Nickel plated spring failures, hydrogen embrittlement, baking

A. Even such components as aircraft landing gear are electroplated, so the potential for hydrogen embrittlement is not a death sentence for a bolt. Nickel plating and baking is probably still a good choice. I haven't heard of making allowance for reduced load rating after plating, but an awful lot has been written on hydrogen embrittlement and I've read only a very tiny portion of it. Sounds like some study is warranted.

A. Hydrogen embrittlement can have disastrous results, but it can be controlled very well in most circumstances.

A prebake, mechanically cleaning (light blast with aluminum oxide), an anodic cleaning cycle, an absolute minimum exposure to the acid etch, plating with low stress sulfamate nickel, baking immediately after the final rinse, baking for a longer period and baking at a higher temp so long as it is below the draw temp after hardening will all reduce the probability of embrittlement.

My experience, zero notch bar failures, is keeping the acid dip to an absolute minimum.

A. I had encountered severe problem of embrittlement when I plated a 0.6 mm diameter phosphor bronze wire. Later I changed to anodic cleaning and Asymmetric A.C. plating. That completely eliminated the problem The cycle was 270 mili seconds plating and 30 mili seconds de-plating.

A. Treat the high strength steel fastener as any other high strength 'aerospace' steel and refer to a suitable electroless nickel plating specification such as MIL-C-26074 [on DLA]E and stress / embrittlement relieve as directed.

A. You might have some embrittlement. If so, the acid etch is about 95% of the problem. Try to use a periodic reverse and a special formulated alkaline cleaner to get the spring as clean (deoxidized) as possible. Then you can use a quick dip in the acid , rinse appropriately and then go into your plate. Cyanide works exceptionally well if you can stand it.

You also need to bake as soon as possible after plate, not when it is convenient or economical.

If you do not use the nickel, you probably would have to use more gold.

A. Have you checked the stresses in your sulfamate nickel coating? For your application I would think you would want compressive stresses in the coating. With compressive stress, when a crack happens in the coating it pulls it together. While tensile stress pushes it open. Are you using nickel chloride for anode corrosion? Chlorides seem to promote tensile stress. Do you monitor the chloride concentration? The stresses can be checked with a spiral contractometer.

A. This gets back to the theory of hydrogen being "baked out" of the plating. Certainly at this thickness the hydrogen isn't diffused "out" of the substrate. Actually it diffuses to "hydrogen traps" where further migration doesn't occur.

This really is a question for an "electroformer"; however, I'd say the length of bake doesn't increase with the plated layer thickness. It's the substrate you are concerned with and the amount of hydrogen in the base metal is dictated by the pre-plate processing and the initial nickel deposit. Whether it's 1 mil or 100 mils, the amount of hydrogen in the surface should be the same.

A. Check AMS or MIL std's for specific information. Times and temperatures are usually based on tensile strength (or hardness) of the base material. Also, application is a consideration. If the hardware is used for a braze repair application, embrittlement relief is not usually required, however at the thicknesses you give, it doesn't look like a braze repair. Time-at-temperature requirements that are thickness related are also usually given in your FAA Standard Practices or elsewhere in the engine service manuals. Also, be mindful that the hardware needs to be baked after strip and before re-plate if you need to rework and that the time between removal from the plating bath and start of baking is important to minimize.

A. Bob I agree with Doug. From tank to bake is defined by some aircraft mfg. Typical I recall we baked for Boeing or Douglas at 375 °F, some parts for 4 hours others for 24. Some within eight hours, some within two hours. I guess you are using Sulfamate nickel. What is most important is the type of metal you are plating to. This will dictate the "rush to bake" and "length of bake". Either case, no offense intended but your best source is going to be from the mfgr. of the part and not this forum where opinions vary and a consensus can be deadly.

Jon, indeed a peek is worth a thousand finesses, and a fact is worth a thousand opinions. So, if there is a spec, that's what people need to go by. However...

It was just a few years ago that -- despite specs -- the fastener manufacturers, zinc platers, and ASTM were in chaos over hydrogen embrittlement relief because people were leaving open the question of responsibility ... so the platers felt: "They told us to plate the parts, and that's all they paid us for; if the fasteners are hardened steel which needs baking, the manufacturer should have told us to do it!". Meanwhile, the fastener manufacturers were saying "What?! How are we supposed to expect that platers are embrittling our parts, doing nothing to fix it it, and not even telling us?".

So I submit that one thing more deadly than an errant consensus is lack of discussion :-)  
I like to optimistically think that if those fastener manufacturers and platers had come to this forum to talk about their expectations, that the readers just might possibly have warned about the problem so that it could be properly resolved by a spec clarification without all the problems that people encountered because they didn't talk publicly about it :-)

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey
2002

A. Hi Jim. Although nickel plating is more "efficient" (generates less hydrogen) than some other platings like conventional hexavalent chrome plating, nickel plating (and possibly the pretreatment for it) and trivalent chrome plating definitely generate enough that baking is required for hardened steel parts. Baking is not required for mild steel. Good luck.

Regards,

Thanks Ted.