I am curious if anyone can share any info on whether it is feasible/workable to rip 24" wide strips of 4" thick XPS and get them wedged between two 24" o/c TJI joists...? Because the joists have the 2x4 endcaps, I will have to angle the sheets in between the TJIs then "punch" it into position. Is 4" too thick/rigid to do this? The Foamular250 comes in 4" but has a compressive strength of 25lb/in so I am curious if I will be able to physically get it wedged in there properly?

won't work. Instead, cut the foam to the width of the center part of the TJI so you can slide it in, then insert a pieces on each side. Tape the joints, and tape it to the sides of the tji, or alternatively, put a healthy amount of spray foam over the joints and seams. (foam to foam and foam to wood seams)

PS: if you cut the side pieces with a slight taper they'll hold everything together tightly.

Oh my this is getting to be difficult. What about two 2" pieces? It seems I can probably wedge a single 24" wide 2" thick piece in there - do that twice for 4" total?

I'm not sure why are you putting high-R/inch foam where it's thermally bridged by R1/inch OSB webbing?

I'm also not sure why you'd put 4" of the worst foam product for the environment ANYWHERE in this house. Three inches of 2lb closed cell spray polyurethane would have the same R value, but only about half the environmental hit, and it's less labor to boot.

This is between TJI roof rafters. Chicago. Zone5. Metal roof. TJI rafters 24in oc. We discussed this already. I want an air gap under the roof to dry it. I also need something to spray against hence the rigid foam. I drew up a section showing what I'm doing...

http://chicagocadcam.com/ChrisHahn/house/TJIsection.jpg

We also discussed the article a couple weeks back that showed the degradation of the Polyiso Rvalue in cold weather, which is exactly what I don't want. So XPS it is...at least for now. If you want to revisit this with me I am all ears and not close minded.

i have not done this method, but have discussed it with my insulator. Our plan is to install 2" polyiso to make the air channel and a flash coat (1"+/-) of ccsf to seal it (or you could use 2" of ccsf) then dpc (cellulose). While there may be some degradation of the R value, IMO, that's splitting hairs and will make virtually no difference in performance. Of course that's only my opinion; I'm not a scientist.

Bob, the article showed (from memory) a 50% loss in R value I believe. It was drastic. In winters where this year we have had many days below zero. I worry about 2 - 4 weeks of condensation in the wall. I am an aerospace/mechanical engineer but don't follow this environmental calculation(s) either. So the concept of drying through ocSPF to the interior...how much condensation can occur if a 3.1in thick polyiso sheet degrades to R-12 instead of R20? Will this be for a few weeks? or for a few months? How long will it take to dry through 9 inches of ocSPF or cellulose or even a HDbatt? Will it dry? Will it mold? This is my house so I want to be sure. I understand some moisture and dewpoint condensation will happen no matter what but I assume if it keeps happening, the wall won't dry enough and will mold. So hence the use of 4in XPS instead of 3.1in polyiso (local supplier has 3.1in thick sheets). XPS actually gets better R value with cold weather per article (my understanding).

do what you are comfortable doing, definitely.

Surfs up, remember the article you reference is about a study that is a work in progress. And only one brand (undisclosed) of polyiso was tested. Dana made a good point in the comment section there stating that similar comparisons show DRASTICALLY lower r-value impacts on polyiso with lower temperature. I did some calculations based on that articles test results and found that for a wall insulated with only that polyiso, the heat loss through the wall was the same if the interior thermostat was set to 80F or 60F! Now the windows, doors, air infiltration, heating system efficiencies all are better at a lower thermostat setting, meaning that you should still lower your to-stat to save energy, but again that was for only one brand of polyiso, and other data sources indicate that the negative relationship of temp to r-value are much less extreme than the one referenced in the article, making my results brand specific at best.

What about 2x layers of foam? 2" of EPS on the outside, then another 2" of polyiso on the inside, then you keep the interior polyiso warmer. XPS is more expensive in general, and as Dana routinely reminds us, is WAY bad for the environment (often causing more global warming potential than it offsets in its lifetime). Or just use EPS and be done with it, unless there is a reason you are avoiding EPS.

My .02

Well my concern is both the wall and roof. This thread is more specific to the roof:

http://chicagocadcam.com/ChrisHahn/house/TJIsection.jpg

in the link I just reposted, what if I

-put 1" EPS wedged in the top

- then blew 2" ccSPF against that to seal it

- then either (1) blew in 8"ocSPF or used a HD Batt to achieve the rest of the Rvalue?

The EPS would give me the airgap and a "backing" for the ccSPF so I don't spray it directly against the roof sheathing. The ccSPF would seal the assembly at the inside edge of the EPS, and the batt or ocSPF would further insulate.

How deep are the TJI's? Depending on the total amount of vapor permeable insulation that is going on the inside, you may want more foam to the exterior for dew point control. Also, since it has yet to be said in this thread, make sure the drywall ceiling is as airtight as you can make it, the drywall is easier to air seal than that foam-TJI connection, and you probably already know to avoid can lights, even the "air tight" ic rated ones.

"make sure the drywall ceiling is as airtight as you can make it"

If I blow ccSPF at the top, making the drywall plane airtight would be a no-no considering if moisture DID get into the cavity, it would have a hard time escaping.

Ok, I'm back to square one on this TJI roof rafters situation. How does one properly insulate ceilings that are cathedral 14in TJIs in Zone5? I assume someone has done this before? Batt products suck, spray foam is expensive and gets blown against the roof sheathing which I'd prefer to avoid, rigid is difficult to install...

Is this still in the planning stages or is the roof on?

If no roof, consider putting the foam on top of the trusses, build and air gap and install the roof.

If roof is on, install osb sheathing where you where going to put the foam, put the foam under the trusses and fill with cellulose.

which ever side you put the foam, make that your air barrier. Install air tight drywall no vapour barrier.

Airtight drywall on a cathedral ceiling is the opposite of a no-no. There is a difference between an air barrier and a vapor barrier. The airtight drywall is,as the name implies, an air barrier. But drywall as a material is very vapor open. As long as you do not install poly, which is the real no-no in this situation, and use vapor open paint (most latex paint is open enough) any moisture that getis into the cavity can diffuse out.

Airtight is KEY as air infiltration in the ceiling can bring in an order of magnitude more water than simple vapor diffusion. And in this case if you use enough foam to the outside of the stackup, then any moisture from the inside that diffuses INTO you ceiling will not get a chance to condense on the inner foam surface. Hence wanting to know the thickness of the TJI. I'll see if I can't scrounge up the code reference that stipulates the minimum required foam thickness. Assuming you are filling the TJI cavity completely with insulation.

I'd be interested in any suggestions here. Summary is Zone5, TJI-14s, metal roof. This is what I was going to do previously...

SS metal roof
#30 felt or synthetic
Sheathing
TJI 14" joist cavity 24" o/c
2" rigid polyiso board ripped to 1' 11-3/4" wide wedged between each TJI leaving 1-3/8" vent (R-12)
1" closed cell SPF for air seal (+R-7 for R-19)
HD batt or spider to fill the rest of the cavity which is 9-1/2" left open
gypsum

R38 is required so 38-19 = R19 batt

With the 9.5" though I can use a R30 HD batt which is only 8.25" thick for a total of R49. leaving the lower nailing span gap alone