Change Prints to accommodate Worn Tooling - There has to be a better way!

[RosieA]

As a corporate SQE, I'm involved in a project with one of our plants and a stamping manufacturer.

We recently moved tooling to a new supplier. We are in the process of doing first articles on 80+ part numbers.

It's my company's process for evaluating these first articles that I am concerned about.

The tooling had been with another supplier for many years. The plant I'm working with, did not do receiving inspection, so there is no record of how well the old supplier was meeting print. They were functionally working.

The first articles from the new supplpier are showing a lot of worn tooling that is not meeting print. A decision was made to accept the product with deviation and change the prints to meet what the tooling could produce, as long as the parts worked in the application.

1. I'm not comfortable with changing the prints this way, as it gets us farther and farther away from the original specification. If we want to remake the tooling, our prints will be considerably different from where we first designed them to be.

2. When the deviations are approved and the prints are changed, the supplier has to resubmit the first articles to the new revision. The more the tool runs, the more wear and it seems like we are in an endless cycle of first articles.

3. Other plants I've worked with have chosen NOT to change the prints, but do a long term deviation that remains in effect until the tooling wears enough to require replacement. That seemed to work better.

Any thoughts on a better process for qualifying the new supplier?

 

[Marc]

I ran into this in 2002. The ramifications were discussed prior to acceptance of contract. The customer agreed that the dies were used and poorly maintained. The customer also agreed that the dies were old and probably could not meet print. The customer promised to change the print because they didn't want to pay for the dies to be repaired and re-qualified. In short, they said they were willing to open a number of tolerances on the print. So - They gave the company a deviation so the company could do some trials and find out the capabilities of the dies. Never did get the changed print. Then at one point the deviation ran out, they insisted the company continue to ship parts and assured they would update the deviation since they hadn't gotten the print changed yet.

Through the magic of business processes, they failed to update the deviation and started checking parts to print. Of course this led to massive rejections of parts. That, of course, had a significant effect on their customer's supplier scorecard. It was a real mess that, last I heard, was not resolved. The customer even put the company "On Notice", sent in a "special" audit team for 2 days, and in short cost the company a whole lot of money. The audit team didn't do a thing about getting the print changed as they promised. The deviation was renewed (again) *BUT* they would not take the rejections off their "scorecard". They also required the company (the supplier) to do a full blown nonconformance investigation!!! The customer screwed things up consistently and wanted the supplier to do a nonconformance investigation. Talk about stupid.

The bottom line is *many* companies refuse to put money into maintaining their dies, but then they bitch because the dies wear and blame the supplier for problems that arise. I've seen it many times.

All I can say is Good Luck. If I was you I would only go for option 3.

People talk about waste in government, business is the same. Lots and lots of waste.

 

[Wes Bucey]

As a corporate SQE, I'm involved in a project with one of our plants and a stamping manufacturer.

We recently moved tooling to a new supplier. We are in the process of doing first articles on 80+ part numbers.

It's my company's process for evaluating these first articles that I am concerned about.

The tooling had been with another supplier for many years. The plant I'm working with, did not do receiving inspection, so there is no record of how well the old supplier was meeting print. They were functionally working.

The first articles from the new supplpier are showing a lot of worn tooling that is not meeting print. A decision was made to accept the product with deviation and change the prints to meet what the tooling could produce, as long as the parts worked in the application.

1. I'm not comfortable with changing the prints this way, as it gets us farther and farther away from the original specification. If we want to remake the tooling, our prints will be considerably different from where we first designed them to be.

2. When the deviations are approved and the prints are changed, the supplier has to resubmit the first articles to the new revision. The more the tool runs, the more wear and it seems like we are in an endless cycle of first articles.

3. Other plants I've worked with have chosen NOT to change the prints, but do a long term deviation that remains in effect until the tooling wears enough to require replacement. That seemed to work better.

Any thoughts on a better process for qualifying the new supplier?

