How to Choose a Material for a Permanently Wet Location

Why this matters

Wet locations do not fail because somebody forgot that water corrodes metal. They fail because the person choosing the material answered "outdoor" and stopped, when the questions that decide the outcome are which water, how often it dries, what else the part touches, and whether it is under load. Those four answers point at completely different materials, and a part chosen for the wrong one of them looks correct on the invoice and fails in a season.

This procedure is a short written specification with nine fields. Fill them in before you buy anything. The last section shows one filled in for a real job.

Before the spec: the hazards that come first

Most permanently wet locations are also rooftops, crawl spaces, mechanical rooms, or pits, and every one of those carries a hazard that outranks material selection.

• On a roof: set up fall protection before you carry a tape measure anywhere near the edge, and stay behind a guardrail or tied off to an anchor rated for fall arrest. Probe soft or ponded areas before putting weight on them; a deck that has been wet for years is the reason you are there.
• Standing water near powered equipment: de-energize at the disconnect, lock or tag it, and prove the circuit dead with the live-dead-live sequence in NFPA 70E-2021, 120.5 before you stand in the water. The de-energizing and lockout duty for electrical work sits at 29 CFR 1910.333(b)(2).
• Anything the part is holding up: put an independent support under the load before you loosen a corroded hanger or bracket. Corroded fasteners do not yield gradually.
• In a pit, sump, or tight crawl space: treat it as a potential confined space and do not enter on the assumption that it is fine because it is shallow. Standing water and decomposing material displace oxygen.

Fields 1 to 3: what the water actually is

Field 1, exposure class. Write one of: continuously immersed, wet-dry cycling, splash and washdown, high humidity with condensation, or buried. This is the field people get wrong most often, and the surprise is that wet-dry cycling is usually more aggressive than continuous immersion. Each dry-down concentrates whatever was dissolved in the water at the wetted line, and each rewetting delivers a fresh charge of oxygen. Immersed metal sees dilute chemistry and limited oxygen. A part in a drip zone sees a concentrating salt bath twice a day.

Field 2, water chemistry. Name the disinfectant, the chloride level, and the pH if you can get them, and if you cannot, name what you do know: softened or not, well or municipal, near salt or not, pool or spa chemistry in the building, condensate rather than supply water. Chloride is the single value that changes the most decisions, because it is what breaks down the passive film on stainless.

Field 3, temperature and whether it cycles. Both halves matter. A steady 60 degrees F and a daily swing from 40 to 120 degrees F are different jobs even at the same average, because the swing drives condensation, drives expansion movement in every joint, and moves some couples across their reversal point.

Fields 4 and 5: what else the part touches

Field 4, contact materials. List everything that touches the part and stays wet against it. Three contacts do more damage than the water itself:

• Copper-based treated lumber. Modern preservative treatments are copper-bearing and are far more corrosive to fasteners than the older chemistry. Aluminum in direct contact with wet treated lumber is excluded by essentially every treated-wood supplier, and ordinary electroplated fasteners do not last. What is accepted is hot-dip galvanized to ASTM A153 or stainless. Confirm against the treatment supplier's own document, because the accepted list is treatment-specific.
• Fresh or wet concrete and mortar. These are strongly alkaline, and alkali attacks aluminum. Aluminum cast into or bearing on wet concrete corrodes and can expand enough to crack the concrete around it. Isolate with a coating or a barrier.
• Wet thermal insulation. Insulation that has taken on water holds it against the pipe and concentrates dissolved chlorides there. On austenitic stainless at temperature, that combination is the classic setup for chloride stress corrosion cracking, generally treated as a concern once metal temperature is above roughly 140 degrees F. Wet insulation is not a comfort problem, it is a corrosion mechanism.

Field 5, potable or process contact. If the material touches drinking water, it needs to be certified for that service under NSF/ANSI 61. This is a hard gate, not a preference, and it eliminates otherwise fine materials.

Fields 6 and 7: load and stress

Field 6, load. Note whether the part carries load and whether that load is steady, cyclic, or shock. Corrosion plus cyclic load is a different failure than corrosion alone: a part that would lose 10 percent of its section harmlessly can fail in fatigue from a corrosion pit acting as a crack starter.

Field 7, sustained tensile stress. Write yes or no. This is the field that separates "stainless is fine" from "stainless will crack." Austenitic stainless under sustained tension, in a chloride environment, above roughly 140 degrees F, is subject to stress corrosion cracking, and the failure is sudden and brittle with almost no visible corrosion beforehand. A stainless fastener torqued into a wet chloride joint on a hot line is exactly that combination. If field 7 is yes and field 2 says chlorides, the answer is not a higher grade of the same family without checking; take it to the fastener supplier.

Field 8: the fastener and coating call

Fasteners get their own field because they are small, they carry load, and a wet location eats them first.

