What to do when the ends of the floor joists are embedded in concrete?
weURPfEL3F
| Posted in Energy Efficiency and Durability on
As an energy adviser in Manitoba, I see a lot of basements with cast-in-place joists. If the header space is filled with concrete up to the underside of the floor deck, the advice is easy: Cut R-5 XPS to fit, caulk or foam it in, and glue the vapour barrier.
While I’m at it – how about the rim joist space parallel to the joists?
GBA Detail Library
A collection of one thousand construction details organized by climate and house part
Search and download construction details
Replies
Geoff
If you are talking about fitting and sealing XPS from the inside, I would worry about the unintended consequences...the embeded ends of the joists will be much colder and therefore wetter and will have no drying potential. Not good. How about XPS, covered with parging, on the outside?
Garth,
How is this any different than wall studs attached to a cold, low R sheathing?
Doug
Couple of differences that I can think of. One is that there is less drying potential. Two is that the concrete is most likely a continuous source of moisture through capillary action.
Garth,
These same floor joists and rim joists have been embedded and in contact with the concrete foundation for the life of the house. If this was a continuous source of moisture it is likely to have shown up in the form of rot by now. Not only that, this area typically lacks insulation when built and would have a considerable surface condensation potential between the above grade cold concrete foundation and the warm air inside the building envelope.
Often, these cast in place joists rest on top of concrete block foundations and the concrete cap actually helps to keep moisture laden air from entering the conditioned space via the cores in the block walls. Now in MN, in the metro at least, if you use concrete block for foundations you must use fill-top blocks for the final course and mortar between the joints. This is also for Radon control purposes.
Doug
Energy flow through the uninsulated concrete may be what kept the wood from rotting. Stopping or reducing this energy flow might just change this. Not saying it will happen, but the potential is there.
Garth,
Time will tell on this as many cold climate homes have used XPS blocking between floor joists to form a continuous air barrier between floors, this is going on 30 years now. If we hear that main floor and second story floor systems are pancaking down to the lowest level because of rotted joist ends, we'll know your are right.
I do not discount what you have said and in new construction the thermal break for rim and floor joists is often on the exterior. This does pose a bit of an issue with regard to thick walls, thermal boundary and continuous air barrier in my opinion, but it may be the safe way to build.
As this was my first post here I didn't realize that my space was limited and a lot of my question was cut off. To finish my thought - what I am wondering about is those situations where the header space between the joists is not completely filled with concrete. I advise customers to remove any fiberglass they have stuffed in there and spray foam the cavity. The question is: how big does that gap between the concrete and the floor deck above have to be before you can advise filling it and bringing the basement wall insulation right up to the floor deck above?
Garth,
To be fair, Geoff had a question about a very specific condition and where I have used interior XPS blocking between joists, I had 2" of XPS from the top of the foundation wall to the footing on the cold side. This would make the foundation wall warm and would not be an exact match to what Geoff was describing. Also the rim and floor joists sat on a treated plate, not directly on or embedded in concrete.
Passive House has avoided this method in some of their homes by supporting the 2nd floor of a 2 story with the strapped wiring chase wall attached to the interior OSB air barrier.
Doug
Fair enough. Geoff also mentioned gluing the foam to the vapor barrier. Geoff, are you referring to poly on the inside of a wood framed, fiberglass bat filled interior basement wall? I know this is commonly done and even approved in some jurisdictions. Here is what Dr Joe has to say about insulating basements.
http://www.buildingscience.com/documents/digests/bsd-103-understanding-basements/?searchterm=understanding basements
Garth
Unfortunately Dr. Joe doesn't address the issue either. What I am referring to regards what is a very common practise hereabouts of casting the joist ends into the top of the foundation wall. The header cavity is sometimes completely filled with concrete, right up to the underside of the main floor deck, sometimes with a gap at the top of varying height. In the case where the cavity is full, I recommend R5 styrofoam in the header space so as not to block all the heat from the basement reaching that lump of concrete, so as to prevent the floor above, at the edge of the, conditioned space, from getting too cold and causing condensation. If there is a gap I always recommend to my clients that they fill that cavity with spray foam before placing any additional insulation in front of the concrete. What I am trying to determine is how big that gap has to be, i.e. what R value needs to go in there, before you can safely extend the wall insulation all the way up to the deck.
I have seen places where overinsulating that header space has caused floors at the edge of the building to sit wet enough to rot - think black oak.
