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

Community and Q&A

Tongue and Groove Ceiling w/ 6mil VB – Enough?

collinp2 | Posted in General Questions on

I am building a bonus room over my garage… above the room is attic space with no current insulation. I have already installed a 6 mil vapor barrier to the joists and then put up the tongue and groove beadboard ceiling. I plan on blowing in R49 fiberglass insulation in the attic. I’m in Nashville, TN.

Will this be adequate to prevent mold and moisture issues in the attic? What about efficiency?

I did not want to put drywall between the joists and the ceiling panels since the overall ceiling is rather low and every 3/4″ matters significantly.

Please let me know of any concerns… I would hate to have to tear the ceiling down after spending so much time and money on it, but I don’t want moisture or mold problems either.

GBA Prime

Join the leading community of building science experts

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


  1. GBA Editor
    Martin Holladay | | #1

    In your climate, you definitely don't want a polyethylene vapor barrier, especially if the room will be air conditioned during the summer. Warm, humid air is likely to reach the back of the poly, and the moisture will condense, leading to mold or rot.

    Moreover, polyethylene is not a durable air barrier. If you want a ceiling finished with tongue-and-groove boards, you have to install gypsum drywall first, followed by the board ceiling.

    I hate to give you bad news, but the ceiling is wrong. I suppose you could remove most of the poly from the attic side of the ceiling -- assuming you have good access -- and then insulate the top side of the boards with spray polyurethane foam for an air barrier. But there are problems with that solution as well -- the spray foam would tend to ooze through the boards.

  2. collinp2 | | #2

    Wow. That is disappointing. I can't sacrifice any ceiling height and simply can't afford spray foam... I looked into that, very expensive in Nashville. I can't believe I've wasted so much money and effort and its a ticking mold time-bomb.

  3. GBA Editor
    Martin Holladay | | #3

    I know that at this point, my advice may not be helpful. But it's too bad you didn't research the topic on GBA before you started the work.

    GBA has long advised against the use of polyethylene in your climate zone. For example, see Vapor Retarders and Vapor Barriers. The article notes, "Unless you’re building in Canada, Alaska, or somewhere close to the Canadian border, you don’t want interior polyethylene or vinyl wallpaper — especially in an air-conditioned house."

    GBA has long advised people installing a tongue-and-groove board ceiling to first install an air barrier of gypsum drywall. For example, see How to Build an Insulated Cathedral Ceiling. The article includes this advice: "The biggest air-barrier blunder is to install tongue-and-groove boards as your finish ceiling without first installing taped gypsum drywall. A board ceiling is notoriously leaky, and this type of ceiling is often associated with roof sheathing rot. (Of course, if you have installed closed-cell spray polyurethane foam insulation, this advice doesn't apply, since you already have a tight air barrier.)"

  4. collinp2 | | #4

    Is there any possibility of ripping out the plastic and then sealing the grooves from above with Great Stuff or caulk? What about putting gypsum on the top side of the T&G boards between the joists and sealing that? I have good attic access, though tight around some of the edges.

  5. collinp2 | | #5

    Also... would it be the same for my 2x4 walls? Is the r13 Kraft backed insulation wrong as well? Do I have to have drywall in under the T&G boards there too?

  6. user-1140531 | | #6

    Why would exterior vapor condense on the top side of a polyethylene vapor barrier and not condense on the top side of 1/2” sheetrock?

  7. wjrobinson | | #7

    Collin you could blow cellulose in and see what happens. You may be fine if moisture is not above normal. If all goes bad pull boards down and redo.

    Dense pack cellulose may be able to deal with the wrong build.

    (this is advice only if you can monitor cellulose moisture and you aren't willing to redo ceiling at this time)

    As to plastic, it is not very vapor open. Kraft on insulation is, The drywall idea is to stop air movement but still let vapor through.

  8. GBA Editor
    Martin Holladay | | #8

    Q. "Is there any possibility of ripping out the plastic and then sealing the grooves from above with Great Stuff or caulk?"

    A. That sounds unlikely to work, because of seasonal movement in the boards due to temperature and humidity changes.

    Q. "What about putting gypsum on the top side of the T&G boards between the joists and sealing that?"

    A. That suggestion makes more sense, and could work, although it would be very fussy and time-consuming. Be sure to tape the seams between adjacent pieces of drywall if you try this technique.

    Q. "Would it be the same for my 2x4 walls? Is the R-13 Kraft backed insulation wrong as well? Do I have to have drywall in under the T&G boards there too?"

    A. The problem is not the fiberglass insulation (although that is the worst performing insulation available). The problem is the lack of an air barrier. You definitely need drywall under the T&G boards on your wall because the boards leak tremendous amounts of air. If you have polyethylene on your walls, that must be removed from the back side of the wall. Don't forget to put an air barrier on the back side of the walls as well as the front side; the fiberglass batts need to be enclosed in a 6-sided air barrier to perform well.

    Q. "Why would exterior vapor condense on the top side of a polyethylene vapor barrier and not condense on the top side of 1/2-inch sheetrock?"

    A. Because drywall is vapor-permeable -- the vapor goes right through it -- while the polyethylene is vapor-impermeable (it stops the vapor and it will be cold in the summer). Try this experiment: on a cold night during the winter, put a 6-pack of aluminum cans of beer and an empty cardboard box on your porch. In the morning, when it is 25 degrees, bring both items into the kitchen. You'll see condensation on the aluminum beer cans, because the cans are cold and the aluminum is a vapor-impermeable material (like the polyethylene). However, you won't see any condensation on the cardboard box, even though it is just as cold as the cans of beer.

