# Screw boss design for plastic electronics enclosures

> Practical screw boss design for plastic enclosures: wall ratios, pilot holes, inserts versus thread-forming screws, and how to test before tooling.

- Author: Milo Perović, Founder & lead engineer
- Published: 2026-11-14
- Topic: Design

Most enclosure problems we get asked to fix are not dramatic. They are a cracked boss after the third service visit, a sink mark on the front face exactly where the logo sits, or a screw that spins freely after a few months on a warm windowsill. Good screw boss design prevents all three, and it is one of the cheapest things to get right before tooling.

This post is for founders, product managers and engineers taking a plastic enclosure from printed prototypes towards injection moulding. It covers boss geometry, the choice between thread-forming screws and inserts, and how we test bosses before anyone cuts steel.

## What a screw boss actually has to do

A boss is a short cylinder that gives a screw something to bite into. It sounds trivial, but it carries several jobs at once: it holds clamp load for years, it locates the PCB or the opposite half, it survives assembly torque from a powered driver, and it must not show on the cosmetic surface on the other side of the wall.

Those jobs pull in different directions. A thick boss is strong but sinks and voids. A tall boss locates well but flexes and fills poorly. A tight pilot hole grips hard but splits the boss. Most of the guidance below is about balancing those trade-offs rather than maximising any one of them.

## Screw boss design rules for moulded parts

These are starting points, not laws. Resin suppliers and screw manufacturers publish their own tables, and those win whenever they disagree with a generic rule. For common enclosure plastics such as ABS, PC/ABS and PC, the typical figures look like this:

- **Outer diameter:** roughly 2 to 2.5 times the screw's nominal diameter for thread-forming screws. Brittle or glass-filled resins sit at the upper end.

- **Wall thickness at the base:** about 50 to 70 percent of the nominal wall where the boss meets it. Going thicker is the main cause of sink marks on the outside face.

- **Pilot hole:** commonly 0.75 to 0.85 times the nominal diameter, depending on screw type and resin. Take the exact figure from the screw maker's table for your material.

- **Thread engagement:** around 2 to 2.5 times the nominal diameter. Beyond that, extra length adds little holding power and increases driving torque.

- **Draft:** about 0.5 to 1 degree on the outside, and as little as the toolmaker will accept inside the hole so the thread stays engaged along its length.

- **Root radius:** around 25 to 50 percent of the nominal wall where the boss meets the floor. Sharp corners here are where cracks start.

- **Counterbore:** a short lead-in at the top of the hole, slightly larger than the screw, stops the first thread from bursting the rim.

Keep the hole cored all the way down to the floor rather than stopping short. A solid plug at the bottom is a thick section, and thick sections sink, void and add cycle time.

### Tall bosses and bosses near walls

Once a boss gets taller than about three times its outer diameter, it becomes hard to fill and easy to knock over. Support it with two to four gussets, each thinner than the boss wall, and stop them short of the top so the screw head or PCB still seats on a clean face.

Avoid merging a boss directly into a side wall. The junction becomes a thick lump that sinks on the outside. Instead, stand the boss slightly off the wall and connect it with a thin rib. The same principle appears in our notes on [designing for injection moulding before you have a mould budget](https://miloperovic.com/journal/designing-for-injection-moulding-before-you-have-a-mould-budget/): uniform walls first, then add strength with ribs.

## Thread-forming screws or threaded inserts

This is the decision that shapes everything else, so we make it early in the design, not after the first tool trial.

OptionStrengthsWeaknesses

Thread-forming screws for plasticsCheapest, no extra assembly step, good pull-out strength when the boss is sized properlyLimited reassembly cycles, sensitive to driver torque, holding power drops as the plastic creeps

Heat-set brass insertsMachine threads, many reassembly cycles, work well in printed and moulded partsExtra assembly step and fixture, boss must be sized to the insert maker's figures

Ultrasonic insertsFast and repeatable in volumeNeeds ultrasonic equipment, more process setup

Moulded-in insertsVery strong anchoringSlower cycles, inserts loaded into the tool, more expensive moulding

Our rule of thumb: if a customer or service technician will open the device more than a handful of times, use inserts. Battery doors, field-serviceable gateways and anything with a replaceable module fall into that group. A sealed sensor node that is opened once at the factory and perhaps once for repair is usually fine with thread-forming screws.

When you do reuse a thread-forming screw, it should be started by hand so it follows the existing thread. A powered driver that cuts a fresh path destroys the boss within a few cycles, which is worth writing into the service instructions.

## Creep, temperature and clamp load

Plastics relax under constant stress. A screw torqued to a sensible value on the assembly line can lose a meaningful share of its clamp load over months, and the loss is faster at higher temperatures. That matters most where a screw compresses a gasket, as on a sealed enclosure, because the seal depends on that load staying put.

Three things help:

- Use enough screws, spaced evenly around the seal, so each one carries less load.

- Let the gasket compress against a hard stop, so clamp load is set by geometry rather than by torque alone.

- Choose inserts or a stiffer resin for enclosures that sit in sun, near heat sources or inside vehicles.

Check the resin's heat deflection temperature against the hottest realistic case for the product, not the room it was designed in. A dark enclosure in direct sun runs considerably hotter than the surrounding air.

## Testing bosses before you commit to tooling

We do not trust boss design until it has been tested in the right material. Printed parts are useful for fit, but SLS nylon and resin prints behave differently from moulded PC/ABS, so screw tests on prints only tell you about prints. For anything structural, we prefer machined blocks or early tool trial parts in the production resin.

The tests themselves are simple:

- **Drive and strip torque:** drive screws until they strip and record the torque. The gap between seating torque and strip torque tells you how much margin the assembly line has. A narrow gap means stripped bosses in production.

- **Reassembly cycles:** remove and refit the screw as many times as the service plan requires, then repeat the strip test.

- **Heat soak:** leave assembled parts at the upper operating temperature for a few days, then check residual torque and look for cracks around the boss.

- **Drop:** bosses are frequent failure points in impact. Our post on [drop testing electronics before you commit to tooling](https://miloperovic.com/journal/drop-testing-electronics-before-you-commit-to-tooling/) covers how we run that.

Section the parts afterwards. Cutting through a failed boss usually shows whether it was a void, a weld line or a pilot hole that was simply too small.

## A checklist before you release the enclosure

Before we hand enclosure CAD to a toolmaker, we walk through every boss with these questions:

- Does each boss have a stated fastener, with the pilot hole taken from that screw or insert maker's data for this resin?

- Is the base wall thin enough to avoid sink on the cosmetic face, with a proper root radius?

- Are tall bosses gusseted, and are bosses near walls joined by ribs rather than merged?

- Does the reassembly count in the service plan match the fastener choice?

- Has clamp load on any sealing screw been considered at the hottest operating temperature?

- Have strip torque and heat soak been tested in the production resin, and is the driver torque on the assembly instructions?

None of this is expensive on its own. Changing a boss in CAD costs minutes; changing it in a hardened steel tool costs days and a noticeable invoice. That is the whole argument for doing the work early.
