Spring Steel Hydrogen Embrittlement

[silentrunning]

There seems to be some disconnect with our platers and ASTM. The platers feel that 3 hours minimum at 350 degrees F should be adequate for spring steel with a Rockwell of 43 C scale. ASTM B850 recommends a minimum of 16 hours at 374 degrees to 428 degrees F. That is a very LARGE difference. The more I am learning, the more I see that this could be a major problem. I would like to solve this once and for all because a lot of our business is steel springs that are plated. :mg:

 

[Golfman25]

There seems to be some disconnect with our platers and ASTM. The platers feel that 3 hours minimum at 350 degrees F should be adequate for spring steel with a Rockwell of 43 C scale. ASTM B850 recommends a minimum of 16 hours at 374 degrees to 428 degrees F. That is a very LARGE difference. The more I am learning, the more I see that this could be a major problem. I would like to solve this once and for all because a lot of our business is steel springs that are plated. :mg:

Yes but, I assume it was working fine before. So what changed? If the plater is doing the same thing as in the past, it sounds like the material might be suspect.

Try a different batch of material. Get the original tested. You may find the material was different. Or it is ok, and the plater needs to change his process. Good luck.

 

[silentrunning]

Golfman, I have sent raw material, heat treated parts with no plating and plated parts to a lab for analysis. I hope they can give me a definitive answer. I am worried that this may be systemic though. The platers say nothing has changed in their processing. In my 40 years with the company we have never had a reject because of Hydrogen Embrittlement and now I want to make sure we go 40 more years with no problem.

 

[Golfman25]

Steel is an interesting animal. Can run without issue for years. Then, seemingly out of nowhere, defects can arise. We once had a part which we put a little embossed bubble into it. No problem for years. Then all of the sudden, cracking. Testing, certs, etc. showed no issues. We eventually found out that there may have been an issue when the mill annealed the material -- it may have gotten stuck in the furnace for an extended time. Good luck.

 

[silentrunning]

Update on Hydrogen Embrittlement

We have finally solved our problem with breaking parts and as is often the case, there was not a single Root Cause. The main culprit did however turn out to be process change. The customer had requested that we go with a NADCAP heat treater instead of the company we had been using for several years. This started the problem. The customer print did not call out any method for the heat treat, only the final Rockwell. The new heat treater used a quench and draw method and the Rockwell came out within range. The parts however had a lot of scale on them when they were finished. The scale caused the plater to have to leave the parts in the acid for a longer time. After much investigation we went back to the old heat treater and found that he was Austempering the parts when he did them for us. This resulted in a much cleaner finished product as well as it being less prone to Hydrogen Embrittlement. We have had two lots heated by the old heat treater and then plated by the same plater. After testing parts from both lots under load for 24 hours we have had no failure. Now I have to study up on heat treating so I can write a factual Corrective Action.

 

[Golfman25]

Re: Update on Hydrogen Embrittlement

We have finally solved our problem with breaking parts and as is often the case, there was not a single Root Cause. The main culprit did however turn out to be process change. The customer had requested that we go with a NADCAP heat treater instead of the company we had been using for several years. This started the problem. The customer print did not call out any method for the heat treat, only the final Rockwell. The new heat treater used a quench and draw method and the Rockwell came out within range. The parts however had a lot of scale on them when they were finished. The scale caused the plater to have to leave the parts in the acid for a longer time. After much investigation we went back to the old heat treater and found that he was Austempering the parts when he did them for us. This resulted in a much cleaner finished product as well as it being less prone to Hydrogen Embrittlement. We have had two lots heated by the old heat treater and then plated by the same plater. After testing parts from both lots under load for 24 hours we have had no failure. Now I have to study up on heat treating so I can write a factual Corrective Action.

Root cause was the customer's request.

 

[Claes Gefvenberg]

Re: Update on Hydrogen Embrittlement

The main culprit did however turn out to be process change.

Which is usually the case, when a problem pops up in an established process. Finding it is often another matter though. Well done.

 

[DRAMMAN]

Re: Update on Hydrogen Embrittlement

Now you nee dot ask yourself how do you go about qualifying changes. Even though it was a customer request, a change like this should go through an approval evaluation.

 

[silentrunning]

Dramman, right you are. Now I have to do a technical Corrective Action to satisfy the customer and also an Internal Corrective Action because I failed to validate a process in accordance with 7.5.2.

 

[DRAMMAN]

I have dealt with hydrogen embritlement a few times. It was always a mystery as everyone always states the process did not change. I never thought to look at material processing or heat treating changes making the materal more susecptible to HE. We woudld always focus in on the post plate baking process. That whole 3-4hrs is a widely used rule of thumb. I always wonde where it came from and how it can possibly apply to everything.