Hello -

We just bought an old victorian house (built 1890) in Denver, CO. The house is made of brick on the first story, then has typical stud/drywall construction for the second story and the attic, which has been finished into a bedroom. While planning the installation of a whole house fan, it came to my attention that the insualtion in the attic and finished bedroom areas is pretty bad (even to my amateur eyes) and needs some help. I am trying to plan the most effective way of doing this and I would appreciate any input. Please excuse my ignorance on some of the terminology, I am still trying to learn about these topics!

The finished attic provides a room approximate 12 feet wide and 26 feet long with the bathroom built in. On either side of the attic bedroom there is a knee wall about 40 inches tall. The roof slope is 8/12, so if you can imagine the attic space behind the knee wall is extremely small. For that reason, we are going to use 3 roof-mounted whole house fans, each of which will be ducted from the roof down to the 2nd floor bedrooms and one to the knee wall itself. The duct is insulated/soundproof and is 16 inches in diameter. There is also a small attic space at the top of the roof- a 39 inch wide flat ceiling exists above the center of the attic bedroom. A very small gable duct is visible on both sides of the house, which would theoretically vent that now tiny attic space at the very top. Again, you can image how small this space is due to the roof pitch.

 Once I have these installed, I want to insulate the attic spaces. I've had a couple of insulation experts come by and the one who I trusted most recommended that we blow cellulose in the whole area. He wants to blow cellulose up the rafters and into that tiny attic space at the top of the room, filling it completely. Then he'll pack the remainder of the attic spaces behind the knee walls on both sides densely so that there won't be settling and close the space. There is some old, poorly installed R13 batting along the underside of the roof and the contractor said it would be fine to just leave it and it would be compressed by the blown cellulose. I like the idea of cellulose and it seems like a good idea.

But I'm not sure about a few things:

1) The two attic spaces behind the knee walls are unvented. I understand that 'always vent' is no longer the mantra, but we won't be adding a vapor barrier or otherwise 'sealing' the space. It won't be vented, but not exactly sealed either. I think the mentality is that the hard-packed cellulose impedes airflow so much that this won't be a problem, but I'm not 100% sure.

2) The attic space on the very top is small, but it does have those two gable vents. Should these be removed or sealed? It seems like it would just allow moisture to come into direct contact with the cellulose.

3) In the tiny attic space at the very top there is some electrical - we have a hanging ceiling fan over the center of the room and a ceiling mounted light fixture. Note that these aren't recessed, in cans, etc. they are simply mounted on the ceiling. Is this an issue? There are also 4 electrical outlets on the knee walls, and a small bathroom is framed into the attic space with it's own electrical, etc. although I assume that none of that is passing through the attic (no bathroom fan). Any issues with this, or is it fine for the cellulose to pack up against the electrical boxes, wires, etc?

4) I installed 3 skylights to the roof, between the ceiling of the finished attic bedroom and the roof, about 3 feet above where the knee wall ends. The installation required tons of shims, etc. to get it perfectly straight. I used lots of drywall screws to ensure that the shims/skylights are well secured in place. Will  the dense packing of the cellulose put excess pressure on these skylights? It would be great to help seal up the space around them but I'm not sure how much pressure we are talking about and I don't want to push them around too much.

I was first shocked at the notion of filling cavities completely with cellulose, but I'm learning that it's done frequently. It's counter intuitive, I suppose, but cellulose isn't like the fiberglass batting we used 30 years ago so I'm trying to get my head around it.  If we fill up the whole space, we should have about 7 inches of packed cellulose between the ceiling and roof, and up to 25 inches in the thicker spaces of the attic.  My goal is to get to R49.

One of the contractors told me that polyurethane would do a better job filling the spaces, and would give a much higher R value in the thinnest cavity between the ceiling and the attic above the attic bedroom. That makes sense, but it's not cheap and not as green from what I understand.

Ok, thanks for reading (if you still are!). Any thoughts, questions, advice, or anything else is much welcomed! :)

Where to begin...

In Denver, most of the attic moisture will be coming from the interior side, mostly during January. Dense-packed cellulose is a good enough impediment to air movement to not suffer R-value losses at high delta-Ts, but isn't an adequate air barrier for preventing air-transport of humidity. As you insulate the attic floor, the roof deck will get colder, increasing the risk of condensation being trapped in the roof deck if the space is unvented. Air-sealing the ceilings below (and the kneewalls adjacent to) the attic spaces is Job-1!

