weavedreamer
Automotive
The intent is to control the extrusion of a spherically tipped rod back to two mounting holes.
This is the proposed solution.
This is the proposed solution.
Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations LittleInch on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Tangent Plane Modifier 4
- Thread starter weavedreamer
- Start date
- Status
powerhound
Mechanical
Sorry. I don't see where I insulted you but since you feel that way, I apologize profusely. I am well aware that insults are the fastest way to turn a thread south and bring the progress to a grinding halt. I don't tend to be so direct with those who really don't know GD&T very well but you do know it, thus I just cut to the chase. Saying that you don't understand the tangent plane modifier is not an insult. An insult is where I would say something that you might find offensive, like calling you names or berating you.
I can't find a single thing that I consider insulting in any of my responses. What did I say?
I'm not trying to insult you, but you are wrong about the tangent plane modifier. Are you willing to address the points I made about para. 6.5? Maybe provide a reference that supports your position?
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
I can't find a single thing that I consider insulting in any of my responses. What did I say?
I'm not trying to insult you, but you are wrong about the tangent plane modifier. Are you willing to address the points I made about para. 6.5? Maybe provide a reference that supports your position?
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
"Where the tangent plane rocks on a convex surface, see ASME Y14.5.1M for methods of verification."
Seems not to be restricted to flat surfaces. The note about not restricting flatness is applicable when it is applied to nominally flat surfaces, not as a limitation of applicability to flat surfaces.
Seems not to be restricted to flat surfaces. The note about not restricting flatness is applicable when it is applied to nominally flat surfaces, not as a limitation of applicability to flat surfaces.
powerhound
Mechanical
Dave,
Are you playing devil's advocate or are you serious? Think about what that sentence is saying. If the intent was to imply that tangent plane could be used on curved surfaces then why does it only refer to a convex surface? I'll tell you why. Because a nominally flat surface that happens to be concave will easily render it's high points with regards to a flat plane thus that scenario is not even addressed.
Now stop the madness. This is getting crazy.
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
Are you playing devil's advocate or are you serious? Think about what that sentence is saying. If the intent was to imply that tangent plane could be used on curved surfaces then why does it only refer to a convex surface? I'll tell you why. Because a nominally flat surface that happens to be concave will easily render it's high points with regards to a flat plane thus that scenario is not even addressed.
Now stop the madness. This is getting crazy.
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
-
3
- #24
Sorry PH. I really can't explain the ability of the committee to muddy the waters by simultaneously adding an example explanation while apparently restricting the applicability; the '94 version the explanation was cleaner by not pandering.
The better argument is that the profile tolerance, in this case, cannot be affected by the tangent modifier. The zone is already defined and the tangent modifier doesn't refine the description. In the case of parallelism the tangent modifier reduces the tolerance zone from applying to all the points on the surface to only the high points on the surface (though any convexity guarantees a pass, at least for parallelism, 14.5.1's lengthy explanations aside.)
I also think the (R) should be (SR) and the spherical radius should have a basic dimension somewhere on the drawing.
If the OP is looking to have some random shape at the top of the pin and not a tightly controlled radius, then use that as a datum and locate the holes relative to the tangent plane established by the datum reference and whatever other surfaces are needed to keep material around the holes.
The better argument is that the profile tolerance, in this case, cannot be affected by the tangent modifier. The zone is already defined and the tangent modifier doesn't refine the description. In the case of parallelism the tangent modifier reduces the tolerance zone from applying to all the points on the surface to only the high points on the surface (though any convexity guarantees a pass, at least for parallelism, 14.5.1's lengthy explanations aside.)
I also think the (R) should be (SR) and the spherical radius should have a basic dimension somewhere on the drawing.
If the OP is looking to have some random shape at the top of the pin and not a tightly controlled radius, then use that as a datum and locate the holes relative to the tangent plane established by the datum reference and whatever other surfaces are needed to keep material around the holes.
powerhound
Mechanical
I actually see that addition as a clarification. Verifying a tangent plane requirement on a concave surface really poses no problem while verifying it on a convex surface isn't as forthcoming. I can see why it would need to be explained in the math standard. I suppose I'll ask about it in Florida later on this month.
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
John Acosta, GDTP Senior Level
Manufacturing Engineering Tech
SSG, U.S. Army
Taji, Iraq OIF II
- Thread starter
- #27
weavedreamer
Automotive
Yes, (SR) should have been the designation.
The application is an assembly. The spherically tipped rod has a retracted and an extended condition. Presumably, the radius is confirmed prior to assembly, thus the desire only to state how far it protrudes.
