I can't with this cantilever! Dana please help

[HydroNick]

Dana ive been following your posts and your a wealth of knowledge. Thank you! I've seen your recommendations on insulating similar cantilevers but none were as ridiculous as mine. I have a 1962 split level ranch transformed and remodeled several times over 60 years. Not a care was given to insulation, heat loss, or logic. Im a plumber so please forgive me if im not great with structural/insulating terminology and practices
Im in Long Island, NY. I believe zone 4A. Im trying to insulate below a bay window which is on my cantilevered first floor.
3×7 joists run over poured cement walls. Brick facade was added at one time. Much later on about 10 years ago the previous owner slapped stone facade onto the brick. Basement and house gets humid in summer and drafty floors in winter. I plan to have dense packed cellulose installed in the rest of the house. I am trying to tackle the bay window in front myself. The backyard facing side of the house has 45ft of overhang over poured concrete wall. Pretty sure no brick.
So under my window I planned to do this:

However based on the following pics I may be wrong going by what i uncovered when i opened it up. So how should I insulate my cantilever?
8ft basement wall interior facing out.



Exterior facing into cantilever


1/4 plywood which is also exterior bottom of overhang and two pieces of .5" what looks like tar paper faced (both sides) sheet rock is slid under joists butting up against sill creating gap. The sheet rock is covering the brick. SO..im going to assume their is a 1" gap for brick. Im also thinking the brick is venting into the interior.





Exterior foundation below. I see no weep holes anywhere


Could it be venting out the sides and not into the cantilever?

 

[John Gayewski]

Foam board to prevent transmission. Fiberglass to prevent infiltration.

Line any exterior contacted areas with foam board. Then use fiberglass or cellulose to prevent air seepage. Packing insulation too tight defeats its intended purpose which is to trap air.

 

[Sponsor]

 

[Dana]

Foam board to prevent transmission. Fiberglass to prevent infiltration.

Line any exterior contacted areas with foam board. Then use fiberglass or cellulose to prevent air seepage.

Fiber insulation has no ability to block air, prevent air infiltration, or air seepage. Packing it tighter makes it more air retardent (it can slow the rate of air movement), but there is no way to make it behave as an air barrier.

Foam board can be detailed as an air barrier, and is generally more vapor-retardent (to vapor diffusion) than fiber insulation.

Packing insulation too tight defeats its intended purpose which is to trap air.

Also counter to the reality. Packing fiber insulation makes it more air retardent and up to a point delivers a higher R per inch. Eg: An R19 fiberglass batt is delivers it's rated R19 performance at it's manufactured loft of ~6". That's (R19/6"=) R3.2/inch. At that density it's air retardency is very low- it's more like an air filter than an air retarder, and in no way resembles an air BARRIER. Fiber insulation does NOT "trap air", even at very high densities. When installed and compressed into a 5.5" 2x6 framing cavity it only performs at R18, or (R18/5.5"=) R3.3/inch. The weight per square foot of an R19 batt is identical to that of an R13 batt designed for 3.5" deep 2x4 framing, and magically when compressed to 3.5" the R19 performs at R13, or (R13/3.5"=) R3.7/inch. At that density it's air redardency is much improved, but still nowhere near that of a much denser R15 batt designed for 2x4 framing (R4.3/inch). When compressed t= into the 2.5" of a 2x3 framing the performance of an R19 or R13 batt is R10, or (R10/2.5"=) R4.0/inch. (Compressing an R15 HD batt to 2.5" delivers R11, or R4.4/inch, still improvement over the R4.3/inch of it's manufactured loft.)

Don't take my word for it- consult the charts provided by the manufacturers.

At low densities cellulose is much much more air retardent than low or mid-density fiberglass. Typical "2-hole method" cellulose runs about 2-2.5'lbs per cubic foot and ~R3.5-R3.7/inch, and is fairly air retardent, enough that convection up through a wall cavity is fairly well blocked (far more air retardent than an R19 batt.) Dense packed to 3.5lbs (a common target density) the thermal performance is still only R3.7, but the amount of air and moisture that can move through it is much reduced, and it is tight enough to not settle under normal seasonal moisture cycling. Blown fiberglass will never settle (at any density) , and hits R3.7/inch even at a fairly low 1.0lbs density, but it is extremely air permeable at that density. Dense packed blown fiberglass becomes roughly comparable in air retardency to 3.5lbs cellulose at about, 1.8lbs density, where it performs at about R4.2/inch of thickness, beating cellulose on R-value. But to totally beat the air retardency of 3.5lbs cellulose takes about 2.2lbs per cubic foot with fiberglass, where it's performance hits R4.3/inch.

See the R-value & density charts of a few blown fiberglass vendors. See also this bit of marketing fluff from on vendor comparing the air retardency of cellulose vs. their own product across density.

There are other relevant differences in characteristics between fiber type, but none will every trap or block air completely at densities that can be installed at pressures that won't blow the drywall & sheathing off the framing.

