I've read lots of posts here, and over at buildingscience.com, but nothing exactly hits our scenario so I thought I'd post to get ideas.

We have a 1937 farm house, 2x4 construction, with tongue & groove sheathing and cedar shake siding.  The interior is completely gutted (empty stud bays).  The exterior is in good shape and I'd like to keep it.  My hope is to insulate the walls to R30, or close, with a near-perfect air seal.  We are replacing the windows too, insulating the roof, etc.  It's a long list.

In my area (Oregon), it seems blown-in cellulose is a non-option, and I'm not particularly interested in spray foam because I'd like to do the work myself.  I'd also like to have wool or cotton batts as the last layer against the living space (for sound-deadening / formaldehyde capturing properties).

Anyway, my plan at the moment is to add unfaced fiberglass into the stud bays (assuming I can find these) for R13, then 1 inch of XPS foam, tongue and grove with seams taped, calked and foamed into place for an air barrior ( and 1.0 perm moisture retarder), then add a new 24" on center 2x4 interior wall with staggered studs, bays filled with wool/cotton insulation for R13.  All services run through the interior wall.  This gives R31, best case.

My concern is how moisture will react with this wall assembly since I can't find a similar case online.  I know heating dominated climates suggest the moisture retarder on the inside wall .. we'll have latex painted drywall.  Anything beyond that I'd be concerned about creating a double barrier.  I was wondering if I could treat that interior XPS as "sheathing" in the wall system (since everything outside it is very permeable).  An article on buildingscience.com suggested that moisture condensation on the inside of foam sheathing was a concern only if it was below the dewpoint temperature for the thermal cross-section of the wall.  This would certainly not be the case, with R18 beyond the interior surface of the foam (outside temperatures never approach 0*F here).

Anyway, I'd love to get suggestions and comments.  Other solutions welcome too.  Most sources I find online don't really address these remodeling situations, especially where someone wants to go beyond the "good enough" R13 sold at the home center.

Why is blown in cellulose not an option?

You are on the right track. I just got done building a similar new construction house, using double 2*3 walls with 2.5" of white beadboard foam between. We used that foam because the budget was very tight and we found that foam used for $2 for a 3.3'*7' board. When you start to account for realistic numbers, your outer 2*4 wall which I am assuming is on 16” centers will probably only actually net you R-10 by the time you account for the thermal bridges of the studs, plates, headers, ect. Your inner wall, which can be done with a single top/bottom plate and 24” centers, can get you an actual R-11. So your final actual R-value will be R-21+R-foam. By the time you go through all the trouble of building the second wall, I would want to use a thicker foam, like a 2” R-max or similar closed cell foam, which has an R-value of 12 or 13, which will net you an R-33 actual wall. If you want to make the wall thinner, you can use a 2*3 inner wall, just be sure you get good material. The 2*3’s we got were so twisted we had to cull so many that it would have been cheaper to get 2*4, but if you monitor quality of the wood, this wouldn’t be an issue for you. If you can’t find unfaced fiberglass, used faced fiberglass and cut slits in the paper to let vapors through. If you are interested in adding to the structural qualities of your home, after you put in the first fiberglass before you put the foam against the outside stud wall, put a layer of plywood over all the inside, glued and nailed off, and it will create a double diagram and add strength to your home. If there isn’t currently an air retarder in the system, taping and sealing the foam will work well, just be sure not to create another air barrier, because you don’t want to trap moisture between two barriers, you want to have one barrier and let the moisture leave in both directions out the wall.

If you have any more questions, shoot me a PM.

Thanks for the comments so far.

Matcartier - Regarding cellulose, I'm not sure. I've talked to some builders and also insulation companies. The consensus seems to be a fear of moisture + cellulose, given the amount of rain we get here. I'm not sure this is a well-founded fear, but it does seem pretty un-available. I live in the Eugene area .. Portland might be different.

aardvarcus - Wow, that sounds very similar to what I'm planning. I agree that my "best case" R values are unrealistic. 2" foam would be better, but I was concerned about the further reduced moisture permeability. If this is not an issue (i.e. if my sheathing assumptions are correct), than 2" XPS is definitely better R/$ based on local prices. I wish there was a used foam supplier around here. The only major difference between the wall systems was that the home you built used EPS foam, which has a higher permeability of 2.0 to 5.0, vs. less than 1.0 for XPS.