I ran into this in 2002. The ramifications were discussed prior to acceptance of contract. The customer agreed that the dies were used and poorly maintained. The customer also agreed that the dies were old and probably could not meet print. The customer promised to change the print because they didn't want to pay for the dies to be repaired and re-qualified. In short, they said they were willing to open a number of tolerances on the print. So - They gave the company a deviation so the company could do some trials and find out the capabilities of the dies. Never did get the changed print. Then at one point the deviation ran out, they insisted the company continue to ship parts and assured they would update the deviation since they hadn't gotten the print changed yet.

Through the magic of business processes, they failed to update the deviation and started checking parts to print. Of course this led to massive rejections of parts. That, of course, had a significant effect on their customer's supplier scorecard. It was a real mess that, last I heard, was not resolved. The customer even put the company "On Notice", sent in a "special" audit team for 2 days, and in short cost the company a whole lot of money. The audit team didn't do a thing about getting the print changed as they promised. The deviation was renewed (again) *BUT* they would not take the rejections off their "scorecard". They also required the company (the supplier) to do a full blown nonconformance investigation!!! The customer screwed things up consistently and wanted the supplier to do a nonconformance investigation. Talk about stupid.

The bottom line is *many* companies refuse to put money into maintaining their dies, but then they bitch because the dies wear and blame the supplier for problems that arise. I've seen it many times.

All I can say is Good Luck. If I was you I would only go for option 3.

People talk about waste in government, business is the same. Lots and lots of waste.

The real horse, of course, is on the supplier. If he accepts the tooling as is without a long-term deviation, he is inviting a similar scenario to the one Marc describes. That deviation needs to be written by a "Philadelphia lawyer" because there is a point where the tooling will no longer meet the deviation. when that happens, the part will probably no longer come close to FFF (form, fit, function) and both supplier and customer are in a bad spot.

Without detailed info on part (and mating parts) and tooling, it is impossible to suggest the one BEST course of action. Here are some things to consider as a starting point:

1. determine the absolute tolerance before parts no longer meet FFF
2. can sufficient safety stock be made with current tooling meeting the looser tolerances to allow the tooling taken out of service and refurbished to original specs?
3. compare net cost after soft costs of interest, storage, etc. plus elapsed time of stockpiling parts with loose tolerances against the net cost of creating new tooling to spec to phase in as soon as possible and end deviation once new tooling is up and running and loose tolerance stock is used up
4. consider whether there is a possibility/probability of a design change so tooling and parts become obsolete (means either making do with loose tolerances or adding additional processing to bring parts to spec)

 

[gbcqc]

My question would be did the tool ever meet print. I ran into this with a mold that was not close to the original print and worked great. Then we needed to build a new mold. I asked the question, do we build it to print or match the print to the parts that we get now that work great.
I was told that I was an idiot for even asking the question. WE WILL HAVE IT BUILD TO THE ORIGINAL PRINT!!!!!!!!!!

One two-cavity mold and $180,000 dollars later, we had a nice boat anchor. Another $150,000 of cut and weld,(yea, we save $30,000 with a mold that would have to be replaced in a couple of years) we had a tool that made a functional part just like the one we had been making for years that worked.

1) So make sure that the original to print part will work. 2) If it does, I would say dev. and when needed make it to print. 3) If maybe not, change the drawing to meet what is working today.

But cover the bases of each possibility.

Just my personal experience.

 

[Marc]

The real horse, of course, is on the supplier. <snip>

Precisely. I learned the value of a robust contract review process 30 years ago. It never ceases to amaze me how many companies accept totally inadequate contracts and end up with problems that cost them so much time and money, and that they will do it again. People discuss risk assessment with respect to products and processes, but I don't remember ever seeing such a discussion of risk with respect to contract review.

 

[Jim Wynne]

As a corporate SQE, I'm involved in a project with one of our plants and a stamping manufacturer.

