Nickel Plating Machined Parts: Tolerance & Hydrogen

Nickel plating a machined part is harder than plating a formed or cast one, and the reason is structural rather than chemical. A machined part is defined by its features — bores, threads, keyways, blind holes, tight tolerances, fine surface finish — and those are precisely the things a plating tank handles worst.

Three failures account for most of the trouble: contamination that survives the pre-treatment because it is hiding in a hole, a dimension that moves out of tolerance because nobody counted the coating, and embrittlement in a high-strength steel that was never baked. This article takes them in turn.

Why machined parts behave differently in a tank

Four properties of a machined part drive everything that follows:

• Machined surfaces carry cutting fluid, not just oil. Cutting fluids are formulated to do a job — extreme-pressure additives containing sulfur, chlorine or phosphorus, and often fatty compounds. Some of those residues are not merely hard to clean; sulfur and chlorine species that carry into a nickel bath act as contaminants that affect the deposit.
• The features are closed. A blind hole or a deep bore traps solution, traps gas, and exchanges slowly with the bulk. Pre-treatment chemistry that reaches an open surface may never properly reach the bottom of a 6 : 1 bore.
• The tolerances are stated in microns. A machined drawing tolerances a bore to a few tens of microns. A nickel deposit is measured in the same units, so the coating is no longer negligible — it is a design input.
• The material may be at risk of embrittlement. Machined parts are frequently made from heat-treated or high-strength steels, and those are the grades where hydrogen is a real hazard rather than a theoretical one.

Pre-treatment: where machined parts are actually lost

Most plating defects on machined parts trace back to cleaning, and most cleaning failures trace back to geometry.

Blind holes and deep bores

A blind hole is a trap for three things: trapped air, trapped solution, and trapped contamination. Each creates a different problem.

The single most useful thing a designer can do here costs nothing: tell the plater what the part does and how it is machined. A drawing that shows a deep bore but not the cutting fluid, the heat treatment or the required finish leaves the plater to guess, and the guess is usually wrong in a way that only appears after the parts are scrapped.

Threads, keyways and sharp edges

Nickel builds up on high-current-density features — edges, corners and thread crests. On a machined part, those are often the features with a tolerance. A thread that has been nickel plated is not the same thread: the pitch diameter grows, the crest rounds, and a nut that was specified to fit may no longer. Internal threads are worse, because the deposit inside a tapped hole is hard to measure after the fact.

Holding dimensional tolerance

Every surface the solution reaches gets coated. That is the whole point, and it is also why a plated machined part does not measure like an unplated one.

The arithmetic that catches people out is the factor of two:

• A bore gets smaller by twice the coating thickness. The coating is on both sides of the diameter.
• A shaft gets larger by twice the coating thickness, for the same reason.

A 20 µm nickel deposit therefore moves a diameter by 40 µm — more than the whole tolerance band on many machined fits. And if both mating parts are plated, both move.

Two consequences follow. First, the coating thickness has to be settled before the part is machined, not after, because it determines the pre-plate dimensions. Second, thickness is not perfectly uniform — it is thicker on edges and thinner in recesses and inside bores — so a tolerance calculation that assumes a single exact number is optimistic. Specify a minimum thickness and a tolerance on the fit that can absorb the variation.

Choosing the nickel thickness

Nickel is used on machined parts for three different reasons, and they call for different thicknesses:

• Appearance — a bright or satin finish. Decorative nickel is thin, and its performance depends on the copper or nickel underlayers beneath it.
• Corrosion resistance — nickel is a barrier coating and it is cathodic to steel, meaning that where the coating is breached or porous, the steel beneath corrodes preferentially and the attack is concentrated rather than spread. Thicker, denser nickel with fewer pores is better; a thin porous deposit can be worse than none.
• Wear and dimensional build-up — engineering nickel used to restore or protect a surface. Here thickness is chosen for the wear requirement, and the tolerance consequence is designed in from the start.

Because the corrosion mechanism is porosity-driven, the process details matter as much as the thickness number: filtration, bath cleanliness and current density all affect how many pores the deposit has. A barrel or rack line running with tired filtration will not produce the same coating as the same line running clean, even at the same thickness. See filtration systems for what that equipment actually does.

Hydrogen embrittlement

This is the failure mode that ends in a broken part rather than a rejected one, and it deserves more attention than it usually gets on a plating purchase order.

Hydrogen is generated at the cathode during plating, and during the acid cleaning and pickling stages before it. On most steels it diffuses back out and does no harm. On high-strength steels it does not: it collects at stress concentrations, and the part can crack and fail suddenly under load, sometimes days or weeks after it was put into service. Failures of this kind are brittle, occur below the material's normal yield, and carry no warning.

The risk rises with strength. The commonly used threshold is around 1000 MPa tensile strength, roughly 32 HRC — above that, hydrogen embrittlement is treated as a genuine hazard rather than a possibility, and the risk continues to climb as strength increases.

