I've recently become aware that Code requires not just the outside of conduit penetrations of exterior walls to be sealed, but also the interior.

I've got 1/2" liquidtite flexible PVC conduit (1) going down to protective "cap" shields on top of buried grounding electrodes, (2) running along exterior of garage wall between a few exterior surface-mounted junction boxes and (3) coming through the exterior garage wall into the unfinished garage interior.


What should I use to seal the inside of the conduit, and where?

I have duct seal "putty" but have read that over time it does not make a good water vapor block especially with wires running through it. I have also read that spray cans of low-expanding closed cell urethane foam make good vapor blocks and are removable for future service.

My current plan is thus to apply some of that spray foam inside each liquidtite neck wherever it enters an electrode cap enclosure or exterior junction box or the first enclosure after going through the exterior wall. I would use duct seal to seal the outside of the conduit where it goes through holes in the wall. Is this reasonable and Code-compliant? (Anecdotes suggest that nobody worries about sealing conduit interiors these days)

2 Answers 2


You can seal the outside of conduit with closed cell expanding foam (as you suggest) or silicone caulking.

You can seal the inside of the conduit with silicone caulking or duct seal putty (as you suggest).

The conduit does not have to breathe or be open if the wiring is rated for exterior/wet, e.g. THWN. Wire that is THHN should not be in a wet location, and therefore the conduit is best kept open at the interior end.

When selecting a silicone sealer, note that

  1. most sealers cure only to a limited depth, perhaps 0.5in. So capping a conduit should be done with at most 1in deep plug, allowing cure depths from both sides, the open end and the tube side. Those with a curing agent or accelerator will indicate so on the label, e.g. for

  2. most silicone type sealers (this in contrast to acrylic or putty sealers) are corrosive curing unless noted otherwise on the label. The fumes released during the curing process may corrode the conduit or the wiring. This likely poses no problem in short open ended stubs of PVC conduit, commonly used to penetrate a house wall from an outside box. But for metal conduit (galvanized steel) and in junction boxes with exposed metal these sealants must be avoided, or -in the case of boxes- left to cure 48hrs with the box lid open.

This is a related and relevant answer, by "ConKbot" (2014):

Your normal every day RTV silicone is 'Condensation cured' where atmospheric moisture is what makes it cure. There are 3 general kinds. Acetoxy cure, which is the cheapest, bonds well, but releases acetic acid on curing (the acid ingredient in vinegar) And corrodes copper and lots of other metal. Even sealing an enclosure with it, the trapped fumes inside can corrode the items. The other two kinds are neutral cure, but still slightly different. Oxime cure which releases a ketone on curing. This can corrode copper on contact, but isnt as bad to use near electronics, just not directly on copper. The most desirable for electronics is Alkoxy cure, which releases methanol. Mind the fumes from all of them, but if you have a large surface area covered, Acetic acid fumes can be pretty nasty. Though your nose will give you plenty of warning that something isnt right.

This is a more thorough contribution by Dr. Barry L. Ornitz of Eastman Chemical Company Research (1997):

The corrosion is entirely dependent on the particular type of silicone sealant used, and most common ones generally available to the public still release acetic acid when curing.

To begin this discussion, however, it should be stated that silicone polymer is a term much like "plastic" - it covers a rather wide range of materials and properties. The chemistry of a particular polymer determines it characteristics, from hard and brittle varnishes, to soft and flexible rubbers. The initial work on polysiloxane chemistry dates back almost 60 years and it continues today, so "getting the formulation right" is hardly a meaningful expression.

Modern silicone elastomers are generally linear polymers of methyl silicone "oils" of extremely high molecular weight which will remain largely independent of one another but which may deliberately be cross-linked at a number of selected points. Two methods are commonly used to obtain such structures:

  1. Benzoyl peroxide or other strong oxidizers can be used to oxidize side groups and so establish cross links.

  2. Some vinyl or other organic groups groups can be incorporated into the siloxane structure, and then a catalyzed condensation polymerization of these is accomplished to establish cross links.

The first of these two methods is common in two-part silicone formulations such as potting compounds. In particular this is used in situation where a deep-section cure must be obtained.

The latter of the above two methods is generally found in the room temperature vulcanized (RTV) sealers generally available to the public. When the organic groups polymerize, a volatile species is generally released - the most common being acetic acid. Other species may be released if appropriate organic groups have been incorporated into the silicone structure; these include alcohols and even amines. These compounds are released as part of the cross-linking (curing) process. Most eventually evaporate from the cured silicone. They are NOT absorbed by the polymerization.