We recently moved tooling to a new supplier. We are in the process of doing first articles on 80+ part numbers.

It's my company's process for evaluating these first articles that I am concerned about.

The tooling had been with another supplier for many years. The plant I'm working with, did not do receiving inspection, so there is no record of how well the old supplier was meeting print. They were functionally working.

The first articles from the new supplpier are showing a lot of worn tooling that is not meeting print. A decision was made to accept the product with deviation and change the prints to meet what the tooling could produce, as long as the parts worked in the application.

1. I'm not comfortable with changing the prints this way, as it gets us farther and farther away from the original specification. If we want to remake the tooling, our prints will be considerably different from where we first designed them to be.

2. When the deviations are approved and the prints are changed, the supplier has to resubmit the first articles to the new revision. The more the tool runs, the more wear and it seems like we are in an endless cycle of first articles.

3. Other plants I've worked with have chosen NOT to change the prints, but do a long term deviation that remains in effect until the tooling wears enough to require replacement. That seemed to work better.

Any thoughts on a better process for qualifying the new supplier?

What you're experiencing is happening a lot these days. OEMs are closing captive plants and outsourcing the work, and when new suppliers get the tools and drawings, what I call the Cracker Jack syndrome begins: there's a surprise in every box.

I know of one case where a major OEM shut down a plant and outsourced the work, and the new suppliers starting making parts that did meet the prints and didn't work in assembly production. The operation was a success, but the patient died.

If parts are working OK in production, it means that there was probably something wrong with the specifications to begin with, and this is an opportunity to review and correct that. I see nothing intrinsically wrong with changing drawings if the changes represent reality.

 

[Golfman25]

As a corporate SQE, I'm involved in a project with one of our plants and a stamping manufacturer.

We recently moved tooling to a new supplier. We are in the process of doing first articles on 80+ part numbers.

It's my company's process for evaluating these first articles that I am concerned about.

The tooling had been with another supplier for many years. The plant I'm working with, did not do receiving inspection, so there is no record of how well the old supplier was meeting print. They were functionally working.

The first articles from the new supplpier are showing a lot of worn tooling that is not meeting print. A decision was made to accept the product with deviation and change the prints to meet what the tooling could produce, as long as the parts worked in the application.

1. I'm not comfortable with changing the prints this way, as it gets us farther and farther away from the original specification. If we want to remake the tooling, our prints will be considerably different from where we first designed them to be.

Then be prepared to pay for tooling revisions. You'll need to do a cost/benefit analysis. You might consider that if the parts are working as is, that the original print was overkill. Ask you supplier for an estimate to bring each tool back to original.

2. When the deviations are approved and the prints are changed, the supplier has to resubmit the first articles to the new revision. The more the tool runs, the more wear and it seems like we are in an endless cycle of first articles.

I would change this process. Have the supplier run samples and submit a first article. If you choose to change the print, use the submitted samples as the approved samples since that is what the print was changed to. It is redundent to have an additional submission to a new revision.

3. Other plants I've worked with have chosen NOT to change the prints, but do a long term deviation that remains in effect until the tooling wears enough to require replacement. That seemed to work better.

I guess that depends on how long you expect the tooling to last. Frankly, I would guess that the real cause is that the tooling was never right to begin with. Worn tooling usually results in burrs and other cutting issues. Very rarely does tool wear become the root cause of a dimensional issue. It would depend a lot on the types of parts. Good luck.

 

[RosieA]

Oh, Jim, you are describing my company! Especially the cracker-jack syndrome.

 

[ProProcess]

...
If parts are working OK in production, it means that there was probably something wrong with the specifications to begin with, and this is an opportunity to review and correct that. I see nothing intrinsically wrong with changing drawings if the changes represent reality.

+1 for me on this statement.
If the product functions (well) with out of spec condition, the spec needs evaluated.
There certainly needs to be a "safety zone" built into the spec, but if this is too large, it cost everyone money!!