only a fireplace with 22-foot ceilings, an R4 roof, and single pane windows gets a little tough after a few days.

          Really grateful to have the members and experts of GBA guide us through our projects, large and small.

          1. begreener | | #33

            Did you get water inside as the result of these "ice dams" ...

            Here in New England, that's the reason we use the word "dam" ...

            The snow melts, turns to water from the underside of the roof heat loss, then the melted snow runs off the edge of the roof where it immediately freezes & turns to ice.

            This ice then starts to build-up in such a way that it forms a "dam" that the melting snow/water then backs up behind & enters the house...

          2. zetetic | | #35

            @begreener

            No leaks. I think the 6.5:12 roof slope helps, as do the 3-foot eaves. We put standing seam on the eaves to help the sloughing process. If you look closely at the photos, you can see the grooves from the standing seam in the underside of the ice dam.

            I think the ice dam forms, but then instead of continuing to fill with water, it slides a bit from the weight of the ice and water, thanks to the standing seam and the steepness of the roof. If it leaks, it may leak through the eave.

            We intend to insulate not only the roof, but the full eave, so that heat loss from the walls below do not melt the snow on the eaves.

          3. Expert Member
            BILL WICHERS | | #36

            I have a few recommendations for you:

            Next time you need new shingles, consider a metal roof. Snow slides off of those much easier, resulting in less buildup (most of the time). That's probably a plus in your area. My first thought seeing all that snow is "what is their snow load number for that area?". I'm sure that was a LOT of weight up there. Be thankful for your very heavy framing :-)

            I have ice dam issues here too, although nowhere near as severe as yours. We get icicles that run from the eave all the way to the ground sometimes. That problem has gotten much better since I improved the air sealing of the upper level (attic floor), and I've added a lot of insulation and improved the attic venting. If you can arrange for a vented roof assembly, that would help with the ice dams too, because some of the heat that gets through your insulation will be exhausted out through the vent channels instead of heating up the roof. The result is a colder roof, so less thawing.

            If you have yet to purchase a generator, I strongly advise you go with a Kohler unit over the much more commonly seen Generacs. I see lots of issues with Generacs, especially when running for extended periods, with heavy average loads, and in hot areas (that last one is probably not applicable to you! :-). The Kohler units are better (note that they are "Rehlko" now, they sold their power division). If you really want reliability, get an 1,800 RPM unit that has liquid cooling. Those are much, much more reliable that the typical 3,600 RPM residential generators, but they cost significantly more. Lastly with generator stuff, size the unit for the entire house and put an ATS right before the panel. Run everything. It's so much nicer to have everything available when running on generator power, and it really doesn't add much cost to do it that way when you're putting in a new system.

            One last thing: put a heat tape in that manhole cover for your propane tank, and wire it to a switch in your house. What is likely to happen is that snowpack will get squished into the cover, and will refreeze in that manhole cover. If you have a heat tape already inside the cover, you'll have an easy way to thaw things out. In the utility world, if the manholes in the street freeze, we can free them by whacking around the edge with a sledgehammer to break up the ice that is in the gaps around the edge of the cover, then we can lift the cover off normally. If you have a plastic cover, you probably won't be able to use the "whack it!" method of freeing things up, because the plastic is likely to break.

            Planning ahead for likely issues down the road can save you loads of effort later.

            Bill

      2. LukeInClimateZone7 | | #40

        I'll admit that I am in a bit of a different climate zone, but I'm not sure R60 Will eliminate the ice dams. It'll reduce them for sure, but with drifts that thick, The top of your roof is going to be still pretty warm. I don't know the range of r values for snow, but maybe someone here can chime in And tell me how off base I am.
        Note, I'm not suggesting that you don't insulate, but rather that maybe additional considerations like that standing seam metal roof that bill was suggesting would be also prudent

        1. freyr_design | | #41

          You need a vented over roof to overcome the insulate value of snow when it gets to something like 60 psf. https://buildingscience.com/documents/insights/bsi-046-dam-ice-dam

        2. zetetic | | #42

          I actually appreciate all recommendations, solicited or not!

          @Luke - agreed, the R-value of the snow might make some melt inevitable if you don't have a "cold roof". My wife is not a fan of the look of standing seam, so composite shingles with standing seam over the eaves it will be. Our home does have some local historic value, so there is an argument for being a bit more period correct (thick roof and standing seam aside).

          @Bill I looked up the snow loads: 386 ground, 324 deck, 297 roof. I'd love a vented "cold roof" - sometimes called a "Boston Ridge" roof. I see these all the time in Wyoming. The problem for us is all the snow - it blocks any passive ridge ventilation, and the lack of an attic means we can't vent through a gable end. We looked into venting the ridge through a tall vertical vent alongside the chimney, but we still couldn't get sufficient airflow volume through the ridge. Just to complicate things, we also have a ridge skylight (trying not to laugh or cry) that we would have to vent around.