  9. toymaker | | #9

    My city inspector in Madison Wisconsin insisted on a poly vapor barrier in my new (2011) home. We are far from the Canadian border so I worried after reading GBA. Much of the poly is exposed in the walkout basement until the day I want to pay additional property taxes and convert the space to "living/taxable" area. So, during our recent very hot Summer with the AC running days on end I checked carefully and frequently for any condensation on the outward face of the clear poly in this the coolest level of the home. There was none. I expected some from the drying studs at least. Made me feel better. Beware - YMMV.

    Cheers, john the toymaker

  10. collinp2 | | #10

    My carpenter told me he has done the same thing and seen it done in our area many times without issue... I may still put drywall in-between the joists as mentioned above, but in his personal experience in our climate over the past 30+ years, my method hasn't caused any issues he has ever seen.

  11. user-1140531 | | #11


    Rather than taking the drastic action of tearing out the poly and the wood ceiling; and adding drywall, and then re-installing the ceiling, why not just leave it and see whether there is a condensation problem before taking such drastic action for a remedy?

    The key point is that there is some amount of R-value in that T&G wood. I think there a very good chance that the R-value of the wood will keep the top side of the poly above the dew point temperature. If so, the dew point temperature will be somewhere in the wood on the conditioned side of the poly.

    Therefore, poly will stop the inward vapor drive before it reaches the dew point temperature of the conditioned space, so there will be no condensation.

  12. collinp2 | | #12

    Indeed, Ron. I think it would be silly to do so much work when i can monitor it and address later if it becomes an issue... which may never happen.

  13. user-1061844 | | #13

    A method that is easier then mounting sheetrock, which is heavy and its seams are prone to cracking over time (airleaks!)- would be to staple INTELLO PLUS membrane on your ceiling, tape the seams with TESCON Vana - install the boards, but leave 1,2 board off to enable the cellulose installation behind it this assembly (the board help support the cellulose) - BLDGTYP used this method in their design/build project very successfully
    see last photo in their blogpost (and attached photo as well -credit BLDGTYP). They hit 0.4ACH50 in their blowerdoor test.

    The vapor variability of the INTELLO will work very well in your climate as it prevents condensation in winter, while becoming vapor open in summer when you AC the house.

  14. user-1140531 | | #14


    I have a few more thoughts about this condensation issue.

    I understand your point about sheetrock being permeable while poly is impermeable.

    However, while it is true that vapor can pass through sheetrock because of the material’s permeability, I would think that inward-moving vapor cannot possibly pass though sheetrock without condensing if it encounters the dew point temperature at any point within the sheetrock. So, if the temperature of the interior is below the dew point of the outside air such that condensation would occur on the poly, then it will certainly occur within the sheetrock if there is no poly.

    I understand your analogy of the cold beer. However, I would say that the reason that no condensation will be seen on the cardboard box is that the moisture will have been absorbed by the cardboard and become invisible. Nevertheless, the cardboard will have become wetter. In a practical sense, the beer in the box experiment will run out of action before the point is made. The box cardboard will not remain at the temperature of the beer for very long. If you could keep the beer cold, and keep the air inside of the box at the same temperature as the cold beer, the cardboard would become soaked unless you could somehow remove the moisture from the inside of the box faster than it condenses on the inside of the box walls.

    In the house example we are discussing: I conclude that if inward vapor drive is occurring because of high outdoor temperature and humidity, and if the vapor encounters the dew point temperature where it meets the conditioned space, it will condense there. If that point is where it encounters polyethylene, the vapor will condense on the impermeable sheet. If that point is sheetrock, the vapor will condense on the surface of, or inside of the sheetrock, depending on where the dew point is located. Once it condenses, it will wick liquid moisture throughout the sheetrock layer. From there, it will dry to the interior conditioned space. So it seems to me that the vapor will enter the sheetrock, condense, and then leave by re-evaporating.

    Perhaps the sheetrock will be able buffer the moisture by spreading it out within the gypsum layer, and evaporate it to the interior faster than it can accumulate to the point of disintegrating the gypsum layer. Paint on the interior of the sheetrock would tend to slow that inward drying, however.

  15. jklingel | | #15

    "Perhaps the sheetrock will be able buffer the moisture by spreading it out within the gypsum layer, and evaporate it to the interior faster than it can accumulate to the point of disintegrating the gypsum layer." •• Now you're talking.
    "Paint on the interior of the sheetrock would tend to slow that inward drying, however." •• Latex paint should be used. More vapor open than oil-based.
    •• You may want to read this. You are playing with fire (mold, actually) by using the poly. Maybe you'll be OK, but what if you aren't?

  16. GBA Editor
    Martin Holladay | | #16

    You wrote, "I would think that inward-moving vapor cannot possibly pass though sheetrock without condensing if it encounters the dew point temperature at any point within the sheetrock."

    I'm sorry, but your impression is flat-out wrong. As always for this type of question, we need to turn to William Rose's book, Water in Buildings. Rose wrote, “The language ‘reaching dew point’ seems to indicate that one could plot a temperature profile through a wall, find the point where that profile intersects a horizontal line indicating indoor dew point temperature, and expect burgeoning water at that location. This impression is decidedly incorrect. If water accumulates, it does so on the surfaces of materials, not within the thickness of materials.”

    Furthermore, Rose wrote, “Capillary materials do not exhibit condensation at the dew point.”

    If you have a hygrosopic material like plywood, cardboard, or drywall, you can't get condensation in the middle of the material. The material can gain moisture, of course, and it can get soggy. (That only happens if it can't dry out in at least one direction.) And if it is waterlogged, and the temperature drops low enough, it can freeze solid. But you can't get condensation in the middle of the material.

