I'm a licensed plumbing contractor, and I've been handling Uponor PEX and radiant heating installations for eight years. In that time, I've personally made and documented 14 significant installation mistakes, totaling roughly $3,200 in wasted budget. This is the 6-step checklist I run before every Uponor compression fitting job — the one I wish someone had handed me back in 2017.
First, let's answer the question that probably brought you here.
Uponor PEX is cross-linked polyethylene tubing made with the Engel method — that's what the "a" in PEX-a means. The cross-linking happens during extrusion, which gives the pipe good cold-weather flexibility and, more importantly, expansion memory. That memory is basically the whole foundation of the proPEX connection system: you expand the pipe and the ring, slide the fitting in, and the pipe snaps back around the fitting like a fist gripping a pole.
Here's something vendors won't tell you: not all PEX is the same. PEX-b and PEX-c are cross-linked in different ways, and they don't have the same expansion behavior. If you pick up a roll of PEX-b thinking "it's all PEX anyway," the fitting system may not seal the way it was designed to. The pipe has to match the system. Period.
The expansion method is standardized under ASTM F1960 (cold-expansion fittings for PEX tubing). It's a proven approach when it's done right — and "done right" requires discipline. That's where my mistakes come in.
Here's what I do now on every job, no exceptions. It takes about 10 minutes, and it's saved us from more problems than I can count.
Look for the print line on the tubing. Uponor AquaPEX has "Uponor" and "PEX-a" printed right on the jacket. If the roll says PEX-b, PEX-c, or doesn't specify, stop before you load it on the truck. The wrong pipe with the right fitting is still the wrong assembly.
In my first year (2017), I ordered what I thought was a direct equivalent from a supply house running a deal. Saved about $0.40 a foot. The pipe looked identical. The expansion behavior was not. A homeowner called about standing water three weeks later — three joints had pulled apart overnight. The $180 I saved cost us about $1,100 in rework plus a damaged relationship.
That was my introduction to total cost of ownership. The unit price was lower. The total cost was way higher.
Honestly, everyone skips this. Expansion tools — even the cordless ones — have a recommended operating temperature range. A tool that's been sitting in a truck at 20 degrees moves slower, the pipe and ring don't expand fully, and the fitting seats partially. You can't see it from the outside, but it will show up in the pressure test or on a service call.
We keep a calibration log at the shop now. Every tool gets checked before each job, and someone initials it. Since we started in early 2024, that log has flagged 47 potential issues that would have turned into on-site failures. Five minutes of checking, a hundred times the savings in avoided rework.
Each expansion ring has its size stamped on it — 1/2", 3/4", 1", and so on. Check it against the pipe diameter every single time. Don't trust the label on the distributor's bin.
In September 2022, I was running a 38-zone radiant job when we realized a batch of rings from our distributor had been repackaged with the wrong stickers. If the crew hadn't been checking the size stamp on each ring, we'd have installed 70+ wrong rings before anyone caught it. The distributor was apologetic. The lesson stuck: trust the stamp, not the sticker.
And don't buy "universal" or "compatible" rings to save $0.30 per piece. The ring's expansion rate needs to match the pipe wall thickness and the fitting design. That's not the place to economize — calculate total cost, not unit price.
This sounds basic because it is basic. But on hour ten of an install, people get sloppy. Use a proper PEX cutter, not a utility knife. A square cut gives you even seating pressure on the fitting's sealing surface. Run your fingertip over the cut end — if you feel a burr, it can tear the O-ring as you slide the nut and sleeve over the pipe. A damaged O-ring doesn't drip right away. It waits until 2 AM when the system is pressurized.
The other classic: someone inserts the fitting, realizes the ring wasn't slid on first, and tries to force it over the seated end. I've watched crews do this. The ring has one job — it can't do it if it's in the wrong position. Ring goes on first. Every time.
This is the counterintuitive step, and it's the one most installers ignore. The depth gauge takes two seconds. Two seconds. It catches the fitting that looks seated and isn't.
In my second year, I had two fittings where I would have bet a paycheck the insertion was perfect. Both looked right. The depth gauge showed the pipe was short by about 1/8 inch. That's the difference between a seal that holds for 20 years and a leak that shows up in 6 months.
Trust me on this one: depth gauge on every fitting, no exceptions. My crew rolls their eyes, but they're the same crew that helped me fix my $3,200 mistake.
Every system gets a pressure test — 30 minutes at 1.5 times the operating pressure, with a reading every 10 minutes. The exact requirement varies by local code (the IRC and UPC have different framing, and some jurisdictions amend it), but the principle is universal: don't close a wall until the gauge holds steady.
My worst leak was behind a shower niche in a custom bath. The tile was finished, and the last thing I wanted to do was run a pressure test — the job felt done. The test caught a bad joint. We opened the wall and fixed it for $340. Without the test, that leak would've sat behind the finish work until the damage was five figures and the client was gone.
Let's put numbers on this. Say genuine Uponor compression fittings run about $4.20 each, and a "compatible" alternative is $3.30. On a 30-fitting job, you save $27 total.
Now, what does one failure cost?
That's $475 of risk against a $27 saving. And failures tend to surface on the one fitting nobody inspected. I calculate total cost of ownership on every job now — unit price plus time plus risk plus rework. The lowest quote on paper is rarely the cheapest in reality.
This one cost me the most, because I knew better. In 2019, I had a 46-fitting job and reused a few expansion rings that had been expanded but never seated. They looked fine.
The upside was $9 in rings. The risk was a joint failure inside a finished floor. I asked myself whether $9 was worth that risk. The answer I acted on was "they look fine." Two of the three rings failed during pressure testing. The testing caught them, but a two-hour job turned into a nine-hour day with the homeowner watching every move.
Expansion rings are one-use parts. The plastic relaxes when expanded; it never fully comes back. Budget them like the consumables they are.
That's the whole thing. It's not clever. Six steps, built on $3,200 of my own mistakes, and it works.
Share this article:
Leave a Comment