I'll never forget the look on the inspector's face. It was March 14, 2023—a Tuesday—and we were staring at a 6-inch weld seam that had just failed pressure testing on our chilled water line. Third rejection of the week. The project manager's phone wouldn't stop ringing. And I was the one who'd insisted we weld every joint "the right way," just like my old supervisor taught me.
I've been in this maintenance role for eight years now. In that time, I've personally made—and documented—12 significant mistakes totaling roughly $18,000 in wasted budget. This particular one stung the worst because the answer had been sitting on my desk the whole time.
How I Ended Up With a Welding Torch
Let me back up. I handle maintenance operations for a mid-size commercial facility—three buildings, about 210,000 square feet of office and lab space. The chilled water loop is older than I am, and in early 2023 the valves on the secondary loop were finally giving up. One had seized completely. Another had a hairline crack running along the body that we caught during a routine inspection.
I submitted the replacement specs in February 2023. NIBCO high performance butterfly valves for the main isolation points, plus genuine NIBCO valve parts for the branch lines—seats, stems, packing kits. I hadn't worked with NIBCO at this scale before, but their catalog was the only one where I could cross-reference dimensions and pressure ratings without improvising. The order came to about $3,200 in materials.
That wasn't the problem. The problem was how I chose to install them.
I come from the generation where you welded or you didn't. My first supervisor could run a bead that looked like a row of dimes, and he drilled it into me that a real installation gets permanently jointed. I didn't question it. Back in 2017, we did an entire boiler room in welded steel. Took eight days. Everyone agreed it looked great.
So when the NIBCO valve parts arrived, I set up the oxy-acetylene rig and got to work. That was mistake number one.
The Sound of a Seat Scraping
Welding on a valve body is not the same as welding on pipe. The heat distorts the sealing surfaces. We didn't realize it until the first pressure test, when the butterfly valve on the 6-inch line wouldn't seat properly. The disc was binding. I could hear it—that dry, grinding sound of metal not meant to touch, scraping anyway.
We pulled the branch valve apart and found discoloration on the sealing surfaces. Classic heat damage. $780 of NIBCO valve parts gone in one afternoon. The most frustrating part? The spec sheet clearly stated the maximum installation temperature and recommended joining methods. I just hadn't read past page two until it was too late.
The Question That Changed Everything
That evening, I was on the phone with a supplier friend, venting about the week. I told him I needed to get better at welding before we tried again.
He asked a question that stopped me cold: "Can you weld without a welding machine?"
I thought he was kidding.
He wasn't. He explained that for most valve installations in our pipe size range, you don't need to weld at all. You press. NIBCO press fittings use a cold mechanical connection—no flame, no hot work permit, no risk of cooking a $400 valve body. The joint seals through radial compression. If you use the right tool and the right jaws, the connection is just as reliable as a weld.
I had no idea. Honestly, I'd been so locked into the "welding is the only real way" mindset that I'd never looked at press technology as a serious option. I figured it was for residential PEX work, not commercial valves.
The Pivot
The next morning, I called our rep and ordered a press tool. It wasn't cheap—the invoice was around $1,400 with the jaws and extra batteries, though I might be misremembering the exact figure—but that's still less than the materials we'd already destroyed.
The rest of the install went completely different.
Instead of dragging the oxy-acetylene tanks across the building, threading hoses, setting up fire watch, and filing hot work permits, we cut the pipe with a standard cutter, deburred it, marked the insertion depth, and pressed each connection in about ten seconds. The tool engages, the jaws close, and you get a visible indicator that the joint is complete.
We finished the 6-inch line, the branch lines, and the bypass assemblies in two days. No fire watch. No burnt fingers. No damaged valve seats. The inspection passed on the first attempt.
Rethinking the Toolkit
That experience forced me to rethink our entire valve maintenance setup. For years, our go-to tools were whatever was lying around the truck. Now I keep a dedicated drawer:
- A NIBCO tools hex keys set—sizes from 1/16" to 3/8"—for packing gland adjustments and set screws
- A pliers wrench XL for the bigger union nuts and flange bolts
- A torque wrench for reassembly
- Press jaw sets for the connection sizes we use most
- Spare seats and stem seals for the valve models we maintain
The hex keys seem like a small thing, but they're the difference between doing a quick packing adjustment on-site and pulling the whole valve out to bench it. And the pliers wrench XL gets used more than I expected—especially on the 3-inch and 4-inch flanges where a standard crescent wrench doesn't give enough leverage without chewing up the corners.
What I Got Wrong
Now, before you think I'm going to tell you to throw away your welding setup—I'm not. Welding still has a place. We still weld the big carbon steel headers. If you're joining large-diameter steel pipe with high pressure and high temperature, welding is still going to be a mainstay for a long time.
But for valve connections in the 1/2-inch to 4-inch range—especially with brass or copper-bodied valves—press is a better choice in most cases. Maybe all cases, if the spec allows it.
What I got wrong was treating every joint like the same problem. A 12-inch carbon steel steam line and a 1-inch brass valve on a chilled water loop are completely different animals. I was applying the same rule to all of them.
That's the embarrassment of it. The NIBCO datasheets had been right in front of me. They clearly state the temperature and pressure limits, and they clearly indicate when welding is not recommended on valve bodies. I just never read them closely because I assumed I already knew the answer.
The Numbers
For anyone tracking costs: that failed first week cost us about $3,200 in damaged materials and rework. We also lost a week of schedule, which meant the lab wing ran without proper cooling for longer than anyone was happy about.
After the pivot, the remaining install ran about $6,800 including the press tool purchase. Whole project wrapped by the end of March. If I remember correctly, the final job took 11 working days total, compared to the 19 we'd projected if we'd kept welding.
We've caught at least 5 similar issues on system walkdowns since then, using the updated checklist I added after this project. The first question on it now: "Does the valve manufacturer's spec allow for this joining method?"
If You're Starting a Valve Replacement Project
Here's the condensed version of what I'd tell my younger self:
- Check the manufacturer's spec for the recommended joining method before you order anything.
- Ask whether the project actually calls for NIBCO high performance butterfly valves, or if a standard lug-style valve would do the job. The high-performance line is worth the cost on larger loops—the stainless steel disc and shaft don't seize the way cheaper valves can—but don't overspec if you don't need it.
- Use genuine NIBCO valve parts for repairs. The packing kits and seat rings are dimensionally consistent. One time a generic stem seal was slightly thicker than the OEM part and it caused the valve to bind.
So, Can You Weld Without a Welding Machine?
The practical answer for most valve installations: you don't need to. Pressing is faster, cleaner, and safer. If you're working with NIBCO valves and fittings in the 1/2" to 4" range, the press system handles the connection without any hot work. You just need the right jaws for the pipe size.
But I want to be honest about the boundary of that advice. This worked for us because our chilled water loop runs around 150 psi and the pipe is copper and brass. If you're on a steam system, or your line operates at higher pressures and temperatures, the calculus might be different—check the ratings and follow the manufacturer's requirements. I can only speak to what we learned in our facility. If your jurisdiction requires welded joints for code compliance, that's a different situation entirely.
Final Thoughts
I used to wear "I can weld it" as a badge of honor. Now I ask a better question: "What's the best way to join this so it lasts 30 years?"
The best way isn't the hardest way. It's the one that's repeatable, reliable, and doesn't put the valve components at risk. For us, that meant switching to press connections for valve installation and saving the welding rig for the jobs where it actually belongs.
Our team's checklist has a line item now that I never thought I'd write: "Verify whether welding has been specified—then ask yourself if it's really necessary."
You don't have to learn this one the expensive way. Ask the press question before you light the torch.
