What Your Surface Texture Really Costs at the Tool

Your ID renders look great. But every surface texture carries a draft angle penalty, a tooling method, and a schedule cost. A contract manufacturer's guide to cosmetic DFM before you cut steel.

DESIGN TO PROTOTYPE

Engineering Team

8/5/202610 min read

Each panel shows a simple cross-section silhouette of a rectangular plastic enclosure wall
Each panel shows a simple cross-section silhouette of a rectangular plastic enclosure wall

There is a specific moment in every hardware program where the mood changes.

The industrial design consultancy delivers the final renders. Everyone in the room agrees they look fantastic. The founder approves them. Then the files land on a toolmaker's desk in Shenzhen, and a week later an email arrives that says the tooling quote has gone up, the T1 date has moved, and by the way, several of the vertical walls need more draft — which means the silhouette everyone just approved is going to change slightly.

Nobody was being difficult. The renders were never wrong. They were simply silent about the one thing that determines whether a cosmetic surface is cheap or expensive: how that surface gets into hardened steel, and how the part gets back out of it.

This guide covers the texture side of that conversation. If you are about to commit to a tool, read it before you sign off on the appearance model, not after.

First question: is it a texture, or is it geometry?

This distinction causes more surprises than anything else in cosmetic DFM, and it is almost never visible in a render.

A texture is a micro-scale surface treatment applied to the mold cavity after the cavity has been machined and polished. Sand grain, leather, stipple, fine matte — these live in the last few microns of the steel surface. They are applied by chemical etch, EDM, or laser. They do not appear in your CAD model at all. Your CAD surface is nominally smooth, and the texture is called out as a specification against that surface.

Geometry is macro-scale form that must be modeled in CAD and physically cut into the cavity by CNC and then hand-finished. Deep ribs, fluting, faceting, a diamond knurl with sharp crests, a rib field that fades out into a smooth surface.

To an untrained eye both read as "texture" in a photograph. To a toolmaker they are completely different line items. Micro texture is a few hundred dollars and a few days. Macro geometry is CNC time on a hardened cavity, sometimes an EDM electrode, and hours of skilled hand polishing on every peak and valley — and it multiplies by cavity count.

The rule of thumb: if you can see the individual features clearly in a render at product scale, it is geometry and it belongs in your CAD. If it only shows up as a change in sheen, it is a texture and it belongs in your spec.

Where this matters most is the fade-out effect that is currently everywhere in consumer electronics ID — the rib field that emerges out of a smooth surface, deepens, then dissolves back into nothing. It looks like a texture. It is not. Every one of those ribs has to be modeled with a varying depth profile, and the polishing at the shallow end, where rib depth approaches zero, is where hand-finishing marks become visible on a Class-A surface. It is one of the most demanding cosmetic features you can specify, and it is routinely specified by people who believe they are asking for a texture.

The trade-off is not simply "laser is better." <cite index="15-1">Chemical etching remains the economical choice for applying organic patterns across large, uniform enclosure surfaces, and it is the industry standard for that job.</cite> Its weaknesses are manual variability and time — <cite index="13-1">the mold has to be taken apart, cleaned, masked and blasted before any pattern is transferred, only one half can be worked at a time, and matching a pattern across the parting line takes repeated iteration.</cite>

Laser texturing works from your CAD data directly, which buys you repeatability across cavities and across replacement tools, and <cite index="17-1">it compresses texturing lead time from a matter of weeks down to days</cite>. It also unlocks effects chemical etching cannot produce — <cite index="16-1">blending two different textures seamlessly into one another on a single surface, for example</cite>. You pay for it in machine hours. And be aware that <cite index="19-1">gradient and shallow patterns need multiple overlaid programs to execute cleanly, which extends processing time and price</cite>. That trendy fade-out grain is expensive in every direction.

One more practical note: <cite index="12-1">deep features are where EDM texturing breaks down, because the spark cannot maintain consistent density and you get bald patches; laser is the answer for holding a texture consistently across complex 3D form.</cite>

The draft angle tax

This is the part that changes your product's shape.

Every crater in a textured cavity is a small mechanical interlock. The plastic shrinks onto it, and when the ejector pins push, the texture has to slide out of those craters cleanly. Under-draft it and the texture shears off, leaving drag marks and shiny streaks down the sidewall. The deeper the texture, the more taper you need.

The widely used baseline table, derived from Bayer's original polymer data, looks like this:

Two things to understand about this table.