  17. user-1140531 | | #17


    If condensation can only occur at the dew point temperature if it is at the surface of materials, why wouldn’t condensation occur at the outer surface of sheetrock if it were at the dew point temperature?

  18. GBA Editor
    Martin Holladay | | #18

    The short answer is "physics."

    The non-technical description is easy: moisture beads up on an aluminum beer can because it has nowhere to go. On a piece of cold cardboard or cold drywall, however, the same beads of moisture are taken up by the adjacent hygroscopic material (the cardboard or the drywall). The result is that the cardboard or drywall become more damp; this is called adsorption. No condensation occurred.

  19. user-1140531 | | #19


    You seem to be reaffirming my original assumption which you previously said was “flat-out wrong.”

    How can the cardboard or sheetrock become damp if no condensation has occurred?

    Does William Rose thoroughly document this claim with science? I am not sure I follow his verbal explanation as conveyed by the language alone. This is the first I have ever heard that vapor can pass into a temperature below its dew point without condensing—IF that is what Mr. Rose is saying.

    He says that condensation does not cause to water to “accumulate” or “burgeon” at the location of its dew point, if that point is within hygroscopic materials. He says that such hygroscopic materials do not “exhibit” condensation at the dew point.

    When he says this, does he mean that no condensation occurs at the dew point,—OR—does he mean condensation does occur, but it does not “exhibit” its presence by burgeoning, or accumulating?

    I can drop water onto a sponge and not see the exhibition of any visible puddled liquid as is the case suggested by the “burgeoning” of “accumulated” water.

    Can you clarify whether or not condensation occurs at its dew point within hygroscopic materials?

  20. GBA Editor
    Martin Holladay | | #20

    Q. "Can you clarify whether or not condensation occurs at its dew point within hygroscopic materials?"

    A. I am happy to clarify. It does not.

  21. user-1140531 | | #21


    I asked a question and you answered thus:

    Q. "Can you clarify whether or not condensation occurs at its dew point within hygroscopic materials?"

    A. I am happy to clarify. It does not.

    Answered by Martin Holladay, GBA Advisor

    And yet you say that the hygroscopic material will become damp from vapor passing through it even though the vapor is not condensing. Water vapor is a gas. It does not make things damp or wet. So if there is no condensation in hygroscopic material, where does the dampness come from?

    Mr. Rose also said this: “If you have a hygroscopic material like plywood, cardboard, or drywall, you can't get condensation in the middle of the material. The material can gain moisture, of course, and it can get soggy.”

    Again I ask: Where does the moisture come from? How can a gas make something “soggy”?

  22. Expert Member
    Dana Dorsett | | #22

    Mayhaps I can clarify it (or muddy the waters further):

    In an air-permeable insulation layer, as soon as the hard cool surface (be it wallboard or poly) is cooled to the dew point of the entrained, air it begins to condense on that surface. As the temp of the condensing surface drops, so does the dew point of the air in the insulation cavity. If sufficient liquid water accumulates some of that water is redistributed by the wicking of the fiber insulation, it may even be copious enough to form a soaking puddle at the bottom.

    When the center of wood or gypsum is at the dew point of the entrained air it will adsorb water without visible condensation, which is what typically happens with wood sheathing in the dead of winter when subjected to interior moisture drives. It never appears as liquid water on the surface, but it can get quite waterlogged nonetheless. Water vapor is a gas, sure, and wood is only somewhat permeable to that gas, but without a sufficient vapor pressure difference to move it out, it'll just hang around in the wood until conditions change (and those changes in conditions can be weeks or months later.)

    Adsorption & release of moisture in hygroscopic materials IS related to the dew point of adjacent air bodies, but it isn't technically condensation, since it isn't beading up as liquid water with it's own independent surface tension- it's adhering to the fiber surfaces, but it's not exactly the same as liquid water. In the cardboard case, the cellulose fibers are hollow, and the adsorption ends up predominantly on inside rather than on the outer surfaces of the fibers, but I digress..

    In the case of the interior drywall/paneling, the same air conditioning that cools the wall to the dew point of the outdoor air also dries out the interior air creating a lower vapor pressure which keeps the moisture that finds it's way into the gypsum moving. But with a foil/vinyl wallpaper or highly retardent paint it can't, and will get soggy, continuing to take on moisture from the insulation layer.

    Solid wood is fairly vapor retardent- about the same permeance of closed cell polyurethane or extruded-expanded polystyrene foam. But t & g is also inherently vented, and moisture in the highly-permeable gypsum layer of the stackup would still find it's way into the conditioned space air. Summertime dew points in Nashville aren't so high that it needs a super-fast drying rate to keep it from developing rot conditions in the wood, but it helps considerably to have at least SOME ability to pass water vapor in to the drier interior.

    With 6 mil poly it's summertime moisture trap, and having a vent channel above the fiber insulation doesn't dry it out until the exterior air dries out. In winter the poly blocks interior moisture from accumulating in the roof deck, but so does a ventilation channel.

  23. collinp2 | | #23

    So is it better to have T&G with no plastic and then cellulose over that? I can easily rip out the plastic from above but am not going to tear down the boards until absolutely necessary.

    I have soffitt vents and a ridge vent, so my attic should be vented properly. I'm either leaving it as-is and blowing in cellulose or I can rip the plastic and then blow the cellulose.

  24. GBA Editor
    Martin Holladay | | #24

    Building scientists recognize that water in buildings exists in at least four states: liquid, vapor, ice, and adsorbed moisture. The moisture that is driven inwards through a wall assembly in summer may hit a layer of cold drywall; the moisture is adsorbed by the drywall. In normal conditions, the drywall never actually gets soggy, because it dries to the interior.

