Not using GD&T? 10

[ctopher]

My last company, and current, have argued with me about using GD&T.
Engineers, purchasing, and managers have argued that it makes the parts more $$.
I tell them no it doesn't.
Last company had parts machined in China, current in Thailand and in USA.
China and Thailand have said not to use it because they don't understand it.
Often I see parts made that don't meet print with GD&T, but are bought off anyway.
I'm at wits end, tired of arguing with everyone.
I also find more people here in USA that don't understand it.
Anyone here run into this? If so, what do you do?

Chris, CSWP
SolidWorks '20
ctophers home
SolidWorks Legion

 

[3DDave]

I'm not calling for a business to spend money. You are. And you want everyone in the supply chain to do so as well because regurgitating the standard is the one thing you think you are good at.

Another straw man - they are qualified to know if the available production capability will produce usable parts. They look at all the years they have been in business and seen their profits go up and customers come back.

Here's how you argue - you make up a scenario that you claim is a sign of failure, claim I said it, and then say "see, it's a failure." It's not an effective strategy.

That ISO example proves a part of it; someone got trained to create that. You have no idea what that set of symbols means. A note would have been clear.

 

[Burunduk]

You do advocate for businesses to spend more money than they should, by keeping many unnecessary tolerance requirements on their part drawings. Or by having to rework parts that don't fit, because the designer never heard of a Virtual Condition.

"production capability will produce usable parts"

You make it sound like every process churns out at least 99.9% good parts. As I already pointed out to you, if you saw a couple of very accurate CNC machines and tools, it doesn't mean that the rest don't exist or shouldn't be used. What is the cost of continuously maintaining a process capability that gives a variance that is only a small fraction of the tolerance value? And when it's all +/- and you try to eliminate stacks, that tolerance value is small. To realize how bad it is, you need to get phone calls on an almost daily basis asking you to come and see if parts can be approved, despite being out of a tolerance that someone thought to be appropriate.

"someone got trained to create that"
I sure hope he was not trained to create THAT. If he was trained at all, maybe "that" could be avoided if management communicated to him on time that he is encouraged to start implementing what he was taught.

"A note would have been clear"
I'm curious what that note should look like.
I'm sure you have a good suggestion.
Assume the specification made sense. For example, replace the '+' with '0.25 UZ -0.05'. Make sure you include the meaning of the datum references and their intended simulation process.

 

[3DDave]

"You do advocate for businesses to spend more money than they should,"

I don't.

"You make it sound like every process churns out at least 99.9% good parts. "

Did not write that either, though if it sounds like that to you that's a problem you have.

"What is the cost of continuously maintaining a process capability that gives a variance that is only a small fraction of the tolerance value?"

That's a great question, because that's how 6 Sigma works.

"I sure hope he was not trained to create THAT."

You can hope all you want. No one just created that random symbol glob out of thin air. They were shown those symbols and set in front of a computer to type that in.

 

[Burunduk]

I was expecting to see the note.
I guess making suggestions is easy, implementing them yourself is less so.

"I don't" (advocate extra spending)

You do.
When useful parts are controlled by 'mostly +/-' method without the use of functional datum reference frames, the company spends excess resources on conforming to tolerances that could be looser, less ambiguous, providing smaller stacks.
The support you give to backward policies and work methods is support for continuing that waste and that difficult struggle for productivity and competitiveness.

"Did not write that either" (on 99.9% good parts processes)

By suggesting that relying on production capability can make geometric tolerancing unnecessary, you imply exactly what I wrote.

"That's a great question, because that's how 6 Sigma works"

I asked about the cost. You stray from the topic.

"No one just created that random symbol glob out of thin air. They were shown those symbols and set in front of a computer to type that in"

You mean that someone once showed him how to do that, because "that's how we do it here".
The end result is a drawing user looking for answers in an internet forum, getting "no comment" replies.
This is your alternative to using standards.

 

[3DDave]

"No comment" was because, unlike you, I don't know about the ISO standards and therefore didn't feel like making a random guess. As you often do. I just dislike when people require other users to download files to see what could be plainly be inserted into the request.

Obviously the problem was based on using a standard; I am also not the company that accepted that work and would not have done so blindly. Did reliance on that standard fix the problem?

You build another straw man. You have a factory for them, apparently.

6 Sigma is a cost control strategy. You cannot be this ignorant and are just putting on an act.

 

[Burunduk]

"6 Sigma is a cost control strategy."

