Key takeaways
- Hydrogen embrittlement is a delayed failure that can occur after plating, pickling or cleaning of high strength wire.
- The risk rises with wire strength and hardness, so a small spring temper part is more exposed than a soft one.
- A bake after plating is a process control, not a cosmetic step, and the delay before baking matters.
- Mechanical plating and zinc flake coatings avoid the electroplating hydrogen path, but they still need a specification.
What hydrogen embrittlement is
Hydrogen embrittlement is a loss of ductility caused by hydrogen absorbed into a high strength steel. The hydrogen collects at grain boundaries, inclusions and stress concentrations, and it can start a crack that grows slowly under a load the part would normally carry. The failure is often delayed: the spring passes a test, sits in storage or in service, and breaks hours or days later.
The fracture is usually brittle, with little visible deformation, so it is easy to confuse with an overload or a material defect. The history matters: if the part was pickled, electroplated or cleaned in an acid bath, and it broke below its design load, hydrogen is a candidate.
Hydrogen can also enter during service, for example from corrosion or cathodic protection. That is a slower route, but it matters for a spring in a wet, galvanic environment. The process controls at the plater address the manufacturing route, not the whole life of the part.
Which springs are at risk
The risk rises with the strength and hardness of the steel. A high carbon music wire at spring temper, a cold drawn stainless spring wire and a high strength alloy spring are all more exposed than a soft, low carbon part.
The stress state matters as well. A spring that is formed to a tight radius, or that carries a high residual stress from coiling, has less margin. A part that is already close to its allowable stress is less able to tolerate a hydrogen-induced crack.
Geometry adds another factor. A sharp change of section, a tool mark, a scratch or a cut end concentrates stress and gives a crack a place to start. A good surface and a generous radius reduce the risk but do not remove the need for process control.
Where hydrogen enters
The most common route is electroplating. In an acid or alkaline plating bath, hydrogen can be generated at the cathode and absorbed into the steel. Zinc, nickel and zinc-nickel plating are all used on springs, and all can introduce hydrogen if the process is not controlled.
Pickling and acid cleaning are another route. A spring that is descaled or cleaned in an acid bath before plating can absorb hydrogen before the plating starts. The acid concentration, the time and the temperature all affect the amount absorbed.
Corrosion in service is a third route. A galvanic couple, a scratch through a coating or a wet environment can generate hydrogen at the surface. That is why a coated spring still needs a sound surface, not just a bake at the end of the line.
Baking: the control
Baking is a controlled heat treatment after plating that allows hydrogen to diffuse out of the steel. It is not a guarantee. The effectiveness depends on the temperature, the time, the delay between plating and baking, the coating thickness and the strength of the steel. A thick coating can slow the diffusion and trap hydrogen at the interface.
The delay matters. Hydrogen moves slowly through steel, and a part that sits for a day before baking may have hydrogen deeper in the section than one baked immediately. The process specification should state the maximum delay, not just the bake time.
The temperature has to be high enough to move the hydrogen and low enough not to soften the spring. For a spring temper wire, that window can be narrow. Agree the bake schedule with the plater and the wire supplier, and record the furnace chart.
Coating routes with lower risk
Mechanical plating deposits a zinc or zinc alloy coating by tumbling the parts with glass beads and a metal powder. It avoids the electrolytic hydrogen generation of electroplating, so it is often preferred for high strength springs. It still needs a specification for thickness, adhesion and corrosion performance.
Zinc flake coatings are applied from a dispersion and cured. They avoid electroplating and can give good corrosion resistance, but the cure temperature and the coating thickness need to suit the spring. A thick coating can bridge a coil gap and change the solid height.
Passivation is the usual finish for stainless springs. It removes free iron and helps the oxide layer form, and it does not involve the same hydrogen path as electroplating. It does not add a sacrificial coating, so it is not a substitute for zinc on a carbon steel spring.
| Route | Hydrogen risk | What to specify | What to record |
|---|---|---|---|
| Acid pickle before plating | High | Acid type, concentration, time and rinse | Bath log and rinse record |
| Electroplated zinc or nickel | High | Bake temperature, time and maximum delay | Furnace chart and plating batch |
| Mechanical plating | Low | Coating thickness, adhesion and coverage | Coating weight and adhesion result |
| Zinc flake coating | Low | Cure temperature, thickness and coverage | Cure record and thickness check |
Illustrative risk ranking. The actual risk depends on the steel strength, the coating thickness and the process control.
What to put in the purchase order
The purchase order should name the coating, the standard, the thickness and the bake requirement. It should also state the maximum delay between plating and baking, the acceptance test and the records that must come with the batch. Without those, the plater will use a house process that may or may not fit a high strength spring.
- Wire grade, temper and tensile strength range
- Coating type, standard and minimum thickness
- Bake temperature, time and maximum delay after plating
- Cleaning and pickling controls before plating
- Acceptance test and sample size
- Records: bath log, furnace chart and coating thickness
- Handling and packing to avoid scratches after coating
Testing and verification
A bend test or a sustained load test can show whether a batch is embrittled, but the result depends on the test method. A sustained load test holds the part at a stress below its yield and checks for failure. It is a screen, not a proof of infinite life.
For a critical spring, agree the test with the supplier and the plater before the first batch. Ask whether the test is on the finished part or on a test coupon, and whether the coupon is from the same heat and the same coating batch. A coupon that is not representative proves very little.
Keep the process records with the test result. If a later batch fails, the records show whether the bake, the coating thickness or the cleaning step changed. That is the difference between a correction and a guess.
A practical route
Start by asking whether the spring needs a coating. A stainless spring in a dry environment may need only passivation. If a carbon steel spring needs a coating, choose a route with a lower hydrogen risk where the corrosion duty allows, and control the bake where electroplating is the only option.
Then write the process into the purchase order, not just the drawing. The drawing says what the part must be. The purchase order says how the coating and the bake are controlled, and what evidence comes with the batch.
Frequently asked questions
How long after plating should a spring be baked?
The safe delay depends on the coating, the steel strength and the process. Many specifications require baking within a few hours of plating, and some require it sooner. Agree the maximum delay with the plater and put it in the purchase order.
Does baking remove all the hydrogen?
It reduces the absorbed hydrogen, but it is not a guarantee of complete removal. The effectiveness depends on the temperature, the time, the delay and the coating. A sound process also avoids unnecessary pickling and chooses a lower risk coating where possible.
Are stainless springs immune to hydrogen embrittlement?
No. High strength stainless spring wire can absorb hydrogen, especially during pickling or electroplating. The risk is generally lower than for high carbon steel, but it is not zero, and the process controls still apply.
What test proves a batch is not embrittled?
A sustained load test or a bend test can screen a batch, but no short test proves infinite life. Use a test that represents the finished part and the coating batch, and keep the process records so a failure can be traced to a change.
Sources and further reading
- ISO 4042, Fasteners, electroplated coating systems
- ASTM B850, Standard guide for post-coating treatments of steel for reducing the risk of hydrogen embrittlement
- Qumek spring coating and bake control notes
Qumek Engineering Desk
Engineering team
The engineering team behind Qumek drawings, DFM notes, tolerances and inspection standards. Every guide is reviewed against the drawings and processes used on live programmes.
Specifying a coating on a high strength spring?
Send the wire grade, the strength and the environment. The engineering desk will set out the coating routes and the bake controls to put in the purchase order.
Send the coating question