    Drywall and wood usually contain some adsorbed moisture. We can measure how much moisture by weighing a sample of a known size, and then baking the sample in an oven to dry out the sample. The difference in weight between the sample before and after it is baked in an oven is the weight of the adsorbed water that was driven out of the sample.

    By the way, one of the quotes you ascribe to Bill Rose was never written by Bill Rose; it was written by me.

  25. GBA Editor
    Martin Holladay | | #25

    It's not liquid water. It's adsorbed water.

  26. user-1140531 | | #26


    I am not sure I follow. I can see vapor condensing on a material surface, and either forming a liquid layer on that surface if the material is impermeable, or being absorbed into the material if it has capillary action. In the latter case, no water liquid would remain on the surface. However, are not both of these outcomes examples of water adsorption on the surface?

    In either case, wouldn’t that water be liquid water?

  27. user-1140531 | | #27


    Thanks for your input.

    You said this:

    “Adsorption & release of moisture in hygroscopic materials IS related to the dew point of adjacent air bodies, but it isn't technically condensation, since it isn't beading up as liquid water with it's own independent surface tension- it's adhering to the fiber surfaces, but it's not exactly the same as liquid water.”

    I do not understand the distinction you are making to qualify the definition of condensation. Since when is condensation limited only to water that beads up with its own independent surface tension; or as William Rose puts it, water that burgeons and accumulates? Steam discharging into the air becomes visible due to its condensation, yet does not exhibit the nature of beaded-up water with its own independent surface tension.

    I understand your point that adsorption and release are not condensation. But are we not talking about the adsorption and release of liquid water? I assume we are because you and Martin have referred to hygroscopic materials becoming “waterlogged” or “soggy.”

    So, assuming we are talking about the adsorption and release of liquid water, where did the water come from? If the liquid water did not manifest by condensation, how did it come about?

  28. GBA Editor
    Martin Holladay | | #28

    Imagine two identical pieces of lumber. Let's say they are two lengths of 2x4 lumber. You put both of them in an oven and bake them dry. You take them out of the oven and weigh them. The two pieces of wood weigh exactly the same -- let's say 2 lbs.

    You take one of these pieces of lumber and you put it in your grandmother's living room for one year. (She lives in Phoenix, Arizona.)

    You take the other piece of lumber and you put it in your brother's living room for one year. (He lives in Juneau, Alaska).

    Now you weigh the two pieces of lumber again. Wow! The one in Juneau weighs more than the one in Phoenix. How come? It has more adsorbed moisture.

    Did any condensation happen? I say no.

  29. user-1140531 | | #29

    Yes I do understand your example, and it gets right to one part that raises the most vexing questions in my mind. I agree that wood absorbs and releases moisture without any condensation happening. But then which one of the three states of water is the “moisture” in the wood? Is it a liquid or is it a gas? I assume it is not a solid.

    Do molecules of water vapor (the gas) adsorb on the surface of wood, coalescing as a film of liquid water, and then absorb as liquid water into the wood by capillary action?

    Are there two types of condensation as follows?

    1) One type where water vapor (the gas) condenses when its temperature falls to its dew point.

    2) Another type independent of dew point where molecules of water vapor (the gas) adsorb on a surface and transform into water (the liquid).

    The dictionary defines “ADSORPTION”: “to gather (a gas, liquid, or dissolved substance) on a surface in a condensed layer.”

    Notice that the definition refers to the layer as “condensed.”

    So the overarching question is this: Can water vapor (the gas) condense into water (the liquid) without its temperature dropping below its dew point?


    Setting aside the above to focus on the following:

    If water vapor (the gas) is chilled below its dew point, it will condense into water (the liquid). The condensed liquid can collect on a surface. Is this coalescing of liquid water on a surface because of dew point condensation called adsorption?

  30. GBA Editor
    Martin Holladay | | #30

    Q. "But then which one of the three states of water is the moisture in the wood?"

    A. As I wrote the last time you asked the question, the water is in a fourth state. It is not a liquid; it is not a vapor; it is not ice. It is adsorbed. Building scientists recognize that water in buildings exists in at least four states.

    I highly recommend that you purchase William Rose's textbook, Water in Buildings. You seem curious about these questions. It's fun to read up and study on these topics. I think you will find that Rose does an excellent job of covering these issues.

  31. user-1140531 | | #31


    To me, saying that a fourth state of water is "adsorbed water" seems like saying that a fourth state of water is falling water, or moving water, or water making waves, or stagnate water. What is the nature of adsorbed water that makes it something other than a gas, liquid, or solid?

  32. GBA Editor
    Martin Holladay | | #32

    Quoting from Water in Buildings:

    "Building materials -- wood, brick, stone, concrete -- all ... have measurable water contents. How should we describe this phase of this water? It's not vapor, certainly. It seems odd to consider it solid like ice or snow. And it is not runny like liquid water. It is best to think of the water contained in hygroscopic and porous materials as being an another phase, that of bound water. ...

    "Water of soprtion in porous and hygroscopic materials should be considered another phase, different from the three phases of pure material. So the conversion of water from vapor to sorbed water is not condensation. Sorption is a better term, covering adsorption and absorption. ...

    "Moisture accumulation occurs in sorptive materials in building envelopes ... any time the surrounding relative humidity rises above the relative humidity to which the materials have adjusted."

    For more information along these lines, buy the book.

  33. user-1140531 | | #33


    I can understand the concept of bound water. If Mr. Rose wants to call that a fourth state of water, so be it. But there are really only three fundamental states of water. Bound water is a type of water storage. Mr. Rose says the bound water is not runny. But does water have to be runny in order to be in liquid state? I assume that the bound water is liquid, but not runny.

