機械社區»論壇 › 求助/理論/基礎/樂園/外語 › 機械樂園 › [2C] 一個老美的貼子：關于垂直度

12 3 / 3 頁下一頁

查看: 10642|回復: 23

[2C] 一個老美的貼子：關于垂直度

火.. [復制鏈接]

十年一夢

電梯直達

1^#

發表于 2012-3-10 00:29:15 | 只看該作者 |倒序瀏覽 |閱讀模式

這是一個美國機械論壇關于垂直度的貼子，地址：http://www.eng-tips.com/viewthread.cfm?qid=317827

很不錯的一個論壇，大家可以注冊一個網名，參與討論。

metaldork (Agricultural)

8 Mar 12 12:53

Can anyone please clarify if a perpendicularity call out is controlled by a basic dimension or not? In other words is the tolerance zone centered on a basic dimension?

CheckerHater (Mechanical)

8 Mar 12 13:03

No, it is sufficient by itself.

It MAY be used together with basic dimensions when used to refine the Position.

powerhound (Mechanical)

8 Mar 12 13:42

No, the tolerance zone moves with the feature being controlled.

Powerhound, GDTP T-0419
Engineering Technician
Inventor 2010
Mastercam X5
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

metaldork (Agricultural)

8 Mar 12 13:56

The surface is dimensioned from a parallel surface ...would the perp callout control the dimension and if so should it be basic?

CheckerHater (Mechanical)

8 Mar 12 14:17

There is a difference between Perpendicularity and Parallelism.
Even in case of Parallelism the dimension can be toleranced; the Parallelism will refine the tolerance.
If I don't understand something, please provide a picture.

Belanger (Automotive)

8 Mar 12 14:35

First, there is an implied basic dim with perpendicuarity. It is the 90 degree relationship to the datum. That's not the kind you were thinking of, but I feel obligated to mention it.

You ask about locating the perp tolerance zone. That should never happen with basic dims. The perp tolerance zone can float freely within whatever other constraints there might be.

IOW, the per tolerance never controls the distance from the parallel surface. There should be something else to do that, such as plus/minus. The perp doesn't get added to that; it must operate within it.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
http://www.gdtseminars.com

dgallup (Automotive)

8 Mar 12 14:38

The surface can be dimensioned from a parallel surface and still have a perpendicularity to a datum that is normal to both surfaces. The dimension will control location and to the extent that the envelope principal applies, form. The perpendicularity must be a refinement of the dimension tolerance, typically half or less.

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

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.

axym (Industrial)

8 Mar 12 15:09

metaldork,

The way I like to look at it is that the Perpendicularity zone can always freely translate - its location is never controlled relative to anything. So the Perpendicularity zone is always completely independent of any linear dimensions (basic or directly toleranced) connected to the considered feature.

dgallup,

In the scenario you described, the Perpendicularity doesn't have to be a refinement of anything. The directly toleranced dimension doesn't control the squareness, so the Perpendicularity tolerance doesn't need to refine it.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

pmarc (Mechanical)

8 Mar 12 15:13

I am with J-P on this, the only basic dimension that applies for perpendicularity is basic 90 degrees to the datum and it is usually not shown on a print due to implied basic 90 deg dimension rule.

dgallup,
You said: "The perpendicularity must be a refinement of the dimension tolerance, typically half or less."
I would be careful with that. If perpendicularity callout is applied only to one "side" of a dimension, its value can be whatever one can imagine (assuming there is no general angular tolerance shown on the print that would control a relationship of the other "side" of the dimension relative to the datum).
The other story is when there are two perpendicularity callouts applied to both "sides" of the dimension. In that case, my vote is their values can be as big as the dimension tolerance (and not half of it).

CheckerHater (Mechanical)

8 Mar 12 15:57

Evan,
Could you please clarify: Do you believe that perpendicularity tolerance zone can "freely translate" OUTSIDE of linear dimension /position tolerance zone?

Belanger (Automotive)

8 Mar 12 16:39

CH, picture a U-shaped bracket -- sort of like field-goal posts. Now take both vertical posts and bend them 20 degrees to the left. Are they within the size (width) tolerance? Yes. Is the post on the left within its perpendicularity tolerance to the ground? No.

So there is no relationship between the tolerance number for the size across the posts and the tolerance number chosen for the perpendicularity.

Sorry for jumping in, Evan. Feel free to modify my explanation

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
http://www.gdtseminars.com

axym (Industrial)

8 Mar 12 17:54

Quote (checkerhater):"Could you please clarify: Do you believe that perpendicularity tolerance zone can "freely translate" OUTSIDE of linear dimension /position tolerance zone?" : ^9 C3 |7 F0 u; H d- C# e

CH,

Quick answer, yes.  But of course there's more to it.

If we think purely in terms of the tolerance zone mechanics, which I always try to do, then the Perpendicularity zone is allowed to translate outside of the linear dimension / Position tolerance zone.

It is true that the feature might not conform to the Position tolerance if the Perpendicularity zone had to translate partly outside of the Position zone.  It definitely wouldn't conform if the Perpendicularity zone had to translate completely outside of the Position zone.  But that is a different consideration.

The way I like to look at it is that the Position tolerance and the Perpendicularity tolerance are independent requirements, that can be evaluated independently of each other.  This is the only way that I am able to make sense of it all.  Some GD&T books state or imply that the orientation zone must float within the location zone, but to me this is an oversimplification.  It is possible for part of the orientation zone to extend outside of the location zone, and still have the feature conform to both tolerances.