 

[John Gayewski]

For the insulation to insulate it needs to accept air. Packing it will make it more dense (increasing the r value) but the air will blow right around it. Hence the polystyrene for transmission and cellulose or fiberglass for infiltration. Line areas that touch the exterior with polyboard. Then "pack" insulation not too tight.

R value measures the rate of temp change which the packing can help, but you can have all the R you want, it doesn't matter if you're deflecting air seepage around the huge R you just packed.

No one is claiming fiberglass insulation is a vapor barrier. It simply issues air to insulate in this case by soaking it up and stopping the movement. Tightly packed fiberglass around piping has another use that is a transmission application with a foil wrap for an air barrier.

 

[Dana]

I've seen your recommendations on insulating similar cantilevers but none were as ridiculous as mine. I have a 1962 split level ranch transformed and remodeled several times over 60 years. Not a care was given to insulation, heat loss, or logic. Im a plumber so please forgive me if im not great with structural/insulating terminology and practices
Im in Long Island, NY. I believe zone 4A.

Welcome to the club!

Air-leaky and uninsulated bay windows & other cantilevers seems par for the course for that era, when energy was comparatively cheap, and codes were as sloppy as the tract-home builders.

(Yes, L.I. is zone 4A.)

3×7 joists run over poured cement walls. Brick facade was added at one time. Much later on about 10 years ago the previous owner slapped stone facade onto the brick. Basement and house gets humid in summer and drafty floors in winter. I plan to have dense packed cellulose installed in the rest of the house. I am trying to tackle the bay window in front myself. The backyard facing side of the house has 45ft of overhang over poured concrete wall. Pretty sure no brick.

A 3x7 is an unusual size. A milled 4x8 would be a nominal 3.5" x 7.25". Could those be the actual dimensions?

The sill plate (the plank that rests atop the foundation, on the under side of the joists) appears to be a double layer of 2x milled lumber (true dimension 1.5" x ___) which is fine, but requires a bit of detailing to make it air tight. Even if there is a sill gasket between the concrete and the foundation sill (probably isn't), it's probably not going to be sufficiently air tight. If there is enough space to slip a knife blade between the concrete & wood a soupy "self leveling" polyurethane caulk of the type used for sealing cracks in concrete would be ideal for sealing up that seam, otherwise use a common polyurethane caulk to seal the bottom board to the concrete, and the seams between the stacked boards.

There also needs to be an air barrier to block the free flow between the basement and the cantilever. Foam board sealed at the edges with expanding foam is good enough for most bays, but the bay where the HVAC duct crosses will also require sealing the duct to the foam board. For the metal to foam board transition use mesh tape and duct mastic. (Seal all the seams & joints in that duct before insulating and closing it in.)

1/4 plywood which is also exterior bottom of overhang and two pieces of .5" what looks like tar paper faced (both sides) sheet rock is slid under joists butting up against sill creating gap. The sheet rock is covering the brick. SO..im going to assume their is a 1" gap for brick. Im also thinking the brick is venting into the interior.

You may have to do a bit of exploratory demolition to determine if there is in fact a ventilation gap between the brick & concrete. If you can't see vent penetrations into the basement or into the cantilever bays there is no venting, but it may also not be necessary. If the surface grading is away from the foundation and protected from direct rain-wetting by the overhang the soil next to the foundation & brick + stone veneers stays pretty dry, and as long as the poured concrete has at least some drying capacity toward the interior closing up and insulating the cantilever isn't likely to create any problems.

Strip the bottom sheathing off to get better access to the subfloor & band joist working from outdoors. Air seal the band joist to the subfloor, caulk any electric penetrations (either can foam or polyurethane caulk), and seal any seams in the subfloor with mesh tape & duct mastic (which will stick to old/dirty/oxidized wood far better than tapes). Be sure to air seal the duct boot to the subfloor as well. (If the space is too tight to seal the duct boot from below, pulling the register and sealing it to the finish floor from above is an acceptable alternative, though often less than perfect.)

Where there is good access from the bottom it's pretty easy to fit R30 rock wool batts tightly into a 7" deep cavity with very few voids or depressions. Common batts are designed for compression fit side to side with 2x framing (1.5" nominal) on either 16" or 24" on centers and you may need to trim the edges with a batt knife to keep it from buckling, creating potential thermal bypass voids. Rock wool batts are manufactured about 1/2" wider than the anticipated framing bay width (to guarantee a full contact compression fit), with slightly less dense fiber at the edges to keep the batt from buckling. With 3" or 3.5" wide joists you may need to narrow them by 1.5-2" to make it fit well. At the correct width it will be self-supporting when installed in the joist bay, with full contact on both sides, but with no caving or buckling.