Job-2 would be applying vapor-retardent paint to the ceiling (or 1-2" of closed cell foam) to the attic side of the ceiling/wall will lower the humidity of the unvented attic space. Weatherstripping & sealing the access hatches may be a weak point. Dense-packing the kneewalls is good, but some amount of vapor retardency on the interior side is also good. An inch of closed cell foam achieves both water vapor & air control, with out making it so vapor-tight that it can't dry toward the inteior in summer.

Dense-pack cellulose won't bow most window frames the way high-expansion foams can. Low-expansion foams purpose-designed for the application will deliver a better air seal.

Dense pack cellulose in the cathedralized ceiling portion may be an issue in your climate though. While cellulose is a very good hygric buffer, and Denver's wintertime humidity is typically low, if you tighten up the house enough that you have 30% RH or higher (or actively humidify to that level for health & comfort reasons) you may end up with sufficient moisture in the roof deck to have a seasonal rot-risk, depending on roofing type (and possibly whether it's N or S facing, etc).

Refer to this:

http://www.buildingscience.com/documents/reports/RR-1001_Moisture_Safe_Unvented_Roofs.pdf

Look at the results for Boulder in the tables, as the most-relevant reference climate.

If you did the cathedralized portion with at least 2" of closed-cell foam on the roof deck, and cellulose for the rest of the R it looks like you'd be good to go though. (They filled out the R with JM Spider superfine spray-fiberglass in the simulation, but dense-packed cellulose & Spider have similar air-retardency, while cellulose offers much better hygric buffering than fiberglass.)

25" of cellulose is more like R90-R95, and it weighs quite a bit, if dense-packed to 3+lbs. Be sure the ceiling can handle the static load (6-7lbs per square foot), which it probably can if the span is short between supporting elements.

If the gable vent openings are wide, with no overhanging roof to block rain from entering, by all means seal them. A bit of occasional rain penetration through the vent in a vented attic is usually no BFD, but if it's all dense-pack cellulose behind it, the cellulose will redistribute the moisture, with a small but real risk of building up moisture over time.

I'm an advocate of using dense-packed cellulose as retrofit but it's not an either/or, and there are often combinations that work better in climates like yours. If your roof lines are straightforward, when it's time to re-roof using panelized iso on the exterior to grab another R12-R25 will keep the structural roof deck warmer==drier when using a dense-pack cellulose only on the interior. (Look at some of the nail-base panelized goods from Atlas, Hunter Panels, et al.) If the roof looks like it'll need it in the next 5 years, you might consider going all dense-pack now, with a plan for panelized iso on the exterior when you re-roof. (With 6" rafters in my antique house in MA that's about the only way I can come close to making R50, even if I did all closed cell foam on the interior. )

Thanks for all that info, which I will read through and digest tonight.

Our roof was just redone a few months ago using ordinary decking so I don't plan to re-roof any time soon.

One thought did enter my mind - you mentioned that it would be possible to do the catheterized section with 2 inches of foam and then pack cellulose in the rest. But how would the installer be able to create a 2 inch layer along that space? I can look up along the rafter and see the gap between the ceiling/drywall and the rood decking, but it's not very much of a space. It seems like they would have to simply fill in the whole space, including the small attic space at the top of the room. Maybe it's possible, though to create a layer.

It certainly seems like a good idea to blow foam in as a vapor barrier. However, the Denver climate is so dry that every installer I've spoken to said that the have no concern that humidity would accumulate and if it did it wouldn't last though the first weeks of spring. If my dry skin is any indicator, it's true.

Thanks again for the advice, I will continue reading and leaning

Posted By Dana1 on 10 Jun 2010 06:23 PM

 While cellulose is a very good hygric buffer, and Denver's wintertime humidity is typically low, if you tighten up the house enough that you have 30% RH or higher (or actively humidify to that level for health & comfort reasons) you may end up with sufficient moisture in the roof deck to have a seasonal rot-risk, depending on roofing type (and possibly whether it's N or S facing, etc).