Baking

The countermeasure is a bake: heating the part after plating so the hydrogen can diffuse out before it does damage. The details matter, and the ones that matter most are the ones most often left off a purchase order:

• Baking must start promptly after plating. The standard practice for high-strength parts is a short window, measured in hours, between the end of plating and the start of the bake. A part that sits overnight has already had time to do what the bake was meant to prevent.
• Baking is a separate process step, not a warm dry. Temperature and time are both specified; the numbers depend on the steel's strength and on the customer's requirement.
• It applies even without plating current. Acid pickling and electrolytic cleaning generate hydrogen too, so a bake may be required on parts that were stripped, cleaned or pickled even if the plating itself was short.
• The specification comes from the drawing. Where a customer specifies hydrogen embrittlement relief, that requirement is normally written as a standard practice rather than invented at the plating shop. Ask what your drawing calls out and plate to it — do not let a supplier assume a general-purpose bake covers a high-strength part.
• The deposit choice interacts with it. Some coatings are more effective hydrogen barriers than others, which is one reason the coating decision and the embrittlement decision are not independent.

If you are buying plated machined parts rather than plating them, hydrogen embrittlement relief is the requirement to put in writing, with the steel grade and its strength, and to ask for evidence of. It is invisible in the finished part and in every visual inspection you can perform.

Masking and selective plating

Machined parts frequently need plating on some surfaces and not others: a press fit, a sealing face, a thread, a bearing seat. Masking is the usual answer, and it is worth designing for:

• Mask the features whose tolerance cannot absorb the coating. It is usually cheaper than compensating in the machining.
• Tell the plater which surfaces are masked and why. A mask specified as "protect the thread" is applied differently from one specified to hold a pitch diameter tolerance.
• Check that the mask survives the process. Some masking materials do not tolerate the temperature or chemistry of a particular bath, and a mask that fails mid-process is not visible until after the fact.

What to check on receipt

For plated machined parts, the useful inspection list is short and specific:

• Thickness on a critical surface, not on a convenient one. A reading taken on an accessible flat face says little about the inside of a bore.
• The fit that matters. Gauge or measure the feature whose tolerance drove the specification.
• Adhesion, on a sample. Blistering and peeling are usually a pre-treatment failure and usually show up first on the parts with the most difficult geometry.
• Coverage inside bores and blind holes, on a sectioned sample if the part is important. This is the defect a visual inspection will never find.
• Bake records, where hydrogen embrittlement relief was specified.

When the part should be barrel plated instead

Machined parts are frequently small and produced in volume — fasteners, turned fittings, small components off a Swiss lathe. Where the quantity is high and the part will tolerate tumbling, barrel plating removes the racking labour entirely and is usually the cheaper route. The trade-off is on the other side of this article: a barrel gives a wider thickness spread across the load and leaves tumbling marks, which is exactly what a tight machined tolerance cannot absorb. That decision is set out in nickel barrel plating.

For the line itself — tank configuration, anode selection, rectifier sizing and the wastewater stages a nickel line needs — see nickel plating equipment.

Frequently asked questions

How much does nickel plating change a bore diameter?

By twice the coating thickness, because the deposit is on both sides of the diameter. A 20 µm deposit reduces a bore by 40 µm. Where the bore is a fit, specify it oversize before plating and state a minimum coating thickness rather than a nominal one.

Do I need to bake machined parts after nickel plating?

It depends on the steel's strength. Above roughly 1000 MPa tensile strength, hydrogen embrittlement relief is standard practice and should be written into the specification. Acid cleaning and pickling also generate hydrogen, so the requirement is not limited to heavily plated parts. The exact bake is normally defined by the customer's drawing or by a referenced standard practice — confirm what applies rather than assuming a general-purpose bake.

Why do plated machined parts blister more than formed parts?

Geometry. Cutting fluid and contamination survive in blind holes, deep bores and the gaps around threads, where cleaning chemistry exchanges slowly. The parts that blister are usually the ones with the most difficult features, which is the signature of a pre-treatment problem rather than a bath problem.

Can threads be nickel plated?

Yes, but not to the same class. The pitch diameter grows and the crests round, so a standard thread class may no longer assemble after plating. Either start from a pre-plate class with allowance, or mask the thread.

Is thicker nickel always better for corrosion resistance?

Not on its own. Nickel is cathodic to steel, so corrosion at pores is concentrated on the steel rather than spread across the surface. A dense, low-porosity deposit is the goal; thickness without control of filtration, bath cleanliness and current density can still leave a porous coating. A very thin porous deposit can be worse than no coating at all.

How thick should engineering nickel be on a machined part?

It follows from what the coating is for — appearance, corrosion resistance, or dimensional build-up for wear — and from the tolerance the fit can absorb. Because the effect on a diameter is twice the thickness, the thickness decision and the pre-plate machining dimensions have to be made together, not in sequence.