On the bays with the duct work it's going to take some careful sculpting to limit the voids, but the goal is to have no empty spaces between the duct and insulation, or the subfloor/band joist/foundation side air barrier and the insultation- full contact and no air channels is the ideal. A purpose-made batt knife is pretty easy to use (requiring nowhere near the manual dexterity of most common plumbing tasks), but an 8" or longer bread knife can also be employed as a batt knife with good effect- it's almost as easy. Utility knives are commonly used by DIYers and handyman-hacks for trimming batts, but are a truly third rate tool for the job compared to a bread knife or purpose made batt knife.

The new bottom sheathing should be either somewhat vapor permeable (say half inch CDX plywood, which is under 1 perm when dry, but about 4-5 perms when damp enough to support mold). Exterior grade fiberglass face gypsum board or MDF can be used, but would need to be painted with an exterior latex primer to avoid being so vapor-open that mold levels of moisture accumulate on the subfloor of an air-conditioned house. Summertime outdoor dew points in your area linger in the high 60s and low 70s for weeks at a time, and a 20 perm gypsum board has the potential become an issue, albeit not a HIGH potential. Again, air sealing the bottom sheathing at the cantilever is more important than it's vapor retardency.

 

[Dana]

For the insulation to insulate it needs to accept air. Packing it will make it more dense (increasing the r value) but the air will blow right around it.

????

Simply not true, on either count. The entrained air is but one part of how fiber insulations work, and not the most important. Dense packing blown insulation (or properly fitting batts) fills all voids, leaving no space for the air to "...blow right around it....".

Hence the polystyrene for transmission and cellulose or fiberglass for infiltration. Line areas that touch the exterior with polyboard. Then "pack" insulation not too tight.

Unlike fiber insulation, polystyrene is EXTREMELY air retardent, and can be detailed as an air barrier. The "...blow right around it..." theory would be the same with foam board too, eh?

It's actually difficult to pack fiber insulation "too tight". The key to getting the performances out of it is a full compression fit with no voids that would allow air to thermally bypass it unimpeded. That's easy to do with blown insulation but takes some patience and care with batt insulation (which can be pretty good, but never as perfect as blown.)

R value measures the rate of temp change which the packing can help, but you can have all the R you want, it doesn't matter if you're deflecting air seepage around the huge R you just packed.

R-value is a steady state measurement (in an ASTM C518 plate to be legal for labeling purposes) and has nothing to do with "...rate of temp change...", which comes into play for measurements such as thermal diffusivity, or "dynamic R" or "DBMS" when assessing thermally massive assemblies.

No one is claiming fiberglass insulation is a vapor barrier. It simply issues air to insulate in this case by soaking it up and stopping the movement. Tightly packed fiberglass around piping has another use that is a transmission application with a foil wrap for an air barrier.

There hasn't been any commentary about the vapor transfer characteristics of fiber insulation, only it's air retardency. Those are not the same thing. Vapor retardency and vapor barriers have to do to how rapidly water vapor (not air) will diffuse through a material or surface when there is difference in vapor pressure (think of it as the absolute, not relative humidity) from one side to the other. A sheet of foil is highly vapor retardent even if it's full of holes and cuts. Air retardency has to do with how rapidly air will move through the material when there is an air pressure difference, which could be induced by wind, stack effect, air handlers, etc. Since air usually has moisture in it (sometimes a lot of moisture), the less air retardent a material is, the more air transported moisture can move into or through the assembly, which in most cases is something to be avoided in wood sheathed wood framed wall/floor/ceiling assemblies. Dense packing slows down the air transported moisture a lot- enough to make the heat transfer of the air flow negligible, but not necessarily well enough to keep moisture transfer within bounds.

In winter in zone 4A indoor air carries substantially more moisture than the outdoor air, so the air barriers on the conditioned side need to be fairly well implemented. It also helps if those air barriers are vapor retardent, but the vapor retardency is far less important. As it happens at less than 1 US perm a 3/4" plywood subfloor is fairly vapor retardent (when dry), and qualifies as a "vapour barrier" under Canadian building codes. But it's only an adequate air barrier if detailed as such, sealing every seam, penetration to prevent indoor air from getting in to the assembly.


The notion that fiber insulation "...issues air to insulate in this case by soaking it up and stopping the movement..." is an odd way of framing it. The fiber doesn't soak up air and it doesn't need to- the total gas volume of the air is constant at any given fiber density. Yes, keeping the entrained air in the fiber from moving is important, but the total gas volume of that air per cubic foot of insulation is pretty irrelevant. Keeping the entrained air from moving is more important for impeding heat transfer than how much air it has. Dense packing reduces the motion of air (via convection or wind driven infiltration, etc.), improving the overall thermal performance of the assembly.

 

[John Gayewski]

????

Simply not true, on either count. The entrained air is but one part of how fiber insulations work, and not the most important. Dense packing blown insulation (or properly fitting batts) fills all voids, leaving no space for the air to "...blow right around it....".



Unlike fiber insulation, polystyrene is EXTREMELY air retardent, and can be detailed as an air barrier. The "...blow right around it..." theory would be the same with foam board too, eh?