I had not thought about increasing the sheer strength of the house, although it certainly couldn't hurt. With the drywall attached to the inner wall it won't be contributing as much sheer as it would otherwise.

Take a look at Eugene's climate stats - the average wintertime temps are in the low 40s, so even if you allowed the interior relative humidity to hit 60% (which the highest you should allow for both health & comfort), with an interior temp of 70F the dew point occurs at around 55F, according to the psychrometric charts, which is roughly the halfway point of your R-value.  As long as the lowest permeance layer occurs on the warm side of that point with limited air exchange to the interior, it'll dry to the exterior just fine.  Outside wintertime relative humidities run around 86% in both December & February, so drying from bulk water incursions can be slow during those months no matter what you use for insulation.

Batts are inherently air-leaky- if you go that route, be sure to air & vapor seal the interior well (vapor retardent paint is fine) or you're likely to get condensation in the insulation.

Blown insulation is inherently less leaky- with fewer gaps an NO compressions convection loops within the insulation are limited (even more so with dense-packed cellulose, but even standard density installation is pretty good.)

Also cellulose redistributes moisture readily, drawing it away from structural timbers, keeping it dry.  When condensation DOES occur in cellulose it'll wick away, but will cause some settling of the material with repeated events (you may have to top it off in 25 years.)  Both the local big-blue & big oragne box stores in Eugene carry cellulose- you might call to find out if they have rental blowers (usually free for a few days if you buy over some minimum quantity.)  It's not rocket science. Other construction supply houses might be able to get you "borate only" or "sulfate free" stuff, which is preferable (keep the sulfated fire retardents away from copper, eh? Be careful about direct contact with the plumbin'... )

If you go with the double stud wall concept, going with 2x3s on the interior and have a 10" wide fill the cellulose itself would form an adequate thermal break at the studs without foam, and deliver a true ~R35 clear wall/R 30 whole-wall performance. 

If you still want to sandwich foam in there, R8, 2" foil or poly faced EPS (taped & caulked) is cheap, and would put the air/vapor retarder the right place, and you won't have to be as fussy about air-sealing on the interior finish wall, and batts on the interior studwall would be fine.  Avoid vapor retardent wall finishes with this layup.  This too would deliver about R35clear/R30 whole wall performance.

The cellulose in double-wall has been analyzed in some detail for thermal & hygric aspects- download the high-R walls document from Building Science, and search the document for discussions regarding Case 4.  Reading it carefully, I think you'll find that with either an interior vapor retarder or placed in the center of the total R, it'll do pretty well in Eugene.  If you want to learn more than you thought possible about blowing celluloses, search the web for [karg cellulose].  (Rick Karg has TONS of web-published tips of the trade.)  This is definitely within the possibilities of a dedicated DIYer with a rental blower.

Addendum: 10" of cellulose would also add substantial thermal mass to the place. If you're up for it, this is likely the cheapest high-performance way to go, and would likely outperform even the best batt layup since it stores & releases heat during diurnal temperature swings, lowering the shoulder-season heat loads. (It's thermal mass is roughly equivalent to 3" of concrete at 10" thickness, 2-2.5lbs per cubic foot density, but with a lower heat transfer speed.)

But if you still want to do it with batts, go high-density "cathedral ceiling" batts for lowest internal convection/highest performance, and DO install at least 2" of poly-faced EPS (best R value foam per dollar, as aardvarcus points out) or foil faced ISO (highest R value per inch) between the stud walls. High density batts are good for about R15 in 2x4 construction, so it'll still be a good performer if with an R8-R12 vapor-retardent air barrier & thermal break between the studwalls to fully block convective bypassing loops within the insulation.

Wow, thanks Dana1 for such a detailed reply. I'm going to have to read it through a few times to digest and follow your references.

Yes, I've seen cellulose and blowers at the big-box stores here. (But not so much at local lumber yards / insulation suppliers.) It was my understanding that the blowers available at big box stores were only rated for distributing loose fill in an attic, not for filling wall cavities to beyond the settling density to ensure a complete fill. If this was not the case, I'll have to look into how I would retain the cellulose before the drywall is up. I'd rather be able to see that it was achieving a good fill. The thermal mass would certainly be a boon. Our primary heat is a high efficiency wood stove. It's good at dumping out a huge amount of heat in a very short period of time.

