FDM 3D Printing

What FDM actually is

Fused deposition modelling builds a part by extruding molten thermoplastic through a nozzle and drawing each layer in turn, roughly 0.08 to 0.35 mm at a time. The nozzle traces the perimeter of the layer, fills the interior to whatever infill density you've specified, then the plate drops and the next layer bonds to the one below it.

That last detail is what makes FDM different from machining a solid block. An FDM part isn't isotropic: it's strong along the layers and weaker across them, because each layer is a thermal weld to its neighbour rather than continuous material. A well-oriented part exploits this; a badly oriented one snaps along a layer line the first time it's loaded. Orientation is a design decision, and we make it deliberately on every job rather than accepting whatever the slicer defaults to.

The trade-off is overwhelmingly worth it. There's no tooling, no minimum order, and a design change costs nothing but the reprint. For anything from a one-off bracket to a 300-piece production run, that economics is hard to beat.

Why an enclosed chamber matters

Both of our machines are fully enclosed, and that isn't a cosmetic choice. When ABS, ASA or polycarbonate cools too fast, each layer contracts as it solidifies and drags on the layer beneath it. The result is corner lift, warping, and in the worst case a clean crack straight through the part days after it comes off the plate.

A sealed chamber holds the ambient temperature well above room temperature and slows that cooling down, so the part shrinks evenly instead of tearing itself apart. It's the difference between "we can technically run ASA" and "we can run ASA reliably on a part you intend to sell".

It also means the higher-temperature filaments are on the menu. Polycarbonate at a 300 °C nozzle, carbon-fiber nylon that needs drying and heat, ABS that you plan to acetone-smooth afterwards, all of these want an enclosure, and all of them are in stock.

Tolerance, and what to expect from it

We hold ±0.20 mm or ±0.3% of the nominal dimension, whichever is larger. On a 20 mm feature that's ±0.20 mm; on a 200 mm span it's ±0.60 mm. Every part is checked against the source model before it's dispatched.

Two practical consequences are worth designing around. First, holes print undersized, because the nozzle path on the inside of a curve leaves a little extra material. A 5 mm hole typically measures 4.7 to 4.9 mm. If a fastener must pass through, either oversize the hole in CAD or plan to drill it out. Second, the first few layers spread slightly under the pressure of the nozzle, so the bottom 0.3 mm of a part sits marginally wider than the rest. On a press fit, that matters.

None of this is a defect; it's how the process behaves. Tell us the mating dimension that has to be right and we'll build the allowance in rather than leaving you to discover it.

How a job runs through the lab

Every file gets opened and looked at by a person. We check wall thickness against the nozzle diameter, look for overhangs steeper than about 50° that will need support, check the part fits the envelope, and decide the orientation that puts the layer lines across the load rather than along it.

Then it's sliced by hand. Profiles are tuned per material and per geometry, not pulled from a generic preset. A tall thin part wants different cooling and different speeds from a flat wide one. The job is scheduled onto whichever cell suits it, printed under a monitored chamber, then de-supported, deburred and inspected.