Skip to content
HiFi MFG

Guides / Chassis / Design for Manufacture

Shielding an Audio Chassis Without Guessing

September 25, 2026

A metal box is not automatically a shield. It becomes one when the current induced on its outside can flow around it without meeting a break, and most audio chassis are full of breaks: seams, screw spacings, ventilation slots, connector cutouts and an anodized finish that insulates every joint.

This guide is about the manufacturing decisions that control that, because almost all of them are made on the drawing rather than in the lab.

The seam is the shield

Fields do not leak through 3 mm of aluminum. They leak through the gap where two pieces of aluminum meet.

What matters at a seam:

  • Continuity of contact, not just proximity. Two flat faces bolted at the corners touch at the corners and bow apart in between.
  • Fastener spacing. The rule of thumb is that the gap between fasteners should be small compared with the wavelength of the highest frequency you care about. For audio gear worried about RF ingress, 50 to 75 mm spacing along a seam is a practical target and 150 mm is too far apart.
  • Surface condition. Bare or conversion coated aluminum conducts. Anodized aluminum does not. Painted aluminum does not.
  • Flatness. A folded sheet cover that is 0.5 mm out of flat along its lip does not contact along that lip.

Anodize is an insulator, and this catches people

This is the single most common manufacturing mistake on an audio chassis that needs shielding.

The finish everybody wants on the outside is exactly the wrong finish on the inside joint faces. Type II anodize is an oxide layer, and oxide does not conduct.

Three ways to handle it:

  1. Mask the joint faces before anodizing. Add a note naming the faces and the earth stud boss. The masked area shows as bare metal, which is invisible once assembled and is the cheapest answer.
  2. Use a chromate conversion coating (Alodine, Iridite) on the mating parts. It is conductive, it protects against corrosion and it accepts paint. Commonly specified as MIL-DTL-5541 Class 3 when conductivity matters.
  3. Add a conductive gasket between the faces so the gasket, not the finish, carries the current. Beryllium copper fingerstock, conductive elastomer or knitted wire mesh. This is the expensive, effective option and it needs a groove or a landing on the drawing.

A drawing note that does the job:

MASK BEFORE ANODIZING:
  ALL COVER MATING FACES, 8 mm WIDE BAND, SEE VIEW C
  EARTH STUD BOSS AND 15 mm DIAMETER AROUND IT
MASKED AREAS TO BE BARE, CLEAN AND FREE OF OXIDE AT ASSEMBLY

Apertures: length, not area

Every hole in a shield is an antenna. The useful mental model is that leakage is governed by the longest dimension of the opening, not by how much area it has.

That has direct consequences for how you draw ventilation:

  • A 100 mm long slot leaks dramatically more than twenty 5 mm round holes, even though the round holes may pass more air.
  • A row of short slots separated by metal bridges is much better than one long slot.
  • Hexagonal or round perforation is better than slotting for the same open area.

If a design needs a long opening for looks, break it with bridges. A 200 mm decorative slot across a front panel, backed by a perforated plate behind it, gets the appearance and keeps the shield.

Connector cutouts and the panel

A connector cutout is a hole whose leakage depends on how the connector body bonds to the panel.

  • A metal shell connector bolted to bare metal bonds well. The same connector bolted to an anodized panel does not.
  • A plastic bodied connector does not bond at all, so the cutout is just a hole. Shielding then has to happen behind the panel or at the cable.
  • Grounding washers, star washers or a scraped landing under the flange all achieve contact through a coating, and any of them belongs on the assembly drawing rather than being left to the assembler.

Where aluminum stops helping

Audio has one shielding problem that aluminum genuinely cannot solve: low frequency magnetic fields from a mains transformer, at 50 or 60 Hz and its harmonics. Aluminum is not ferromagnetic, so it does not provide a low reluctance path for that flux.

Practical answers, in the order most people should try them:

  1. Distance and orientation. Move the transformer away from the input stage and rotate it. This is free and it usually helps more than any material.
  2. A steel or mu-metal can around the transformer. Local shielding at the source beats shielding at the boundary.
  3. A steel partition between the power section and the audio section, bonded to the chassis.
  4. A toroidal transformer instead of an EI, which has a far lower external field to start with.

Mu-metal has a manufacturing catch worth knowing: its permeability is destroyed by mechanical work, so it must be annealed after forming, and it cannot be bent or drilled afterwards without losing performance. Specify it as formed and annealed to final shape.

What to put on the drawing

  • Which faces are masked before finishing, and the width of the masked band
  • The finish on internal parts, conversion coating rather than anodize where conductivity matters
  • Fastener spacing along each seam, as a dimension rather than a count
  • Flatness on cover lips and mating faces
  • Perforation pattern with the individual hole size, not just the open area percentage
  • Earth stud location, thread, and the bare landing around it
  • Gasket groove dimensions if a gasket is used, and the gasket part number
  • Any conductive hardware: star washers, grounding washers, scraped landings

The honest summary

Most hi-fi does not need a Faraday cage. It needs a box whose seams are electrically continuous, whose transformer is far from the input stage, and whose ventilation is holes rather than long slots. Those three decisions are made on the drawing and cost nothing extra to manufacture, as long as somebody remembers that the beautiful black anodize on the outside must not be on the joint faces inside.

Frequently asked questions

Does anodizing stop a chassis from shielding?

Anodize is an electrical insulator, so an anodized joint face does not conduct. The box still shields as a set of plates, but the seams stop being electrically continuous, which is where leakage happens. Mask the joint faces and the earth boss before anodizing, or use a conductive chromate conversion coating on the mating surfaces instead.

Is aluminum or steel better for shielding?

For electric fields and high frequency, aluminum is fine and lighter. For low frequency magnetic fields, from a mains transformer for example, aluminum does almost nothing and you need a ferrous material, usually steel or mu-metal, close to the source. Most audio chassis need both: an aluminum box plus a local steel or mu-metal can around the transformer.

How big can a ventilation slot be before it leaks?

It is the longest dimension that matters, not the area. A long thin slot leaks far more than a round hole of the same open area, because the slot radiates like an antenna at frequencies whose half wavelength approaches its length. Many small round holes beat a few long slots for the same airflow.

Need a quote for this part?

Send the drawing. We match you with up to 3 shops and reply within 24 business hours.

Get matched with an audio parts shop

Keep reading

This guide is part of Audio Chassis Manufacturing: The Whole Picture, one of four sections in the guide library.