It's actually difficult to pack fiber insulation "too tight". The key to getting the performances out of it is a full compression fit with no voids that would allow air to thermally bypass it unimpeded. That's easy to do with blown insulation but takes some patience and care with batt insulation (which can be pretty good, but never as perfect as blown.)



R-value is a steady state measurement (in an ASTM C518 plate to be legal for labeling purposes) and has nothing to do with "...rate of temp change...", which comes into play for measurements such as thermal diffusivity, or "dynamic R" or "DBMS" when assessing thermally massive assemblies.



There hasn't been any commentary about the vapor transfer characteristics of fiber insulation, only it's air retardency. Those are not the same thing. Vapor retardency and vapor barriers have to do to how rapidly water vapor (not air) will diffuse through a material or surface when there is difference in vapor pressure (think of it as the absolute, not relative humidity) from one side to the other. A sheet of foil is highly vapor retardent even if it's full of holes and cuts. Air retardency has to do with how rapidly air will move through the material when there is an air pressure difference, which could be induced by wind, stack effect, air handlers, etc. Since air usually has moisture in it (sometimes a lot of moisture), the less air retardent a material is, the more air transported moisture can move into or through the assembly, which in most cases is something to be avoided in wood sheathed wood framed wall/floor/ceiling assemblies. Dense packing slows down the air transported moisture a lot- enough to make the heat transfer of the air flow negligible, but not necessarily well enough to keep moisture transfer within bounds.

In winter in zone 4A indoor air carries substantially more moisture than the outdoor air, so the air barriers on the conditioned side need to be fairly well implemented. It also helps if those air barriers are vapor retardent, but the vapor retardency is far less important. As it happens at less than 1 US perm a 3/4" plywood subfloor is fairly vapor retardent (when dry), and qualifies as a "vapour barrier" under Canadian building codes. But it's only an adequate air barrier if detailed as such, sealing every seam, penetration to prevent indoor air from getting in to the assembly.


The notion that fiber insulation "...issues air to insulate in this case by soaking it up and stopping the movement..." is an odd way of framing it. The fiber doesn't soak up air and it doesn't need to- the total gas volume of the air is constant at any given fiber density. Yes, keeping the entrained air in the fiber from moving is important, but the total gas volume of that air per cubic foot of insulation is pretty irrelevant. Keeping the entrained air from moving is more important for impeding heat transfer than how much air it has. Dense packing reduces the motion of air (via convection or wind driven infiltration, etc.), improving the overall thermal performance of the assembly.

Writing several paragraphs to prove how insulation works isn't a hobby of mine. In fact the more simply you can explain something is an indicator of how well its understood. Arguing the difference between "soaking up", our actual insulator, which in this case is air, and "holding entrained" air doesn't really serve a purpose, other than to scratch an itch you may have. I'll stand by my statements and go about my day. Feel free to offer a better solution for the person asking a question.

My objective is to paint a clear picture of the objective and passivley explain the process of how things work. Going into detail with charts and graph and using language soup gets in the way of conveying ideas.

We line the space with a transmission reducer/ vapor barrier (since you'd like to include that use). Then we make a big fluffy pillow the size of the space and try not to over stuff it. We'd like to keep our air "entrained" and not moving through our around our pillow.

 

[HydroNick]

Welcome to the club!

Air-leaky and uninsulated bay windows & other cantilevers seems par for the course for that era, when energy was comparatively cheap, and codes were as sloppy as the tract-home builders.

(Yes, L.I. is zone 4A.)



A 3x7 is an unusual size. A milled 4x8 would be a nominal 3.5" x 7.25". Could those be the actual dimensions?

The sill plate (the plank that rests atop the foundation, on the under side of the joists) appears to be a double layer of 2x milled lumber (true dimension 1.5" x ___) which is fine, but requires a bit of detailing to make it air tight. Even if there is a sill gasket between the concrete and the foundation sill (probably isn't), it's probably not going to be sufficiently air tight. If there is enough space to slip a knife blade between the concrete & wood a soupy "self leveling" polyurethane caulk of the type used for sealing cracks in concrete would be ideal for sealing up that seam, otherwise use a common polyurethane caulk to seal the bottom board to the concrete, and the seams between the stacked boards.

There also needs to be an air barrier to block the free flow between the basement and the cantilever. Foam board sealed at the edges with expanding foam is good enough for most bays, but the bay where the HVAC duct crosses will also require sealing the duct to the foam board. For the metal to foam board transition use mesh tape and duct mastic. (Seal all the seams & joints in that duct before insulating and closing it in.)




You may have to do a bit of exploratory demolition to determine if there is in fact a ventilation gap between the brick & concrete. If you can't see vent penetrations into the basement or into the cantilever bays there is no venting, but it may also not be necessary. If the surface grading is away from the foundation and protected from direct rain-wetting by the overhang the soil next to the foundation & brick + stone veneers stays pretty dry, and as long as the poured concrete has at least some drying capacity toward the interior closing up and insulating the cantilever isn't likely to create any problems.