Before I got this message I was working on finding a reasonable source for a whole lot of XPS foam sheets. It occurred to me that I could do a better job in the outside bays by fitting 2" foam into each stud bay and leaving a 1.5" dead air space / drainage plane toward the outside. Our stud spacing isn't regular enough to make batt installation very pleasant, and just working with two types of insulation would be simpler for the walls. That would make it 2" XPS in the studs (~ R10), 2" XPS continuous between the walls (R10), and the 2x4 wall on 24" centers with wool/cotton (R13). About R30 with a true thermal break.

If you can get the foam to cut and fit nice and tight that would work. We had considered using foam in our stud bays, but by the time we factored in the labor of cutting them "just right", it would have been much harder. Here is an idea to consider. First get some faced fiberglass, usually the big boxes have 15"*3.5"*25' rolls really cheap. Run the batts horizontally, not vertically, and when you come to a stud, score the fiberglass, but not the paper, and pull the fiberglass to create a little gap, to where you can put the kraft paper backing right against the stud. I know this is hard to explain, so see the attached picture. Be sure you install them tight against one another any you won't have any problems.

I have been planning a insulation re-fit on my cottage and will be using EPS Foam.

2" foam between the studs won't perform anything LIKE R10 (more like R5-6), due to the ~R2 thermal short-circuits of the studs.  A cellulose fill will (thermal shorts included) give you more than twice that  at ~R12.

If you filled the stud bay completely with EPS (beadboard) you'd get about the same R as with cellulose, but without the thermal mass, and a higher liklihood of gaps & convection paths around the EPS.  XPS is good stuff, but EPS is more appropriate on the outer layers for thick layups with a dries-toward-exterior profile, and cheaper too.

Box-store cellullose blowers can be used for cavity fill (even dense packing, if you want to fabricate your own reducers & hoses).  The worst that happens when you don't get it to a high enough density is that over time it will settle, eventually performing only about as well as batts did on day 1.  It may take a 2-stage blower to hit 3.8-4lbs/ft^3, but even 3lbs ft^3 is a sufficient dense pack from most practical points of view. (In 50 years you can top it off, eh? ;-) )  Most manufacturers recommend densities of 3.0-3.5lbs/ft^3 to meet their dense pack performance specs- decent single-stage blowers can easily deliver 3, and even 2.5 won't settle very much.  (Open blow is more like 1.4-1.8lbs/ft^3). It's not tough math to figure out if you're hitting your density goals either, and if you need to, drilling a set of holes near the top to pound it a bit harder after the fact can bring your average density up where it counts the most.

If you plan to dense-pack it, either go with a double-layer of 1/2" (min) sheet rock, or better still, 7/16 OSB under the sheet rock layer or it'll likely bulge a bit over time.  Dense-packed or not, going with a layer of OSB on the interior wall and blowing the insulation would give you ample opportunity to drill & fill, inspect & touch-up as-needed before putting on the finish layer.

jleejj and Dana1,
Experience and in-depth research of air movement in fiberglas insulated walls will show you to use something else!!! (I think Dana 1 knows way more than me about it!) I came across this yesterday after someone had it in their blog on this site. Very good info on cellulose installation!

http://www.builditsolar.com/Projects/SolarHomes/LarsenTruss/LarsenTruss.htm
RichM

Quote:"2" foam between the studs won't perform anything LIKE R10 (more like R5-6), due to the ~R2 thermal short-circuits of the studs. A cellulose fill will (thermal shorts included) give you more than twice that at ~R12."

I'm sorry but that is WAY off.

Wood framing on old 16" center homes takes up about 25% of the cavity area.
R-12 means a U value of 0.083333333 or 0.0833333 btu's lost per hour per degree of temperature differential.
The BTUs lost by the wall equal the BTUs lost by the framing plus the BTU's lost by the cavity insulation.
The wooden studs have a R value of 1 per inch, so the will have a U value of .28571, and account for 25% of the wall area, so:
Wall U = (0.25 * Wood U)+ (0.75 * Fill U),
0.083333333=0.25*.28571+0.75*X,
X=0.01587,
R fill=1/X=62.9943
To get a R-12 with thermal shorts the cellulose fill 3.5" thick would have to have an R-Value of 62.9943.
That would mean cellulose would have an R/inch of 17.9983.
If this was true all the fiberglass and foam manufacturers would long be out of business.

aardvarcus: Your simple-arithmetic model is giving the wood siding in the assembly an R value of zero when it's actually about R1, and overestimating the stud coverage by almost a factor of 2. (15% is a widely cited number.) Even with fire blocking (not common in 1937) it's not anywhere near 25%. (It isn't anywhere near that in my full-dimension lumber circa 1923 bungalow, let alone a depression-era structure.)