• Name the grade, not the family. "Stainless" is not a specification. In fresh water and ordinary outdoor exposure, 304 is usually adequate. Where chlorides are present, from coastal air, road salt, softened water, or pool chemistry, 316 with its molybdenum content is the normal step up.
• Hot-dip galvanized is a coating with a thickness and a life, not a material property. Electroplated zinc is a thin coating and does not belong in a permanently wet location.
• The small part is the noble part. Whatever else the spec says, the fastener should not be the metal that is intended to corrode.
• Coat the noble half or both halves, never only the active half. Coating only the active member turns every pinhole in the coating into a concentrated pit.

Field 9: access, inspection, and the acceptance line

Where is it, can you see it, and when will somebody look? A wet-location material decision that assumes perfection is a bad decision, because the water always turns out to be worse than the survey said. Put the part where it can be inspected, and write the acceptance line now: what you will look at on the next visit, and what reading or observation means it is going wrong. A specification with no acceptance line produces no data, and in two years nobody can tell you whether the choice worked.

The filled-in spec: a rooftop unit stand

A packaged unit sits on a flat roof on treated timber sleepers over a membrane. A condensate line drips at the stand's downhill leg. The building is close enough to open water that salt shows on glass.

• Field 1, exposure class: wet-dry cycling, not immersion. Condensate drips whenever the unit runs, then the roof dries between cycles. Severity is set by the concentrating cycle, so this is graded as the harsh case, not the moderate one.
• Field 2, chemistry: condensate is low in dissolved solids on its own, but it washes airborne salt off the roof surface and carries it to the leg. Chlorides present, source is airborne rather than the water supply.
• Field 3, temperature: ambient, cycling widely with sun load. No hot metal, so field 7's chloride cracking threshold is not reached here.
• Field 4, contacts: copper-based treated sleepers, the roof membrane, and standing condensate. Aluminum is out at the sleeper contact. Standard electroplated fasteners are out.
• Field 5, potable: no.
• Field 6, load: steady dead load plus wind and compressor vibration. Cyclic component present.
• Field 7, sustained tension: yes at the hanger bolts, but at ambient temperature, so the chloride cracking gate does not close.
• Field 8, fasteners: 316 stainless at the sleeper contact and at the hanger bolts, chosen over 304 because of the airborne chloride in field 2, and over hot-dip galvanized because the drip zone will consume a coating. Isolating washers where the stainless meets the galvanized stand frame, and the coating goes on the stand frame, which is the noble half here relative to nothing, so both halves get coated at the contact.
• Field 9, acceptance: at each scheduled visit, look at the two legs in the drip zone and the two sleeper contact points under them. Going-wrong signals are white powder at a fastener head, a rust bloom running down from a bolt, or any fastener that turns by hand.

The time arithmetic that justifies field 9. This roof is already visited about twice a year for filter and coil work. Adding a two-minute look at the four contact points costs about 0.07 hours a year, and it rides on visits you are already billing rather than adding a trip. Replacing one corroded hanger later means supporting the unit, cutting a seized fastener, and remaking the connection, which runs roughly 1.5 hours plus the mobilization, and it is warranty or emergency time, not scheduled billable time. The inspection is about 20 times cheaper in hours than the repair, and the two are not even the same kind of hour, which is the real argument.

What would flip this spec: move the same job inland with no airborne salt and 304 replaces 316 at every position, because the only reason for the molybdenum was field 2. Put the sleepers on a membrane that is being replaced next year and the whole stand becomes a temporary installation where hot-dip galvanized is defensible. And if the condensate line is rerouted to a drain so the leg never gets wet, field 1 collapses to "high humidity" and most of the spec relaxes, which is worth pricing before you buy the stainless.

The two fields people fill in from habit

Field 1 gets answered "outdoor." That is a location, not an exposure class, and it hides the entire wet-dry question. If the answer to field 1 does not contain the words immersed, cycling, splash, condensation, or buried, it has not been answered.

Field 8 gets answered "stainless." Then a 304 fastener goes into a chloride environment because that is what the truck had, and it pits at the washer face inside two years. The grade is the field. Write the number.

References

• NSF/ANSI 61, Drinking Water System Components - Health Effects, for materials in potable water contact
• ASTM A153, Standard Specification for Zinc Coating (Hot-Dip) on Iron and Steel Hardware, for fasteners in contact with copper-based treated lumber
• 29 CFR 1910.333(b)(2) with NFPA 70E-2021, 120.5 for de-energizing and proving circuits dead before working in standing water near powered equipment
• Treated-lumber supplier documentation for approved fastener and flashing materials, which is treatment-specific
• See related: Galvanic Corrosion and the Metals That Fight; Reading Rust and Corrosion Patterns