More Musings of an Energy Nerd
Basements and crawlspaces can be damp. That’s why many homeowners assume that a basement or crawlspace needs at least one vapor retarder.
They’re right. But where does the vapor retarder go? Do you need one for your walls? The ceiling? The floor? And if some of these assemblies include insulation, you probably have another question: on which side of the insulation does the vapor retarder belong?
These are all good questions.
Two preliminary points
Before providing advice on vapor retarder placement in basements and crawlspaces, it’s worth answering two questions: Why are basements and crawlspaces damp? And what type of performance can we expect from a polyethylene vapor retarder?
There are three basic mechanisms that explain moisture accumulation in basements:
- Water can enter a basement or crawlspace in liquid form—through cracks in the walls, cracks in the slab, or the crack at the perimeter of the slab.
- Water vapor can enter a basement or crawlspace by evaporating from the interior surface of a damp wall or slab.
- Moisture in the indoor air can condense on cold surfaces in your basement, forming drips and puddles.
For more information on these three mechanisms, see Dealing With Basement Water From the Inside.
The next question is: What performance can you expect from a vapor retarder—for example, 6 mil polyethylene?
- You can expect polyethylene to greatly limit the migration of water vapor (that is, evaporation from a damp surface).
- If warm moist air ever contacts cold polyethylene, you can expect condensation to form on the surface of the polyethylene that faces the warm air.
- You can’t expect 6 mil polyethylene to stop liquid water, since water will almost always find a small hole or seam to seep through.
For example, let’s say you have an old-fashioned damp basement with uninsulated…
Weekly Newsletter
Get building science and energy efficiency advice, plus special offers, in your inbox.
This article is only available to GBA Prime Members
Sign up for a free trial and get instant access to this article as well as GBA’s complete library of premium articles and construction details.
Start Free TrialAlready a member? Log in
52 Comments
It seems like this is mostly talking about new construction. Retrofits can be trickier--you might need a vapor barrier above the slab, followed by insulation and flooring. And for retrofit walls, there's an added incentive to keep the concrete dry at the top, because you might have wood resting directly on the concrete with no capillary break, or even have wood embedded in the concrete. Some people recommend some level of vapor permeability for interior foam insulation for that reason, even open-cell spray foam. One concept is to use open-cell spray foam for the top few feet and foam boards or ccSF for the lower parts. Any comments on that scenario? See for example this previous question of mine: https://www.greenbuildingadvisor.com/question/crawl-space-open-cell-spray-foam
One strategy is to use mineral wool boards as exterior foundation insulation above grade to keep the wall warmer in winter and promote drying to the outside--at least in a cold climate. That could potentially backfire in a hot, humid climate, where that insulation would help keep the foundation cold in the summer, and allow humid outdoor air to condense on the concrete.
Charlie,
Thanks for your perceptive comments. You're right, of course, that in an older house that has no vapor barrier under the basement slab, it's impossible to install the polyethylene where it belongs without demolishing the slab. The usual solution is to install a layer of polyethylene above the slab followed by a continuous layer of horizontal rigid foam, followed by plywood or OSB. For more information on this approach, see "Installing Rigid Foam Above a Concrete Slab." (After reading your comment, I edited my article to include information on this issue. Thanks for the suggestion.)
Concerning your second concern -- an older house with an untreated mudsill resting direction on top of the concrete wall, or an older house with embedded joists -- I have provided several recommendations in my article, "Insulating Basement Walls With Embedded Joists." The best solutions tend to be site-specific, and depend on an analysis of the relevant risk factors.
Hello Martin,
In the case of a slab that doesn’t have a vapor barrier under the slab, does the use of rigid insulation with an integrated vapor barrier facing the slab obviate the need for the polyethylene?
Thanks for your insight.
Sofiane,
Q. "In the case of a slab that doesn’t have a vapor barrier under the slab, does the use of rigid insulation with an integrated vapor barrier facing the slab obviate the need for the polyethylene?"
A. Yes.
I wish I could afford to "lay some sheets of OSB" in my crawlspace - or as I used to throw one into the mud on site to keep my boots clean. Now I'm considering piling a few sheets under my bed as an inheritance for my nephews when I'm gone.
Joking aside, a very useful blog.
