If a Blueprint calls for ASME Y14.5 wouldn't this be the standard rules apply. How would I know if and when to apply ISO? I have yet to see anything on a blueprint stating this. The reason I bring this up is for the good old Profile callout. Before I make an argument, I want to cover all my resources. Customer already claims in an email that PC-DMIS is not capable of doing this. I and everyone on here already that's horse_S_H_I_T. But its Friday and I'm ready to P_I_S_S someone off
The actual profile reading is the Max and Min. Whatever the tolerance zone is. The only question is "do the max and min fall within the tolerance zone?"
2x max could only apply to bilateral equally distributed profiles (hypothetically, if it were the ASME way).
I agree Reporting Min and Max would be helpfully descriptive. In fact you could report a bunch of different point locations to more fully describe the lay of the surface. But if the question is, what number do you report so that you can say if its in tolerance or not, we discussed that to a draw further up this thread and I dont really care to reopen that discussion. We have to agree to disagree and resist the temptation to consider the other deficient in intelligence because they wont agree with us.
Just to show I'm not crazy, here are a couple photos from a training book of our local GD&T guru, Dr. Greg Hetland.
He has impressive credentials. see here:
http://www.iigdt.com/Content/Affiliations.htm The example shows the worst reading being multiplied by 2 for the "actual profile value".
Thats just to show that I have good reason for holding to this method of reporting profile.
I know Kirb will still disagree...
The cover lists both ASME and ISO. Where does it differentiate, in your book, between whether that snippet is talking about ISO or ASME? I think someone is confusing ISO and ASME.
Why wouldn't Kirby disagree? I do. The standard does.
Well now that makes more sense. Lets create a book of both standards and pick and choose what standards, That I (Greg Hetland) want people to follow. I don't disagree the Standards do. That's awesome.
Equally disposed bilateral profile of .020 fully constrained to datums. This is basically a straightwoward +/-.010 dimension. Right?
So, you measure it in pc-dmis and the deviations range from +.0095 to + 0099. pcdmis reports the measured profile as .0099, not even using up half the profile tolerance. If you report this result in something like an FAI report, it's downtight misleading (maybe even fraudulent). BUT, if you double that worst value you report .0198 for the profile result. This represents the true nature of the result - on the ragged edge of the tolerance.
Of COURSE the worst deviation should be doubled. And of COURSE this is my opinion. There are no specific instructions for how to do it. And frankly, the doubling approach falls apart when dealing with an unequal bilateral profile. So it isn't a universal solution.
The bottom line is that pc-dmis fails bad parts and passes good parts. And that's probably what you should tell your customer.
I've made peace with the pcdmis method. But it isn't a matter of ASME vs ISO. It's just their way.
Y14.5 is a tolerancing standard, not a reporting standard. It does not absolutely require a part to be held in a certain way when it's machined. All it does is describe the size and shape of tolerance zones and datums. Nothing that I'm aware of says to report anything in a specific fashion. It's convenient to try and reduce things down to a single useful number. However, most things in the standard, including profile and position, simply can't be reduced to a single useful for all things number. Identical position numbers by the formula can have a near infinite number of XY locations. Doubling deviations might make sense on an equally disposed profile callout but fall apart on unequal disposed profile callouts. I've even seen ++ and -- profile requirements. WIth flatness and straightness, you will want to know how and where the part is not flat or straight. The single numbers are nice but context in the form of XYZ points is what's required to make decisions on.
Bingo! And the same goes for ISO - it defines the tolerance zones, i.e. defines PASS/FAIL condition, but doesn't say anything about how to present actual deviations.
Your example is not a bad part, regardless of how it's reported. So, where's the fraud? Nobody that I've seen has said "PC-DMIS has it right", so let's dispose of that strawman.
It is a matter of ISO vs ASME. ASME does not specify how TO report, but it does specify how NOT to report. Very clearly.
Doubling doesn't tell you anything more than just reporting the Max and Min deviation do, plus Max/MIn works in ALL cases. Win-Win.
nevermind...