A quick question, please. The humidity in Denver is typically low, and my assumption was that the interior of the house would have about the same humidity. We have no AC, swamp cooler, or anything like that. In fact, we're installing whole house fans so the inside will be well vented (and hopefully dry) in the summer. Winter heat comes from a gas furnace.

What would make the interior humidity rise once the house is well sealed?

thanks again !

The humidity of the outside air is a small secondary effect. It's all about the dew point of the INTERIOR air, and where in the assmbly it stays at that temp or lower. The dew point of 68F air that is 30% relative humidity (the lowest recommended for comfort & health) at 5500' of altitude is about 35F. The roof deck is on the exterior, an runs approximately at the outdoor temp (the S-facing pitch will average somewhat higher than the average air temps, the N-pitches slightly less. The monthly average temps in Denver for the 13 weeks of December/January/February are below 35F:

http://www.climate-zone.com/climate/united-states/colorado/denver/

This means that unless that roof deck is ventilated, any interior AIR that reaches it will deposit moisture there, since without a modest flow of dry exterior air to purge that moisture it can only dry to the exterior by vapor-diffusion through the roof deck & shingles. This is a slow process, and a process that halts completely during periods when there's snow, ice, or liquid water on the roof. During warmer days there will be condensation & re-evaporation cycles, but that water isn't going anywhere but into the wood until it has far more hours ABOVE 35F per week than below (and the further above 35F, the higher the vapor pressure==the more rapidly drying by vapor diffusion takes place.) Water reaching the interior of the roof deck November might still be there in April, but by Jun significant drying should be occurring.

If the interior surfaces are too water-vapor permeable (even if air-tight), some accumulation can occur too, but the vapor permeation is at least 1-2 orders of magnitude lower than air-leakage as a means of moisture transport. Cellulose is somewhat protective, since it can store a significant amount of the humidity diffusing or air-leaking into the assembly without damage or losing thermal performance, but it's reservoir capacity isn't infinite. (Fiberglass & rock wool won't store any humidity, and temporarily loses R-value under condensing conditions, whereas the hollow fibers of cellulose wicks & redistributed moisture and retains R value until it's something like 30-35% water by weight.) Air sealing the interface really IS job-1: a square inch of air leakage is worth more than a whole ceiling of vapor diffusion through unpainted (but air-tight) plaster or gypsum board. But with an unventilated roof deck this lower level transport via vapor diffusion counts too.

Standard latex paints are very modestly vapor retardent at about 2-3 "perms" An inch of closed cell foam is a bit over 1 perm, 2 inches a bit under 1 perm. Vapor-retardent latex is about 0.5 perms, which is still more than 10 times the vapor permeability of 6-mil polyethylene "vapor barrier", but slow enough that in a climate like Denver's the cellulose can absorb the seasonal humidity loads without saturating, slow drying toward the interior in the spring, and both to the interior & exterior during the summer. (Sun on dew-or rain-wetted shingles can also result in high vapor drives into the roof deck, but Denver's dry climate makes that less of an issue than it is in some other locations.) Getting the ceilngs & kneewalls air-tight, and the vapor permeance down to about a perm or less (but not so low that it'll never dry) is key. The roof deck is protected whether that vapor retarder is at the roof deck or the interior surface, but putting it at the interior limits the amount that ends up in the cellulose. An inch or two of closed cell foam applied between the ceiling (or kneewall) and the cellulose achieves both air-tightness and the approximate vapor retardency you're looking for.

When you seal up an occupied house air-tight the interior humidity sources that raise the humidity are those pesky humans: respiration, taking showers, cooking, etc (and to some degree ground-water moisture diffusion through the foundation, but that's a lesser effect in CO unless you're barely above the local water-table.) In Denver it's pretty easy to control indoor RH year round by adjusting the ventilation rates since the outdoor dew points are typically well under 50F even in summer. (Here on the left-coast some amount of mechanical dehumidification is necessary during the warmer parts of the year.) A nighttime ventilation cooling strategy using whole house fans at my house would RAISE the indoor humidity to over 60% RH (the highest recommended for health & comfort) most of the summer, but works fine in most of the country from the Rockies & west.

It turns out, I forgot most of my college physics and it took me a while to understand your post. I think I understand it now, though - thank you.