One: treat it as a floor, not a target. Many production shops work to more conservative numbers than the Bayer baseline. <cite index="3-1">A common working guideline is that a VDI 24 finish wants at least 1.5–2.0°, while a coarse VDI 36 needs 4.0° or more.</cite> Glass-filled resins need more still. If your part has a deep sidewall, add margin — the deeper the draw, the more a missing half-degree hurts.

Two: the imperial rule of thumb agrees with it. <cite index="10-1">The conventional rule is to add roughly 1.5° of draft per 0.001" of texture depth, or about 1° per 0.0005–0.0006" for Mold-Tech patterns.</cite> Run the numbers on VDI 30 and you land in the same neighbourhood as the table. Either way, the arithmetic is not optional.

Here is why this matters to your ID, not just your tooling engineer: draft is a change to the visible outline of the product. A 60mm-tall enclosure wall at 0.5° draft and the same wall at 3° draft are not the same shape. On a squared-off, minimal-silhouette design, that difference is visible. If you specify a coarse texture on a tall vertical wall and only discover the draft requirement at the tooling stage, you are choosing between a texture change and a form change, late, under schedule pressure.

Decide the texture grade during ID, and model the draft that grade demands into the CAD from the start.

Reading a texture reference board like a manufacturer

Design inspiration boards circulate constantly in this industry — six-panel grids of gorgeous close-ups, tagged with names like mesh, fish-scale, honeycomb, fade-out, raised dot, herringbone. Here is what each family actually implies once someone has to build it.

Mesh and lattice patterns. Usually not a texture at all — these are through-openings, which means core pins, weld lines behind every opening, and a whole separate DFM conversation. If it is a blind lattice (embossed, not open), it is macro geometry: CNC time plus difficult polishing in every recessed cell.

Fish-scale and overlapping-facet patterns. Directional geometry. The critical question is orientation relative to the draw direction. Scales that face "into" the pull are effectively small undercuts; scales oriented along the pull release cleanly. Same visual, entirely different tool. If the pattern wraps a curved corner, the draw direction changes across the surface and part of your pattern will be fighting ejection. This is the classic reason a texture gets a side-action added to the tool.

Honeycomb and repeating dimple fields. Well suited to laser texturing, which handles sharp-edged repeating geometry with perfect registration. Shallow versions are inexpensive. Deep versions carry the full draft penalty on every cell wall, and on a curved surface the cells will distort visually unless the pattern is mapped in 3D rather than projected flat.

Fade-out ribs. The most demanding item on any of these boards, for the reasons above: it is modeled geometry, not applied texture; the fade-out zone is a hand-polishing challenge; and if you execute it by laser instead, you pay for multi-layer programs. Beautiful, and worth it on a hero product. Just budget for it deliberately.

Raised dots and braille-scale bumps. Deceptively easy-looking. The issue is that very shallow raised features sit right at the threshold where injection pressure, resin flow and venting determine whether they fill sharply or come out soft and rounded. They also show gate-to-far-side variation across a large surface. Ask for a texture plaque in your actual resin before you commit.

Herringbone and chevron patterns. Strongly directional, which makes them excellent grip surfaces and unforgiving of draw-direction mismatch. The same caution as fish-scale applies, plus a cosmetic one: directional patterns make any mismatch across the parting line glaringly obvious. Plan where the texture stops and where the split line falls at the same time.

Five failure modes we see repeatedly

  1. Drag marks from under-draft. Diagnosed at T1, expensive to fix. The remedy is to remove steel — which changes the part — or reduce the texture grade, which changes the appearance.

  2. Texture mismatch across the parting line. Two mold halves textured separately, joined, and the grain does not line up. Far more visible on directional patterns than organic ones.

  3. Weld lines showing through. Texture is often used specifically to hide flow marks and weld lines — that is one of its real functional benefits. But a fine, low-grade texture on a glossy resin will not hide a weld line downstream of a hole or boss. Match texture grade to how much you need it to conceal.

  4. Inconsistent density in deep features. The bald-spot problem in EDM texturing. Catch it at the tooling-method decision, not at first shots.

  5. Assuming texture is reversible. It is not, in one direction. Going coarser later is straightforward — you re-texture. Going smoother or removing texture means welding the cavity and re-polishing, with a real risk to the surface and a real hit to schedule. Start finer than you think you want.