Clear wall will be about R12, whole wall, depends on the number size of openings and how beefy the headers are. ORNL sez whole-wall R (with the rest of the framing included) is ~ R10.

Play with the ORNL sim a bit:

http://www.ornl.gov/sci/roofs+walls/AWT/InteractiveCalculators/NS/SimCalc.htm

The ORNL sim is based on test data from actuall assmblies. The 2x4 assemblies are 1.5" x 3.5", but even with full dimension lumber the thermal short of the stud is increased by only ~30% (33% wider, but also 13% deeper) but the fill depth would be increased by 13%, so it's mostly a wash. ORNL stands by their numbers- I tend to accept numbers from folks who actually measure stuff for a living.

With 2" of foam between the studs the depth of the short circuiting stud is about half that of a full fill, with about 1.5-2x the U value that it would have in a fully filled cavity. The stud-short is no longer easily represented by a simple model with the 3-D geometries involved. The extension of the stud beyond the foam has a different internal temperature profile than if it were fully between the foam to the full depth. For the sake of argument, call it 1.5x or ~U 0.4 in the R10 foam case, compared ~U 0.3 in the cellulose fill case.

Even if it was the world's crappiest cellulose, and it took the full 3.5" depth to hit R10, the R10 foam wall would take a much larger clear/whole-wall hit than cellulose due to the 50%+ increase in U-value of the 2" of stud depth vs. 3.5" or 4" of stud depth.

Most cellulose achieves ~R12.5 in 2-hole method, ~R13 dense packed for center cavity R, at 3.5" of depth, which is how the clear wall values can hit R12 with wood siding & sheathing included.

The calculator verifies what I just said, that the whole wall R value with the thermal shorts included isn't R-12. According to the calculator its more like R-10. That calculator also states a R-13 fiberglass outperforms dry cellulose and comes within R0.25 of wet blown cellulose.

Yes my simple calculation ignored the sheathing.

15% stud coverage is frequently cited for OVE stack framed buildings. When you start adding double top and bottom plates, 16" centers, and other things like how many jack studs they put under the headers, what they used for headers, ect. the % wall coverage really can jump up there. Around where I live, old houses usually have a plethora of extra studs, to the point where 25% is an underestimate.

It really doesn't matter because putting 2" of foam in a wall isn't a good idea anyway.(Trying to cut it out anywhere close to fitting in a stud cavity is a bear, I know I have done it before.)

You are correct, to truely model the short stud, we do need a 3-d model, but you can estimate what it will be.

When you consider the reduced depth of stud around the 2" foam you fail to account for the films of air that are going to develop on the back of the drywall and the surface of the stud.

The following chart can be found on Wikipedia.
Non-reflective surface R-values for air films
Surface position   Direction of heat flow   R (h·ft²·°F/Btu) 
Vertical (eg: a wall) Horizontal                 0.68

If the air is going to flow through the drywall, back into the air gap between the drywall and the foam, and then into the stud at the point where it meets the foam for the minimum stud traveling distance, then the R values encountered are as follows. I am ignoring the tiny R-values that will be shared in both scenarios.
2" Foam Thermal Shorts-
Ignoring film between the drywall and the inside.
Ignoring R value of drywall
R0.68 between drywall and air gap
Ignoring R value of air in the cavity
R0.68 between air gap and stud
R-2 bridge of stud
R-1 value of sheathing
Ignoring R value between sheathing and outside
The r-value of the thermal short is R-4.36.

3.5" Full cavity useage thermal shorts-
Using a 3.5" stud and the R-1 sheathing, with no air gaps involved and ignoring the same factors,
the R value of that thermal short is R-4.5

There isn't really a lot of difference here. R-0.14 isn't going to change the numbers much.




Aside- I don't trust anyone who measures stuff for a living without looking over their shoulder. At the last home show I visited a man who worked at the TVA energy savings booth told me that I needed between 300-400 air changes per hour in my house. No he didn't misuse terms and mean cfm, he actually thought the air should turn over this much. I wonder how many other people he told that to that didn't know better.