    But the question is how this applies to the case of the Tennessee house we are discussing in which inside temperature is below the outdoor dew point temperature.

    You have said that water vapor will condense as liquid water on the exterior side of the poly vapor barrier.

    You say that if the poly vapor barrier is replaced with a layer of sheetrock, water vapor will pass through the sheetrock and not cause any problem due to wetting. You say that no wetting problem will occur because no condensation will occur— even though the vapor is passing a point where the temperature falls below its dew point.

    But then both you and Dana have referred to the potential for the water vapor passing though the sheetrock as having the potential to make the sheetrock soggy or waterlogged. Clearly, this can only happen if the water is in its liquid state.

    So, from your explanation, I must conclude that the water vapor passing through the sheetrock does turn into water (the liquid), but not due to dew point condensation. But then the water re-evaporates from the conditioned side of the sheetrock fast enough to prevent a buildup of wetting in the sheetrock sufficient to cause damage.

  34. GBA Editor
    Martin Holladay | | #34

    Either you aren't reading my answers, or you just like to contradict.

    If the drywall is covered on the inside with vinyl wallpaper, it could certainly get soggy. But as I already wrote in post #25, "In normal conditions, the drywall never actually gets soggy, because it dries to the interior."

  35. user-1140531 | | #35


    I have read every word of your explanations more than once, and my objective is not to contradict. If you read what I said, you will see that I never said the sheetrock would get soggy. I explained it can dry faster than the water accumulates just as you have explained it. In fact, if you read further up, you will see that I actually detailed the interior drying as the desirable and probable outcome before you even brought it up.

    My only point in mentioning your reference to the potential for sogginess (#33) was to clarify and establish that we are talking about water in its liquid state.

    Specifically, we agree that water vapor (the gas) passes into the sheetrock, and changes to a liquid as it does so. From there, it re-evaporates to the interior space.

  36. Expert Member
    Dana Dorsett | | #36

    With water in a liquid state the vast majority of the molecules are in contact only with water, and move freely. In an adsorbed state the molecules are primarily in contact with the other materials to which it is adsorbed, becoming a film at most a few molecules thick primarily in contact with on the other material, without sufficient surface tension to separate itself into something that behaves a s a liquid.

    Adsorbed water doesn't flow like liquid, doesn't suspend inorganic ions in solution, etc., it can move in response to vapor pressures, but it is definitely not a liquid, nor is it a gas. It more closely resembles water molecules that have been dissolved into a solid. Without the presence of the absorbing solid it would self-attract in to a liquid form with a surface tension with an ability to flow and dissolve other materials, rather it's the material that is dissolved into the solid. It doesn't "re-evaporate" form a liquid form, rather it vaporizes from adsorbed state as the vapor pressure differences allow.

    Ever considered looking it up?

    Sure at some higher level if the path to the low vapor pressure side is blocked (say, by vinyl wall paper) the adsorption can eventually saturate the solid, resulting in actual liquids filling any voids- a truly sopping wet soggy condition, but that is WAY beyond the moisture level at which mold & rot can set in, which is what we are concerned about in building assemblies.

    The hollow-tube form factor of cellulosic fibers (be it in the structural wood ,or in paper/cotton insulation) give it quite a bit of surface area onto, and can buffer quite a bit of adsorbed moisture without it becoming a liquid. In the insulation case the adsorbed moisture has essentially no effect on the insulating characteristics. Only when it is saturated and forming liquid does R-value suffer. Mineral and glass fibers have dramatically less adsorption surface area onto which to distribute the water, which makes them less protective of the structural wood, which is another reason why some of us prefer cellulosic insulation as cavity fill when appropriate- it adds resiliance to the assembly.

    Gypsum board can buffer quite a bit of adsorbed water with out damage too, but there again it's possible to reach a saturation point at which liquids begin to form. And again, that's well beyond the moisture level at which mold could start growing on the facers.

  37. user-741168 | | #37

    Always risky to enter a conversation like this late in the game. Anyhow...

    I figure the three phases refer to pure materials--elements or compounds. There are many sorts of impure materials--solutions, mixtures, emulsions, wet stuff, sponges, soil. The vapor pressure (saturation vapor pressure) which gives us dewpoint, was calculated over pure liquid water at a given temperature. Add salt to the water and the apparent dewpoint goes down. Add anything to the water and the water vapor pressure goes down. Add sawdust. Psychrometric dewpoints only pertain to liquid water, since the only bonds that matter are the hydrogen bonds holding water molecules to one another. Throw in hundreds of other bonding sites, all of different bonding energy, and it's a different ball game.

    I hope this helps. I try to avoid use of the term "condensation" except where liquid water is present. We still have the burden of describing how wet is too wet. I'll try to include a figure from my book.

  38. GBA Editor
    Martin Holladay | | #38

    Thanks, Bill.

  39. GBA Editor
    Martin Holladay | | #39

    I vote for option #2 -- which I realize is a lot of work. However, it's the best of the three options.

  40. user-1140531 | | #40


    Thanks for your response and diagram. The diagram actually presents a good framework for asking my question.

    Your diagram indicates that there are two routes for water to become bound water. One is the intake of water vapor by sorption, and other is the intake of liquid water by suction.

    Question: Say the mixed material is below the dew point temperature of the water vapor in the pure material. Will water vapor in the pure material condense on the surface of the mixed material, and then be sucked into the mixed material to become bound water?

  41. user-741168 | | #41


    We tried it. Two things happen. One, the weight of the block of cold wood goes up, even though there's nothing to be seen. Two, the surface warms up quickly. Gas molecules are much more energetic than liquid or bound molecules so they warm any surface where they stop being vapor molecules (see, I didn't call it condensation). Its the reverse of evaporation cooling a surface.