The resin changes the answer

Two parts with an identical VDI spec, molded in different resins, will not look identical. Amorphous materials like PC-ABS and ASA come out noticeably glossier than polypropylene at the same steel finish. Glass fill raises the required draft and can produce visible fibre at the surface. Colour and pigment loading shift perceived gloss again.

This is why a texture decision made on a rendered image, or on a plaque in the wrong resin, is not a decision. <cite index="12-1">The professional practice is to cut a steel coupon from the same heat-treat lot, apply the intended texture with production electrodes, mold a short run in the real resin to confirm the texture replicates, and lock the specification before texturing the full tool.</cite> On a cosmetic consumer product, that step is worth the two weeks it costs.

What to hand your manufacturer

If you want a tooling quote that will not move on you, supply these six things alongside your CAD:

  1. The texture standard and grade in writing — VDI 3400 #24, Mold-Tech MT-11020, or a named supplier's pattern code. "Matte" is not a specification.

  2. A textured zone map — which surfaces are textured, where texture stops, and how it terminates at edges and parting lines.

  3. Draft already modeled to suit the grade you chose, not the minimum draft that made the render look sharp.

  4. The production resin and colour, not the prototype material.

  5. Gloss level and acceptance criteria — how the part will be inspected and under what lighting, and whether a master plaque will govern approval.

  6. Which features are geometry and which are texture — stated explicitly, so nobody quotes the wrong process.

Do that and the texture conversation takes one round instead of four.

The takeaway

Surface texture is treated as a late-stage aesthetic decision and it is actually an early-stage engineering one. It sets your draft angles, which sets your part geometry. It sets your texturing method, which sets several weeks of your tooling schedule. It interacts with your resin, your parting line, your gate location and your ejection strategy.

None of that means you should design a boring product. It means the interesting cosmetic decisions should be made with the tool in mind, at the point where they are still free to change.

At Peakingtech we run cosmetic DFM reviews on appearance models before tooling is committed, precisely because this is the cheapest possible moment to find these issues. If you have an ID package and you are not sure which details are free and which ones cost you three weeks, send it over — we will tell you which is which.

Frequently asked questions

What is the difference between SPI and VDI 3400? SPI is the North American standard, organised around polishing method and running from A-1 (mirror) to D-3 (coarse blast). VDI 3400 is the German standard, organised around EDM texture depth and running on a numeric scale from roughly 12 to 45, where higher numbers are rougher. SPI is the better reference for gloss and transparent parts; VDI is the better reference for matte and tactile finishes. There is no official conversion between them — published cross-reference charts are approximations based on Ra values and do not guarantee that two grades will look the same.

How much draft angle does a textured surface need? It depends on texture depth and resin. As a baseline, VDI 24 in ABS wants about 1°, VDI 30 about 2°, and VDI 36 about 3°, with polycarbonate needing more than ABS and glass-filled grades more still. The common imperial shortcut is to add about 1.5° of draft for every 0.001" of texture depth. Treat any published figure as a minimum and add margin on deep sidewalls.

Should I use chemical etching or laser texturing? Chemical etching is more economical for large, uniform surfaces with organic patterns and remains widely used for enclosures and appliance housings. Laser texturing is better for sharp geometric patterns, gradients, blended textures and complex 3D surfaces, gives far better repeatability across cavities and replacement tools, and typically has a shorter lead time — at a higher hourly cost.

Can I change the texture after the mold is built? You can make it rougher relatively easily by re-texturing. Making it smoother or removing texture requires welding the cavity and re-polishing, which risks the surface and costs schedule. Specify a finer grade than you think you want and coarsen later if needed.

Why does my texture look different on the actual part than on the sample plaque? Almost always resin, colour or geometry. The same steel finish produces different gloss in different polymers, pigment loading shifts the result again, and curved or deep surfaces replicate texture differently than a flat plaque. Approve texture on a plaque molded in your production resin and colour.

Does texture increase part cost or only tooling cost? Primarily tooling cost and tooling schedule. The per-part impact is usually small, though texture can slightly affect fill behaviour and, if draft is marginal, can increase scrap from ejection defects. The larger and more variable cost is on the tool.

Is a fade-out or gradient texture a texture or a modeled feature? Usually a modeled feature. Ribs that vary in depth and dissolve into a smooth surface have to exist in your CAD and be machined and hand-finished into the cavity. Some gradient effects can be produced by laser texturing, but they require multiple overlaid programs and cost more than a uniform pattern. Either way, budget for it as a premium feature rather than a free styling choice.