I wasn't even trying to tell him the whole wall number (which is a VERY squishy number, too dependent on the particulars), only the clear-wall- which seemed to be where he was going, but missing the all-important thermal shorts. (And even if you were thinkin' whole-wall, R10 isn't "...WAY off", it's "17% off", which is only "kinda-off" in my book, eh? ;-) )

I'd need to see a better of model of the air gap performance than this- simple arithmetic doesn't work well here, but you might be close (or not). Construction faults can wreak havoc with performance when air can flow freely, even if the model works well in the perfect world.

BTW: Did the guy advising you to live in a wind tunnel put it in writing? (Better yet, did he publish his model, put it on the web as a paradigm reference? :-) ) Even smart people come up with stupid stuff all the time, which certainly keeps it interesting (even if it makes it seem hopeless for the rest of us.)

I hope to goodness he didn't have it in writing anywhere. My father who was with me asked him where we could buy jet turbines to get that much air flowing, and he just looked at us funny. It was even funnier because he was dressed up in a white lab coat and he looked like he was a mad scientist. We tried to dicscuss it with him better, so he wouldn't confuse any more homeowners, but we might as well have been talking to the wall.

Wow, I check my e-mail and there's a whole lot more discussion on this thread. Thank you for all the replies back regarding whole wall vs. clear wall, various mathematic approaches to estimating it, and the calculator link. Way back in my first post about foam between the studs I think I threw an approximately symbol ("~ R10") in there and then knocked R3 off the ideal value at the end. Either way, after looking into it some more I agree that foam between the studs would be way too much of a headache. Gaps of only a few mm can seriously degrade thermal performance, and I couldn't hope to achieve much better even carefully ripping each sheet on a table saw. I'm going to have to use this approach when filling the 2x4 rafter bays above the second floor knee walls (the other 1.5" forms a bypass for attic ventilation), but otherwise I'm going with high density fiberglass between the outside bays. It's 1/3rd the cost of XPS and at 3.5" thick will net a higher R value too.

Dana1 - I completely agree with cellulose as an insulation product with many advantages. If I was building a new house today, I would be seriously considering a Larsen Truss design with dense-pack cellulose. However, after much research I've concluded that for a live-in remodel with limited local availability of cellulose, my proposed fiberglass+foam+cotton will be easier to do incrementally and will still yield a satisfactory result. Your comments and other research have eliminated my concerns about moisture, although I'm still debating EPS vs. XPS. I do plan for blown-in cellulose in the attic cavities. You mentioned the thermal mass of cellulose in a wall assembly. Where would I find the thermal mass rating of cellulose, BTUs per lb per *F. It occurred to me that cellulose blown in the floor cavity between the two levels might help even our the big temperature swings created by wood heat.

aardvarcus - I'm going to follow your fiberglass suggestions for the exterior wall, either running them horizontally or vertically depending on how even the stud spacing is. I can get high density R13 batts easy here, and since the vapor barrier will be against the foam (and more permeable than it), it probably wouldn't even be necessary to slit them. (Probably still will.) Thanks for putting up photos of your double wall house in a separate thread. I almost replied to that thread too before I realized it was the same person. :) How funny that I'm pondering this design just days before you post pictures of such a house being built. I'm sticking with 2x4 interior walls though ... more room for services. I hate those shallow electrical boxes, and 3.5" x 24" batts are standard even for cotton and wool. For the outer layer of a knee wall exposed to the attic, would you recommend that I nail on OSB to create 'sheathing' (facing the attic). I know fiberglass batts require dead-air space to be effective, but I'm not sure how people usually detail this area.

Finally, some quick math for my little house suggests that achieving 300ACH would require moving air through a 6" duct at 850MPH. It would be simpler to just build a house with no exterior walls.

Some folks never check to see where the decimal point lives, eh?

I've seen stuff that wild written up as application notes for electronic components though- obviously by "tech writers" without so much as a clue...

Did he calculate the wind chill factor on that for ya too?

I'm going to have to use this approach when filling the 2x4 rafter bays above the second floor knee walls (the other 1.5" forms a bypass for attic ventilation)
Thats what we did on the double stud house you see pictured. Works well, but they are a bear to cut. We ended up using a circular saw with a throwaway blade held tight to a 2*4 board, because the saw likes to wobble and the foam offers no resistance to the blade turning sideways.