Strip the bottom sheathing off to get better access to the subfloor & band joist working from outdoors. Air seal the band joist to the subfloor, caulk any electric penetrations (either can foam or polyurethane caulk), and seal any seams in the subfloor with mesh tape & duct mastic (which will stick to old/dirty/oxidized wood far better than tapes). Be sure to air seal the duct boot to the subfloor as well. (If the space is too tight to seal the duct boot from below, pulling the register and sealing it to the finish floor from above is an acceptable alternative, though often less than perfect.)

Where there is good access from the bottom it's pretty easy to fit R30 rock wool batts tightly into a 7" deep cavity with very few voids or depressions. Common batts are designed for compression fit side to side with 2x framing (1.5" nominal) on either 16" or 24" on centers and you may need to trim the edges with a batt knife to keep it from buckling, creating potential thermal bypass voids. Rock wool batts are manufactured about 1/2" wider than the anticipated framing bay width (to guarantee a full contact compression fit), with slightly less dense fiber at the edges to keep the batt from buckling. With 3" or 3.5" wide joists you may need to narrow them by 1.5-2" to make it fit well. At the correct width it will be self-supporting when installed in the joist bay, with full contact on both sides, but with no caving or buckling.

On the bays with the duct work it's going to take some careful sculpting to limit the voids, but the goal is to have no empty spaces between the duct and insulation, or the subfloor/band joist/foundation side air barrier and the insultation- full contact and no air channels is the ideal. A purpose-made batt knife is pretty easy to use (requiring nowhere near the manual dexterity of most common plumbing tasks), but an 8" or longer bread knife can also be employed as a batt knife with good effect- it's almost as easy. Utility knives are commonly used by DIYers and handyman-hacks for trimming batts, but are a truly third rate tool for the job compared to a bread knife or purpose made batt knife.

The new bottom sheathing should be either somewhat vapor permeable (say half inch CDX plywood, which is under 1 perm when dry, but about 4-5 perms when damp enough to support mold). Exterior grade fiberglass face gypsum board or MDF can be used, but would need to be painted with an exterior latex primer to avoid being so vapor-open that mold levels of moisture accumulate on the subfloor of an air-conditioned house. Summertime outdoor dew points in your area linger in the high 60s and low 70s for weeks at a time, and a 20 perm gypsum board has the potential become an issue, albeit not a HIGH potential. Again, air sealing the bottom sheathing at the cantilever is more important than it's vapor retardency.

Thanks for your reply. The joists are nominal 2.5 x 7-1/4. The ground is kept dry do to the overhang. So having said that can I go ahead with insulating according to the plan pictured above? R30 Rockwool is all sold out here. So i was going to stack R15 Should i add xps or polyiso sealed to subfloor ?

Seal 5"sill plate
Air barrier xps or polyiso interior side
Seal xps or polyiso to subfloor?
R30 or layer r15 in joist bay
And can i use GP plywood?
Is this what system should be? Correct me where im wrong. You have any nice picture diagrams to illustrate?
Thanks,
Nick

 

[Dana]

We line the space with a transmission reducer/ vapor barrier (since you'd like to include that use). Then we make a big fluffy pillow the size of the space and try not to over stuff it. We'd like to keep our air "entrained" and not moving through our around our pillow.

I don't advocate the use of vapor barrier, but strongly advocate the use of AIR barriers to control both heat and moisture flows.

In general when an assembly can be made to work from a thermal & moisture perspective without vapor barriers it's generally more resilient, since it has the ability to dry if it ever accumulates moisture, making it more resilient, less prone to mold & rot. Since the beginning of widespread uses of true vapor barriers in the 1970s missapplication of vapor barriers has created as many or more moisture problems as they have solved.

In this case and this climate zone (IECC zone 4A) as well as this specific case (insulating a cantilevered floor) using materials with variable or only moderate vapor retardency for the air barriers would be best practice, and true vapor barriers would only make the assembly less resilient. As a heating dominated climate it's useful to have the air barriers on the interior at least somewhat lower permeance than the exterior cladding. But given the summertime outdoor dew points in that location using extremely vapor open exterior cladding has at least some potential for creating summertime moisture issues if the home is highly air conditioned.

In a home that is air conditioned to 68F the floor and subfloor can dwell below the outdoor dew point on Long Island for weeks at a time, and during those weeks the subfloor takes up the moisture from the entrained air in the insulation. If the exterior cladding is extremely vapor open that moisture is drawn in via diffusion from the outdoor air at a fairly high rate, and the moisture content of the subfloor can rise to 20% or higher, making it more mold prone. But with half-perm cladding (half inch CDX with latex paint) or even as vapor open as 3-5 perms (latex paint on fiberglass faced gypsum board) the rate of moisture entering the assembly via diffusion is low enough that high moisture levels in the subfloor are never reached, and when outdoor dew points drop, the moisture flow reverses, and any accumulated moisture in the subfloor gets released. Also if the HVAC duct that penetrates into the cantilevered joist bay is used for air conditioning entrained moisture in that joist bay will condense on to the duct when the system is running, drawing even more outdoor moisture in when outdoor dew points are high. But with even modest vapor retardency on the air barriers it will never be enough to accumulate and drip (an all too common problem with uninsulated ducts in vented attics in that region.)

