Every Perforated Core Has a Bald Stripe — And It Changes the Numbers

By Felicia November 5th, 2026 40 views
Every Perforated Core Has a Bald Stripe — And It Changes the Numbers

What the frame shows

Cores standing on end across a workshop floor: long large-bore ones at the back, short small-bore ones clustered in front, all of them the same component at different sizes. They are grouped by diameter and length rather than by order, which is a scheduling decision rather than housekeeping, and it says something about how the plant runs. But the detail worth looking at is smaller than that. On several of the cores you can see a plain vertical line running the whole length: a stripe of solid wall where the perforation stops.

That stripe is the subject here.


A core is flat strip before it is a tube

Perforation is a flat operation. A sheet lies on a bed, a tool punches or a beam cuts, and the pattern is laid out in two directions on a plane where everything is square and every pitch is measurable. The pattern runs in rows, it can be staggered or straight, and the flat sheet can be checked with a rule.

A tube is a bad place to make holes. To perforate a finished tube you either rotate the tube under a head or carry a head around the tube, and the pattern has to close on itself when it comes back to where it started. If the last row does not land exactly, you get a seam of narrower or wider material that nobody designed, or a doubled hole.

So the normal route is: perforate the flat strip, then roll it, then weld the longitudinal joint. Which means the holes are made before the part has a shape, and the shape is decided afterwards.


The seam is where the pattern has to stop

The longitudinal weld closes the tube. You cannot put holes in it. A hole in a weld zone is a leak path, a stress concentration, and an invitation to a crack that starts at the edge of the hole and runs into the seam.

So the pattern is laid out with a deliberate gap. Leave a band of solid strip where the joint will be, roll the strip so that band becomes a stripe running the length of the tube, weld inside it, and dress the weld. The result is a perforated tube with one bald line down its side.

None of this is a compromise somebody made on a bad day. It is the only way to have both a closed tube and a perforated wall.

The alternative route exists: perforate the finished tube with a laser, working around the circumference. It has its own version of the same problem, because the weld is still there and the beam still has to stop short of it. The stripe does not disappear; it just moves from being a layout decision to being a programming one.


The stripe is the strongest part of the tube

It is worth saying what the stripe does for the part, because it is usually described only as a loss.

A perforated wall is a wall with a large fraction of its section removed. It is weak in hoop, it dents under a thumb, and it will ovalise if you look at it wrong. The unperforated stripe is a continuous strip of full-thickness material running the entire length, and it behaves like a spine. It resists ovalising, it gives the tube something to hold its shape with while it is being handled and fitted, and it is the natural place to take a support or a weld to the outer shell.

That is why thin cores can be stacked and handled at all. Take the stripe away and the same wall thickness would be a sagging sleeve.


The open area a drawing specifies is not the open area the part has

Here is where the stripe stops being an interesting detail and starts changing numbers.

A perforation pattern is normally specified as a percentage: open area over total area. On a flat pattern that is easy to compute and easy to check. But the tube is not all pattern. The stripe is part of the circumference and none of it is open.

So the effective open area of the finished core is the pattern's ratio applied to the perforated fraction of the circumference, not to the whole of it. The shortfall is not a rounding error, and it scales the wrong way: the stripe width is set by the weld and does not change with tube size, while the circumference does. On a large-bore core the stripe is a modest slice of the perimeter. On a small-bore core it is a much larger share, so the same nominal pattern leaves a small core further from its stated open area than a large one.

This is the sort of thing that never appears on a drawing because the drawing shows the pattern, and the stripe is a consequence of how the tube is closed.


Which means the core is not symmetrical

A perforated core with a solid stripe is not the same in every direction. Point the stripe at the inlet and the sound meets a solid wall first. Rotate it a quarter turn and the same sound meets open holes.

Two cores from the same batch, same pattern, same length, fitted at different clock positions, are not acoustically identical. That difference is small next to the difference between two different patterns, but it is the kind of small difference that makes one finished muffler sound slightly unlike the next one, and it is entirely avoidable.