Evan Janeshewski

Axymetrix Quality Engineering Inc.
www.axymetrix.ca

CheckerHater (Mechanical)

9 Mar 12 6:32

Thank you Evan,
Appreciate straight "Yes" - rare thing nowadays.
Unfortunately I am not convinced. Imagine the hypothetical situation:
You are using some sort of measuring machine. You find out that the axis of certain hole (or boss) is laying WITHIN perpendicularity tolerance, but OUTSIDE of position tolerance.
Would you suggest the part to be accepted or rejected? (Another straight Yes or No will be appreciated)
I personally believe that your position and perpendicularity zones should at least partially overlap. (It will create some interesting conditions that probably were never fully documented, but nevertheless).

Joan-Paul,
Your argument appears to be far stretched and borderline cheating (Sorry)
You said it yourself: "Is the post on the left within its perpendicularity tolerance to the ground? No." So what you describe is "bad" part that should never be accepted in the first place. The purpose of GD&T is to describe the parts we will accept, right?
Another stretch: "tolerance number for the size across the posts". Most people would consider "fork" being two features of size, not one. Is space between two holes feature of size?
So your argument basically is: "If we can make bad part from ambiguous drawing, then there is no relationship between the tolerances". Sorry, but I am not buying it.

Belanger (Automotive)

9 Mar 12 7:03

CH -- I don't quite understand your concern. First, yes, the space bewteen the posts is a feature of size; so is the outside of the posts. But my example was really meant to discuss surface perpendicularity (I just used the posts as a visual example. Sorry for the confusion.)

"Is the post on the left within its perpendicularity tolerance to the ground? No."

The reason I wrote that is that many people think that a size dimension also controls orientation -- that's not true! So I'll rephrase it differently: Two vertical surfaces can lean in the same direction and still be within perpendicularity tolerance. They are also within the size tolerance. But there is no connection between the tolerance values for the perpendicularity and the size.

John-Paul Belanger
Certified Sr. GD&T Professional
Geometric Learning Systems
http://www.gdtseminars.com

CheckerHater (Mechanical)

9 Mar 12 7:54

Thank you JP, we are getting somewhere.
Let say, we have FOS with perpendicularity requirement added. The size tolerance is .001, the perp. .100. But they both affect VIRTUAL CONDITION, right? Now, what if we add position requirement to our virtual condition?
You see, I never said perpendicularity is a refinement of size. But I still insist on the following conditions: Perp. tolerance zone being outside of position tolerance zone makes no sense. Perp. tolerance zone being larger than position tolerance zone makes no sense. It may be legal though.
When you have 2 FCFs applied to same feature, their requirements have to be met together (I didn't say "simultaneously"). If perp. tolerance is .100 and position is .010 position requirement controls perp. indirectly and makes perp. requirement useless.
So, when used together with position requirement, perpendicularity only makes sense when perp. tolerance zone is smaller than position tolerance zone, and both tolerance zones overlap at least partially. To me it means "refinement".

pmarc (Mechanical)

9 Mar 12 8:57

Apologies for jumping into main topic.
Can you have a look at attached picture?
http://files.engineering.com/getfile.aspx?folder=a7879b3a-d0d7-459f-b9ac-829eb08891da&file=surf_perpendicularity.JPG
How big can those perpendicularity tolerance values be, assuming they both have to be equal? Is there any limit or not? Thanks.

fsincox (Aerospace)

9 Mar 12 9:15

I believe it is acceptable to use a space between to pins say as a feature of size, right?
Frank

CheckerHater (Mechanical)

9 Mar 12 9:29

Frank,
Yes, but it will make them more ambiguous, than 2 FOS with position requirement, right?

pmarc,
What do YOU think?

http://files.engineering.com/getfile.aspx?folder=83017c19-26f6-45c9-8d71-70
- u: s- _6 A- k: J. l/ M

CheckerHater (Mechanical)

9 Mar 12 10:03

pmark,
You are a genius - I think this is exactly what OP was asking: do you use basic dimension to control perpendicularity of parallel elements.
I would use basic dimension with profile to control the whole contraption. That will also make it less ambiguous.

pmarc (Mechanical)

9 Mar 12 10:04

CH,
You are asking what do I think about my sketch or about your modified version of my sketch? Or is your sketch showing only one of situations that may occur for my tolerancing scheme?

fsincox (Aerospace)

9 Mar 12 10:20

Sure ideally all should be controlled, that is right out of the book. Frank

CheckerHater (Mechanical)

9 Mar 12 10:22

Actually both (or all three?).
I feel like using direct dimensioning may create ambiguity here. There is a reason ISO calls it "two-point dimension".
(See my post about using Profile)

CheckerHater (Mechanical)

9 Mar 12 10:36

Frank,
Off-topic, but since you brought it up; have you ever seen round hole dimensioned to the edge rather than center?
I did, I was even forced to do it myself, I still don't like it.

fsincox (Aerospace)

9 Mar 12 10:55

CH,
Are you kidding? I cut my teeth on that. It is what I was looking to GD&T to help me escape from. ;)
This thread has confused me a bit with the whole perpendicularity zone centered, center implies a location. Seems like at least part of the perpendicularity zone must be in the location tolerance zone or you would just move it.
Frank

美的, 機械