Malcolm,
Use shipping pallets. That's what I have in my dirt-floored basement. After 10 years, you can replace them if they get a little soft (assuming, of course, that you know a local business that gives away old pallets).
Martin,
Several of my neighbours have started using them for infill in the walls of their pole built sheds.
However recently, rather than ask whether I minded if they left the pallet with the materials for me to dispose of, my lumberyard charged me a deposit which would be refundable when I returned the pallet to them. Crazy times!
Great article, but I am curious what the process and assembly is when using Roxul rigid insulation under slab and exterior of poured concrete basement walls? Also curious if there are other foamless solutions for new and existing basements and how to achieve moisture control with those?
User 7568882,
GBA has published at least three article focusing on sub-slab mineral wool insulation. Here are the links:
"Sub-Slab Mineral Wool"
"Insulating a Slab with Rigid Mineral Wool"
"Urban Rustic: Prepping for a Basement Slab"
I have written at least two articles with ideas for foam-free foundations and foam-free houses. Here are the links:
"Foam-Free Insulated Foundations"
"Building a Foam-Free House"
Great! thank you Martin.
What about retrofitting a old rubble basement? Specifically, isn't there value of an air barrier putting down a poly layer on the basement walls to tie into a new subslab poly layer bringing it up to tie into an interior air barrier up through the house?
Steve,
If you are retrofitting an old rubble-stone basement, the best interior air barrier is a layer of closed-cell spray foam. The spray foam will act as (a) an air barrier, (b) an insulation layer with decent R-value per inch, and (c) a structural layer that helps tie loose stones together. Note that you will need a thermal barrier (a fire-protection layer) on the interior side of the spray foam.
What if you can't get the new ccf formulations with the low green house gases in them? Wouldn't it then make sense to either pull up and connect the poly from the subslab up? Or alternatively tie the subslab poly into a dimple board that is used for interior waterproofing?
Steve,
If you need to address the problems associated with rubble foundations, you're in a situation where the best possible solution is closed-cell spray foam. Of all the suggested uses for closed-cell spray foam -- which admittedly is a controversial product -- this application is one of the ones where there are few good substitutes.
I would search long and hard for a spray-foam contractor who uses one of the blowing agents that has a relatively low global warming potential. If you simply can't find such a contractor, you should still consider using closed-cell spray foam. But of course, the decision is still yours.
Thanks Martin!
To clarify, we started with a 1inch concrete slab on top of heavy clay and high water table.
We are currently underpinning with 3inch compacted Glavel with embedded weeping tile to a sump pump. Followed by 2inch of type IX EPS > Poly > 3 inch concrete slab. Will be thermally breaking the slab with 2inch EPS all around as well.
Our underpinning contractor usually puts a delta membrane against the new underpinning footers to just above the slab. And then tyvek all the way up the wall.
I was thinking of asking them to bring the delta up up grade, and then connect poly the rest of the way up to create a "inverted bathtub"? Where the subslab poly ties into the delta that ties into poly up higher in the wall. Bulk water would be managed by the interior weepers to sump.
Finally insulate the whole thing with unfaced EPS against the delta/poly walls?
Hypothetically would this assembly work? (clearly it seems more labour intensive and doesn't help to tie the stones together like you mentioned about CCF solution)
Steve,
I'm not an expert in this kind of retrofit. But most rubble walls have gaps between the stones and all kinds of cracks that easily allow liquid water to find its way to drainage tiles at the base of the wall. That's why dimple mat is usually unnecessary on the interior side of a rubble wall.
So my preferred approach is to assume that the wall includes its own drainage paths, and to install closed-cell spray foam on the interior--depending on the closed-cell foam to be the air barrier, insulation layer, and a structural element that helps tie the stones together.
All of that said, your approach may work well. Good luck.
I was taught, and experience has borne out, that if you are going to settle for poly over dirt floor and nothing else that you should ballast it with something that exerts even pressure all over the surface of the plastic, like round pea-gravel. The reasoning being that even a small tear or gap in the poly can allow large amounts of moisture to escape through the hole in the poly. The moisture can migrate under the loose poly and find the holes. Whereas if you ballast it with something that keeps the poly in solid contact with the earth, and there is a hole or a tear, the only moisture that will migrate through the holes is moisture from the immediate area of the hole itself.
