Is Heat Treatment for Metal a Special Process?

[arin_23]

Someone should tell the master sword-smiths making samurai swords for centuries that they need enclosed furnaces with calibrated thermocouples and the profile checked every 3 months or so and engineering tables to achieve the end result. And, golly, they better have records of those validation steps.

I've exhausted my argumentation. I think it is very useful that readers know very clearly where some of the covers (and possibly their employers) stand on this issue.

Dear Sidney,

I understand that you have added a bit of in your post and it is very relevant to think that the samurai sword smiths have achieved mastery in their production as far as the quality of the product is concerned.

But it is also for sure that while we talk about the mastery, we never talk about the rejections during the product realization process as the flip side. They used to prepare gem of a sword after may hits and trials.

Just let me share one more though with you, were there many competitors competing for a chunk of business??? Were there laid down specifications those days? IMO - no!!!!!

Just consider a modern age scenario where two weapon manufactures are supplying the same gun, one having a well validated process with less rejection and quality of product as conforming, and other having excellent product quality with more rejection figures due to lack of process conformity.....I would undoubtedly go with the first.

Regards,

Arin

 

[piney]

As in the Welding - Is welding always a special process per ISO9001:2000 Clause 7.5.2 thread, my conclusion is that all encompassing generalizations bring a degree of risk.

Any process where you cannot validate the end result without performing a destructive test is a special process. Welding and heat treating are included in this, no thinking outside of the box required.

 

[Sidney Vianna]

I understand that you have added a bit of in your post and it is very relevant to think that the samurai sword smiths have achieved mastery in their production as far as the quality of the product is concerned.

You missed my point. According to Mr. Grunfeld, heat treatment only happens in modern, thermo-profile verified industrial ovens. My example of samurai swordsmithing was to show the fact that heat treatment of metals has been in existence for millennia, preceding modern equipment.

Any process where you cannot validate the end result without performing a destructive test is a special process. Welding and heat treating are included in this, no thinking outside of the box required.

The piece about destructive testing is your own addition to the definition of "special processes" and unsupported by the ISO 9001 standard. Many times, you can validate processes using non-destructive techniques. I've already offered (what I believe are) examples of a welding and heat treating process that do not need to be validated. You are obviously entitled to disagree, but let me offer you this: In the sliding gate scenario, the hardening of the rail would be a definitive product enhancement, minimizing the wear on the rail, increasing it's life cycle. Now, you come as an auditor, observe the fact that the gate maker is doing this heat treatment and demand the process be validated. The gate maker will have to develop a process specification, make samples, have samples destructive and non-destructive tested, etc., all non-value adding steps in order to comply with your interpretation of the standard. Or the gate maker might say: forget this non-sense. I will no longer heat treat the rail, which will make the product inferior, but the system will now be in compliance with the auditor's interpretation of what special processes are. A typical example of how ISO 9001 mis-implemented and mis-audited goes against the goal of improvement.

 

[Jim Wynne]

You missed my point. According to Mr. Grunfeld, heat treatment only happens in modern, thermo-profile verified industrial ovens. My example of samurai swordsmithing was to show the fact that heat treatment of metals has been in existence for millennia, preceding modern equipment. The piece about destructive testing is your own addition to the definition of "special processes" and unsupported by the ISO 9001 standard. Many times, you can validate processes using non-destructive techniques. I've already offered (what I believe are) examples of a welding and heat treating process that do not need to be validated. You are obviously entitled to disagree, but let me offer you this: In the sliding gate scenario, the hardening of the rail would be a definitive product enhancement, minimizing the wear on the rail, increasing it's life cycle. Now, you come as an auditor, observe the fact that the gate maker is doing this heat treatment and demand the process be validated. The gate maker will have to develop a process specification, make samples, have samples destructive and non-destructive tested, etc., all non-value adding steps in order to comply with your interpretation of the standard. Or the gate maker might say: forget this non-sense. I will no longer heat treat the rail, which will make the product inferior, but the system will now be in compliance with the auditor's interpretation of what special processes are. A typical example of how ISO 9001 mis-implemented and mis-audited goes against the goal of improvement.

In addition to all of this, there's an assumption that "heat treating" is limited to processes intended to harden metals. Where does this leave a process such as the baking used to prevent hydrogen embrittlement of electroplated parts? It's not unusual at all to see customer drawings that just say (as part of a plating spec) "Bake for [x amount of time] at 400 degrees F within one hour of plating." Doing what the specification says to do fulfills the requirement.

The need for destructive testing isn't, by itself, a criterion either. If I choose to accept the financial burden of destructive testing of a statistically valid sample of heat-treated parts, I've done "subsequent monitoring and measurement," haven't I?

Validation of a heat treating process relieves the organization of the need for destructive testing, and is presented as an option to destructive testing, not a mandate against it.

 

[Big Jim]

The need for destructive testing isn't, by itself, a criterion either. If I choose to accept the financial burden of destructive testing of a statistically valid sample of heat-treated parts, I've done "subsequent monitoring and measurement," haven't I?

I believe you have not. You cannot sell the parts you have destructively tested. You have not monitored and measured what you are selling and that is why validation is needed.

 

[Jim Wynne]

I believe you have not. You cannot sell the parts you have destructively tested. You have not monitored and measured what you are selling and that is why validation is needed.

So you're saying that a statistically valid sample doesn't represent the lot? If that's the case, how can I ever validate anything? Why would you accept the idea of a statistically valid sample of almost anything else, such as machined parts?

 

[Big Jim]

So you're saying that a statistically valid sample doesn't represent the lot? If that's the case, how can I ever validate anything? Why would you accept the idea of a statistically valid sample of almost anything else, such as machined parts?

I'm certainly not saying that.

I'm saying that you have not performed "subsequent monitoring and measuring". If you could, then you would not need to validate. Validation is not "subsequent monitoring and measuring".

A statistically valid sample that confirms the product can be used as intended CAN BE PART OF VALIDATION.

 

[Jim Wynne]

I'm certainly not saying that.

I'm saying that you have not performed "subsequent monitoring and measuring". If you could, then you would not need to validate. Validation is not "subsequent monitoring and measuring".

A statistically valid sample that confirms the product can be used as intended CAN BE PART OF VALIDATION.

Define "subsequent" in this context. Or more to the point, subsequent to what? I think it means "after manufacture," (which includes secondaries like heat treating) so if I do legitimate lot sampling subsequent to heat treating, how is it that I have not performed subsequent monitoring and measuring?

ETA: Whether or not heat treating is a special process depends wholly on the specifications.

 

[Sidney Vianna]

ETA: Whether or not heat treating is a special process depends wholly on the specifications.

Stone him....

 

[Big Jim]

Define "subsequent" in this context. Or more to the point, subsequent to what? I think it means "after manufacture," (which includes secondaries like heat treating) so if I do legitimate lot sampling subsequent to heat treating, how is it that I have not performed subsequent monitoring and measuring?

ETA: Whether or not heat treating is a special process depends wholly on the specifications.

The context is right there in 7.5.2. No word-smithing will get around it. With your train of thought there would never be a need for validation of processes.