Particle Counter Housing Die Casting

• 1
  What a particle counter housing has to do at the same time
• 2
  Why high-pressure die casting suits this enclosure
• 3
  Choosing an alloy for a cleanroom instrument
• 4
  Where die casting tolerance ends and CNC begins
• 5
  Porosity, leaks, and how they show up downstream
• 6
  From CAD model to volume-ready housings
• 7
  Die casting versus the other routes for a counter housing

A particle counter housing is not a box. It holds a laser diode, a photodetector, a sample pump, and the flow path between them. The optical axis has to stay aligned to within tens of microns after the unit is dropped, shipped, and warmed up in a cleanroom. That requirement alone rules out a lot of thin-wall enclosures.

The housing also has to seal. Ambient dust entering the optical chamber shows up as false counts, and a customer who sees false counts sends the instrument back. Door gaskets, cable glands, and connector ports all need a flat, machined land to compress against.

Heat is the third job. A laser diode and a small diaphragm pump together can put 10-20 W into a housing the size of a shoebox. Die-cast aluminum spreads that heat well, but only if the ribs and bosses connect to a surface that actually sees air.

So the part is a structural frame, a seal carrier, and a heat spreader in one casting. Each of those roles pulls the design in a different direction, and that tension is what makes particle counter housing die casting worth planning early.

High-pressure die casting injects molten aluminum into a steel die at 400-700 bar. The metal fills thin sections fast, cools against the die wall, and comes out with as-cast walls of 1.5-3.0 mm. For a housing with internal ribs, mounting bosses, and a connector cutout, that is close to the shape you want before any machining.

The alternative routes each lose on something. Sheet metal gives good flatness but needs welded or riveted corners, and those joints are hard to seal. Plastic injection molding is cheap at volume but needs a conductive coating for EMI, and the coating wears at screw threads. Billet machining gives the best tolerance but removes most of the material as chips and costs far more per part above a few hundred units.

Die casting wins when the annual volume sits between roughly 500 and 100,000 units and the part has 3D geometry that a press brake cannot form. Below that range, tooling cost dominates and a machined or vacuum-cast prototype is usually the better first step.

Porosity is the known weakness. Gas trapped during fill leaves small voids, and a void under a sealing face becomes a leak path after machining. Good die design, vacuum-assisted shot profiles, and a controlled gate position are how that gets managed.

• 1
  Thin walls fill better1.5-2.5 mm cast walls flow more reliably than 5 mm sections with hot spots.
• 2
  Draft matters1-2° on outside walls, 2-3° on inside cores, otherwise parts drag and gall.
• 3
  Radius every corner0.5-1.0 mm fillets reduce stress and improve metal flow at the gate.

ADC12 and A380 are the workhorses. Both are Al-Si-Cu alloys with good castability, low shrinkage, and decent stiffness. ADC12 is common in Asian tooling and machines cleanly. A380 behaves almost the same in the die and is the usual choice when the tool is cut in Europe or North America. Either is fine for a housing that stays indoors.

If the housing needs to be welded or anodized for appearance, A360 or a low-copper alloy is a better base. High copper content gives anodized parts a darker, less even color. For conductive anodizing that grounds the housing to the chassis, the copper level also shifts the surface resistance.

Magnesium AZ91D is worth a look when weight is critical, such as a handheld or cart-mounted counter. It casts thinner and lighter than aluminum, but it needs a proper coating system and it is not the right choice for a wet or humid environment without extra protection.

In our shop the cast blanks are often finished on the same floor as the machining. That lets us machine ADC12 castings to ±0.005 mm on the optical mounting face and to Ra 0.8-1.6 μm where a gasket seats, without shipping the part between vendors and losing the datum.

• 1
  ADC12 / A380Default for indoor instrument housings; best cost at volume.
• 2
  A360Pick it when the part will be anodized or welded.
• 3
  AZ91D magnesiumWeight-critical builds only; needs a coating plan.

An as-cast die casting holds about ±0.1 mm on a stable dimension and looser across a parting line. That is not enough for an optical mount or a gasket land. The trick is to cast the feature with 0.3-0.5 mm of stock and machine only what the function needs.

The machined faces usually are the optical mounting plane, the gasket groove floor, the connector flange, and the thread bosses. Datum strategy matters more than machine choice here. We normally pick the machined optical face as the primary datum, then locate the gasket land and the connector bores from it in one setup.

Flatness on the optical mount is the number that decides image quality. A gasket land that is not flat leaks. On a 5-axis center we hold ±0.005 mm on those features and Ra 0.8-1.6 μm on the sealing surface, which is smooth enough for a molded gasket to seal without a separate coating.

Do not machine the whole housing. Every face you cut adds cost and removes the skin that carries the as-cast finish. Aim for 15-25% of the surface area under the cutter, no more.

• 1
  Cast stock on machined faces0.3-0.5 mm is enough for cleanup without long cycle times.
• 2
  One datum, one setupLocate gasket land and connector bores from the optical face.
• 3
  Sealing faces get the finishRa 0.8-1.6 μm; cosmetic faces can stay as-cast.

The most expensive defect in a die-cast housing is a subsurface void that only opens after machining. The part looks fine at goods-in, then leaks at the sealing land after the gasket groove is cut. By then the casting lot is committed.

Two things reduce that risk. First, keep the gasket groove away from the last-filled region of the die, which is usually the far corner opposite the gate. Second, specify a leak test, not just a visual check, on the machined sealing face. A pressure decay test at 0.2-0.5 bar over 10-30 seconds catches most through-wall porosity.

Heat treatment can help or hurt. A stress-relief cycle after casting reduces the chance of the housing moving over time, which matters when optics are aligned to 50 μm. But a full T6 solution treatment on a die casting risks blistering from trapped gas. That is why high-integrity castings are usually stress-relieved rather than fully aged.

Dimensional creep is the slow version of the same problem. A housing that holds alignment at first article but drifts 30 μm over a year will fail a field recalibration. Stress relief plus a stable alloy is the practical answer.