Last updated: 2026-09-04 · 5 min read

Heat Staking Boss Design: Why Does the Joint’s Quality Start in the Mold?

Heat staking melts a molded thermoplastic boss and forms a head that clamps another component — and the staking press only shapes what the boss gives it. A boss with the right diameter, fillet and material health stakes into a strong head; one with sink at its root or a sharp corner stakes into the crack the press gets blamed for. Boss design belongs to DFM, not the assembly floor.

Key Facts

The boss’s outside diameter runs near twice the shank or hole it serves, giving the staked head enough area to clamp without crushing.

Five head shapes serve different duties: flat for clamp area, dome for centering, hollow for thick bosses on cosmetic surfaces, serrated for lock against rotation and pull-out.

The melt the boss supplies is molded, not staked — sink, knit lines and fill at the boss are DFM items the staking press cannot fix.

Parting line — The line where the two mold halves (A-side and B-side) meet. Parting line location affects: gate placement, ejector pin marks, witness lines, and cosmetic appearance. A well-designed parting line minimizes visible witness marks. The related guide covers adjacent topics.

What Are the Design Rules for a Stakable Boss?

Five elements, each earning its geometry by what failure looks like:

Design Element	Rule of Thumb	Why It Matters
Boss diameter	Roughly twice the captured shank or hole	Head area clamps without crushing
Root fillet	Generous radius where boss meets wall	Kills the stress riser at the base
Boss height	Supplies the melt volume for a full head	Starved heads under-clamp; tall bosses sink
Sink control	Core the boss; limit root thickness	The opposite surface stays visible
Head shape	Flat, dome, hollow, or serrated per duty	Clamp, center, save material, or lock

The diameter rule is about the head’s leverage: the staked head spreads clamp force over the captured component, and a head twice the shank’s width has the area to hold without punching through. Narrower heads concentrate the load and cold-flow over time; wider heads waste cycle and invite sink.

The root fillet is the joint’s longevity: the stake’s pressure and the product’s vibration both concentrate at the boss’s base, and the sharp corner that mills cheaply becomes the crack starter that fails in the field. The fillet costs nothing but attention and carries the load around the corner.

Height balances two failures: too short, the boss cannot supply the melt the head needs and the head forms thin; too tall, dwell lengthens and sink deepens at the root.

The sink row is the cosmetic contract: a solid boss is a local thick section pulling sink into the opposite surface — coring the boss or restraining root thickness keeps it flat. Head shape is where duty speaks: flat heads maximize clamp area; domes self-center; hollow stakes recover excess melt; serrated heads lock against rotation.

How Do You Specify and Prove a Staked Joint?

Four steps, in order:

Design the boss as a joint, with the stake’s duty in mind. The job picks the head and the head picks the boss. Clamp duty favors flat heads sized for bearing area. Locate-and-center duty favors dome heads dropping into clearance holes, tolerant of drift. Lock duty favors serrated heads whose teeth bite the surface. The joint’s service sets the margins: thermal cycling moves plastic and metal at different rates, so head geometry absorbs the travel; vibration checks the root against fatigue; service access favors domes over permanent serrations.
Keep the plastic healthy around the boss. The staked head forms from melt the boss supplies — that melt’s condition is molded, not staked. Core the boss where duty allows, restrain root thickness, and check the opposite surface with the sink prediction. Move gates or the boss so knit lines don’t land at the root. A boss at the end of flow starves and stakes into a cold head. The boss needs enough draft to release and real ejector support — drag-ejected bosses bend before they see a stake.
Specify the staking on the drawing, not in memory. The joint exists twice — molded boss and staked head — and the drawing owns both. Molded state: shank diameter, height, fillet radius, surface condition. Staked state: head diameter, head height, captured clearance. Process callouts: method, tip reference; temperature, pressure and dwell are qualified by the supplier, the drawing holding the result. Multi-stake parts need uniformity: all tips matched, because one tired tip among eight makes one weak stake.
Prove the joint with tests tied to its duty. Clamp duty: pull-off and push-through tests sampled across lots. Lock duty: rotation and pull-out on serrated heads, plus vibration testing per the product’s environment. Thermal duty: cycle between temperature extremes and re-test — the joint’s strength after its environment is the number that matters. Production lot sampling catches tip wear, resin drift and dimension drift while they are still trends.

Cycle time — The total time for one injection molding cycle: injection + packing + cooling + mold open/close + ejection. Typical range: 15–60 seconds. Cycle time determines production rate and cost per part — reducing cycle by 5 seconds saves 10–15% annual production cost. The related guide covers adjacent topics.

The staking press only shapes what the boss gives it. A boss designed as a joint component — with duty, head form and dimensions on the drawing — stakes into a repeatable assembly. A boss designed as a leftover cylinder stakes into a defect that got molded in on schedule.

Key parameter identified — critical variable confirmed against spec.

Tolerance within achievable range — commercial ±0.15mm or tighter as needed.

Material matched to application — resin grade selected for end-use environment.

Process window validated at T1 — parameters locked after first trial.

Inspection method defined — CMM, visual, or functional test specified.

Frequently Asked Questions

When is a hollow stake the right choice instead of a solid one?

The hollow stake — a cored boss whose walls collapse outward as the tip forms the head — solves three solid-stake problems. The void problem: a thick solid boss sinks at its core, visible on the opposite surface; the hollow design removes the thick core, so the boss cools like a tube and the surface stays flat.

The cycle problem: the hollow stake needs less melt and less dwell. Costs are bounded: less pull-through strength, a core pin per boss. Selection: solid for modest non-cosmetic bosses, hollow wherever the boss runs thick, the surface matters, or many stakes add cycle.

Heat staking versus ultrasonic staking — how do you choose?

Both form the same head from the same boss; they differ in energy delivery. Contact-heat: heated tips melt directly — gentle on surroundings, tolerant of variation, able to form large heads, slower. Ultrasonic: vibration melts by friction — fast and clean, but vibration travels: sensitive components can suffer, and the window is narrower.

Populated boards lean contact-heat; bare housings lean ultrasonic. Large heads favor contact heat; high-volume small stakes favor ultrasonic. Some crystalline resins challenge ultrasonic windows. The boss design barely changes between them — designing for either is common practice.

How do you stop sink marks appearing opposite a boss?

Sink opposite a boss is the thick-section problem. Design fixes do the heavy lifting: core the boss — removing the thick core is the most effective move; restrain the root where boss meets wall, with a controlled fillet keeping section near nominal.

Tooling supports with venting at last-to-fill regions and uniform cooling. Process tunes what design left: pack-and-hold pressure feeds shrinkage while the gate is live — but packing cannot compensate a section made too thick. Programs that core at design time ship flat surfaces; the rest buy flatness with reject rates.

Key Takeaways

The staking press shapes what the boss gives it — joint quality is a mold decision, not an assembly one.

Five design elements define a stakable boss: diameter, root fillet, height, sink control and head shape.

Specify both the molded boss and the staked head on the drawing — the joint exists in two states.

Prove the joint with duty-specific tests: clamp, lock, thermal — and keep sampling in production.

Heat staking is where a molded boss becomes a joint — but the quality was decided in the mold.
Diameter, fillet, height, sink control and head shape are the five elements. Specified in both states on the drawing and proved by duty-specific tests, the joint is permanent, fastener-free and repeatable.