Steven,
For the polyethylene to perform as a decent vapor barrier (to limit diffusion), it need not be perfect. Even if the polyethylene consists of 98% plastic and 2% holes, it will still be 98% effective as a vapor barrier. (For more on this issue, see "Polyethylene Under Concrete Slabs.")
So your concern has nothing to do with vapor diffusion. When it comes to air leakage, however, you have a valid point. The amount of air leakage into a house through a hole in polyethylene installed over a dirt floor varies widely, but such air leakage can occur, especially in a house with a very leaky ceiling on the top floor (in other words, a house depressurized by the stack effect), or a house that is depressurized by exhaust fans. (Believe it or not, the source of the air is air in the soil, which is connected to the atmosphere through pores and gaps in the soil.)
Ideally, your house will be relatively airtight, and you will try to limit depressurization caused by fans. But you're basically right -- sealing seams and holes in polyethylene laid on a dirt floor is a good idea. The more robust the poly, the better.
If you plan to install a concrete slab on top of the polyethylene, don't worry about air leaks. The slab is your air barrier (although caulking the perimeter of the slab after the concrete has cured is probably a good idea).
We have a basement slab that has 2″ of XPS under it and on the sides, and a Stego-wrap vapor barrier between the insulation and the concrete. It has had 13 years for the concrete to dry out since construction. I plan to finish the floor with engineered wood flooring (or old-growth fir). Should the subfloor assembly be vapor permeable -- should I avoid putting something like Delta-FL under the plywood subfloor that would sandwich the concrete between vapor barriers? Does something need to go between the plywood and the concrete? We probably will open the windows in the summer and use air conditioning only if it gets really hot. I have looked through GBA articles extensively, and everything about constructing a basement subfloor seems to assume a concrete slab that is not insulated and has no vapor barrier. We are in Portland, Oregon. Thanks!
Adam,
Q. "Should I avoid putting something like Delta-FL under the plywood subfloor that would sandwich the concrete between vapor barriers?"
A. Sandwiching the concrete between two impermeable layers will do no harm. Concrete doesn't mind being damp forever. (That's why you see concrete footings and columns to support a bridge across a river.) For more information on Delta FL, see "What’s the Best Basement Flooring System?"
Q. "Does something need to go between the plywood and the concrete?"
A. I would install polyethylene in this location.
Q. "We probably will open the windows in the summer and use air conditioning only if it gets really hot."
A. Since your slab has 2 inches of XPS between the slab and the soil, you're fine. The slab will be at room temperature, not soil temperature, so condensation or moisture accumulation are not major concerns.
Why do all slab floors need poly? If we have a crushed stone layer drained to remove liquid water and acting as a capillary break seems we shouldn't ever have liquid water in contact with the slab -with our without poly.
If moisture diffusion always moves from warm to cool, then it would seen moisture would move downward in all conditions experienced in the house.
Yes, poly is easy/cheap to install. Yes, its required. Just wondering why moisture wouldn't diffuse downward without it?
David,
Moisture diffuses with not much regard as to whether the two spaces are cold or hot. The relative temperatures have more to do with whether it will condense or not. There is no mechanism to cause moisture to diffuse downwards into the moist substrate - hence the need for poly under slabs.
I am certainly not an expert, but these articles seem to suggest diffusion is almost always from warm to cool. But the distinction you make about "whether it will condense or not" might be the reason the diffusion through the slab doesn't follow this pattern?
https://www.buildingscience.com/conversation/cup-joe-second-law-thermodynamics
https://www.sbcmag.info/news/2018/aug/video-thermodynamics-simplified-heat-flows-warm-cold
https://www.buildingscience.com/documents/insights/bsi-021-thermodynamics-its-not-rocket-science
David,
I'm of course not saying that isn't true, but the dominant drive is to equalize moisture. The damp soil isn't diffusing downward, or it would always be dry. It is trying to make the slab the same moisture content as it is, and the slab is playing the same game with the air above it.
David,
Malcolm is right. If you put a wet towel in the refrigerator, and then bring the wet towel out of the refrigerator into a warm room, the moisture in the towel will evaporate into the room, even though the towel was cold to start.
What happens is that the interface layer changes temperature to balance to cool side with the warm side, and once that interface layer warms up a little, evaporation (diffusion) occurs.