The fix is not a better tool. It is a specification: say where the stripe goes. If the drawing does not say, the answer is whatever the welder found convenient, and it will not be the same on Thursday as it was on Monday.


One tool, several sizes

The plant in the photograph is making one component across a range of diameters and lengths, which raises a question a buyer rarely asks: does the same perforating tool make all of them.

The pattern itself can be shared. Hole size and pitch are set by the tool, and the same pattern can be laid on a wider or narrower strip. What does not scale is the stripe, because the stripe is decided by the weld rather than by the pattern.

So a change of diameter is not a change of perforation, it is a change in how much of the circumference the perforation gets to use. The same tool, run honestly, produces cores whose effective open area drifts with size. That is not a fault in the tool; it is a property of the geometry, and it is why the percentage on a drawing should be read as a starting point rather than a promise.


What you can check in a stack of cores

You cannot count holes. Nobody counts holes, and a supplier who claims to have counted them is describing a sample.

What can be checked quickly, on any core in the stack:

  • Pitch over a known span. Ten pitches measured with a rule catches a tool that has drifted far faster than counting would.
  • Rows across the stripe. The row count either side of the weld should match the drawing, and the stripe width should be consistent from core to core.
  • The stripe itself. Straight, uniform width, and the weld sitting inside it rather than beside it.
  • Dents and ovality at the ends. Thin perforated tube is weakest where it has been cut, and a stack that has been laid down and walked past will show flats at the open ends.
  • The pattern stopping short of the weld. A hole that touches the seam is a future leak, and it is visible from outside.

Standing the cores on end, as they are here, is the right way to store them. Laid in a pile they take a set from their own weight, and a core with a flat is a core that will not sit concentric in its shell.

Why repetition is the real product

A single good core proves that the pattern can be made once. A stack of hundreds proves something harder, which is that the same pattern can be made on a Tuesday and again in three weeks, at a different diameter, by a different operator, and still land in the same place relative to the seam.

The acoustic design of a muffler is decided on a drawing and in a test cell. Whether the five hundredth unit sounds like the first is decided on a workshop floor, by a stripe nobody specified and a clock position nobody wrote down.


FUPOWER has been building exhaust components in Shanghai since 2009, across four facilities, with pressing, forming, perforation, mandrel bending, TIG welding, casting, CNC machining and silicone hose production under one roof. If a part has to be repeatable rather than merely good, that is the conversation worth having.


Key Features

🪡 A bald stripe in every core — the longitudinal weld cannot be perforated, so a band of solid wall runs the full length by necessity.

📄 Flat first, tube second — holes are made on a sheet where the pattern is square and measurable, then the strip is rolled and closed.

🔒 The seam sets the limit — a hole touching the weld is a leak path and a crack starter, so the pattern stops short of it deliberately.

💪 The stripe is a spine — full-thickness material down the length is what stops a thin perforated wall from ovalising in handling.

📐 Open area is not one number — the ratio belongs to the pattern, the stripe belongs to the circumference, and the part only gets the second applied to the first.

📉 Small cores lose more — stripe width is fixed by the weld while circumference shrinks, so the same pattern drifts further on a small bore.

🧭 A core is not symmetrical — clock position relative to the inlet changes which wall the sound meets first.

🕒 Two identical cores can differ — same batch, same pattern, different stripe orientation, subtly different result.

🔧 One tool, a range of sizes — the pattern is shared, the proportion it covers is not.

📏 Check pitch, not hole count — ten pitches with a rule finds a drifting tool faster than counting ever would.

🧱 Stored on end for a reason — a core laid in a pile takes a set from its own weight and will not sit concentric.

🏭 Repetition is the product — the five hundredth unit matching the first is a floor discipline, not a design feature.

The Straight Edge Is the Expensive Feature
Previous
The Straight Edge Is the Expensive Feature
Read More
Fupower Exhaust S-Bend TIG Welding | Mandrel Bent Pipe for Smooth Exhaust Flow
Next
Fupower Exhaust S-Bend TIG Welding | Mandrel Bent Pipe for Smooth Exhaust Flow
Read More