Wrong answer, again.
It is a quality control methodology.
It requires great efforts in monitoring, analyzing, and upgrading the manufacturing process.
Maybe you think that implementing Six Sigma is free, and only assimilation of drawing standards requires any investments?

"Obviously the problem was based on using a standard"

Not obvious at all.
A couple of times I saw a perpendicularity feature control frame with the degree symbol after the tolerance value.
Was it the fault of standards?
Or maybe the result of a company management not paying enough attention to the qualification of its engineers?

So maybe you don't write the note because you don't know the ISO standards?
Suppose that the intent was per Y14.5.
The "expensive" specification would be:
|Profile|0.25(U)0.075|A|B|
The tolerance zone is the ASME version of what I suggested earlier.
The controlled feature is a groove in a stepped shaft. Datum feature A is two coaxial diameters having the same size specification and qualified for coaxial position as a pattern. Datum feature B is a flat shoulder. Now write the cost-effective note to convey the requirement for cheap vendors who are against geometric tolerances.

 

[3DDave]

https://www.google.com/search?q=6+Sigma+is+a+cost+control+strategy

"How Can Six Sigma Lower Costs and Improve Quality?"
"Ways to Save Your Company Money with Lean Six Sigma"
"Six Sigma Costs and Savings - iSixSigma"

Feel free to study this on your own. There are plenty of case studies discussing making the process variation as small as possible.

Suppose the company I sent that part to has hit within 0.00001 of nominal on their Swiss screw machine for the last 750,000 parts at $0.03 each. There is no need for a note of any kind. See how straw works?

Stop sending drawings to child-labor camp producers and you won't have so much trouble.

 

[Burunduk]

Google all you want, Six Sigma is primarily about improving product quality.
I have nothing against it or against the advanced manufacturing technologies that a process upgrade requires,
But there is no guarantee of reducing any costs. A company implementing it successfully may increase its revenue, therefore covering for the extra spending of implementation of the method, but only if the customer cares for the additional quality of the type of product being offered. At some occasions the costs can be reduced, indeed, if scrap rates are brought down in a relatively short time without investing a fortune in new equipment.
If your product doesn't benefit from the tight control on the process, you may end up wasting money on the effort.

A different thing, which is more on the topic of this thread, is simply examining what is the function of the part, and what should be the corresponding requirements from the manufacturer, and conveying just these requirements in a rigorous manner, by using datum reference frames and geometric tolerancing. Then you can produce the part on the same average CNC machine as you did earlier, inspect it on that same average CMM you've been using all along, while reducing scrap AND improving the quality. The only thing you spend on is knowledge and competence for your employees.

"0.00001 of nominal"... "$0.03 each"..."sending drawings to child-labor camp producers"...
What's next?
Your evasive tactics become more and more hallucinatory and grotesque.
No need for a note?
Then why did you recommend writing them?
You said "a note would have been clear", despite all the previous explanations you got on why not notes.
Either write it or admit it was a bad idea.

 

[3DDave]

So you now admit several things - that 6 sigma is about having far tighter controls of process than the product demands, that it saves money, that it makes more money, that there are studies that prove all these things

and

that you have not one shred of support for your position.

Always a good day to have the truth exposed so plainly.

 

[Burunduk]

I didn't "admit" all that. Read carefully.
In addition, "saves money" and "makes more money" are different claims, both are way too absolute in the context of Six Sigma. For the third time - it is a quality control methodology.
Any financial benefits are by-products that may or may not be realized. A company may achieve the direct goals of Six Sigma without any costs reduced or income increase. It may even lose money.

As I said I'm not against Six Sigma, but when a company starts implementing it, it defines quantitative goals such as reducing scrap. At this phase, the premise is that the definitions that govern what passes and what fails are well thought through, so that no useful parts fail and no useless parts pass. The documentation of those definitions is the first thing that should be reviewed, before even starting. The requirements must reflect and correspond with the part's function. This is why the company should primarily care about what's in the print. If the specification for geometry is ambiguous such as when +/- tolerances are used for location, and that ambiguity is "compensated" by tightening the tolerance values, that product definition is a horrible basis for Six Sigma - it may multiply the spending on optimizing the process. Will the customer be willing to bear the cost of that waste? Only if he agrees to pay the 3DDave tax.

If everything could be made to "hit within 0.00001 of nominal" there would be no need for Six Sigma, right?
Congratulations on providing another point to support my position, you have done well at this so far.