I have a 200 year old farmhouse with a field stone foundation and gravel floor. The water table is just below the floor so the basement is always extremely damp. So far I have installed an interior curtain drain out to daylight. I am jacking the house off the foundation and installing EPDM between the sill and the foundation. I plan to lay down poly on the floor and roll it up the wall to connect to the EPDM. After that, I will pour a concrete floor with a gap at the walls to allow water that comes through the walls to enter the curtain drain, but the poly will be continuous. My thought is this will create a reverse swimming pool, keeping the moisture "outside" the basement interior.
I am skeptical of the approach you outlined. A reverse bathtub might even fill with water.
See article quote below.
"You can’t expect 6 mil polyethylene to stop liquid water, since water will almost always find a small hole or seam to seep through."
I am using reinforced 10 mil. Water will come through the foundation, I am just trying to give the water a path to the curtain drain while keeping out as much of the humidity as I can. There wont be any hydraulic pressure against the poly as there is a gap between it and the wall.
John,
Unlike David, I see no problem with your approach. The key is a functioning drainage system. As long as your drain pipes aren't clogged, and lead to daylight, that's where the liquid water will go.
I’ve read a variety of articles on this subject over the past several years. The advice should also include an explicit description of the assumed yearly climate, particularly temperature and humidity changes. Then the reader can decide what would be helpful for their structure. One size does not fit all.
Craig,
I firmly believe that there is no need to complicate advice with irrelevant factors, since the irrelevant factors make it hard for builders and homeowners to follow offered advice. I think my general advice applies to U.S. climate zones. That said, I am willing to stand corrected if you want to suggest a climate zone where my advice would be bad.
I'm an old-house-retrofit, would Rockwool be preferable to foam board on the interior, with no vapor barrier? Then the wall assembly can dry either direction, any season.
Chris,
Q. "In an old-house retrofit, would Rockwool be preferable to foam board on the interior, with no vapor barrier?"
A. No.
Q. "Then the wall assembly can dry either direction, any season."
A. It's never a good idea to invite moisture into your basement. You don't want inward drying.
Martin, another great article. I appreciate your logical, science-based and concise writing. Regarding the combination of a vapor barrier and rigid foam boards on the walls of a crawlspace, I agree that the combination of the rigid foam and vapor barrier on the walls is redundant if diffusion is the mechanism for moisture movement into the crawl space, but if there is any risk of bulk water intrusion, a vapor barrier on the wall (with a gap at the top for termite inspections) would do a good job of directing bulk moisture to grade below the floor's vapor barrier. This of course assumes that all seams are sealed between floor and wall portions of the vapor barrier. With rigid foam on the walls, I would suspect that lapping the floor vapor barrier up a foot or so, onto the inside surface of the foam, and sealing the seams, would work as well. Regardless, I believe that the detail of the connection between the vapor barrier on the floor and either the vapor barrier or foam on the wall is an important detail. I'm very interested in your take on this.
Armand,
The way to handle bulk water is with a drainage system. You can't depend on polyethylene to stop bulk water.
If your have free-draining backfill on the exterior of your building, or an exterior dimple mat to facilitate drainage, and a footing drain that works, then you've addressed the bulk water issue. Expecting polyethylene on the walls to play a role in this function is a fool's errand.
You are of course correct Martin, the detail that I was suggesting has limitations, but I continue to believe that there is value in forcing whatever bulk water enters the crawlspace, under the poly on the floor. Even houses with good gutters, downspouts, grading, etc., occasionally experience excessive water at the building perimeter, say from a clogged gutter. If the detail of the poly on the floor of the crawlspace is reversed from my suggestion, you can have a swimming pool in the crawlspace.
Dimple mat is a great product, but I don't often have the luxury of installing it. I always focus on the gutters, downspouts and grading.
Armand,
You are absolutely correct that focusing on gutters, downspouts, and grading is the first step in addressing a damp crawlspace or basement.
Thank you, Martin, for this very helpful article which we have read with great interest. We own a house near the tristate NH-VT-MA intersection. A few questions related to the “Basement Walls” section:
1) You say that outward of the foundation there should be crushed stone or gravel. Roughly how many feet outward should this loosefill extend: 3 feet? 6 feet? Whatever the width of the excavator’s bucket happens to be?
