Radii Verification

[KP1SP]

Hello,

I'm designing a spec for a tyvek(CR27 1073B Coated) lid for a thermo formed plastic tray, nto at liberty to say material for that. What is the best way to verify that the radii on the lid are within tolerance? Raidus 1 is 0.38" ± 0.06", the other 3 radii are 0.86" ± 0.06". I have a customer that wants to know if I can lower the tolerance and i simply just don't know what my rollerpress/rotary dies' capabilities are.

I want to verify with past products to confirm whether ±0.06" can be lowered and held or if I need to leave the ±0.06".
Attached is a rough draft.

Regards,

-Kyle Paolino

 

[MikeHalloran]

You need to be talking to the people in your shop, and measuring the leftovers from past production.



Mike Halloran
Pembroke Pines, FL, USA

 

[KP1SP]

MikeHalloran,

I have, they are just operators. They don't validate the radii, just width x length. I have measured leftovers from past production. All I can measure are print location markers and width x length dimensions. Other than printing a 1:1 of the spec and placing samples on top, we don't have a system in place to validate the precision of the radius.

Upon measuring past samples, I found that the width x length dimensions are within tolerance. Being anywhere from ±0.01" to 0.03", so the 0.06" tolerance covers that.

We are all fairly new to this industry here, and unfortunately because of the material and the fact that is for medical use, we need to be as precise of possible.

-Kyle Paolino

 

[pmarc]

Other than printing a 1:1 of the spec and placing samples on top, we don't have a system in place to validate the precision of the radius.

Although it looks crude, it is a method, especially if not other measurement equipment is available. For sure it makes more sense than trying to find centers and actual values of these radii. Just keep in mind that tolerance zones are crescent-shaped boundaries, similar to shown in fig. 2-22 in Y14.5-2009 standard or fig. 2-18 in Y14.5M-1994.

 

[dgallup]

Generally we use an optical comparator for measuring radii although in some cases it is possible to trace with a profilometer. CMM should also be able to measure radii. In my opinion radii are one of the hardest measurements to make accurately and should not be tightly toleranced unless it is important to function. I tend to make radii reference dimensions when possible to avoid the need to inspect them. If you are certified to a quality system you will need a capable measurement system.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[drawoh]

KP1SP,

This sounds like a good case for profile tolerances. You actually do not care what the radii are. You care that your lid fits.

Model the lid with a slight clearance. Apply a profile tolerance all around. Make an inspection go-gage that conforms to the MMC of the lid. No-go gauges are more complicated, but in your case, probably less critical. This all is easily used by semi-trained people.

--
JHG

 

[pmarc]

dgallup made very interesting comment: "In my opinion radii are one of the hardest measurements to make accurately and should not be tightly toleranced unless it is important to function. I tend to make radii reference dimensions when possible to avoid the need to inspect them. If you are certified to a quality system you will need a capable measurement system."

For your information, per ISO 14405-2:2011 directly toleranced radii, chamfers, angles, distances between stepped surfaced are named AMBIGUOUS specifications. Basic linear and angular dimensions together with geometrical tolerances like position, profile of surface, angularity, etc. shall be used instead. I really hope that something similar will appear in future editions of Y14.5 and that it won't happen in 2050.

 

[CheckerHater]

Where does it say that “AMBIGUOUS” automatically means “BAD”?

 

[KP1SP]

All,

I have spoken with some people around here and we have decided to have a film proof(clear plastic material with print) made for each lid. Just to verify our standard tolerance can be held, I was mainly trying to avoid any non conformancies with prior customers and also not to have to change their already approved and produced specs.

Thank you all for your input.

-Kyle Paolino

 

[pmarc]

AMBIGUOUS doesn't automatically mean BAD - that is true.
From drawing specification point of view, and not only from that perspective, AMBIGUOUS can be way worse than BAD.
Not to mention that none of the two at any stage is tolerated by GD&T.

 

[CheckerHater]

Practical question:
If I am to attach basic dimension and profile control to every fillet on my drawing, should I use datums as well?

 

[pmarc]

I would say NO, you do not have to use datums. It depends on type of geometrical characteristic you want to control.

 

[Belanger]

Where does it say that “AMBIGUOUS” automatically means “BAD”?

That would be paragraph 1.4(b) of Y14.5:
"Dimensioning and tolerancing shall be complete so there is full understanding of the characteristics of each feature."

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems

 

[CheckerHater]

Without datums the tolerance zone is free to translate / rotate in any direction.
What prevents me from accepting part like the one shown on the picture as long as profile is falling within the tolerance zone?

 

[CheckerHater]

John-Paul,
The paragraph you quoted simply says “everything should have a dimension”.
Now, let’s say I made a drawing.
I put dimension on the drawing.
According to some purists, my dimension may have 10 different interpretations.
What if all 10 interpretations will result in good parts?
Is my dimension ambiguous? Sure.
Should I care?

 

[pmarc]

What prevents me from accepting part like the one shown on the picture as long as profile is falling within the tolerance zone?

The fact that profile defines true (basic) form of the arc which is convex, not concave.

The paragraph you quoted simply says “everything should have a dimension”.

I recommend para. 1.4(d) in Y14.5-2009.

 

[CheckerHater]

1. I don’t see words “convex” or “concave” anywhere on the drawing.
2. In ISO absolutely no relation between the features is IMPLIED just because it looks certain way: convex, concave, tangent, whatever.
3. The only way to say if the arc is convex or concave is to specify arc center location in relation to the rest of the part, and this is exactly what ISO is trying to avoid by introducing profile.
4. To be fair, forever and ever radius was measured with radius gages and optical comparators, and no one cared about the center anyway.
5. So far, I don’t see ISO approach being any less clumsy than the ASME one. Maybe in the future… But I am planning to retire before that.

PS How can you comply with Para 1.4(d) when the the standard is missing or leaving unclear several important definitions like "radius" or "size" for that matter?

 

[pmarc]

I have a feeling that we could discuss about it for hours and reach absolutely nothing, thus I am simply giving up.

 

[Belanger]

CH -- the paragraph I quoted doesn't "simply" say that everything should have a dimension. It says that "dimensioning and tolerancing shall be complete so there is full understanding of the characteristics of each feature."

Quite a difference, yes?

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems