Application note

Water Leak Sensor Case Plastic Mold: Where the Seal Actually Lives

This page is for engineers and tooling buyers who are about to release a water leak sensor case plastic mold. It covers the features that decide whether the housing stays dry, the tolerances that matter and the ones that do not, and how to judge a shop before you cut steel.

Get Instant Quote Talk to an Engineer

±0.005 mm toleranceDFM in 12 hoursNo MOQISO 9001 / IATF 16949

On this page

1
Key takeaways
2
What a water leak sensor case plastic mold has to hold
3
Wall stock, ribs and the warp that kills a seal
4
Material and shrinkage: the numbers that move
5
Tooling: how the water leak sensor case plastic mold is cut
6
Which tooling route fits your leak sensor housing

Quick read

Key takeaways

The seal face sets the specA 0.02 mm step on the mating groove leaks before any wall thickness does.

Core and cavity in one setupSingle-setup 5-axis work keeps groove-to-boss position honest across the tool.

Shrinkage is not uniformRibs, bosses and thin walls cool at different rates; compensate per region, not per part.

Steel choice follows volumeP20 suits pilot runs, 1.2343 or S136 holds up past 500,000 shots.

Ask for the CMM reportFirst-article data on the groove, not just a visual check, is the real gate.

Section 1

What a water leak sensor case plastic mold has to hold

A leak sensor housing is a small pressure vessel that never gets treated like one. It sits in a basement, a server room or under a sink, and it has to keep electronics dry for years while the plastic creeps, the gasket relaxes and the ambient temperature swings. That combination is what makes this mold harder than its size suggests.

The functional surfaces are few. The gasket groove, the lid mating face, the cable entry boss and the screw or snap bosses that pull those faces together. Everything else, the outer shell, the mounting ears, the label recess, is cosmetic or structural and can live with looser numbers.

So the drawing should split tolerances by function. A groove width of 1.50 ±0.03 mm and a groove depth of 1.20 ±0.02 mm do more for seal integrity than tightening the overall envelope. Chasing ±0.005 mm on the outer wall buys nothing and pushes tool cost up.

1
CriticalGasket groove width, depth and radius; lid mating flatness; boss position.
2
ControlledWall thickness, cable entry bore, PCB standoff height.
3
FreeOuter cosmetic surfaces, label recess, mounting ear profile.

Section 2

Wall stock, ribs and the warp that kills a seal

Most leaking units we see are not a groove problem. They are a warp problem. A nominal 2.0 mm wall next to a 4.0 mm boss creates a thick-to-thin transition, and the thick side holds heat longer. The part curls as it cools, the lid face stops being flat, and the gasket has to absorb a gap it was never sized for.

Keep nominal wall between 1.5 mm and 2.5 mm for ABS or PC housings, and core out thick sections rather than accepting them. Where a boss is unavoidable, tie it to a nearby wall with a rib instead of leaving it standing alone. Rib thickness at 0.5 to 0.6 of the nominal wall keeps sink marks off the visible face.

Draft matters here too. 1° to 1.5° on the outer walls and 0.5° to 1° on internal features is enough for a textured cavity. Deep gasket grooves want 2° or more on the side walls, otherwise the part drags on ejection and you scuff the exact face that has to seal.

1
Thin-to-thickTransition over at least 3× the wall difference.
2
Ribs0.5–0.6 × nominal wall, 0.5° minimum draft.
3
Sharp cornersAdd 0.5 mm minimum internal radius to avoid stress risers.

Section 3

Material and shrinkage: the numbers that move

Housing material is usually chosen for chemical resistance and cost, not for molding ease. ABS is the default for indoor units and runs at roughly 0.4% to 0.7% shrink. PC gives better impact and temperature range but shrinks around 0.5% to 0.7% and needs more drying. Glass-filled grades drop to 0.2% to 0.5% and are far more anisotropic.

That anisotropy is the trap. Flow direction and cross-flow direction shrink differently, often by 0.1% to 0.3% in filled grades. On a 90 mm long housing that is 0.09 mm to 0.27 mm of difference between the two axes. If the tool is cut with a single shrink number, the groove ends up oval and the gasket load varies around the perimeter.

Gate placement decides which way the flow runs across the seal. Put the gate so the melt front sweeps the gasket groove in one direction and reaches both ends of the groove at similar pressure. A gate that fills the groove from the middle outward leaves a weld line on the seal face, and a weld line on a seal face is a leak path.

1
ABS0.4–0.7% shrink, easy flow, good for indoor enclosures.
2
PC0.5–0.7% shrink, dry 3–4 h at 120 °C before molding.
3
PA6-GF300.2–0.5% shrink, 0.1–0.3% anisotropy in flow vs cross-flow.

Section 4

Tooling: how the water leak sensor case plastic mold is cut

The gasket groove is where toolmaking shows. It is narrow, often 1.2 to 2.0 mm wide, and it has to hold depth within ±0.02 mm along its whole length. On a 3-axis machine, the groove is cut in multiple setups and each setup adds its own positional error. Stack those errors and the groove floor steps.

We cut core and cavity on 16 simultaneous 5-axis machining centers in a single setup, which keeps groove-to-boss and groove-to-screw-hole position tied to one datum. The 127-machine shop floor includes 12 four-axis mills, 27 three-axis machines and 16 mill-turn centers, so electrodes, inserts and the mold base can run in parallel instead of queuing behind one spindle.

Steel choice tracks volume. P20 pre-hardened is fine for a pilot tool and a few thousand shots. For production past 500,000 shots, 1.2343 or S136 through-hardened to 48–52 HRC holds groove geometry longer. Core pins and the cable-entry slide should be nitrided regardless of the base steel.

Cooling is not optional on a part this small. A 12 mm pitch conformal or baffled circuit around the groove keeps the seal face at even temperature, which is what stops the warp described above. We machine the cooling layout as part of the same 5-axis setup so the circuit follows the groove rather than the mold plate.

1
Single setupGroove and bosses tied to one datum, no stacked offsets.
2
SteelP20 for pilot, 1.2343 or S136 past 500,000 shots.
3
Cooling12 mm pitch around the seal face, baffled where space is tight.

Selection

Which tooling route fits your leak sensor housing

Match program size and sealing requirement to the tooling route.

Route	Best for	Typical tolerance	Watch out for
CNC-machined aluminum tool	Pilot runs under 5,000 parts	±0.005 mm on groove	Soft steel wears at gate and groove edges
P20 pre-hardened mold	5,000 to 200,000 parts	±0.01 mm on groove	Needs re-polish on the seal face mid-life
1.2343 / S136 hardened	200,000 to 500,000+ parts	±0.01 mm on groove	Higher tool cost, longer first cut
Aluminum bridge tool	Design freeze not final	±0.02 mm on groove	Not for production volume or filled resin

The call we would make

If the housing seals against a molded gasket groove, spend the money on a hardened single-setup tool and hold the groove at ±0.01 mm. If the unit uses a die-cut foam gasket and the groove is only a locating channel, a P20 tool with ±0.02 mm is enough and gets you to market faster.

Get Instant Quote Talk to an Engineer

FAQs

Questions engineers ask before cutting steel

How tight does the gasket groove really need to be?

For a compressed elastomer gasket, groove depth within ±0.02 mm and width within ±0.03 mm usually holds a consistent compression ratio. Going tighter than that does not improve sealing on a plastic housing, because the plastic itself moves more than the groove under load.

If the gasket is a molded silicone ring rather than a die-cut sheet, the groove can be looser. The ring absorbs the variation. The face flatness matters more than the groove number in that case.

Should the housing be a single part or a two-piece lid design?

Single-piece housings with a snap or ultrasonic weld joint remove the gasket entirely, which removes the leak path. They need a clean weld energy director and a uniform joint around the perimeter, which is a mold feature, not an assembly feature.

Two-piece designs with a screw-down lid tolerate looser tooling and are easier to service in the field. The trade is a longer bolt circle and a gasket that has to be sized for the full face.

What causes flash on the seal face and how do you prevent it?

Flash on a seal face usually comes from one of three things: insufficient clamp force for the projected area, a parting line that is not flat, or a vent that is too deep and sits on the seal land.

The first fix is tooling. Keep vents 0.02 to 0.03 mm deep and move them off the seal land. The second is process: raise clamp tonnage or reduce injection speed near the end of fill so the cavity pressure does not spike.

Can you mold the housing in a flame-retardant grade?

Yes. Flame-retardant ABS and PC/ABS grades are common for sensor enclosures and they mold much like the standard grades, with two differences. They are more corrosive to tool steel, so use a hardened or plated cavity, and they have a narrower processing window.

Tell us the grade at quote stage. Corrosive resin changes our steel recommendation and the surface finish plan for the groove.

How do you verify the groove before shipping the tool?

We run a first-article inspection on the molded sample and report the groove width, depth and radius at four or more points around the perimeter, plus the lid mating flatness. Reports go out on request.

For the tool itself, we check the groove geometry on the CMM before assembly. If the groove is out, no amount of process tuning saves the part.

What is the smallest order you will run?

There is no minimum order quantity. We run from one prototype to 10,000+ part runs. For a leak sensor program, the usual path is a machined aluminum bridge tool for the first functional samples, then a hardened production tool once the design is frozen.

Quotation and a free DFM analysis come back within 12 hours, and production can start within 24 hours of approval.