Geoff
I know what you are referring to. Around here we call it "beam filled" and there are many older homes like this. I am no expert, but my advice would be to seal the perimeter of the joist area with canned foam or caulk to reduce any air leakage as much as possible. If insulating is causing condensation on the floor above, then it is probably best to not insulate the beam filled area at all. Better to lose some heat than to cause mold or rot. Hope this helps.
Doug,
This is a situation I run into every day in a variety of houses, and I see every possible insulation configuration in them. Probably at least half of the houses in Winnipeg have the joists cast into the foundation wall. In some cases there is a space between the concrete in the header area and the floor deck above. My question remains how big does that space, filled with blown in polyurethane foam, have to be to effectively isolate the floor above from that cold hunk of cement? Will 2", i.e. R10 suffice? Or is 3" necessary, or 4"? At what point can I safely advise the customer that it is now okay to continue the wall insulation up to the underside of the floor deck between the joists?
Geoff
I think that i finally get what you are asking...sorry for not getting it sooner. I would think that even an inch of foam would be enough to prevent condensation on the floor above. After all, R5 is probably more than is in the walls of these homes in many places (headers, through studs etc) and is more than most windows...
Geoff,
Is the foundation wall currently insulated? Inside or out? Describe the wall construction above the rim/floor joist area. Is the house a rambler or multistory?
Garth
Now we're on the right track! But I'm still not sure about the next part.
On the outer side of that lump of concrete we have 1 1/2" of lumber and, probably, some parging. Above it we now have R 5-6 of foam. If I then block most of the heat from the basement side by putting R24 batt insulation against the inner face, is the floor above going to be warm, or just barely above that condensation threshhold?
Geoff
It will be as warm as anything else in the house that is insulated to R5...in other words, not very warm....but still, probably quite an improvement. I still think that the ends of the joists may be compromised somewhat, but it is so dry in the winter that it will not likely be an issue. I'm sure that you know this, but the R24 batts will need an vapour permeable air barrier on the interior to perform well. No poly.
Geoff,
The problem raised by several readers -- that any interior insulation makes the ends of the joists colder, and therefore wetter, and therefore more likely to rot -- is a real one. Moreover, concrete is a conductor, so it acts as a thermal bridge. You can't just insulate the rim joist area if you want the insulation to be effective; you need to bring the insulation down to the footings.
Really, in this situation the best solution is to insulate the foundation and rim joist area from the exterior.
The problem of embedded joists and beams was discussed in Insulating Old Brick Buildings:
“If the building is insulated on the interior, the ends of these embedded beams get colder — and therefore wetter. Moreover, less energy is available to help them dry out.
“ ‘Embedded wood timbers can rot,’ says [John] Straube. ‘... There are a number of techniques to address embedded beams. You can inject the wood with borate salts to preserve the wood. You can insert metal wedges to conduct heat to the end of the beam. You can install hot water pipes to heat the end of the beam. Finally, there’s the practical Yankee solution: you build a load-bearing wood wall to support the beam, and then you fire up your chainsaw and cut off the end of the beam.’
“Almost all of the solutions to the embedded beam problem have drawbacks except the chainsaw solution.”
Garth,
The basement insulation method favoured by Manitoba Hydro is as follows: Install a moisture barrier of 6 mil poly from grade level down and let it run out onto the basement floor. Then build a 2x4 stud wall and install it 3 1/2" away from the exterior wall. Drop R12 batts in behind the stud wall, then fill the stud cavities with a second layer of R12. Finally, join the vapour barrier to the moisture barrier at the foot of the wall and run it up to the ceiling as usual. The idea of this is to encapsulate the fiberglass, so that if there is any wall seepage it will run out under the structure onto the floor and not trash your insulation.
In Manitoba we get a great variation in relative humidity between summer and winter, so a vapour permeable barrier would tend to lead to soggy insulation in the summer months.
Geoff,
I strongly advise you not to follow the proposed insulation method using fiberglass and poly. Many research programs have shown that this system doesn't work. Many such installations develop mold.
This advice has been obsolete for at least 20 years. To insulate a basement wall, stick to closed-cell spray foam or rigid foam. Never use fiberglass batts or polyethylene.
Martin
To be fair, most batt and poly basement installations do not include the first layer of poly as described in the Manitoba Hydro method. Might this method be more acceptable if the second (interior) layer of poly was eliminated and if humidity was controlled? I do agree that exterior insulation is the best solution in this case, as it protects the embedded joists, but this is difficult to impossible in some cases.