This question has provided me with a new appreciation for what the tangent plane modifer is, and a degree of frustration for trying to express what I would have thought should be confirmable with just a height gage.
The application is an assembly. The spherically tipped rod has a retracted and an extended condition. Presumably, the radius is confirmed prior to assembly, thus the desire only to state how far it protrudes.
This question has provided me with a new appreciation for what the tangent plane modifer is, and a degree of frustration for trying to express what I would have thought should be confirmable with just a height gage.
WD,
Backdriving the holes is the best bet. Because a datum plane is, practically by definition, tangent to a part, there is no question of its location on the part.
Thinking more on it, I'd go with datum A as-is, add a width datum to the diameter of the rod, and add a datum (nominally tangent) to the end of the rod. With these three datums immobilizing the part, control the position of the holes in the direction perpendicular perpendicular to the tangent plane. An alternative is the width of the block instead of the rod as the secondary datum. The width datums can be referenced RFS or MMC.
The basic explanation is you want the end of the rod at a particular location and you want the holes to line up with the matching holes in another part when the end of the rod is in the correct position. The end of the rod is the functional interface and should be a datum reference.
If you were to create a fixture to assemble this part, it would make sense to push the end of the rod against a stop and then set the depth of the block until the holes were positioned over pins. The alternative is to adjust and measure, which is inefficient.
Backdriving the holes is the best bet. Because a datum plane is, practically by definition, tangent to a part, there is no question of its location on the part.
Thinking more on it, I'd go with datum A as-is, add a width datum to the diameter of the rod, and add a datum (nominally tangent) to the end of the rod. With these three datums immobilizing the part, control the position of the holes in the direction perpendicular perpendicular to the tangent plane. An alternative is the width of the block instead of the rod as the secondary datum. The width datums can be referenced RFS or MMC.
The basic explanation is you want the end of the rod at a particular location and you want the holes to line up with the matching holes in another part when the end of the rod is in the correct position. The end of the rod is the functional interface and should be a datum reference.
If you were to create a fixture to assemble this part, it would make sense to push the end of the rod against a stop and then set the depth of the block until the holes were positioned over pins. The alternative is to adjust and measure, which is inefficient.
-
1
- #29
Hi All,
I would say that this is not a legitimate application of the tangent plane modifier. I don't think that there is any text that specifically forbids it, but I don't think it's a logical extension of principle either.
Unless otherwise specified, Profile of a Surface applies to the entire surface. There is no mention in the standard of controlling other geometry derived from the surface. The derived geometry in this case is not just a tangent plane, it would need to be a tangent plane that is constrained in orientation to the ABC datum reference frame (the spherical end cap has many possible tangent planes).
There might be another way to get something close to the required control without the use of a tangent plane. Another way of looking at it is that the extremity of the spherical surface needs to be at certain distance from the B and C holes. Perhaps specify a surface profile tolerance on a limited area of the spherical surface. The limited area would be graphically indicated with a hatched area with a certain (small) diameter, in the "middle" of the spherical tip. This would require that the horizontal location of the pin be quite accurate relative to ABC - there would probably have to be a position tolerance for the pin axis. If the form of the pin is reasonably good, then the extremity of the spherical tip should be reasonably in line with the axis of the pin.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
I would say that this is not a legitimate application of the tangent plane modifier. I don't think that there is any text that specifically forbids it, but I don't think it's a logical extension of principle either.
Unless otherwise specified, Profile of a Surface applies to the entire surface. There is no mention in the standard of controlling other geometry derived from the surface. The derived geometry in this case is not just a tangent plane, it would need to be a tangent plane that is constrained in orientation to the ABC datum reference frame (the spherical end cap has many possible tangent planes).
There might be another way to get something close to the required control without the use of a tangent plane. Another way of looking at it is that the extremity of the spherical surface needs to be at certain distance from the B and C holes. Perhaps specify a surface profile tolerance on a limited area of the spherical surface. The limited area would be graphically indicated with a hatched area with a certain (small) diameter, in the "middle" of the spherical tip. This would require that the horizontal location of the pin be quite accurate relative to ABC - there would probably have to be a position tolerance for the pin axis. If the form of the pin is reasonably good, then the extremity of the spherical tip should be reasonably in line with the axis of the pin.
Evan Janeshewski
Axymetrix Quality Engineering Inc.
- Status
Similar threads
- Question
- Locked
- Question
- Locked
- Question
- Question