I know fiberglass batts require dead-air space to be effective, but I'm not sure how people usually detail this area.
Fiberglass doesn't need dead air space as in blank cavities on either side of it, because it creates thousands of little air spaces between the glass fibers when it is fluffed and put into place. I believe you are thinking of radiant barriers, because they need an air gap on the outside to work.

 For the outer layer of a knee wall exposed to the attic, would you recommend that I nail on OSB to create 'sheathing' (facing the attic).
From a structural perspective, it would help, but first, how are the rafters attached to this knee wall and how is this knee wall attached to the wall below it. If the rafters are just toe nailed into the top plate of this wall, you need to firmly attach them to the wall below using some sort of metal bracket or tiedown. What you are really trying to accomplish is a system to where as the wind puts pressure on the roof, the rafters are anchored to the kneewall plate, the kneewall plate is anchored to the kneewall studs, the kneewall is anchored to the wall below, and the wall below is anchored to the foundation.

Assuming your house is "standard construction", this is how you might go about it.
Use those metal ties with nail holes and the 90 degree twist to nail firmly to the side of the rafter and then firmly into the side and bottom of the plate.

Assuming that the lower wall just ends and that the kneewall is built seperate, use the coils of metal band with the holes drilled in them to bridge from the studs of the lower wall to the studs of the knee wall. If the knee wall is just the same studs of the lower wall sticking up in the attic, nothing else is needed here.

Anchor the bottom plate of the wall into the foundation, using metal straps, blue screws and washers, ect. I can't really predict what this will be like without more information.

Once you have gotten all of these connections reinforced, now you can begin to put plywood diaghrams onto your walls. The most important part of all this is connecting the plates to the studs firmly. For example, if you had a 5' kneewall, you might just rip the plywood into two 2' by 8' pieces, then put one on the lower 2', firmly capturing the bottom plate and the lower parts of the studs, one piece on the upper 2', firmly capturing the upper plate and the upper parts of the studs, and ignore the part in the middle.  The key is if you have to patch in small pieces, do it in the middle, not at the top or bottom.

Before you put any plywood on this wall, stuff some insulation over and around the top of the wall on the other side of the kneewall, to help keep heat flowing up the outside wall inside.

I am going to try to MS paint you a diagram really quick. It is really out of scale, but you should get the idea. If your house isn't built like I had tried to draw it, just let me know and I will correct the drawing.

Wow, I wasn't expecting this thorough of a bracing discussion but I'm all for advice.  Our house is odd, "built by a logger" as the expression goes around here.  (Simple things like window & door headers were outside their understanding.)  We've had a bit of work done to resolve the most serious structural issues, but there isn't any wind or seismic bracing to speak of.  I've just started looking into the necessary retrofits.

The knee wall doesn't have a top plate.  Each stud is cut to fit below a corresponding rafter, and a steel plate (recent addition) holds them together.  The knee wall is supported by a beam.  Beyond it are the joists for the first-floor ceiling.   Attached below is a scale cross-sectional drawing.

When I said "dead air space", I meant that fiberglass is supposed to be installed in a cavity with 6 solid surfaces around it ... not exposed to drafts.  If it's installed between the studs in our kneewall, then at the top it will be exposed to the drafts of attic ventilation while at the bottom it will have loose-fill cellulose pilled up against it.  I may be reading too much marketing comparisons from the other insulation types.

Yes, I was planning on installing strapping between the two floors on the outside walls.  Not possible for the knee-walls, because they are all over open spanned areas of the house.  I'll need to find a fastener to positively tie the beams into the exterior walls.

I believe the way the rafters sit on our exterior wall is decidedly non-standard.  It's good from an insulating standpoint ... plenty of room for in-fill cellulose.  It seems like through-bolts with blocking would be the best way to positively tie the plate the rafters sit on to the double top-plate of the wall below.  The straps you mention with the 90* twist could tie down the rafters.

The pink rectangle indicates the position of those ripped foam sheets between the studs I was talking about.  They'll extend into the upper attic space far enough to keep loose-fill insulation from getting blown down them.  I plan to add similar foam baffles at the exterior wall to protect the vents.

There still a few areas of bracing I'm probably missing, or unclear on.  Also, is any metal hardware used to connect the corners?  Seems like a weak point, if everything else is braced down to the foundation.