The Common Plastics and Where Each One Fails

Why this matters

Plastic parts almost never fail from the thing people expect. They do not usually wear out, and they rarely fail at the pressure they were rated for. They fail because sunlight got at them, because somebody raised a temperature, because a product from a different trade touched them, or because a stress riser got molded or threaded into them. Each of those has a signature, and the signature is what separates a warranty claim from a repeat failure at the same spot.

Reconstructing a crack from the paperwork

A hot water branch in CPVC cracked at a wall penetration and flooded a finished space. The pipe was intact everywhere else. There was no way to interrogate the failure directly, because the failed section had already been cut out and replaced by the emergency call before anyone thought about cause. What was left was the file.

The emergency crew had done the right first thing: isolated the branch, relieved pressure at an open fixture, and let the line cool before cutting, because a hot water line at storage temperature scalds on contact and the pressure behind it drives the water at you when the pipe opens.

Three more hazards belong with any work that cuts, cements or heats plastic, and they are the ones that get skipped because nothing about them looks dangerous. Solvent primer and cement are flammable and their vapors are heavier than air, so they pool in trenches, crawl spaces and closed risers: ventilate the space, keep ignition sources out of it, and wear the glove class named on the product's safety data sheet rather than thin disposables, which those solvents go through. Heating or burning PVC and CPVC releases hydrogen chloride, so never bring a torch to them and treat any fire involving them as a respiratory exposure. PTFE above roughly 500 degrees F releases decomposition products that cause polymer fume fever, which is why torch work near PTFE tape or seats needs the threads cleaned first. Dry cutting or grinding any of these throws dust that belongs behind eye protection and extraction.

The record gave three dated entries.

Month zero: the branch was installed and hydrostatically tested. The test passed and the pressure and duration were written down. So the pipe left the installer sound, and installation damage as a cause has to explain how the pipe survived a test at above working pressure.

Month fourteen: a separate work order from a different trade shows work at that wall, sealing the penetration where the branch passes through. The line item names the task. It does not name the product.

Month twenty-one: the failure. That is seven months after the penetration work and twenty-one months after installation.

The shape of the timeline does the diagnosis. A material or installation defect fails early or fails under test. A pressure or temperature excursion fails at the moment of the excursion and there is no record of one. A failure that begins at a specific location seven months after a specific product was applied at exactly that location is a chemical exposure with a lag, and the lag is characteristic of environmental stress cracking rather than of chemical attack, which shows up as swelling or softening much sooner.

CPVC is known to be susceptible to environmental stress cracking on contact with plasticizer-bearing products. The list that catches people includes some thread sealants and cutting oils, some spray-applied insulation and foam, some firestop products, and some flexible tapes and coatings. Manufacturers publish their own compatible-products lists precisely because the incompatible ones are things a different trade applies without ever thinking about the pipe.

What the record could not tell us, and why that is the finding

The one thing the file did not contain was the name of the product applied at month fourteen. Without it, the cause is a class rather than a specific product, and no claim can be made against anyone.

That gap is not a footnote, it is the actionable output. The corrective action was not "use a different sealant," because nobody can say which one was used. It was a change to the job template: any product applied within contact distance of plastic pipe gets its product name written on the work order, and the pipe manufacturer's compatible-products list gets attached to any job that penetrates a wall carrying plastic pipe.

The failure mode of not doing that: every subsequent crack at that building is also uninvestigable, and the shop pays for each one while learning nothing.

The four ways a plastic fails

Every entry in the table below is one of these four. Naming the mechanism first is what tells you whether to change the material, the location, or the neighboring trade's product list.

Ultraviolet exposure. Sunlight breaks polymer chains at the surface. The signature is chalking, fading, and a surface that crazes into a fine crack network, worst on the sun-facing side and often perfect on the shaded side of the same pipe. Pigmented and carbon-black compounds resist it; unpigmented and light-colored ones do not.

Heat, which does two separate things. It softens and it derates, and the derating is the one that surprises people. See the next section.

Chemical attack and environmental stress cracking. Straight chemical attack shows as swelling, softening, tackiness, or dissolution, and it shows up fast. Environmental stress cracking needs sustained tensile stress plus a chemical that is only mildly aggressive, produces a brittle crack with no softening and no discoloration, and it takes months. The second one is far more common in the field and much harder to attribute, because by the time it cracks nobody remembers what touched it.

