Vlog 05: The Pipe Bending Production Line — What 1D, 1.5D and a Size Range Actually Mean
What the line actually is
Frank walks the exhaust factory floor and stops at the bending section. Four CNC mandrel benders, sitting in a row, each loaded with a different diameter of tube. Two workers move between them through a twelve-hour shift. The machines run every day.
That description contains three separate claims, and they are not the same kind of claim. Four machines is a capacity statement. One inch to three inch is a tooling statement. 1D and 1.5D is a capability statement, and it is the one that decides whether your part is routine or difficult.
The video also shows the output rather than the input: the parts finished that day, stacked together, in different sizes. That is a more honest thing to film than a single perfect bend, because a day's production is a run, and a run is where consistency either exists or does not.
The number that decides everything is the radius ratio
A bend radius on its own tells you almost nothing. Three inches of radius is generous on one-inch tube and severe on three-inch tube. What matters is the radius expressed as a multiple of the outside diameter, which is the D-ratio.
At 1D, the centreline radius equals the tube diameter. A three-inch pipe bent to 1D has a centreline radius of three inches. At 1.5D, the same pipe is bent to a radius of four and a half inches.
That single number governs how hard the bend is, because it sets how much the material has to deform. The inside of the bend is compressed into a shorter arc; the outside is stretched over a longer one. Tighten the ratio and you compress and stretch both harder, in a shorter distance, with less room for the tooling to control what happens.
This is why "we do 1D and 1.5D" is a more useful sentence than "we bend pipe". The first is a specification. The second is a category.
Why 1D is the harder bend
At 1D the inner arc is very short. The material there has to compress into much less length than it started with, and compressed metal with nowhere to go does one of two things: it thickens evenly, or it folds. Even thickening requires the compression to be distributed along the arc rather than concentrated at one point, and at 1D there is very little arc to distribute it along.
The outside has the mirror problem. It is being stretched over a long arc, which thins the wall, and the thinner that wall gets the less margin remains for the welding and the pressure the part was designed for.
Meanwhile the mandrel, whose job is to hold the bore round from the inside, has less space to work in. A tighter bend means the mandrel nose is doing its most difficult job in the most confined geometry.
None of that makes 1D impossible. It makes it the bend that separates a shop with matched tooling from a shop with a bender. Many exhaust systems are drawn around 1.5D for exactly this reason: not because 1.5D is better, but because 1.5D is where a competent line reliably holds quality, and 1D is where it has to be set up deliberately.
A hundred and eighty degrees in one sweep
The part shown in the video is a 180-degree bend, and a U-bend is the geometry that tests a bending line most directly.
To make one in a single sweep, the mandrel has to be supported through the entire arc and then withdrawn along a bore that is now curved, while the clamp holds the tube at a point that is not where it started. There has to be enough straight length ahead of the bend for the clamp to grip and behind it for the mandrel to travel, which is why a drawing can be geometrically correct and still not be bendable as drawn.
The alternative is two 90-degree bends welded together at the middle. That is what a shop does when it cannot make the sweep, and it leaves three things behind: a weld sitting inside a formed region, a heat-affected zone in metal that was already worked, and a joint whose internal mismatch sits directly in the flow path. A single continuous 180 has no joint to align and nothing to weld in the bend.
So the interesting claim in the video is not that the U-bend is smooth. It is that it is one bend.
The size ladder, and where tooling runs out
The line covers one inch, two inch, two and a half inch and three inch.
It is worth understanding what adding a size actually costs, because it explains why small shops stop where they do. Bending tooling is per diameter, not per machine: a mandrel sized to the bore, a bend die cut to the radius, a clamp die, a pressure die, and usually a wiper die for tight work. Moving from 2.5 inch to 3 inch is not a setting change. It is another set of tooling, and a larger one, on a machine that has to be stiff enough to push it.
That is the real content of a size range. A shop that lists four diameters has bought four families of tooling and learned four sets of programs. A shop that lists a range above what its tooling covers is either farming the work out or welding bends from straight sections.
What "full-size mandrel" means, and why to ask
The video says the line supports full-size mandrel bending at its maximum dimension, and that phrase is worth translating because it is a real distinction rather than a marketing one.
A mandrel is a shaped tool drawn through the bore ahead of the bend, and its length relative to the bend matters. A short or partial mandrel supports part of the sweep; a full-length mandrel supports the whole of it. The difference shows up exactly where you would expect: at the tightest ratio and the largest diameter, which is where the section most wants to flatten.
The question to ask is concrete. At three inch, is the mandrel full length for the bend being quoted, and is a wiper die used at 1D? A line that answers both without hesitating has tooling matched to its size range. A line that answers vaguely has a bender and a catalogue.
Four machines is a capacity claim, not a quality claim
It would be easy to read four benders as four times the quality. It is not. Four machines do not bend better than one good machine. They do something else.
Each machine holds a setup. A bend die at a given radius for a given diameter, a mandrel, a program, and a set of leg lengths. Changing that over is the slow part of the job, so four machines mean four setups can be live at once. A production run can continue on one machine while the next job is being set on another, and a mixed order — two and a half inch one week, three inch the next — does not require tearing the line down between them.
That is what the number buys: parallel setups and uninterrupted runs, which is a scheduling property. It becomes a quality property only indirectly, through the fact that a job which does not have to be rushed between changeovers tends to be set up properly.
Setup is the bottleneck, not the bend
A mandrel bend takes seconds. Everything around it takes longer: loading the tube to the right depth, confirming the program matches the drawing, running a first article, measuring the radius and the leg lengths, correcting the compensation, and then running the batch.
