Commercial
Every TPO roof lives or dies at its seams. Here is what a proper heat-weld actually is, why it beats glue, and how you can tell a good one from a gamble.
Why are welded seams important on a TPO roof?
On a TPO roof, adjacent membrane sheets are fused with a hot-air welder that melts the surfaces together into one continuous piece, rather than relying on adhesive that can peel over time. A correct weld is as strong as the membrane itself and is the single most important factor in whether the roof leaks. Poorly welded seams are the most common cause of TPO failure.
When we weld TPO, a robotic or hand welder blows hot air, typically well over 1,000 degrees Fahrenheit, into the gap between two overlapping sheets while a pressure roller presses them together right behind the heat. The top surface of the lower sheet and the bottom surface of the upper sheet melt and flow into each other. When they cool, they are one material. There is no separate substance holding them; the membrane has become continuous.
That is fundamentally different from a glued or taped lap, where a third material sits between two sheets and has to keep bonding for decades through heat cycling, UV, and moisture. Adhesives age. A proper thermoplastic weld does not have a separate bond line to fail; it simply is the membrane.
Coastal roofs get hit with two things that punish weak seams: persistent marine-layer moisture and daily thermal cycling as the roof heats up and cools down. A glued lap that is fine in a dry, stable climate can fatigue and open up under that constant expansion and contraction, and once a seam opens even slightly, wind-driven rain from a Pacific storm drives water straight in.
A fully fused weld handles that cycling as one piece, because there is no interface to work loose. This is a big part of why we lean on thermoplastic membranes for flat roofs from Oceanside down through Del Mar. The welded seam is the feature that answers the coastal moisture problem most directly.
A good weld is a Goldilocks problem. Too little heat or too fast a pass and the sheets do not fully fuse, leaving a cold weld that looks fine but peels apart. Too much heat or too slow and you scorch and thin the membrane, weakening it. The welder's temperature, speed, and the pressure of the follow roller all have to be dialed in for the day's conditions.
Conditions matter because a cold, damp coastal morning welds differently than a hot dry afternoon in Escondido. A crew that knows what it is doing runs test welds at the start of the day, adjusts the machine to the actual membrane and weather, and confirms the settings before they weld a single production seam.
Good crews do not trust their eyes alone. The standard field check is a probe test: once a seam has cooled, a rounded metal probe is dragged firmly along the edge of the weld. If the probe catches and opens the seam anywhere, that spot was not fully welded and gets re-welded on the spot. It is simple, cheap, and it catches problems before they become leaks.
For a higher level of assurance, destructive test cuts can be taken and pulled apart to confirm the weld tears through the membrane rather than separating at the seam, which is what you want to see. We treat seam probing as a non-negotiable part of the job, not an optional extra, because the seam is where a TPO roof actually fails.
When we get called to an aging TPO roof that is leaking, the seams are the first place we look and usually where we find the answer. Fishmouths, which are little open puckers along a lap, are a classic sign of a rushed or cold weld. Edges that have lifted or peeled tell the same story. Dirt streaks tracking away from a seam often mark where water has been finding its way under the lap.
The good news is that thermoplastic seams are repairable. Because the material re-melts, we can clean, re-heat, and re-weld a failing seam, or weld a cover strip over it, and restore a continuous surface. That repairability is only possible because of the welded-seam design in the first place.
We were brought onto an Oceanside warehouse roof that was only a handful of years old but leaking in several bays. The membrane itself was in good shape and plenty thick. When we probed the seams, the long straight welds held, but nearly every T-joint, where the corner of one sheet crosses two others, opened under the probe.
The original crew had welded the running seams with a machine but rushed the T-joints by hand without patching them, and those three-layer intersections had never fully fused. Water was wicking in exactly at those points. We re-welded and added target patches over every T-joint, then probed the whole roof.
The roof stopped leaking and did not need a new membrane, only competent seam work. It is a clean example of the core lesson: on TPO, the membrane can be perfect and the roof can still fail if the seams, especially the fiddly intersections, are not welded and tested properly.
Ask us during a free inspection. Straight answers, no pressure.