Why surface finish, and the part that had to feel right

Designing for touch rather than for a dimension.

Surface finish had to be treated as its own design requirement, separate from every dimension on the drawing, because a part can measure perfectly on every axis and still feel wrong the instant a hand closes around it, and a hand judges a surface by a kind of feedback no caliper is built to record.

What is actually happening

A drawing specifies length, width, diameter, every dimension a caliper or a gauge can confirm, and a part that matches every one of those numbers exactly can still fail the only test that actually matters for a feature a hand is meant to touch directly, whether it feels right. Texture, grip and the subtle drag or slip of a surface against skin are not dimensions in the ordinary sense, they are a separate category of information a hand reads instantly and a specification sheet has no column for, which means a part can pass every formal check on the drawing and still be rejected the moment someone actually holds it.

This is easy to overlook precisely because dimensional checks are the ones a design review naturally reaches for, since they are the ones that can be written down, measured with an instrument and signed off with a number. Surface feel resists that same treatment, since there is no single agreed unit for how grippy or how slick a surface should be, which pushes it toward becoming an afterthought purely because it is harder to put on a checklist, not because it matters any less to whoever eventually holds the finished part.

The sanded-spoon comparison

Two wooden spoons cut to identical length, width and thickness, one left rough straight from the saw and one sanded smooth afterwards, would satisfy an identical written specification listing every one of those dimensions, and yet almost nobody reaching into a drawer for a spoon to stir a pot would pick up the rough one twice. The hand makes that choice instantly and without consulting any measurement, responding purely to how the surface feels against skin, and the sanded spoon's advantage exists entirely in a property the original written spec never described at all. A printed part meant to be gripped, turned or pressed directly by a hand is judged by exactly the same unwritten standard, and no amount of dimensional accuracy substitutes for getting that unwritten standard right.

Why the three processes feel completely different in the hand

A resin print straight off the machine has a naturally smooth, almost glassy surface, since curing a liquid produces none of the visible ridges an extrusion process leaves behind, which makes it the strongest starting point of the three when a genuinely slick, refined feel is the goal. A filament print carries obvious layer lines a fingertip can feel immediately, ridges that read as unfinished or cheap to a hand expecting something smoother, though those same ridges can be sanded, filled or chemically smoothed afterwards at the cost of extra time a resin part rarely needs. A PolyJet print sits closest to resin in raw surface quality straight from the machine, with the added ability to vary how firm or soft a given surface feels by choosing a different material for it, letting a single part carry a hard shell in one region and a genuinely soft, grippy texture in another without any secondary process at all.

The number that matters here

Asked to judge two dimensionally identical parts purely by touch, with eyes closed, people sort a smooth surface from a ridged one almost immediately and with near-total agreement, a far higher rate of consistent judgement than the same people typically show when asked to eyeball two similarly close dimensions and guess which is larger. Touch, in other words, is not a vague or subjective sense standing in contrast to a precise measurement, it is its own reliable instrument, simply one that reads a different property than a caliper does, and dismissing it as a soft or secondary concern gets the actual reliability of the two judgements backwards.

What follows from this

Designing a part meant to be handled directly means deciding on its surface finish as deliberately as its dimensions, choosing a process and a level of post-processing that matches how the part is actually meant to feel rather than treating surface texture as whatever a given machine happens to leave behind by default. This sometimes means accepting a slower, more expensive process purely for the feel it produces, or adding a sanding or smoothing step to an otherwise finished part, a cost that looks unnecessary on a drawing full of dimensions already met and turns out, the moment a real hand picks the part up, to have been the entire point. The practical habit worth carrying out of this whole set of three printers is the same one that opened it, naming what a part actually needs before choosing how to make it, only here the requirement being named is how the part will feel rather than how strong or how multi-material it needs to be.

What this does not explain

A pleasant surface finish cannot rescue a part whose underlying geometry is wrong, a grip shaped for the wrong hand, a knob turned the wrong way to operate its mechanism, since touch judges texture and comfort, not correctness of function. Getting the feel right and getting the mechanism right are two separate achievements, and a part that only manages one of them, however well, still has not finished the job the other one was always meant to cover.

Surface finish preference is also genuinely personal in a way pure dimension rarely is, since one person's pleasantly grippy texture is another person's uncomfortably rough one, and a part built for one named user, exactly as with force and reach, benefits from being checked against that specific person's actual preference rather than against a general assumption about what a smooth or textured surface ought to feel like to anyone at all.