Single part silicone sealers generally contain a cross-linking catalyst (such as a tin compound) that is activated upon exposure to moisture in the air. These single-part sealers should not be expected to cure deeper than about a centimeter as the diffusion of moisture through the cured silicone, and the diffusion of the acetic acid or other compound out of the curing silicone is limited by the thickness. To demonstrate this, leave the cap off a tube of silicone sealer for a few days. The silicone around the neck of the tube will be hard, but the rest of the tube will be uncured.

Since the acetic acid is released during curing, it can attack the underlying substrate material. This can cause corrosion of certain metals and prevent the proper adhesion of the silicone. However, on other materials, the acid can etch the surface slightly and increase the adhesion. Aluminum is one such material. Copper and zinc, however, are corroded by the acid. Thus brass and galvanized steel should not be used with silicones which release acid. Dissimilar metals can form electrolytic couples and corrode severely underneath a covering of acetic acid releasing silicone. Silicones do not adhere well to all other plastics either.

General Electric's Silicone Division can provide information about the compatibility of their products with different surfaces. GE manufactures silicones for industrial as well as domestic use. The following condensed table is from a GE publication on silicone sealants for domestic use. N-R means not recommended.

            Adhesion Characteristics of General Electric Silicones 
                    Commonly Available to Homeowners*

                        Silicone II             Silicone        Paintable
                                                Rubber          Sealant 5091


Alclad Aluminum         Excellent               Excellent       Excellent
Anodized Aluminum       Excellent               Excellent       Excellent
Brass                   Good                    N-R             Good
Chrome                  Excellent               Excellent       Excellent
Copper                  Good                    Good            Good
Galvanized Steel        Fair                    N-R             Fair  
Carbon Steel            Excellent               Excellent       Excellent
Stainless Steel         Excellent               Excellent       Excellent


ABS                     Excellent               Excellent       Fair
Fiberglass              Excellent               Good            Fair
Formica                 Excellent               Excellent       Good
PVC (rigid)             Excellent               Fair            Good
LEXAN polycarbonate     Good                    Fair            Good
Vinyl (soft)            Good                    Fair            Good
Acrylic/Plexiglas       N-R                     N-R             N-R
Glass                   Excellent               Excellent       Excellent
Rubber (any type)       N-R                     N-R             N-R

* Note: The contractor grade silicone is similar to the Silicone II adherence properties, however, contractor grade silicone differs from Silicone II by having a longer tooling time, faster curing time and different formulations and ingredients.

GE Contractor Bath/Plumbing products, GE6040 (clear) and GE6070 (white) are specifically formulated to be non-corrosive to metals. In conclusion, using common silicone sealants around antenna connections is asking for trouble. Much like the Plasticizers in the vinyl jackets of coaxial cable, if you do not know they are corrosive, assume they are as the special non-corrosive grades are normally labeled as such.

  • Actually planning on sealing the conduit inside with the urethane foam - is there a downside to that?
    – Armand
    Dec 30, 2022 at 18:31
  • 1
    @Armand for me, as a DIY, it's messy, expensive (unless you know how to close off the can cleanly for re-use) and more difficult to remove if ever needed since the foam expands into the conduit. If the conduit penetrates a fire wall, make sure to get fire rated foam. The normal foam seconds as excellent fuel...
    – P2000
    Dec 30, 2022 at 18:49
  • 1
    Note that I would not recommend ordinary RTV silicone caulking for inside conduits, lest you create a corrosive atmosphere inside the conduit thanks to the RTV silicone releasing acetic acid as it cures Dec 30, 2022 at 20:53
  • @ThreePhaseEel excellent point, updated my answer after some more digging
    – P2000
    Dec 30, 2022 at 23:21
  • Superb answer - As an occasional visitor to this group I started to scroll down to see if I had written it some years back then realised that it had been answered in the last day. I learned a fair bit of this from scratch when working on sealing equipment that we were having manufactured in China. I've seldom seem such a wide ranging and useful commentary. Dec 31, 2022 at 10:01

I found a supposedly independent test of 4 conduit sealing methods at http://www.hixson-inc.com/_images/ConduitSealEffectiveness.pdf .

Their motivation for testing:

enter image description here enter image description here

Their test results summaries:

enter image description here enter image description here

Their conclusions:

enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.