Apcemmert Metal Fabricators All articles
Quality Assurance & Procurement

Proof in the Workshop: What a Trial Assembly Must Demonstrate Before Finishing

Apcemmert Metal Fabricators
Proof in the Workshop: What a Trial Assembly Must Demonstrate Before Finishing

Photo: Leadingtopunion, CC0, via Wikimedia Commons

Multi-piece bespoke metalwork rarely fails because a single component is badly made. It fails because components that were each within tolerance do not agree with one another when brought together. A trial assembly, carried out before finishing, is the point at which that disagreement becomes visible while it is still a workshop matter rather than a site dispute. For UK trade buyers, the question is not whether trial assembly is good practice in principle, but when it is proportionate, how it should be controlled, and what evidence should survive it.

When a Dry-Fit Earns Its Place

Not every fabrication order needs a full trial assembly. A run of identical brackets or a simple single-piece frame gains little from being built twice. The case for a dry-fit strengthens as the number of interfaces grows and as the consequences of misfit move away from the workshop.

A trial assembly is usually justified where several of the following apply:

Stair and balustrade packages, architectural screens, complex support frames, machine guards and replacement plant items commonly meet several of these conditions. A dry-fit is not a substitute for dimensional inspection of individual components. It is a check on the relationship between them.

Define the Purpose Before Building the Set

A trial assembly without a stated purpose tends to become an expensive rehearsal. Before the first component is laid out, the parties should agree what the exercise is intended to prove. Typical objectives include confirming hole alignment at bolted connections, verifying overall envelope dimensions, checking that welded sub-assemblies do not clash, and establishing that interface points match the survey or the adjoining structure.

It is equally important to agree what the trial will not prove. A workshop dry-fit does not replicate foundation tolerances, site settlement, erection sequence or the flexibility of a frame under load. Nor does it confirm that a galvanised or coated assembly will still fit after surface treatment. Those limitations should be recorded so that the trial report is not read as a guarantee of site fit.

Where the client, principal contractor or designer requires witness attendance, that should be arranged at the outset. A trial that is inspected only after dismantling provides far less value than one observed at the point of fit-up.

Match-Marking as a Controlled Record

Match-marking is the mechanism that carries the result of the trial from the workshop to site. It should be planned with the same care as the fabrication drawings. Marks need to be unique, legible after finishing, and referenced on the erection drawings. Painted or stamped marks applied before galvanising may be obscured; a durable tagging or marking method should be agreed where the finish would otherwise erase the identification.

The marking scheme should identify not only the component but its position and orientation. Handed parts, asymmetric bracing and connections that will only bolt one way are common sources of error. A mark that says which piece it is but not which way round it goes leaves the site team to rediscover the problem that the trial was meant to solve.

During the dry-fit, shims, temporary packers and clamps should be recorded rather than left as undocumented adjustments. If a connection closes only with a packer, that is a finding. It may be acceptable, but it should be a agreed and drawn consequence, not a workshop habit that disappears when the assembly is dismantled.

Recording Results While the Assembly Is Together

Evidence gathered after dismantling is weaker than evidence gathered at the point of fit. The trial should be recorded as it happens: overall dimensions, diagonal checks, connection fit, gap measurements at interfaces, and photographs of each marked node. Where a measurement is taken against a survey benchmark or a setting-out datum, that reference should appear in the record.

Non-conformances identified during the trial should enter the same disposition route as any other workshop non-conformance. Some will be corrected by rework, some by remaking a component, and some by agreed adjustment at site. The important point is that each is closed out with a named action and a revised drawing or instruction where necessary. A trial assembly that generates a list of observations but no decisions has merely postponed the problem.

What the Trial-Assembly Report Should Contain

Before dispatch, the trade buyer should expect a report that stands on its own. As a minimum, it should include:

For UK projects, the report should also align with the inspection and test plan and, where applicable, the execution class documentation for the order. It is a quality record, not a marketing document. Its value lies in being sufficiently specific that someone who was not present can understand what was checked and what was concluded.

The Buyer’s Role Before Dispatch

A trial assembly is most effective when the buyer engages before it takes place. That means confirming the report format, agreeing witness points, and ensuring that any client-supplied interface information is current. It also means resisting the temptation to treat a successful dry-fit as permission to skip the final dimensional check after finishing. Coating and galvanising can change fit at close tolerances, and handling can introduce distortion.

The sensible position is that the trial assembly proves the geometry of the bare assembly; the post-finish inspection confirms that the finished articles still reflect it. Both records belong in the dispatch pack.

Conclusion

A trial assembly is not a ritual. It is a controlled experiment in fit, carried out at the stage when correction is still a workshop task. When its purpose is defined, its match-marking is disciplined and its results are recorded against the drawings, it gives trade buyers something more useful than reassurance: a documented basis for expecting the pieces to go together on site. In bespoke metalwork, that is a reasonable definition of proof.

All Articles

Related Articles

Left Is a Decision, Not a Default: Fixing Orientation on Handed Metalwork

Left Is a Decision, Not a Default: Fixing Orientation on Handed Metalwork

The Substitution File: Approving Steel Changes Before the Saw Starts

The Substitution File: Approving Steel Changes Before the Saw Starts

Close-Out by Agreement: Snagging Rules for Bespoke Metalwork Handover

Close-Out by Agreement: Snagging Rules for Bespoke Metalwork Handover