One Fixture, Many Sizes: Fence Component Welding
Direct answer: One fixture covers a family of fence-type assemblies when its clamping points sit where every size in the family has material, when the clamp positions control distortion in thin sections rather than merely holding the part, and when torch access still works after the fixture tilts the assembly. Changing size should move locators, not the fixture body; if the body has to change, that is several fixtures sharing a base plate.
Who this is for: Production engineers running high-mix, low-volume fabrication of guard rails, fences, grilles and similar framed assemblies.
Scope: This EVST guide is written from a single filmed station: a fence-type frame clamped on a tilting fixture, with a robot welding the frame and the fixture turning the assembly to bring the opposite side into position. It is one station observed continuously, so nothing here states a line rate, a changeover time or an acceptance result for the filmed work.
What decides coverage in fence component robotic welding
Fence-type assemblies look similar and measure differently from batch to batch. Whether they can share a fixture is not settled by comparing outlines; it is settled by finding the positions every size in the family has in common.
In the reference footage the fixture holds the frame at the junctions between uprights and rails – positions that exist on every variant of that family. The more such shared positions, the wider the coverage.
If a clamping point only exists on one size, the fixture serves that size alone. Overlaying the drawings for the whole family shows very quickly which positions repeat and which appear once.
Clamp position controls distortion
Thin sections move as they cool. Where the clamps sit matters more than how hard they press: clamping at mid-span and clamping at a node produce different straightness on the same assembly.
The filmed fixture distributes several clamping units against the structural nodes of the frame, which is what keeps distortion inside a usable band once heat goes in. ISO 13920 provides the tolerance framework these targets are normally written against.
Clamping pressure proves a pneumatic state, not that the part is seated. Whether seating detection is required follows from the tolerance the family has to hold, and after any stop or human entry the seating and datum have to be reconfirmed rather than resumed by program line.
According to ISO 13920:1996, straightness, flatness and angular deviation on welded constructions are specified as tolerance classes, which is how a target for this family should be stated rather than as a single number.
| Question | What decides it | What to prepare |
|---|---|---|
| Can one fixture cover the family? | How many clamping positions exist on every size | An overlay of all drawings in the family |
| Will thin sections stay straight? | Clamp position relative to structural nodes | Section thickness and the straightness requirement |
| Does the torch still reach after the tilt? | Fixture frame and part geometry in the tilted attitude | Seam positions on the second side |
| Is changeover acceptable? | Whether only locators move, and how they are exchanged | Batch structure and changeover frequency |
Tilting brings the opposite fillet into position
The fixture turns the whole assembly over. That is not presentation; it brings the fillets on the other side into a position where they can be welded with the pool supported.
After the tilt, torch access has to be verified again. The same program faces a different spatial relationship before and after, and assuming symmetry is how interference gets discovered on the floor.
Load, offset and locking for the tilt mechanism are sized on the largest variant in the family, not on an average one.
What actually changes at changeover
The intended behaviour is that the fixture body stays put and only locators – and a few clamp positions – move. That keeps changeover predictable and spreads the fixture investment across the family.
If the body has to be reworked for each size, this is not one fixture covering many sizes; it is several fixtures sharing a base plate, with a very different cost profile.
How locators are exchanged – pins, quick-change plates or bolted blocks – is a design decision that decides whether changeover is measured in minutes or in hours, and it should be judged against the least experienced shift rather than a commissioning best case.
Size the family before sizing the equipment
List what has to be covered: dimensional range, section thickness, seam distribution, batch structure and changeover frequency. The fixture concept follows that list.
A common gap is validating only the extremes. Intermediate sizes are often the awkward ones, because they sit against neither the upper nor the lower stop.
In high-mix, low-volume work the generality of the fixture usually affects output more than the speed of the robot does.
According to EN 1090-2:2018, execution requirements for steel structures are tied to an execution class, so the tolerance a fence-type assembly has to hold follows from the class its application requires.
According to ISO 5817:2023, weld acceptance is stated as a quality level for each imperfection type, which is what a family acceptance criterion should reference.
What the footage cannot establish
Changeover duration, achievable cycle and how many sizes one fixture finally covers are not visible in the footage; they come from measurement on the floor against the family list.
Weld quality level is the same kind of statement: the footage shows equipment running, not that a family of assemblies has been accepted.
The material used here is a continuous observation of one station and does not represent the output of a line.
The order EVST works in is the same each time: count the shared clamping points, then check reach after the tilt, then measure changeover.
Frequently asked questions
How do I know whether one fixture can cover my range?
Overlay the drawings for the whole family and count the positions that carry material on every variant. Those are the candidate clamping points; the number of them, not the visual similarity of the parts, sets the coverage.
Does higher clamping force reduce distortion?
Not by itself. Position matters more: clamping at structural nodes restrains movement, while clamping at mid-span can even introduce it. Force only has to be enough to keep the part seated.
Why re-check reach after the fixture tilts?
Because the geometry the torch faces is different after the tilt. Access that works flat can be blocked by the fixture frame or by the part itself once the assembly is turned.
What makes changeover slow in practice?
Usually the locator exchange method rather than the program. Bolted blocks are slower than pins or quick-change plates, and the honest figure is the one a normal shift achieves, not the commissioning best case.
Project inputs for an application review
To assess fixture coverage across a size family, an EVST application review opens from these inputs:
- the size list: dimensional range, section thickness and material
- seam distribution and welding position requirements per size
- batch structure and changeover frequency
- straightness and dimensional acceptance requirements
Send the size list and the batch structure, and we will map the shared clamping points and tell you where the coverage of one fixture actually ends. Related reading: robot welding workstation, robot product range.