Martin,
As far as joist ends in concrete are concerned, thousands of houses in Winnipeg were built this way from the 30s right into the 70s. I have looked at hundreds of them and I have yet to see evidence of rot unless exterior details were done wrong and the rim joist is rotten. This is not to say that I think it's a good building method - it's just the way it is.
With regard to the moisture barrier/batt/vapour barrier method, the only problems I have seen have resulted when the builder put up the moisture barrier and then took off for the weekend before coming back to install the framing and insulation. Naturally there was condensation trapped behind the poly in those cases.
A large part of the reason this method is recommended is that it is about the cheapest way to get to R24, which is what our federal and local programs stipulate for basements in order to receive incentive grants. Government being 20 years behind the times is pretty much par for the course!
A lot of our homeowners are achieving the R24 standard by sticking 2" of rigid foam to the wall and then framing with 2x4s and adding R14 Roxul batts. The moisture barrier against the wall is foregone in this situation, but an interior vapour barrier is still applied. This method is difficult, however, when walls are not smooth and even, as in a rubble foundation.
Garth,
My own advice is always: no poly and no fiberglass batts on basement walls.
Some consultants, including those at the Building Science Corp., say it's OK to build a stud wall and fill it with fiberglass batts, as long as there is a layer of rigid foam between the stud wall and the concrete wall.
I think it's better to just add the amount of foam you need to achieve the desired R-value, and keep the stud bays empty. In any case, the BSC consultants and I agree on the "no poly" rule.
Geoff,
You wrote, "Thousands of houses in Winnipeg were built this way." There's no arguing with the results you see in the field, so it's possible that Winnipeg basements are dry enough, and your joist lumber is rot-resistant enough, that you don't have problems.
Nevertheless, the only relevant data is that gathered in homes with insulated rim joists. If you're telling me that joists don't rot when there's no insulation, that's not news. The escaping heat is keeping everything dry.
The real question is, if you examine 100 homes with interior insulation around the joist ends, are any of those joists beginning to rot after 5 or 10 years?
Martin,
A great number of the houses I see with the cast in joists have fiberglass stuffed in the cavity above the concrete, if there is one, and more fiberglass in front of the concrete. I always try to remove the fiberglass from one or two header spaces to see if moisture is being trapped or if the area is overinsulated. Up to now, and I've been doing this nearly 6 years, no rot in the joists, although I have seen, as I mentioned earlier, mold in the floor decking and floor finish material on the upper side.
Maybe it is all due to our extreme climate. Winnipeg in winter is so dry you can just about stick the cat to the wall with static electricity.
Geoff
Love the cat on the wall comment. I know all about that kind of dry. You might want to suggest to Manitoba Hydro that they hire a good building consultant to check out their recommended methods. Talk to John Straube or someone else at BSC...might prevent some future liability issues...
Garth
The issue there, of course, is that you are dealing with a large bureaucracy with many departments. They employ a great many engineers and consultants already, so the difficulty is in achieving consensus. Plus they publish this information in booklets, on disc and on the internet, so changing it would require altering all those resources plus numerous forms, and then a lot of retraining.
When you consider that, as yet, there have been no major problems, especially since probably the majority of these retrofit insulation projects have been undertaken by the homeowners themselves, there is considerable inertia to overcome.
This is not to say that I will not continue myself to give customers the best advice I can that will fulfill as well as possible the requirements of the program as well as the clients needs. If you or Martin could kindly direct me to the best, latest thinking on this?
Hi Geoff,
I am looking for some advice on embedded floor joists in Winnipeg. Have you determined the best insulation method for the joist space? We are wanting to remove the current frost walls, rebuild to code, and add 3" of sprayfoam to the foundation walls. My question is should we fill the embedded joist cavity with sprayfoam or just apply 1" of sprayfoam as a vapor barrier, or caulk and add rigid?
Geoff,
Q. "If you or Martin could kindly direct me to the best, latest thinking on this?"
A. I'm not sure what "this" is. However, here are some links:
GBA: Insulating basement walls
Building Science Corp.: Basement Insulation Systems
Renovating Your Basement
For information on embedded joists, you might want to look at Assessing the Durability Impacts of Energy Efficient Enclosure Upgrades Using Hygrothermal Modeling. That document notes, "One-dimensional analyses ignore details of penetrations through the enclosure wall. In older buildings, floor joists (often wood) and floor slabs (often wood or concrete) are often embedded in the wall. The temperature and moisture conditions at these locations may need to be analyzed using 2-D methods, although if the results of the clear wall are dry enough, or the embedment depth small enough, no further analysis may be required."