With good air barriers on all sides of the "pillow" there is no reason to not "...over stuff..." the pillow, only a gain in performance. Detailing solid wood sheathing (or where appropriate, edge sealed sheet foam) as the primary air barriers is more robust than relying thin sheet goods such as 4 mil polyethylene as the air barrier, which are less rugged, more easily punctured & torn.

 

[wwhitney]

We line the space with a transmission reducer/ vapor barrier (since you'd like to include that use). Then we make a big fluffy pillow the size of the space and try not to over stuff it. We'd like to keep our air "entrained" and not moving through our around our pillow.

A couple quick points: when you say "transmission reducer" that's non-standard terminology, are you referring to an air barrier or a vapor barrier?

And on the "try not to over stuff it," that's not a useful goal in this context. R-value per inch goes up with insulation density, so for a fixed volume and plenty of available insulation, stuff as much as you can in the volume to maximize total R-value.

When you have a fixed amount of insulation (per unit area) and plenty of available volume (e.g. a batt whose nominal depth is not greater than the stud bay depth), then you want to keep the insulation uncompressed to maximize R-value. Compressing the fixed amount of insulation increases R-value per inch, but not enough to make up for the decreased number of inches.

Cheers, Wayne

 

[wwhitney]

Vapor retardency and vapor barriers have to do to how rapidly water vapor (not air) will diffuse through a material or surface when there is difference in vapor pressure (think of it as the absolute, not relative humidity) from one side to the other.

Quick question: is vapor pressure the only determinant of vapor drive? E.g. in a steady-state closed system with an externally maintained temperature gradient, the vapor pressure will end up constant throughout?

Thanks,
Wayne

 

[Dana]

Thanks for your reply. The joists are nominal 2.5 x 7-1/4. The ground is kept dry do to the overhang. So having said that can I go ahead with insulating according to the plan pictured above? R30 Rockwool is all sold out here. So i was going to stack R15 Should i add xps or polyiso sealed to subfloor ?

Ah- so it's a milled 3 x8 a fairly non-standard lumber to find on the shelves these days!

Stacking pairs of R15s should work, even though the nominal stacked depth is only 7.0", a quarter inch shy of 7.25", they have enough loft to fill that space, and even if it developed a 1/8-1/4" gap at the top over time (not likely) it would not be a very big thermal bypass, a bypass that WOULD be sealed off if the caulking job between the subfloor & joists is done well.

When installing the top batt take care to tuck in the corners and edges to ensure it's in full contact with the subfloor, then tug it back gently to give it a hint more loft. When installing the bottom cladding, tugging gently on the bottom batt to make it slightly proud of the joist bottoms will ensure a compression fit, eliminating any thermal bypasses the bottom of the cavity.

There is no good reason to install foam board on either the subfloor, the band joist, or the bottom sheathing of the cantilever. If you like it's fine to use foam board as the interior side air barrier on top of the sill plate. Any foam will do fine, but polyisocyanurate is safer in a fire. Polyiso chars in place rather than melting sending a burning stream of molten plastic down the wall, the way polystyrene would.) Foil faced polyiso is fine, as is fiber faced roofing polyiso. Foil facers are true vapor barriers, but you're not really relying on drying toward the interior here, given the relative size of the bottom sheathing area, the main drying path.

Check your local craigslist- there are often people selling used roofing foam (from commercial building re-roofing and demoliton) at a steep discount from box store or distributor pricing. Since you are going to be cutting it all to size it dinged corners & facers or irregular thickness or dimensions (due to shrinkage over decades of use) don't really matter in your application.

Seal 5"sill plate
Air barrier xps or polyiso interior side
Seal xps or polyiso to subfloor?
R30 or layer r15 in joist bay
And can i use GP plywood?
Is this what system should be? Correct me where im wrong. You have any nice picture diagrams to illustrate?
Thanks,
Nick

To the question marks- no you don't need to cover the subfloor with XPS or polyiso- just seal all the seams & penetrations. (In air conditioned homes in the hot humid gulf coast states a half-inch of foam board on the subfloor may be useful, but not on L.I..) Any plywood (from 3/8" to 5/8" thick) will do for the bottom sheathing- be sure to seal it's edges & seams to the framing too (sometimes glueing the periemeter of each sheet to the framing as gets installed works, sometimes not, depending on whether the edges of the plywood are directly at the framing or whether it's hanging off by a bit).

