Made to specification - working with a components maker on custom parts

It starts with the drawing - and the application behind it

A good custom part begins with a clear drawing, but the drawing alone is rarely enough. We want to know what the part does, because the application decides things the drawing may not say. A brass insert that gets moulded into plastic has different demands from one that gets pressed into metal. A fitting that carries water needs an alloy that resists dezincification; one that carries current needs a different priority again. When we understand the job, we can often suggest a better way - a small change to a feature that makes the part cheaper to machine without hurting how it works, or a finish better suited to where it will live.

This is also the stage to separate the dimensions that are critical from the ones that are not. Almost every drawing has a few features that truly matter - a thread, a seat, an outside diameter that has to fit something else - and many that have generous room. Telling the maker which is which is one of the most valuable things a designer can do, because it lets us put the tight control where it counts and not waste cost chasing tolerances that do not affect the part. A maker who works to your drawing without ever asking what it is for is not really engaging with your part.

The first sample - proving the part on paper is the part in hand

Once the drawing is agreed, we make a first sample. This is the moment the part stops being lines on a screen and becomes brass you can measure and fit. The sample exists to answer one question: does the part as drawn actually work in your assembly. Often it does. Sometimes the sample reveals something neither side could see on paper - a feature that interferes with a mating part, a tolerance that is tighter than the function needs, a shape that is awkward to make and could be eased.

I always encourage customers to test the sample in the real assembly, not just inspect it on a bench. A part can measure perfectly to drawing and still be wrong, because the drawing missed something about how it sits in the product. Better to find that on a sample than on the first production batch. Expect a round or two of revision here. A sample that needs a tweak is not a failure - it is the process working. The cost of changing the part at sample stage is tiny next to the cost of changing it after tooling, so this is the right place to be fussy.

Tooling and tolerances - committing to repeatability

Once the sample is approved, we move toward production, and depending on the part that often means tooling - dies for a forged component, fixtures and the right setup for machined ones. Tooling is the investment that turns one good part into thousands of identical ones, and it is worth understanding that it carries a lead time and a cost. A part that will be forged needs a die made first; that die is what makes every piece the same. Rushing this stage to save days at the start usually costs more later.

Tolerances become a firm commitment here, not an aspiration. On the sample we proved the part can be made to the drawing; for production we agree the tolerance band we will hold on every piece, on the features that matter. This is where the conversation about critical dimensions pays off. We set our process, gauging and checks around those features so the variation stays inside the band batch after batch. A realistic tolerance, held reliably, beats an ambitious one missed half the time - and a good maker will tell you honestly which is which for your part rather than promising the tightest number and hoping.

First-article approval and the production gate

Before a full run, the right step is a first-article approval - we make the first production pieces using the actual production tooling and process, and you sign off that they meet the drawing. This matters because a part made the production way can differ subtly from a hand-made sample, and you want to catch that on a handful of pieces, not a full batch. The first article is the formal handshake that the part, the tooling and the process all agree.

Get this gate right and the rest of the programme runs quietly. The first article becomes the reference everything is measured against - the standard the ten-thousandth piece has to match. From here, our quality discipline is about holding that standard: checks at incoming material, at machining, at finishing and at packing, so a drift is caught at the stage it happens rather than discovered by you. A customer who has approved a clear first article and a supplier who checks at every stage have between them removed most of the ways a programme goes wrong.

Scaling from one part to a full programme

A relationship that starts with one custom part rarely stays there. Once a customer trusts that we can take a drawing and turn it into a reliable, repeatable component, the natural next step is more parts - a family of related fittings, a fastener in several sizes, the brass content of a whole product. Scaling well is its own discipline. The volumes grow, the schedule gets tighter, and the consistency that was easy to watch over a few thousand pieces has to hold over far more. This is where owning the process end to end, from raw material to finished part across our plants, earns its keep - we can plan material, machining, finishing and dispatch as one flow rather than juggling vendors.

The thing that makes scaling possible is the same thing that made the first part work: clear specification, honest tolerances, a real first-article standard, and checks at every stage. Build those into the first part and the second, fifth and fiftieth follow on the same foundation. That is how a single custom component grows into a programme that ships to customers across the more than thirty-five countries we serve - not by being clever once, but by being consistent every time.

The practical takeaway

If you have a custom brass part to make, treat the maker as a partner from the drawing onward. Bring the application, not just the geometry; tell them which dimensions are critical; test the first sample in the real assembly and expect a revision or two; respect the tooling lead time as the investment that buys repeatability; and insist on a first-article approval before the full run so the production part, not just the sample, is the one you signed off. Do that, and the gap between a drawing and a reliable programme of tens of thousands of identical parts becomes a path you can walk with confidence - which is exactly what made-to-specification work should be.