2) You say to use crushed stone / gravel OR a dimple mat. Would a good “belt and suspenders” approach include both?
3) We want very much to wetproof the *outside* our Victorian era fieldstone foundation. We already have a full perimeter interior drainage system with motorized sump pumping to daylight. Nevertheless, we feel it’s much better to stop water from entering in the first place, as opposed to collecting it into the system and having the electric sump pump it out. Another factor in our desire for *exterior* wet proofing is the fact that our fieldstones are extremely heterogeneous in projection. As a result, only some of the water entering through the foundation wall does indeed trickle into the collection flange: Any trickling near an especially protrusive rock merely trickles down the face of this proud one, misses the collection flange, and forms puddles on the floor. We have extended the collection pockets as much as possible but the heterogeneity of our stones continues being a headache. For this reason, once we find an excavation contractor who is willing to take on our job (no luck so far) we will need to apply a parge coat of cement to exterior of foundation to create flush surface.
In this scenario, would the correct order be: apply asphaltic dampproofing coat directly to fieldstones, then apply parge coat cement, then dimple mat, OR apply parge coat cement to fieldstones, then paint cement with asphaltic dampproofing, then dimple mat?
4) After waterproofing outside the foundation, we are hoping to insulate there as well (even though we will most likely need additional insulation on inside to reach R20, which we are fine with). Problem is, the only way we will be able to insulate outside the foundation will be if our dimple mat ends up being either flush with the suprafoundational (plankboard) wall sheathing (unlikely) OR proud of the latter by < 4 inches. Why 4 inches? Because
we also hope to strip walls above sill plate to bare sheathing, press/roll on a fully adhered membrane, then attach 4 inches rigid insulation (which is the thickness a building scientist recommended for our house). We want to avoid having to flash the top of the exterior basement insulation if possible. Assuming the face of the dimple mat ends up proud of the sheathing by < 4 inches, we will add exterior insulation to the basement until it reaches 4 inches proud (or nearly so). We will then protect the above-ground basement insulation using one of the products listed in your 2012 article “How to Insulate a Basement Wall”. Does the above sound like a reasonable approach to avoiding the need to flash above the exterior basement insulation?
As always, thank you for your time and help and for this excellent article.
Anonymous,
Q. "You say that outward of the foundation there should be crushed stone or gravel. Roughly how many feet outward should this loosefill extend: 3 feet? 6 feet? Whatever the width of the excavator’s bucket happens to be?"
A. My answer is "whatever the width of the excavator's bucket happens to be." That said, if you have a very big excavation hole, you don't want to pay more for crushed stone than necessary. Common sense applies here. About 16 inches is plenty, although placing the material uniformly is tricky. Landscape fabric may be useful to separate the crushed stone from the soil, and you may need a worker in the hole to keep everything where it belongs during backfilling.
Q. "You say to use crushed stone / gravel OR a dimple mat. Would a good “belt and suspenders” approach include both?"
A. We are entering the field of human psychology here rather than building science. If it helps you sleep at night, and you can afford it, install both. I wouldn't. If I could afford dimple mat, I'd be satisfied that the dimple mat will handle the task without any crushed stone as backfill.
Q. "Would the correct order be: apply asphaltic dampproofing coat directly to fieldstones, then apply parge coat cement, then dimple mat, OR apply parge coat cement to fieldstones, then paint cement with asphaltic dampproofing, then dimple mat?"
A. The latter.
Q. "Does the above sound like a reasonable approach to avoiding the need to flash above the exterior basement insulation?"
A. If I understand your description accurately, your approach will work, as long as you include a durable flashing (for example, 16-ounce copper with soldered seams) that connects the housewrap under your siding with the exterior surface of your exterior foundation insulation. This flashing will have a vertical leg that slips behind the housewrap, and then a sloped surface to shed water, and then another vertical leg that conveys water to the outside surface of the material you intend to use to protect your exterior foundation insulation.
Thank you, Martin. Your response is extremely helpful. We appreciate your time and clarification
Any tips on ways to figure out if an older home has poly under the concrete floor?
Davin,
Find an inconspicuous area of your basement and drill a 1.5-inch or 2-inch-diamter hole in the concrete with a concrete bit. Use your finger to feel the edges of the hole for polyethylene.