          @Bill Your comment about the Kohler genset couldn't be more timely. Our builder is pushing Generac. We originally spec'd Kohler after I did my research and reached out to Generac and Kohler to dig into the two. There's some wisdom to go with what your builder suggests because of their familiarity and experience, but it sounds like I need to revisit this. The Kohler water-cooled motors are great, and you can argue that the motor is the heart of the generator.

          At the time, we were looking at the Kohler 24RCLA (inline 4 liquid cooled, 1800 RPM, $13k) and the Kohler 26RCAL (V2 air-cooled, 999cc, 3600 RPM, $7k). Some argue that the Generac engine and electronics are easier to maintain DIY (DIY is my style), but this is against the backdrop that the Kohler is more reliable (reliability beat DIY most days). Generac offers their 22kW 7403 (V2, air-cooled, 999cc, 3600 RPM, $6k). The Kohler generates 18% more power at a 17% price premium. The Kohler and Generac break down about the same way for water-cooled.

          @Bill - great tip on the manhole cover! It's that, or a steel manhole cover, but I don't know if the remote tank level sensor will transmit through steel (?)

          1. freyr_design | | #43

            You should really read the bsc article I linked.

          2. zetetic | | #44

            @freyr_design. Yes, I'm familiar with that article. I always enjoy reading and watching Lstiburek. He has contributed so much to building science. I have many of this books, lots of post-it notes throughout. I've been researching cold roofs for years, taking photos wherever I see them in ski/snow country.

            Have you seen the Fine Homebuilding article: "Does a roof need a ridge vent" by Scott Gibson? (it's behind a paywall, first article is free, though):

            https://www.finehomebuilding.com/project-guides/roofing/does-a-roof-need-a-ridge-vent

            Quote from the article:

            "Dick Russell admits that under certain conditions, a layer of ice-encrusted snow can cover and block ridge vents. But overall, he thinks they remain functional most of the time." (OK, MAYBE GOOD)

            and

            Kankan writes that his well insulated roof has no problem with ice dams, as long as the ridge vent is visible. “But when it gets buried,” Kankan adds, “bam, dams.” (NOT OK)

            and

            “I have seen with my eyes my ridge vent get completely buried in snow and saw results of lack of ventilation instantly,” Kankan says.

            No details or nuance on either the Russell or the Kankan roofs or snow loads, so I have to take both quotes with a grain of salt. I think they are in Minnesota.

            In another FHB article, titled "Framing a Cold Roof" by Steve Kearns (Idaho), I actually hunted down Steve and called him up to ask about the "Boston Ridge", a kind of raised ridge that creates larger vents under the ridge. The snow where he uses them near ski resorts does get deep, but it's a bit lighter and fluffier. I've seen these raised "Boston Ridges" in Wyoming as well - but again, the snow doesn't get as deep, wet and heavy as it does in the Sierras.

            I've seen a few Boston Ridge designs at ski resorts (deep roof snow) that vent through a chimney that exhausts above the snow. The chimneys are wide and tall, and back the length of the ridge, almost like another ridge rather than what you'd normally think of as a chimney. A special ridge-length chimney is not really a cost-effective thing for us to do, and would probably look weird on our house.

            That said, from the BSC Lstiburek article that you linked, Photograph 8: "Note the air inlet at the fascia. Yes, the ridge vent is covered with snow, but snow is not an air barrier. I only need ***enough ventilation to compensate for the thermal resistance of the snow layer***, and that is not much if I have an airtight roof and lots of insulation, say R-60."

            How much ventilation is enough to compensate for 8-10 foot deep snow weighing tons (297 lb snow load at the roof, 386 lb at the ground)?

            ***If you cannot compensate for the thermal resistance of the snow layer***, then does the intake air from the eaves get warm in the over-roof and not exhaust quickly enough, creating a not-so-cold roof under an insulating blanket of snow, until the snow melts and causes an ice dam. Defeating the cold roof? Kankan would say "yes". Russell would probably say "maybe."

            Per the Lstiburek article you mentioned, we intend to insulate the full eave.

            This article is a true gem, references by Lstiburek. I'm going to try to quantify our snow thermal resistance, and the size vents needed (and hence the free space around the ridge vent):

            https://www.erdc.usace.army.mil/Portals/55/docs/CEERD-RV/CEERD-RR-H/BuildingTechnology/ResearchPapers/MP-02-5778,%20Guidelines%20for%20Ventilating%20Attics%20and%20Cathedral%20Ceilings%20to%20Avoid%20Icings%20at%20Their%20Eaves.pdf

            As another wrinkle, we are located in the WUI and need to make sure wildfire embers cannot enter the roof at the eave or the ridge (or any vent). Code may require the cold roof to be non-flammable internally. All of this drives up the cost of the cold roof, until unvented R49 or R60 look more practical, ice dams and all.