And yes that system should work. It looks a lot like this:

^^^This diagram shows a cantilever off a framed wall rather than a foundation wall, but the top plate of the framing is essentially the same as the sill plate in your application. Since you don't have pre-existing blocking over the top of the sill plate, sealing rigid foam board as the interior side air barrier is used in lieu of the fiber insulation on the interior side of the blocking that is depicted here.

Another take:

^^ The green dots indicate seams with sealants.^^

Are you planning to insulate the interior side of the foundation as well? (R10 continuous insulation would meet IRC 2018 code in your location, easily/cheaply implemented with 2" reclaimed roofing polyiso, with some caveats on several installation details.)

 

[John Gayewski]

A couple quick points: when you say "transmission reducer" that's non-standard terminology, are you referring to an air barrier or a vapor barrier?

And on the "try not to over stuff it," that's not a useful goal in this context. R-value per inch goes up with insulation density, so for a fixed volume and plenty of available insulation, stuff as much as you can in the volume to maximize total R-value.

When you have a fixed amount of insulation (per unit area) and plenty of available volume (e.g. a batt whose nominal depth is not greater than the stud bay depth), then you want to keep the insulation uncompressed to maximize R-value. Compressing the fixed amount of insulation increases R-value per inch, but not enough to make up for the decreased number of inches.

Cheers, Wayne

Transmission of radiant heat. Transmission is the point of the polyboard barrier. The vapor barrier aspect is secondary but relevant. Transmission is a pretty standard term for radiant heating. It's actually the largest portion of figuring heating loads. Air infiltration is secondary. I don't know why I got dragged back into this one. We already went over this. I'm not in favor of compressing batt insulation. Which is an opinion I have come to with my eyes wide open and learned in the ways of heat exchange.

 

[wwhitney]

Transmission of radiant heat. Transmission is the point of the polyboard barrier.

If you mean foil faced polyisocyanurate, that can be a radiant barrier if it is oriented towards an air gap and under conditions where the foil won't get covered in dust. But radiation is only a relevant heat transfer mode where there's an air gap. Otherwise conduction and convection (apparently even within fiberglass batts) are the dominant heat transfer modes.

And again, on compressing batts, it's about the context: compress a batt and leave voids or air spaces, a significant loss. Stuff even more insulation into a fixed cavity with no voids or airspaces, you get a small benefit.

Cheers, Wayne

 

[John Gayewski]

If you mean foil faced polyisocyanurate, that can be a radiant barrier if it is oriented towards an air gap and under conditions where the foil won't get covered in dust. But radiation is only a relevant heat transfer mode where there's an air gap. Otherwise conduction and convection (apparently even within fiberglass batts) are the dominant heat transfer modes.

And again, on compressing batts, it's about the context: compress a batt and leave voids or air spaces, a significant loss. Stuff even more insulation into a fixed cavity with no voids or airspaces, you get a small benefit.

Cheers, Wayne

Ok. I'd stop the conduction of heat with pink board. Then there's nothing to radiate. Those are both the transmission factors I/we speak of when I/we say transmission. Transmission is meant to cover those. Convection ( related to infiltration) is stopped with bat's laid in perpendicular with one another unpacked, but snug. Then you can rely on air being as the insulator it is, as there's no way it can be heated.

Keep in mind we're talking relatively here. Of course some heat transfers on some small level. No way to be perfect unless we made a prefect vacuum.

The objective is to stop the floor from giving its heat away. Packing batt insulation tight doesn't really stop the radiation. Unless you can get it impossibly tight up against the bottom of the floor.

 

[Dana]

Ok. I'd stop the conduction of heat with pink board. Then there's nothing to radiate. Those are both the transmission factors I/we speak of when I/we say transmission. Transmission is meant to cover those. Convection ( related to infiltration) is stopped with bat's laid in perpendicular with one another unpacked, but snug. Then you can rely on air being as the insulator it is, as there's no way it can be heated.

With out an air gap (or vacuum, in the event you live in outer space ) there is no radiated transfer of heat. With the materials in contact the heat transfer if primarily conducted. In the presence of a gas and a gravitational field (like here on earth) there can be convected heat due to the higher density of cold air relative to hot air, assuming the gas is not constrained and allowed to move.

Keep in mind we're talking relatively here. Of course some heat transfers on some small level. No way to be perfect unless we made a prefect vacuum.

In a vacuum heat transfer by radiation is pretty effective- it is how the sun can warm the earth even from millions of miles away.

The objective is to stop the floor from giving its heat away. Packing batt insulation tight doesn't really stop the radiation. Unless you can get it impossibly tight up against the bottom of the floor.

Packing the batt insulation tight to the subfloor eliminates heat radiation, making the primary heat transfer mechanisms conduction & convection (of the entrained air within the batt). It doesn't have to be "...impossibly tight..." to the subfloor for that to happen- any reasonable contact will do.