    So yes, you're right. the water gets sucked in, and gets distributed inward. But the warming effect is strong. A minute after pulling a block of wood out of the fridge it' looks close to room temperature on an IR image, from both conduction and latent warming.

  42. user-1140531 | | #42


    What you describe with the weight of the block of wood going up indicates that is absorbing water. I assume the block was below the dew point temperature of the water vapor since that was a condition of my previous question. You indicate that there are two mechanisms for the material to take in water to become bound water. Which mechanism do you conclude is responsible for the wood taking on water and becoming heavier?

    From what Martin quoted of your explanation earlier, it was not clear to me if you were saying that condensation can or cannot occur at the surface of the wood or sheetrock, assuming that the temperature of that surface was below the dew point temperature of the water vapor. I thought you may be saying that such surface condensation can occur, but it will be immediately sucked into the material, and therefore will not be visible or apparent as liquid on the surface. I am still not clear on that detail.

    I wonder what would happen if you ran this experiment:

    Use two identical blocks of wood exposed to the same saturation of water vapor at the same temperature. Then chill one block of wood to maintain its temperature below the dew point of the water vapor. Maintain the temperature of the other block of wood to match the temperature of the water vapor. Then compare the amount of weight change and its rate between the two blocks.

  43. collinp2 | | #43

    Thanks Ron and Martin... sure wish I would have found this site before I started. I did quite a bit of research but obviously not enough.

  44. user-741168 | | #44


    The moisture content of wood goes up and down all the time. The driver is Relative Humidity. The FPL Wood Handbook, Chapter 4, gives the wood sorption isotherm. It's the curve that relates wood moisture content to RH. If a certain wood moisture content is x%, which corresponds to 50% RH, then whenever the RH goes over 50% it gets wetter and when the RH is under 50% it gets drier. Since dewpoint is 100% RH it will certainly wet the wood, but so will 80% and 60%RH air, in our example. Here's why I don't like using "condensation" for sorption processes: water may go from 60%RH air into the wood--do we imagine there is liquid water or condensation anywhere in that process? I don't think so.

    The two mechanisms I mentioned were sorption (sucking water out of the air under appropriate conditions) and latent heating--heating the surface because water is being sorbed there.

    You pose an experiment of chilling one block and keeping the other one at (room?) temperature. The colder one will get wetter. At freezing temperatures and below, the mechanism can involve forming frost on the surface (not bound water), which, when it melts, gets sucked into the wood. Or not, if the surface is fully saturated.

    All this is the flip side of lumber drying processes.

    Here's another reason I don't like imagining liquid phases, even if instantaneous: mold is not a fish. The water activity that is optimal for mold is not 100% RH, but more like 80%. If we want to "think like mold" we need to concern ourselves with wetnesses that are less than condensation. That's where the real action is.

  45. user-741168 | | #45

    Perhaps I sound picky regarding use of terms, like "condensation". The one that really bugs me is "trapped!". I like pointing out to people that wood, gypsum, concrete all have measurable moisture contents and that that water is TRAPPED in these materials. When someone warns me I'm gonna trap moisture in there I say horray! the laws of physics are still at work.

  46. user-1140531 | | #46


    I understand your point about mold being more associated with RH at lower levels than 100%. And I understand your point that vapor is taken into wood by sorption without action of dew point or dew point condensation on the surface. But I am trying to get an answer to another question.

    This is my question: Say you have a sheetrock ceiling in which the layer of sheetrock is say 65 degrees F on the top side. The sheetrock is being held at that temperature by air conditioning in the space below. Say outdoor air above the sheetrock has a dew point of 80 degrees F. Will water vapor in the outdoor air contacting the 65 degree surface of the sheetrock condense on that surface?

    Dana says it will, and Martin says it won’t.

  47. wjrobinson | | #47

    Ron, Bill is saying the surface will not be wet like a hose running on it. And not like a can of soda covered in water from being cold in hot humid air. Try it yourself. Put a piece of drywall in and out of your fridge. I may have to play with my fridge and some samples.

  48. user-1140531 | | #48


    I understand your point. If water condenses on glass it will bead up. If water condenses on a sponge, you won't see it. All I want to know is: Does it condense?

  49. collinp2 | | #49

    So all this liquid vs. gas, etc is interesting... but it looks like all that aside, I have 3 options...

    1. Remove all boards and start over... took me 2 weeks and $700 worth of wood to install so that isn't a likely option. Can't afford it. We have taken out loans for materials so stretched pretty thin.

    2. Remove plastic from above and painstakingly add drywall in between the joists and seal.

    3. Do nothing at all and hope for the best.

    I still don't know what I will do.

  50. GBA Editor
    Martin Holladay | | #50

    Q. "All I want to know is: Does it condense?"

    A. To quote Bill Rose, "Here's why I don't like using 'condensation' for sorption processes: water may go from 60% RH air into the wood -- do we imagine there is liquid water or condensation anywhere in that process? I don't think so."

  51. user-741168 | | #51


    In 1937, when Teesdale and the Forest Products Laboratory were trying to figure out if there's condensation on sheathing when indoor air dewpoint is above sheathing temperature, they discovered something that we don't hear much about. They discovered that the vapor pressure in the cavity (which replenishes water slowly from diffusion and leakage), at any instant, depends mostly on what the moisture content of the sheathing is--not on what the permeances of the materials are. In the long run, permeances are important, but in the short run, we can always tell what the cavity vapor pressure is if we know the sheathing temperature and its moisture content. The wood (or gypsum) tells the local air what humidity to be (in the short run), not vice versa. Over weeks or months we can see the influence of permeances and air flows.