I'm getting the impression that a good approach may be to have them blow in some foam along the attic deck and knee wall to create a good barrier, then blow in cellulose for the rest. Of course, some installer say that the air flow is so reduced by the cellulose that this is unnecessary. Better safe than sorry.

One thing that concerns me is that with the new fans going in, we'll have 3 six foot acoustical/insulated flex-ducts running from the floor of the attic where the grills to the 2nd floor bedrooms are to the roof. This shouldn't be a problem except I'm concerned about squeezing/damaging them when we insulate. Packing the entire cavity with cellulose could cramp the ducts, so foam could be better on that one.

Posted By Dave-H on 11 Jun 2010 11:13 AM
It turns out, I forgot most of my college physics and it took me a while to understand your post. I think I understand it now, though - thank you.

I'm getting the impression that a good approach may be to have them blow in some foam along the attic deck and knee wall to create a good barrier, then blow in cellulose for the rest. Of course, some installer say that the air flow is so reduced by the cellulose that this is unnecessary. Better safe than sorry.

One thing that concerns me is that with the new fans going in, we'll have 3 six foot acoustical/insulated flex-ducts running from the floor of the attic where the grills to the 2nd floor bedrooms are to the roof. This shouldn't be a problem except I'm concerned about squeezing/damaging them when we insulate. Packing the entire cavity with cellulose could cramp the ducts, so foam could be better on that one.

It's more than air flow- it's VAPOR permeation.  Water molecules are smaller the diatomic nitrogen & diatomic oxygen that makes up the bulk of air. Water vapor molecules are roughly half the size:  Since hydrogen atoms have only one electron shell that they're sharing with the oxygen atom, H2O is barely larger than mono-atomic oxygen, since those electrons are just filling in the remainder of the oxygen's outer shell, and the (very small) naked proton that makes up the hydrogen nucleus is just hanging along for the ride.  At half the size of the averag air molecule H20 as water vapor gets through many materials where air doesn't.  Various methods have been developed for measuring the vapor-permeability of otherwise air-tight construction materials.

Dense-packed cellulose slows the air by 99% but doesn't STOP it, and it's vapor permeability is sky-high, wide open (which is good in some situations, not so much others.)  With dense-pack the air-transport is about the same order of magnitude as the vapor transport through the wall, but still too high to reliably protect an unvented roof deck in your climate.

Also note- the vapor permeability of open-cell foam is also pretty high- only closed-cell will do in this app, and it'll do more good on the warm-in-winter side of the the assembly where you can (but at least 2" on the roof deck, if that's the only option.)  With an inch or more of foam as an air-barrier & vapor retarder on the warm side you can fill in the rest with cellulose and it'll all stay dry (enough) & happy year-round (until the roof leaks, of course. )

Another vendor just came and took a look. He specialized in blown fiberglass but also did cellulose and foam. He had an obvious bias towards the blown fiber.

He said that nobody could possibly blow the cellulose up the rafters to close the space, especially without pulling the r13 batts first. The cellulose installer said this was no problem and that the hard packed cellulose would compress the batt. He said that there is no guarantee that the cellulose will fill the top cavity, the little vented area at the very top of the roof. Cellulose guy said he woiuld just feed his hose up to the top and fill it, no problem.

He said that the foam would basically explode in the ceiling and was unusable in this circumstance. It kept coming back to blown fiber.

There are a lot of strong feelings and bias in the insulation world. Hard to get an objective story.

Posted By Dave-H on 11 Jun 2010 01:12 PM
Another vendor just came and took a look. He specialized in blown fiberglass but also did cellulose and foam. He had an obvious bias towards the blown fiber.

He said that nobody could possibly blow the cellulose up the rafters to close the space, especially without pulling the r13 batts first. The cellulose installer said this was no problem and that the hard packed cellulose would compress the batt. He said that there is no guarantee that the cellulose will fill the top cavity, the little vented area at the very top of the roof. Cellulose guy said he woiuld just feed his hose up to the top and fill it, no problem.

He said that the foam would basically explode in the ceiling and was unusable in this circumstance. It kept coming back to blown fiber.

There are a lot of strong feelings and bias in the insulation world. Hard to get an objective story.