Hi, thanks for the article. I'm retrofitting a 1940 basement on a existing slab (no insulation or polyethelene underneath). Unfortunately I don't have the headroom for insulation or OSB on top. Given the lack of these items is there still an advantage to adding a layer of poly or something like DMX one-step (seems taped) underneath LVP flooring. I'm on the east coast with humid summers. Main objective is to reduce condensation and associated mold issues. Thanks!
- Thomas
Thomas,
To reduce condensation on a cold concrete floor, the only solution is insulation. The insulation can be under the concrete or on top of the concrete, but you need the insulation. A layer of poly does nothing -- all you end up with is cold poly with puddles on top.
If you don't have headroom for one inch of insulation, it's time to do some demolition. Demolish the slab, dig down a few inches, and do it right.
Love this article and all the links. Such an important topic.
While I appreciate that Joe L. will know more than I do, I have always worried that the poly under the slab would have thousands of holes from being sandwiched between very heavy slab and the relatively sharp crushed stone.
My other worry are concrete installers. On one job, they used a wheelbarrow to place the concrete for the slab. They did not place anything under the metal wheel guard at the front of the wheelbarrow when dumping the concrete over the poly. There was one hole in the poly for each wheelbarrow of concrete dumped. I was new to construction and didn’t say anything. The concrete installers need to either use a concrete pump to distribute the concrete around the floor or place a board under the wheelbarrow wheelbarrow guard. A conversation to have with the concrete contractor before the pour.
For basement walls; how thick is thick enough for EPS to act as a vapor barrier? I'm in the process of insulating my poured foundation walls. I have immediate access to some free left over materials, and had planned a few walls sections as follows;
2.5" + 1.5" of EPS Type 2
2" + 2" of XPS
2"XPS + 1.5" EPS Type 2
These sheets are all lapped, and I plan to run the first layer horizontal and second layer vertical. I'm less concerned with the configurations above but I'm against a budget and for the rest of the walls, I had hoped to just use two layers of 1.5" EPS Type 2. Assuming a perm rating of 2.8/inch, are my calculations correct that 3" would have a perm rating of 0.9333?
Overall, the plan is to install a service wall in front of the foam with R12 batts, then OSB to cover. Am I going down the right path? Atlantic Canada (US Zone 6?), ~noted as 4,200 heating degree days, but that seems to be trending lower past decade.
Hi, great article! I am in Florida in a flood zone and my house has a 8’ high crawl space. The previous owner ‘finished’ the crawlspace. He installed a french drain and sump pump around the perimeter to deal with hydrostatic pressure, put a 4” concrete slab on a 6 mil vapor barrier, painted the inside of the foundation walls and installed drywall on the ceiling. Since the house is in a flood zone the crawlspace has flood vents. A unique setup that I as the new homeowner need to fix. My goal is to air seal the attic as much as I can, dehumidifier the space but keep it completely separate from the living space. So my question is - what would be the best way to separate the crawl space from the living area. The floor joists are engineered web trusses. I don’t want to use spray foam.
The options is see are
1) Rigid foam (foilfaced polyiso) on the underside of the joists. Seams taped
2) airtight membrane stapled to the underside of the joists. Seams taped
3) liquid applied air barrier such as ProClima Visconn on the underside of the plywood subfloor
4) Zip R sheathing on the underside of the floor joist. Seams taped - not sure if there are fire protection issues with this
Some of these would be air and vapor barrier, others would be vapor open.
I’m not sure if I need both since the crawlspace would be humidity controlled.
I’d love your feedback on my options and if there are other ways to get to the goal of creating a dehumidified crawlspace that completely separated from the living space - please let me know.
Thank you so much.
Niko
When using EPS on a basement wall, are you using something that is just pure foam, with no lining? All I can find locally has a plastic lining on the board.
B Tremp,
In many areas of the country, it's easy to buy EPS without any facing. If your area of the country only sells faced EPS, you can go ahead and use it without worrying.
Thank you Martin!
It comes with plastic facing on one side and foil on the other. Should the foil face into the room? Or should I just peel them off?
You can have the foil facing the room. If you have an air gap between the foil and drywall, you can gain a little bit of R value from the reflective barrier.
Log in or become a member to post a comment.
Sign up Log in