            If someone has successfully built a cold roof in the High Sierras (i.e. zone 6b or 7) where they get up to 400 inches of annual snow or more, I'd love to hear from them.

          3. freyr_design | | #45

            You should do both r50 and a vented over roof. With good air sealing at your roof deck and the insulation, this vented over roof should not have issues cooling, and at worst if it gets covered you will not be worse off. Venting in wui is not that difficult, you just need to use Vulcan vents or sim. They make a profile that is basically a 1” continuous that you can use at both your fascia vent and your ridge. This profile is a bit gmhard to find so you might have to call them but it is ideal for rainscreen and these applications.

            If you are doing shingles you will already be adding a second roof deck so you might as well vent it.

          4. Expert Member
            BILL WICHERS | | #47

            The traditional way to vent a roof "above the ridge" is to use one or more cupolas, which might even fit the "period correct" requirement you mentioned. That might be something to consider. Snow in thinner layers is air permeable, so ridge vents still work, but your snow levels are in a different league, and I'd consider such thick levels to be air barriers. Even if nothing else, the very deep snow will likely pack together in the lower layers and significantly restrict airflow if it doesn't stop it altogether.

            None of the controllers are easy to maintain DIY. Generac's is entirely propreitary, and you can't do much of anything without their special software (only available to dealiers AFAIK). Kohler's residential air cooled gensets aren't much different in that regard.

            The liquid cooled small commercial gensets Kohler makes use a more standard type of controller, leaving much of the rest of the unit more "normal" and maintainable. The mechanics of those liquid cooled units are vastly superior as well. I'd go that route (and that's what I have at my own home -- a Kohler liquid cooled 1,800 RPM unit) if you have the budget for it.

            The builder is pushing Generac because they are likely a dealer. Generac will make pretty much anyone a dealer, Kohler is a lot more restrictive. One of my contractors services both, and says "I keep a lot of Generac parts on my truck, because I NEED those", because he has to fix their stuff more often :-) I would strongly advise against the Generac unit here, especially if you expect to depend on it frequently for long periods of time.

            You might be able to get the sensor signal through that manhole cover if the manhole's mounting ring isn't all the way to the tank. Basically if you have a concrete structure or other less-than-RF-proof material, you might be OK. If the level sender can use a remote antenna though, that would be the way to go. Better yet would be a wired sensor. I'm not a big fan of wireless things when I can avoid them.

            Bill

  11. Deleted | | #34

    Deleted

  12. zetetic | | #46

    @freyr_design Very good point that we won't be worse off with a cold roof. Another possible benefit of the cold roof, depending on the construction, the cold roof gap might reduce thermal conduction. Also, thank you for the WUI Vulcan tip - I'll look into that.

    Do you have a favorite / preferred way to frame your cold roof? We are trying to be super deliberate in our spending and the cost-benefit. Our builder wants to blow cc spray foam. I've been looking at the Raycore SIPs (R52 costs about $11/sqf, last I checked). The builder is concerned about panel-to-panel gaps and feels the extra labor offsets the SIP savings. I like the thermal uniformity of the SIPs. Always trade-offs.

  13. zetetic | | #48

    @Bill - my gut says, given the choice, that our money will be better spent on adding more insulation (guaranteed to improve the ice dams) rather than adding a cold roof (may or may not improve the ice dams). Or maybe move some insulation budget over to the genset.

    Sounds like the air-cooled gensets can turn into expensive bricks (special software to service them), which is not ideal if you need the genset and service can't be scheduled. Of course, regular maintenance and the weekly or bi-weekly test run is meant to catch problems before there's a power outage. But I like having a Plan B (DIY) in case there's a perfect storm.

    Your insights on the gensets are invaluable and will help us make a more informed decision.

    1. Expert Member
      BILL WICHERS | | #49

      With very thick snow loads, the vent channel IS going to help. How much, I can't say for certain.

      No problem with the gensets. Feel free to post new questions about that if you have any other questions. I work with generators professionally at work (I do a lot of work with critical power systems for things like telecom sites and hospitals). Happy to help you out here.

      Bill

      1. zetetic | | #50

        @Bill Thank you for putting a finer point on the cold roof, that it's not entirely a lost cause with all of our snow. FWIW, we live less than 5 minutes from where the 1982 Alpine Meadows avalanche occurred (Netflix movie, "Buried").

        I truly appreciate your offer to answer genset questions. I'll create a new Q&A:

        https://www.greenbuildingadvisor.com/question/air-vs-liquid-cooled-standby-generator-for-winter-outages

Log in or create an account to post an answer.

Community

Recent Questions and Replies

  • |
  • |
  • |
  • |