It was a Tuesday in late February 2023. I remember because we had just wrapped our Q1 quality audit, and the report was sitting on my desk—mostly green, a few yellows. Nothing alarming. Then the warehouse manager called my extension. “You need to come see something.”
What I saw was a pallet of thirty-two Burnham baseboard radiators—the Model 67B, standard contractor-grade, nothing exotic. On paper, they matched our spec sheet perfectly: length, fin spacing, copper-aluminum construction. I even pulled the tape measure myself. They were sized exactly as ordered.
But here’s the thing: they were the wrong model series.
The Trigger Event: A $22,000 Mismatch
The specification called for the 67B series with a ¾-inch supply. What arrived was the 67B—same face, same mounting pattern—but with a ½-inch supply. That half-inch difference in the fitting meant they wouldn’t connect to the pre-installed manifold system on a commercial retrofit project we were running. The installers had already laid out the pex loops. The manifolds were mounted. The job was scheduled for the next Monday.
I’ll be honest: I didn’t catch it at first either. I measured length, height, fin count. It wasn’t until one of the project managers, a grizzled guy named Tom, came over and pointed at the inlet and said, “That’s half. We spec’d three-quarter.”
We had a decision to make: cut the manifolds and re-pipe for ½-inch, which would create a bottleneck and invite future complaints about low BTU output, or send everything back and order the right ones. The vendor claimed the ½-inch was “within industry standard for this application.” I rejected that on the spot. Not because I’m a hardass—I’m actually fairly reasonable about field adjustments—but because the owner of the building was a detail-oriented facilities director who would catch the undersized connections during a walkthrough. We would own that deviation for the next twenty years.
We rejected the batch. The vendor redid the order at their cost, but we ate the delay: two weeks of idle labor, re-tooling of the manifold schedule, and a $22,000 redo to the piping layout we’d already built. That was the cost of one wrong fitting.
The Real Problem: What That Spec Didn't Say
The vendor failure in March 2023 changed how I think about spec writing. Before that incident, I treated specifications as a list of dimensions and material grades. After, I realized they are a contract about intent.
The spec we sent the vendor said: “Burnham Model 67B baseboard radiator. ¾-in supply.” Simple, right? But the vendor’s system interpreted the 67B as a series number, and their default supply size—unless explicitly overridden in the line-item notes—was ½-inch. Our order went in, their automated pricing tool read “67B,” matched it to the standard warehouse stock, and thirty-two units hit the shipping dock. The ¾-inch spec was buried in a field that only triggered a manual check if the price exceeded a threshold.
Here’s the part that still bothers me: we had a verification protocol implemented in 2022 that required a second person to review every purchase order over $5,000. But the review was about price, not technical alignment. Nobody was looking at the spec detail column. We were checking for budget overruns, not installation compatibility.
I should add that this was a reputable vendor. We’d used them for years. They had a dedicated customer portal and a good track record on Burnham parts. The mistake was entirely in the handoff between our spec language and their system default. In my opinion, the extra cost is justified if it means including a spec review step that compares the order line item against the detailed installation blueprint—not just the model number.
The Process After: What We Changed
In Q2 2023, I updated our verification protocol. Now every order that involves a Burnham component—boiler, baseboard, indirect water heater—goes through a three-point check:
1. Model-to-blueprint match. The person who creates the purchase order has to pull the specific sheet from the project drawing and physically check the supply size, mounting configuration, and clearance specs against the vendor listing. Not the model number alone. The fitting details.
2. Default trap review. We added a line in our PO template that says: “Is the vendor default size correct for this application? If yes, confirm. If no, override and flag.” It sounds obvious. It wasn’t obvious before.
3. Pallet-level inspection on arrival. Previously, we spot-checked one unit per pallet. Now we check every unit until we confirm consistent batch quality. In 2024, I rejected 12% of first deliveries across all vendors. That’s down from 18% in 2023, because we caught more issues earlier.
We also added a clause to every vendor contract: “Vendor must confirm spec deviations in writing within 24 hours if auto-populated defaults differ from PO line details.” That’s saved us two potential redo situations since.
The Broader Lesson: Assumptions Are Expensive
People think expensive vendors deliver better quality. Actually, vendors who deliver quality can charge more. The causation runs the other way. The price reflects the reliability of the system, not the physical product.
This was true five years ago when digital ordering systems were less sophisticated. Today, the opposite is often the case: automated systems inherit defaults that might have been correct for a different project. The same vendor portal that speeds up checkout creates a blind spot for non-standard specs.
Switching to a structured spec review process cut our project turnaround from 5 days to 2 days for routine orders, oddly enough. The extra upfront time eliminated the redo cycle that used to eat the savings. Reducing manual data entry errors also eliminated the discrepancy we used to have between the PO and the packing slip.
I also think about the Burnham Alta combi boiler we spec’d for a different job around the same time. That unit is more complex—condensing technology, multiple heat exchanger circuits—and we treated the spec writing process with a lot more caution because it was unfamiliar. That job went smoothly. The baseboard job, which seemed routine, was the one that bit us. Familiarity made us sloppy.
The assumption is that rush orders cost more because they’re harder. The reality is they cost more because they’re unpredictable and disrupt planned workflows. Our $22,000 redo wasn’t a rush job. It was a standard lead-time order with a default error.
How to Avoid This Kind of Problem
If you’re specifying Burnham equipment—baseboard radiators, an Alta combi boiler, an indirect water heater—do yourself a favor and look at the fitting details before you hit “order.” The model number is the start, not the end. Every manufacturer has defaults that may not match your application.
For existing systems, especially retrofits, verify the existing piping size. I see a lot of contractors assume a ¾-inch manifold system can take any Burnham baseboard radiator. It can't unless you check the supply port on the unit.
And if you’re cleaning a condenser or doing maintenance on a hydronic system, check the manufacturer literature for the specific year of installation. Burnham has changed fittings and control board layouts over the years. The manual you have from 2018 might not match the 2023 unit.
I still use Burnham equipment. I still think it’s a solid, reliable brand for both residential and commercial hydronic heating. That was never the issue. The issue was our process, and the assumption that a standard spec would be interpreted the way we intended. The cost of that assumption was $22,000 and a delayed project launch.
In the end, the installation went in. The radiators arrived, were checked individually, and matched the blueprint. The building heated properly. The facilities director never knew about the redo. But I do, and I made sure the spec template changed so the next person doesn’t have to learn this lesson the same way.