    Long story short, there is always a gradient in vapor pressure from a "thirsty" piece of gypsum or wood and the air that surrounds it. If the rate of replenishment is slow, as in a cavity, or an insulated assembly with low air change, then the gradient can occur over a space of inches. If the humid air is blowing directly over the thristy material, then the gradient occurs in micro-space.

    When stuff gets wet, there will be some water molecules very tightly bound to charged parts of sorptive materials. Then other water molecules will bind to the bound water molecules, and it makes a film many molecules thick, with each molecule less tightly bound than its inboard neighbor. What do we mean “liquid”? At what water film thickness does water stop being a bound film and start being liquid? It’s a matter of assigning a definition to a particular bond strength. Same with trying to draw a line between sorption and condensation.

    In the case you describe, looking microscopically, we have 1) water molecules attached almost permanently to the gypsum crystals at 65 degF, 2) water molecules less and less tightly held in a film around the attached molecules, 3) an equilibrium interface where molecules come and go from the surface, randomly, or reacting to minor changes in temperature or concentration, 4) a gradient of water molecule concentration in the air from the surface out to where the dewpoint is measured. The surface where the exchange is occurring may or may not be “soggy”.

    You seem to be trying to find a home for the word “condensation” in all this, and I’m trying to say that you can get a clearer understanding by letting go of the word “condensation” and taking it for what it is.

    Hope that helps.

  52. user-1140531 | | #52


    I understand the point of letting go of the word or concept of condensation in order to understand sorption. But my question is entirely, and only about condensation. For my question, I am setting aside the process of sorption and the thirstiness of material. I am not asking whether condensation is part of sorption. The only connection between sorption and my question is the material being permeable, thus making sorption possible. But I am not asking about sorption.

    We all agree that condensation will form water on the surface of metal if the metal is below the dew point temperature. I am asking if condensation will form water on the surface of permeable material if it is below the dew point temperature. I realize that water formed on the surface of permeable material will not accumulate and puddle on the surface. I am only asking if it can originate on the surface by the process of condensation, if the surface is below the dew point temperature.

  53. wjrobinson | | #53

    Ron, Bill has given great clarity to the subject. Bill prefers not to use the term condensation in your example. You really should study all that you can next and maybe come back to this?

    What we all really need is to not build moldy rotting homes.

  54. user-741168 | | #54

    Hi Ron,

    I guess I wasn't too clear in my answer. Sorry about that. When I said there is a gradient of vapor pressure, that means that there is a gradient of dewpoint. The dewpoint you measure in the room is not the "dewpoint" that the exchange surface sees. If a surface is sorbing, it is gobbling up water molecules from the air, essentially lowering the local water vapor concentration. And as this happens, the exchange surface is heated by the water molecules going from their most excited state--vapor--to a lower-energy state. And if the material is thermally resistive like wood, then that surface gets warmed up quite quickly.

    You're wondering about a cold sorptive surface meeting high dewpoint air. I just reviewed a video of me taking a 2x4 block out of the refrigerator, next to a can of beer and a brick, when it was humid outdoors. The wood warmed up in a matter of seconds, largely from latent heat of sorption, far faster than either the brick or the can of beer. If we had molecular-sized dewpoint sensors, we'd see that the air dewpoint at the sorbing surface is lower than the dewpoint we measured further away in the room.

    When a material is sorbing, there is a water surface within the material that has a net gain of water molecules in its exchange with the air. Perhaps you wish to call that condensation. You may. But you should note that, if that's true, condensation is occurring in all sorptive materials exactly half the time. Because sorptive materials are getting wetter half the time and getting drier half the time.

    I prefer to reserve "condensation" for non-porous materials, where condensation leads to dewdrops, and where the only bonds in play are the well-defined hydrogen bonds in a drop of water. I would argue that the bond strengths in sorptive processes are higher than those of simple hydrogen bonds, and if we call that "condensation" then, theoretically, all sorption is condensation. That would not be a helpful outcome. That's like saying that whenever a water vapor molecule stops being vapor and forms a bond of any kind, then condensation occurs. Actually, you'd be in good company--physicists call "condensed matter" in the universe anything other than free atoms.

    I think we all agree on allowing a difference between sorption and condensation in building science. I hope so anyhow.

  55. user-1140531 | | #55

    Thanks Bill. I am in thinking mode right now. I have carefully reviewed your explanations, and will continue to do so as I think this through. I have some ideas about experiments that I will post tomorrow for your review.

  56. Expert Member
    Dana Dorsett | | #56

    A key difference between condensation and adsorption is that condensation responds to gravity, and can create a puddle at a remote location downhill from the condensing surface. Adsorbed water is bound to the material moves in response to vapor pressure differences, and capillary draw when the bound moisture is sufficient saturated in the material to develop a surface tension to the water films.

    In Collin's case you want to avoid the prospect of condensation slowly rolling down the poly to concentrate at the top plate of the studwall & eaves or soaking into the rafter edges (both are likely scenarios over a long hot muggy summer.)

    I'm going to have to go with Martin on this one- get RID of the poly, and install gypsum as the interior air-barrier between the wood and rafters, as painful a solution as that is. Interior poly in air-conditioned space is sometime a problem even in Ontario where it's required by code (particularly on brick-clad houses), and it's all but guaranteed to be a problem in Nashville where the summertime dew points are even higher.

    Hijacking Collin's thread with all of the theoretical definitions stuff isn't exactly what this forum was designed for. Maybe there needs to be heading for a "building science" forum for taking Ron's technical issues to.

