Which finishes grow zinc whiskers, and what should electronics enclosures specify instead?

What are zinc whiskers, and how do you recognize the problem?

Zinc whiskers are straight or slightly kinked metal filaments, typically a few microns thick and up to a few millimeters long, that sprout from a zinc-coated surface with no outside cause. The problem usually shows up first as unexplained, intermittent electrical failures, not as a visible defect on the part.

NASA’s whisker research group describes metal whiskers as having typical diameters of a few microns and lengths that reach several millimeters, in rare cases more than 10 mm (NASA NEPP, basic info on tin whiskers). That size is the trouble. A whisker is too thin to see in ordinary light, conducts like the metal it is, and is long enough to bridge closely spaced conductors on a circuit board, a connector or a power supply.

The classic case is the raised computer-room floor. NASA Goddard documented whiskers, some longer than 2 mm, on the zinc electroplated steel underside of floor tiles in a computer room more than two decades old, and reported that whiskers shed during tile removal ended up inside servers and routers and caused short circuits (NASA Goddard, zinc whiskers on floor tiles). The same failure pattern applies to any zinc-plated sheet metal near electronics: enclosure panels, chassis, mounting brackets, cable trays, bus rails, DIN rails and connector shells.

The symptoms that point at whiskers rather than a design fault are distinctive. Failures cluster after construction, cabling work or floor-tile lifting, and they hit unrelated equipment that shares the same cooling air. NASA’s zinc whisker awareness briefing explains why boards often come back “no fault found”: a whisker can carry tens of milliamperes before it melts, so where more current is available the short is intermittent and the evidence burns away, while low-current circuits can be shorted permanently. Its library data center example, traced to a floor installed about 20 years earlier, lists 10 catastrophic server failures in two to three months, recorded as power supply shorts, and about 12 router failures in 18 months (Brusse, NASA Goddard, 2003).

What makes a zinc coating grow whiskers?

Compressive stress in the coating is the leading explanation, and electroplated zinc is where most failures have been recorded. Bright, additive-rich plating baths and extra stress from forming, bending, clamping or fastening after plating raise the risk.

No growth mechanism is confirmed. NASA’s whisker group, writing mainly about tin, names relief of compressive stress within the plating as the leading theory, and lists residual plating stress, externally applied compression, bending, scratches and thermal expansion mismatch among the contributing factors (NASA NEPP); its zinc briefing says the same mechanism may apply to zinc. For zinc plating specifically, a finishing column by Frank Altmayer of the NASF Foundation names residual stress in the plated coating, compressive stress applied to the part and thermal expansion mismatch with the substrate (Altmayer, 2025). On a zinc-coated part, the drivers to look for are:

1. Electroplated zinc in general. Electroplated zinc on steel is the coating in NASA’s floor tile, bus rail and connector shell cases. Hot-dip coatings are generally treated as lower risk, but the VTT study found whiskers on hot-dip zinc too, mostly in dull areas with fine grains and compressive stress.
2. Bright plating chemistry. The same VTT study notes that brighteners produce the small crystals behind a shiny zinc plate, and it links small zinc crystals to whisker growth. Altmayer adds that baths carrying more organics, organic breakdown products or metal contamination give more stressed coatings.
3. Stress applied after plating. Bending, forming, clamping or torquing a plated part adds external compressive stress, which both NASA and Altmayer list as a contributor.
4. Time. NASA’s zinc briefing describes an unpredictable incubation period of months or even years, then growth of typically less than 1 mm per year, so a part that passes inspection at receipt can fail years later.
5. Disturbance. Whiskers are harmless while attached. Lifting tiles, pulling cables, vacuuming or vibration releases them into the airflow.

Chromate and trivalent passivates are thin conversion films for corrosion resistance, and the record shows they do not prevent whiskers: NASA’s bus rail and connector shell had yellow chromate, and the VTT samples had a blue passivate.

What should electronics enclosures specify instead of zinc plating?

Specify a finish with no electroplated zinc on any surface the cooling air can reach: painted or powder-coated steel over a phosphate pretreatment, anodized or chromate conversion coated aluminum, or unplated stainless steel. If zinc-coated steel is unavoidable, use hot-dip material and overcoat every surface.

Choose by the job the finish has to do, not by habit. The decision usually comes down to four cases:

• Painted steel enclosure or chassis. Start from bare cold-rolled steel, phosphate it and powder coat it. This removes the metallic zinc layer entirely. If the design also needs corrosion protection at cut edges, discuss a zinc-rich or hot-dip base with full overcoat as a conscious exception; the powder coat thickness guide covers what to specify for coverage.
• Grounding and bonding surfaces. Where a bare conductive surface is required, aluminum with a chromate conversion coating or stainless steel gives conductivity without zinc. Mask the bonding pads on a painted part rather than leaving them zinc plated.
• Brackets, rails and small hardware inside the box. Stainless steel or aluminum parts avoid the question. Zinc-plated screws are common; if they are in the airflow, ask whether a stainless screw costs less than the risk.
• Structure outside the equipment. Floor pedestals, cable trays and frames away from air intakes can use hot-dip galvanizing, which has a smaller whisker record than electroplate, as explained in the zinc plating vs galvanizing guide.

Zinc-nickel is often proposed as a drop-in upgrade because of its corrosion performance, covered in the zinc-nickel vs zinc plating guide. Altmayer reports that researchers tend to agree alloy plating helps avoid or inhibit whiskers, but published whisker evidence is much thinner for zinc-nickel than for pure zinc, so for a part that sits inside electronics, treat it as unproven unless the plater can show test data. Tin and cadmium plating are not alternatives: NASA lists both among the metals that grow whiskers.

How is the finish handled on a Saola order?

Machined parts are cut at our lead CNC plant, while stamping and all surface finishing are done by partner plants we manage, and the quote names the production site and country of origin for each. Your drawing’s finish callout and any zinc prohibition are passed to the finishing partner as written.

Machined enclosure parts, heat sinks, brackets and covers are made by CNC milling and turning at our lead CNC plant in Vietnam. Formed sheet parts come through stamping at partner plants we manage. The surface finishing menu at our partner plants includes powder coating, anodizing and electroplating, including zinc plating with trivalent chromate. Blue trivalent chromate is the topcoat quoted when a drawing asks for zinc plating without naming one, which is exactly the whisker-prone finish this guide warns about, so a part going into electronics needs the finish and the prohibition written on the drawing. Hot-dip galvanizing and zinc-nickel are not on that menu, so name them on the RFQ if you need them and they will be quoted separately. List the finish documentation you need on the RFQ so it is quoted with the parts. ISO 9001 certification is in progress, not held. Send the drawing with the finish notes through the quote form.