If you're talking about a full cavity fill of slow rise  2lb foam, yes, that would have a risk of blowout (and fixing it would be a major PITA.)  But if there's line of sight to be able to gun-shoot a 1-2" layer on to the sloped ceiling side you'd be fine. (I've dealt with contractors with sufficient talent to pull that off even in 9' tall wall cavities as narrow as full-dimension 2x4s.)  To foam-shoot a layer you'd have to pull the batts though.

Dense-packing cellulose over R13 batts is awkward-but-doable in a 2x4" cavity, but dead-easy if it's 2x6" or greater. All you need is enough space to snake in a semi-rigid ~ 1.25-1.5" o.d. plastic tube in there, which would be no problem at all in your case.  (Same story for dense-packing up to the gable peaks.)  I'm not a professional installer, but even I have successfully done dense-pack cellulose over batt. While it may be possible to pull the batts, unless they're so balled up in there that you can't punch the dense-packing tube through it's not worth taking the time, since the compressed batt will have about the same R-value as filling that volume with cellulose.

Your celllose guy is probably telling you a pretty straight story.  But he probably hasn't run a WUFI simulation on what happens to dense-pack cellulose in a cathedralized ceiling.  DO read, and try to understand what they're doing in that RR-1001 paper I linked to.  While dense-pack cellulose by itself works fine in climates somewhat milder than yours, Denver just a bit on the cool side to not end up with rot-risk at some point over any 25 year period. Some year's it won't be an issue, but back to back winters that are colder than average might be enough to get problems rolling.  But if you're planning on re-shingling at some point in the next 5 years, dense-pack-only now, with sheet insulation above the roof deck later can work with relatively low risk. Without a perfect air barrier it

Or, your house may run drier than recommended by ASHRAE & others. At your altitude, for every ~5% RH you can drop it below 30% the dew-point drops another ~4F, so at 20% inteior RH (which feels pretty damned dry), the dew point will be in the mid-20s, and your monthly average winter temps will be above that, with more drying time than condensing time at the roof deck.  You'll be more susceptible to flu & colds though.  The tighter you make the place, the more humid it'll be- indoor humidity meters are a good idea for people in tight homes.  30-50% RH is the healthiest, 60% is OK in summer, but increases dust-mite populations. Over 65% the mold hazard starts to explode.

Air sealing with spot-foam in the harder to seal areas, and painting the interior with vapor-retardent paint is probably enough to for dense-pack-only to work as long as the interior plaster or gypsum has no cracks and will be maintained. The air-tight integrity of your duct runs in that area are also a bit of a concern.  The WUFI simulation for Boulder showed a ~4-week/year moderate-risk window with a cellulose-only solution, but a 0 (ZERO) risk with 2" of closed cell applied to the roof deck, with the rest filled with high-density fiberglass (roughly equivalent to 3lb density cellulose for air movement.)  Appyling 1.5-2" of XPS or EPS above the roof deck at some later point, you'll be in even better shape, since the roof deck will then be running several degrees warmer, with many fewer condensing-hazard hours.

The interior has been painted with ordinary paint twice in the last 3 months, and I'm not planning on doing it again
The roof was just completely redone, new decking and all so I'm not planning on doing that again either!

I'm still looking for a good recommendation on a way to insulate the attic that minimizes risk and is as economical as possible. Another contractor who claims to offer foam, fiber, polyurethane, cellulose, and just about everything else is coming in 30 minutes. Hopefully I'll get a coherent offer from this one

Print out that RR-1001 document from Building Science Corp and discuss it with them. 2" of closed cell foam on the underside of the roof deck keeps the definition of the project simple can cures a lot of ills when combined with dense-packed cellulose for the rest. With that you also get about R12 out of the foam itself which will add about R5 to what it would have been as an all-cellulose show.

But with a composite-shingle roof, dense-pack only, it's only a moderate, not a high risk in your neighborhood.  It'll still be there in 10 years- it's the 25 year picture where you might take notice.

Ok, I will continue learning and read that document again. I appreciate the input and I've already learned a lot. More and more it seems like a layer of closed cell foam to seal it + blown cellulose will do the trick. But, if I can get away with just cellulose I'd save some money so I'm still considering it. I'm kind of risk averse so the foam is sounding good to me!