This is why "our machines run every day" is a statement about scheduling rather than stamina. A line running daily is a line whose setups are known, whose programs exist, and whose first article is a confirmation rather than an experiment. A line that runs occasionally spends its first hour rediscovering things it should already know.
It also explains the two-workers-to-four-machines ratio honestly. Two people cannot hand-feed four machines at once, and nobody is claiming they do. The machines cycle automatically; the people load, unload, inspect and set. A shop that needed four operators for four benders would be describing a different kind of production entirely.
Three metals, three different springs
The video offers stainless steel, aluminium, titanium, and carbon fibre among the materials the group works in. On a bending line, the first three behave like three different problems, and the reason is springback.
Every bent tube gives part of the angle back when the tool releases. Austenitic stainless springs back noticeably. Aluminium, being much less stiff, springs back more for the same geometry. Titanium springs back significantly and, worse for tooling, tends to gall and seize against surfaces it is dragged across. The same program does not transfer between them; each material gets its own compensation, measured rather than assumed.
That is why material breadth in a bending line is a tooling and programming claim rather than a marketing list. Three metals means three sets of springback data, three sets of lubricant decisions and three sets of wear patterns on the same dies.
Carbon fibre is a separate matter and the video does not claim it is mandrel-bent. Carbon components are produced in-house through the group's carbon facility, which is a different process. Being clear about that distinction is more useful than blurring it.
What the end of a shift actually proves
The last shot is the day's finished pipes, mixed sizes on the floor.
A stack like that is evidence of a run, not a sample. It shows that several diameters were produced in the same working period, on the same line, by the same two people — which is the situation your order will actually be made in. A single perfect part photographed on a bench proves that the shop can make one perfect part. A day's mixed output proves the line can hold several jobs at once without losing any of them.
The useful version of this when you are evaluating a supplier is to ask for parts from the same setup rather than the best part of the week. Consistency is a property of a run, and only a run can demonstrate it.
The most popular sizes tell you what the market builds
Two and a half inch and three inch are the volumes on this line, and that is not arbitrary.
Those are the diameters that performance exhaust work converges on: two and a half inch covers a very large share of cat-back and midpipe work, and three inch is where higher-flow systems, downpipes and turbo-back builds sit. A bending line whose volume is concentrated there is a line tooled for the work its customers actually order, rather than for the sizes that happen to be easy.
It also tells you what to expect on lead time. The popular sizes are popular because they run often; the uncommon ones wait for a setup.
What to send with a drawing
If you are sending a project, four things decide whether it can be bent well, and none of them is the material grade.
The radius ratio, stated as a multiple of the diameter, not as a bare radius. The leg lengths and the straight between bends, because the clamp needs somewhere to hold and the mandrel needs somewhere to travel. The wall thickness, because at a tight ratio the outside of the bend is where the margin lives. And where the joints will be, because a weld placed inside a bend is a decision you want to make deliberately rather than discover.
A drawing that carries those four is a drawing a bending line can quote honestly. A drawing that carries only a shape is one that gets a price and then a conversation.
Four factories, five materials
FUPOWER runs four owned workshops — exhaust, silicone hose, carbon fibre, and pre-filter — and between them they cover five material families: titanium, stainless steel, aluminium, carbon fibre and silicone. Established in 2009, the group keeps 3D scanning, 3D printing, CNC machining, laser cutting, mandrel bending and TIG welding inside the same chain of accountability, which means a bend that does not match its fixture is one conversation rather than three suppliers pointing at each other.
That breadth is unusual. Most exhaust shops bend one metal and buy the rest in, because each material carries its own tooling, its own springback behaviour and its own failure modes. Holding five at once is not a slogan about size; it is a statement about how many sets of tooling and how many sets of process knowledge sit in the same building. It is why we can claim, without overstating it, to be among the few factories that handle this range under one roof.
The video's claim is that the bends are smooth, that the parts are good, and that the range of materials is unusual. The first two are things you can check with the part in your hand. The third is checkable too, but the check is different: ask what tooling exists at the largest diameter, ask what the springback compensation is for each metal, and ask how many setups can run at once. A line that answers those three is describing itself accurately.
Key Features
⚙️ Four CNC mandrel benders on one line — four setups live at once, so a run continues while the next job is being set.
🌀 Bend radii from 1D to 1.5D — the tightest ratio, where a bend stops being routine and tooling has to be matched to the job.
📏 Tube from 1 inch to 3 inch — including the sizes performance exhaust work actually orders, with 2.5 inch and 3 inch carrying the volume.
🔄 180-degree bends in a single sweep — a continuous U-bend with no weld sitting inside a formed region and no joint to align.
🧰 Full-size mandrel support at the maximum diameter — the bore is held round through the whole sweep, not just part of it.
🏭 Four owned facilities, five material families — exhaust, silicone, carbon fibre and pre-filter production under one roof.
🔥 Stainless, aluminium and titanium, each on its own program — three different springbacks, three sets of compensation data, measured rather than assumed.
🗓️ Running every day — setups are known and programs exist, so a first article is a confirmation, not an experiment.
👥 Two operators across a twelve-hour day — the machines cycle automatically; the people load, unload, inspect and set.
📦 Mixed-size output in a single shift — evidence of a run rather than a sample, which is the condition your order will be made in.
🧭 Radius stated as a ratio, not a bare number — so a drawing can be judged bendable before anyone quotes it.
🔍 Three questions that separate a line from a bender — what tooling exists at the largest size, what the springback figure is per metal, and how many setups run at once.