This subject has bothered me for some time and I would like some input on it. At my previous position I had to help my supervisor with a PPAP for 4 new parts. The process of said parts were built on a robomac. The Robomac would bend a straight rod into the part we needed. All of the contour fixtures were checked within tolerences and all samples measured within tolerence. However these parts then were sent off to get heat treated. Once this process was complete, all parts measured out of tolerence and would not fit any fixture due to a .25" bow.

Now heres the part that bothers me...

We decided to remeasure these parts and all tolerences were extremly out. However the part would still work during assembly to it's coresponding part. Once my supervisor informed the plant manager about this, the plant manager and owner decided to look into this. They came up with the idea that all parts would be sent directly to the customer right after being heat treated and told us to use our previous numbers.

I admit that I have very little QC experience (3yrs) but I'm purrty sure this was wrong. Does this type of thing happen often with other companies? I'm concered because it's part of a seat assembly and I would hate to see something on the news about a baby being propelled out of a car during an accident because of faulty parts.

This subject has bothered me for some time and I would like some input on it. At my previous position I had to help my supervisor with a PPAP for 4 new parts. The process of said parts were built on a robomac. The Robomac would bend a straight rod into the part we needed. All of the contour fixtures were checked within tolerences and all samples measured within tolerence. However these parts then were sent off to get heat treated. Once this process was complete, all parts measured out of tolerence and would not fit any fixture due to a .25" bow.

Now heres the part that bothers me...

We decided to remeasure these parts and all tolerences were extremly out. However the part would still work during assembly to it's coresponding part. Once my supervisor informed the plant manager about this, the plant manager and owner decided to look into this. They came up with the idea that all parts would be sent directly to the customer right after being heat treated and told us to use our previous numbers.

I admit that I have very little QC experience (3yrs) but I'm purrty sure this was wrong. Does this type of thing happen often with other companies? I'm concered because it's part of a seat assembly and I would hate to see something on the news about a baby being propelled out of a car during an accident because of faulty parts.In my opinion (not backed by specific citation to any published material), the proper, ethical course of action would have been to discuss the situation frankly with the customer. Distortion of parts in heat treating is common. It is a particular problem for cold formed parts which tend to "relax."

It would seem the most equable solution would be a "function" test rather than specific measurement. You don't indicate the diameter of the shaft, but I get the impression it may be small enough to deform to fit during assembly. The question is whether the stress ("spring") will affect the final assembly function.

This subject has bothered me for some time and I would like some input on it. At my previous position I had to help my supervisor with a PPAP for 4 new parts. The process of said parts were built on a robomac. The Robomac would bend a straight rod into the part we needed. All of the contour fixtures were checked within tolerences and all samples measured within tolerence. However these parts then were sent off to get heat treated. Once this process was complete, all parts measured out of tolerence and would not fit any fixture due to a .25" bow.

Now heres the part that bothers me...

We decided to remeasure these parts and all tolerences were extremly out. However the part would still work during assembly to it's coresponding part. Once my supervisor informed the plant manager about this, the plant manager and owner decided to look into this. They came up with the idea that all parts would be sent directly to the customer right after being heat treated and told us to use our previous numbers.

I admit that I have very little QC experience (3yrs) but I'm purrty sure this was wrong. Does this type of thing happen often with other companies? I'm concered because it's part of a seat assembly and I would hate to see something on the news about a baby being propelled out of a car during an accident because of faulty parts.

This is a variation on the old "Just ship 'em and cross your fingers" strategy, one that's commonplace in job shops. It's clear that the chain of failures reached a predictable conclusion:

Failure #1: The designer failed to take basic material properties into account, or did take them into account and assumed that whoever made the part would recognize the issue and deal with it.

Failure #2: Your company failed to predict the problem in the contract review process. As Wes pointed out, post-heat treating distortion is common and should be expected.

Failure #3: After what should have been predicted happens, management resorts to prevarication and blame-shifting rather than actually dealing with the problem.

The result of all of this, which is repeated ad infinitum in job shops everywhere, is that the quality manager and regrettably, even inspectors, are put on the horns of a nasty ethical dilemma: Tell the truth and risk getting pounded over the head, or acquiesce and become an accessory to whatever bad things happen as a result of the nonconformance.

As for the case at hand, you mention PPAP. Normally a PPAP submission will include a process flow diagram, PFMEA and control plan. The flow diagram should show the parts going to the heat treater and either coming back to you or going directly to the customer. In either event, the PFMEA document should include evidence of (A) the verification process for heat treated parts after they're returned to you (and appraisal of the risks associated with heat treating) or (B) the risks associated with shipping the parts to the customer directly from the heat treater. Associated controls, in either case, should be included in the control plan. What does your PPAP documentation say?

If you're sure the parts are functional, that means that the specifications aren't. A functional gage is the best solution in that case, but you have to understand that the customer, before approving a PPAP submission, is probably going to measure the parts. The PPAP will then be dead in the water, and the problem will have to be dealt with. The question is, why wait for something bad to happen? In my experience, it's much better to apprise the customer of the reality of the situation, and suggest functional gaging. Your company accepted the specifications, so you're going to be stuck with supplying parts that meet them, even if that includes the unrecompensed expense of a straightening operation. Your bosses are doing themselves no favors by sending out time bombs.

Jim,

Your advise is very sound. :applause:

Thank you Jim and Wes. =) I had the same thoughts during all of this and it really bugged me until I read your replies. I no longer work at this company but I do wish them the best of luck. No effort was made on the owners part to notify said company of this problem. I guess this was due to the failed deadline. The PPAP however was still submitted as the parts would be returned to us however this is not the case. It is my thoughts that they will have to make changes to their PPAP and resubmit more sample data.