Pack the fiber insulation tightly makes it more air retardent, reducing convective heat transfer within the batt, and eliminating large voids that might allow large convection currents to go around the batt. Thus a full, snug fit is key to getting the thermal performance out of fiber batts (at any density). As a general rule, the denser the fiber insulation, the higher the performance. Crummy R19s are fairly low density and low performance and loses some performance at higher temperature differences across the insulation layer, even with good air barriers on all sides. Compressing it to R13 density (or installing an R21-R23 high density batt) delivers a density high enough to overcome most of the convective losses, and it's performance then improves with temperature difference across the insulation layer.

 

[HydroNick]

Ok so air seal but vapor permeable got it.
Do I caulk this gap? The cantilever floor is made of that drywall with water resistant face on both sides. It covers the brick. It sits in their like a mudsill. So water didnt drain out the bottom of cantilever nor did water get inside. I guess it all vented into my basement through poured wall and/Or into cantilever. The vapor from that gap might actually do just that...emit vapor wetting the rock wool and thats ok since it will dry out bottom.. However Im concerned then about that gap emitting air..
This pic is interior facing out into cantilever. The joists are in contact with sill plate but also run over that gap.

For better reference you can see the gap at 22.5" on my tape. That gap is created by cantilever floor sandwiched between brick and joist. It butts up to the sill plate. When i removed cantilever floor from exterior i left that section in tact. I didnt want to dig it out from under joists or disturb.

 

[Dana]

Quick question: is vapor pressure the only determinant of vapor drive? E.g. in a steady-state closed system with an externally maintained temperature gradient, the vapor pressure will end up constant throughout?

Thanks,
Wayne

The temperature gradient has no affect on the vapor pressure. If the dew point of the air on the cold side is the same as on the warm side there is no vapor pressure drive.

The reasons vapor retarders are useful in cooler climates is that the exterior sheathing can dwell for weeks or even months at temperatures below the dew point of the indoor air. When that happens the moisture in the entrained air in the insulation collects in the wood sheathing in the form of adsorb (a thin molecular layer of water over the microscopic fiber surfaces in the wood), raising it's moisture content. The dew point of the remaining entrained air then tracks with the temperature of the exterior sheathing, so there is a vapor pressure across the interior sheathing (wallboard, or in this case, a wooden subfloor), drawing water vapor through the material. The rate of that moisture transfer per unit area is a function of the differences in dew point between the conditioned space air & the entrained air in the fiber insulation, and the vapor retardency of the material.

The indoor RH of most homes on L.I. runs about 35-40% @ 70F in winter, which corresponds to a dew point of 41F- 45F. So whenever the exterior sheathing is 40F or cooler (which is about half the total hours of the coldest 10 weeks of winter in that location) there is a vapor pressure induced at the interior drywall layer slowly pulling moisture from the indoors that ends up in the exterior sheathing. A couple layers of standard latex paint on paper faced gypsum drywall runs about 3-5 perms, which is pretty good. In that climate that reduces the accumulation rate low enough to protect the exterior sheathing from excessive wintertime moisture build up (so long as there aren't AIR leaks into the insulation layer from the indoors.) In Ontario or Quebec standard interior latex paint alone isn't quite good enough, but half-perm "vapour barrier latex" primer is. In the case of the bottom sheathing of a cantilevered floor, there is also some drying to the exterior since the bottom sheathing doesn't get substantial rain/snow/dew wetting, far less moisture drive coming from the exterior to deal with.

A 3/4" subfloor is pretty vapor retardent at ~0.5 perms when bone dry, but rises when the humidity content of the wood rises. The vapor permeanace of OSB rises in a similar way to plywood, but at humidity levels that support mold plywood is substantially higher, making it something of a "smart" vapor retarder:

 

[Dana]

Ok so air seal but vapor permeable got it.
Do I caulk this gap? The cantilever floor is made of that drywall with water resistant face on both sides. It covers the brick. It sits in their like a mudsill. So water didnt drain out the bottom of cantilever nor did water get inside. I guess it all vented into my basement through poured wall and/Or into cantilever. The vapor from that gap might actually do just that...emit vapor wetting the rock wool and thats ok since it will dry out bottom.. However Im concerned then about that gap emitting air..
This pic is interior facing out into cantilever. The joists are in contact with sill plate but also run over that gap.
View attachment 77369

For better reference you can see the gap at 22.5" on my tape. That gap is created by cantilever floor sandwiched between brick and joist. It butts up to the sill plate. When i removed cantilever floor from exterior i left that section in tact. I didnt want to dig it out from under joists or disturb.
View attachment 77371

Large gaps on wood sheet goods like plywood or OSB can usually be sealed with mesh tape + duct mastic, not caulk. Asphalted gypsum board and asphalted fiberboard might need to use an asphalted roofing patch + mesh tape rather than duct mastic. (I don't have much experience with detailing those types products as air barriers.)

 

[HydroNick]

I found a post by someone with the same issue hes also in 4A. I guess this changes my above insulation plan?
https://terrylove.com/forums/index.php?threads/bay-window-rodents-and-mold.77107/