Mechanical, meaning notch sensitivity and creep. A scratch, a threaded root, a molding gate, or a tool mark concentrates stress, and rigid plastics are much more sensitive to that than metals are. Creep is the slow permanent deformation of a plastic under sustained load, which is why a plastic part that is fine on day one can be loose or distorted after a year of constant pressure.

Heat derates the pressure rating, steeply

A plastic pipe's pressure rating is stated at a reference temperature, conventionally 73 degrees F, and it falls sharply as the temperature rises. This is not a safety margin quietly eroding, it is a published multiplier.

The published derating for PVC at 140 degrees F is roughly 22 percent of its 73 degrees F rating. Work through what that means for a line somebody thought was conservatively specified: if it operates at 60 percent of the cold rating and then runs at 140 degrees F, it is at 0.60 divided by 0.22, which is about 2.7 times its actual rating at that temperature. That line is not near its limit, it is well past it, and nothing about the installation looks wrong.

CPVC exists to solve exactly this. Its familiar rating point is 180 degrees F at 100 psi, against roughly 400 psi at 73 degrees F for the same product, which is the same shape of derating, about a quarter, applied from a much higher starting point. Use the manufacturer's own derating table for the specific product and schedule rather than a remembered number, because the factors differ by material and by wall.

And never use PVC or CPVC for compressed air or any compressed gas. The prohibition is not about the pressure rating, it is about what stored gas energy does when a brittle pipe lets go: a water line splits and sprays, a gas-filled plastic line fragments and throws pieces. Use materials rated for compressed gas service, and check what the local code accepts.

The common plastics and where each one fails

Two rows in that table deserve a second look because they contradict habit. Acetal is the plastic a tech reaches for when a part needs to hold a dimension, and chlorinated potable water is the service it should not be in. And PTFE, the material everybody treats as chemically bulletproof, fails mechanically: a PTFE seat or seal under constant load keeps deforming, so a joint that was tight last year is not tight now, and re-torquing it just restarts the process.

The contact list nobody keeps

The reconstruction above turned on a product that a different trade applied, and that is the general case rather than an unlucky one. Plastic pipe and plastic components live in walls and ceilings where insulators, firestop crews, painters, pest control, and cleaning staff all apply chemicals within contact distance, and none of them have any reason to think about the pipe.

Three practical moves:

• Get the pipe manufacturer's compatible-products list and keep it with the job, not in a drawer. These lists exist because the manufacturers have been paying for this failure for decades.
• Record product names on the work order whenever anything is applied against a plastic component. A product name written down at the time is the difference between an attributable failure and a mystery.
• Treat any wrap, tape, foam, or sealant touching a plastic pipe as part of the fluid list, in exactly the same way as the fluid inside the pipe. It is in continuous contact at the same temperature, and it is there for the life of the installation.

How to verify you got this right

• The material name in your record identifies the compound, not just the family. For rigid vinyl that means the compound cell classification under ASTM D1784 rather than the word "PVC," and for PEX it means the chlorine-resistance classification under ASTM F876 rather than the word "PEX." Two fittings that look identical can be different compounds with different chemical lists.
• The operating temperature has been checked against the derated pressure rating at that temperature, not against the rating on the pipe stamp, which is stated at the reference temperature.
• Everything in continuous contact with the part is on a list somewhere, including insulation, sealants, and tapes.
• Any exposed run has either a UV-resistant compound, a coating rated for it, or a documented exposure limit from the maker with an install date to measure it against.
• Where a plastic part carries sustained load or sustained pressure, someone has thought about creep, which means the joint gets rechecked on a schedule rather than assumed to hold.

References

• Pipe and fitting manufacturer documentation for pressure-temperature derating tables and published lists of compatible and incompatible contact products
• ASTM D1784, Standard Specification for Rigid Poly(Vinyl Chloride) (PVC) Compounds and Chlorinated Poly(Vinyl Chloride) (CPVC) Compounds, for compound cell classification
• ASTM F876, Standard Specification for Crosslinked Polyethylene (PEX) Tubing, for chlorine-resistance classification of PEX
• See related: Chemical Compatibility and the Checks Worth Running; How to Choose Between a Metal and a Polymer