Parametric vs Modelling Your Own: Why Sized-to-Fit Files Win

You need a hose adapter. The old one cracked, the chandlery is closed, and you have a 3D printer on board or in the garage. At this point you have two options: open a CAD program and model the part from scratch, or download a parametric file and type in your measurements.

Both work. But after building and testing several hundred marine parts, we can say with some confidence that for the vast majority of boat jobs, the parametric route wins — not because modelling is hard, but because boats are not standardised and the first print is almost never the last one.

The real problem: nothing on a boat is a round number

Land-based DIY has the luxury of standards. Plumbing is 15 mm or 22 mm. Screws are M4 or M6. On a boat you inherit whatever the yard used in 1987, plus three decades of previous owners' improvisations. A single vessel can carry BSP and NPT threads, metric and imperial hose, French deck fittings and American pumps.

That means the part you need is rarely a catalogue item. It is a 3/4" BSP male on one end, a 19 mm hose barb on the other, offset by 40 mm because the bulkhead is in the way. A parametric file handles that in thirty seconds. A fixed STL does not handle it at all, and modelling it yourself takes an evening.

What "parametric" actually means

A parametric file is not a finished shape. It is a set of rules — a recipe — that generates the shape from the numbers you supply. Change the thread size and the wall thickness adapts. Change the hose diameter and the barb ridges rescale. Change the length and the internal bore follows through cleanly instead of leaving a stepped, leaky transition.

The important consequence: the geometry stays valid. When you stretch a fixed STL in your slicer to make a fitting 10% longer, you also stretch the thread pitch, the wall thickness and the sealing face. The part looks right on screen and fails on the boat. Parametric regeneration recalculates every dependent dimension, so a 12 mm barb and a 32 mm barb both come out with correct ridge geometry and correct wall thickness for pressure.

Time and skill: an honest comparison

	Model it yourself	Parametric file
Time to first print	1–4 hours (plus learning CAD)	2–5 minutes
Time to revision 2	15–45 minutes	Under a minute
Thread accuracy	Depends on your thread library and tolerances	Built in and print-tested
Wall thickness for pressure	Your judgement	Scales with diameter automatically
Print settings guidance	Trial and error	Supplied with the file
Flexibility beyond the design intent	Total	Limited to exposed parameters

Note the last row. Modelling your own part wins on one axis, and it is a real one: if you need something genuinely unusual, a parametric file that was never designed for your case will not stretch to cover it. This is not an argument that parametric always beats CAD. It is an argument that for the recurring, dimension-driven parts that make up most boat jobs, parametric is the faster and safer default.

Where the iteration cost really bites

Here is the part people underestimate. Almost nobody gets a threaded or press-fit part right on the first print. Printers vary, filament varies, and a 0.2 mm difference in extrusion width changes whether a thread turns freely or binds solid.

So the realistic workflow is: print, test, adjust clearance, print again. Maybe three cycles. With a parametric file that is three number changes. With your own CAD model it is three trips back into the sketch, hunting for the dimension that drives the thread clearance and hoping the rebuild does not break a downstream feature.

This is also why we expose a clearance or fit parameter on threaded parts rather than baking in one "correct" value. Your printer is not our printer. If you want the detail on how those tolerances work in practice, we covered it in how to 3D print watertight boat fittings that hold pressure.

Measure once, generate once

The parametric advantage collapses if your input numbers are wrong, so measurement is the step that actually deserves your attention. A few habits that pay off:

• Use calipers, not a tape measure. A 19 mm and a 20 mm hose look identical by eye and behave very differently under a hose clamp.
• Measure the thread, do not guess it. Count threads over 25 mm and check the crest diameter. BSP and NPT of the same nominal size are close enough to start, and far enough apart to leak.
• Measure the hose internal diameter for barbs, and the outside diameter for anything that clamps around it.
• Write the numbers down before you walk back to the computer. Everyone thinks they will remember. Nobody does.

We have a fuller walkthrough in how to measure a hose and thread for a perfect 3D-printed fitting.

When you should model it yourself

To be even-handed: model your own part when the geometry is genuinely one-off and not dimension-driven. A bracket that has to wrap around a specific moulding, a cosmetic trim piece matching a curve unique to your hull, a mounting plate for an instrument nobody else owns. These are shape problems, not size problems, and parametric files are built for size problems.

It is also worth learning CAD regardless. Understanding how a part is constructed makes you better at judging whether a downloaded file will actually survive the load you are about to put on it.

Practical starting points

If you want to see how this works before committing to anything, the free files collection is the cheapest way to test the workflow on your own printer. From there, the plumbing and pumps range covers the fittings that most often go wrong at the worst moment, and deck and rigging covers the organisers, guides and hardware that take a beating above decks.

Whichever route you take, print a short test section of any threaded part before you print the whole thing. Ten minutes and two grams of filament will tell you whether your clearance number is right, and save you a 90-minute print that ends up in the bin.

A note on safety

3D-printed parts are not certified marine equipment. Nothing on this site is type-approved, and no printed part should be treated as equivalent to a certified fitting. Before use, satisfy yourself that the material, wall thickness and print quality are appropriate for the job — and be particularly conservative with anything below the waterline, anything carrying fuel or gas, and anything in a load path where failure could injure someone or sink the boat. Through-hulls, seacocks, standing rigging and structural fastenings should be commercial certified components. Printed parts are excellent for convenience items, organisation, mounts, non-critical plumbing above the waterline, and temporary repairs that get you home.

Questions about a specific fitting or a parameter you cannot find? Write to info@marinelab3d.com — if a part needs a parameter it does not currently expose, that is useful for us to know.