  57. collinp2 | | #57

    I've settled on and have come to peace with the idea of drywall between the rafters... 24" spacing and 375sq ft, so it shouldn't be too terrible to accomplish.

  58. user-1140531 | | #58


  59. user-1140531 | | #59


    Regarding your comment:

    "Hijacking Collin's thread with all of the theoretical definitions stuff isn't exactly what this forum was designed for. Maybe there needs to be heading for a "building science" forum for taking Ron's technical issues to."


    I am sorry if this discussion amounts to hijacking the thread. My questions about the theory were not just for the sake of an academic exercise. They pertain to the key element of the GBA recommendation here to Collin wherein he is being warned of destructive consequences of a building strategy and told how to modify that strategy to avoid those consequences.

    I have seen other reference sources that offer a conflicting viewpoint on this matter. They could be wrong. So I am simply asking for an explanation of the science behind the recommendation. Otherwise, without a clear demonstration of the practical performance, I would not use that recommended strategy in a building project of my own.

    I would be leery of pushing vapor through a layer of sheetrock across the dew point threshold on a full time basis, and counting on just the right outcome where moisture passes, but does not accumulate at the interface.

    Earlier, I had told Collin what I would do if I were faced with the question he asked about in his original post regarding his ceiling insulation project. In reconsidering this, I have no recommendation on the matter pertaining to Collin’s original question, and do not wish to offer any advice.

    For my own curiosity and conclusions, I will build a test apparatus that will determine the outcome of using gypsum board as an air barrier across a dew point threshold.

  60. homedesign | | #60

    Thank you Collin for asking a good question...
    and thank you Ron for "Hijacking"

    I don't think the Q&A was designed for any particular purpose
    I think it's kinda like an "Open Source" Forum

    you never know what's gonna happen

    or who might answer your questions

  61. Expert Member
    Dana Dorsett | | #61

    "I would be leery of pushing vapor through a layer of sheetrock across the dew point threshold on a full time basis, and counting on just the right outcome where moisture passes, but does not accumulate at the interface."

    This is how it works in the vast majority of houses built, and has been for millenia: As long as the drying rate keeps up with the adsorption rate nothing bad happens. With the advent of space heating, air conditioning, and insulation the problems get exaggerated, and we find out what does/doesn't work. This has been studied to death over the past 30 years, and in general, inserting true vapor barriers like poly or foil into building assemblies causes as many or more problems than they solve, since the direction of the moisture drives change with season & temperature.

    Moisture transfer through building materials has also been very carefully modeled, and models rigorously verified in academic settings, both in laboratories and on full assemblies in the field. If you want to know the moisture performance across seasons for any layer in any assembly stackup and indoor humidity & temp condition/profile, download a copy of WUFI and learn how to use it. (Caution: garbage in ==> garbage out, this is not a dumbed down hacker tool, but it's WAY more informative than any single steady-state condition model.)

    Which is NOT what Collin should be doing, proving what's dead obvious to those who have been at this for awhile.

  62. Brenda514 | | #62

    Hello Martin,

    I am in the process of designing a house to be built in Pagosa Springs, Colorado (southwest corner of state), and we’re installing a T&G vaulted ceiling in the great room. I asked the builder about putting some kind of backing like drywall or osb behind the T&G, and he told me it’s not needed and the T&G goes directly on the trusses.

    I’m not a builder, but that seemed strange to me and I thought that the heat in my house would escape through the spaces between the boards? Wood expands and shrinks with weather, so how could this be different? Is it normal to attach the T&G directly to the trusses? He uses an R-40 batt insulation in the roof. Shouldn’t there be something between the insulated trusses and the T&G?

    Please let me know your opinion as soon as possible. Thanks.


    1. Expert Member
      BILL WICHERS | | #67

      I have this exact problem in my own house, and recently posted a question here about a potential “quick and dirty” fix to last for a short period of time until I can rework the ceiling completely.

      You ABSOLUTELY NEED some kind of air barrier behind any tongue and groove boards used on a ceiling! They leak air like crazy! Airflow through the boards and then through the fiberglass batts will negate much of the insulating value of the batts. Don’t let them tell you the facing on the batts is enough either. There are multiple ways to do it right, but definitely do NOT let your contractor just install the T&G boards directly to the trusses!


  63. NormanWB | | #63

    Your contractor is following convention, but not best practices. While you can nail the T&G directly to the trusses, it will leak air and water vapor, which will impact your heating and cooling costs, as well as potentially damage your roof by causing rot and mold. Put up sheetrock and make sure it is properly taped. This will provide and effective air barrier. You can apply the T&G over that.

    His R-40 batt insulation will allow water vapor to pass through directly to the roof sheathing, where is will condense in the winter, causing it to eventually fail. Make sure he has ventilation baffles between the insulation and the roof sheathing, and vent path from the soffit to the ridge, or an adequate amount of rigid foam above the sheathing to prevent the water vapor from condensing .

  64. user-2310254 | | #64


    For more on cathedral ceiling best practices, see

  65. GBA Editor
    Martin Holladay | | #65

    Response to Brenda (Comment #62),
    Your contractor doesn't know what he is doing. Run, don't walk, away from this contractor.

    You are smarter than he is. Every ceiling needs an air barrier. For more information on this issue, see "Questions and Answers About Air Barriers."

  66. Brenda514 | | #66

    Thank you Norman, Steve and Martin for your reply and advice on how to handle my project. Its hard to find a good contractor that you can trust, when you don't have the knowledge of the "correct" way to build a home.

    Thanks again and wish me luck.


Log in or create an account to post an answer.


Recent Questions and Replies

  • |
  • |
  • |
  • |