Applying concentricity to a cylinder

[GuyNamedKevin]

I have a drawing of a simple cylinder that has to be dynamically balanced. It uses the diameter of the cylinder as a datum feature to establish an axis, then it specifies concentricity on the cylinder with respect to that axis. Is that allowed? Thanks.

 

[dingy2]

The simple answer is no.

Concentricity is applied on a cylindrical feature relative to another cylindrical feature that is coaxial. In your example, concentricity is applied to the same feature that is used to create the axis.

I would suggest that you use cylindricity. It will control the roundness, straightness of the axis and size changes along the feature combined. Make sure that you cylindricity tolerance is no larger than 1/2 the feature size tolerance range.

Dave D.

http://www.qmsi.ca

 

[Belanger]

Concentricity isn't what you want, so I agree with Dave. (Except for the last statement -- a cylindricity tolerance must be less than the total size tolerance.)

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

http://www.gdtseminars.com

 

[PaulJackson]

GuyNamedKevin,

Instead of designating the entire cylinder as the primary datum feature set up two circular element target lines at some optimally wide or feature stabilizing location and specify their basic distance apart form each other and relative to one of the the cylinder ends (or central to the ends) then go ahead and specify concentricity of the cylinder to the axis established by the circumferential targets.

In so doing even cylinders the exhibit greater but acceptable uniform non-circular form with minimal cumulative concentricity deviation may achive your desired rotational balance.

Paul

 

[dingy2]

J-P

I knew the last statement might be somewhat controversial.

Dave D.

http://www.qmsi.ca

 

[GuyNamedKevin]

Thanks guys, much appreciated.

-Kevin

 

[Belanger]

I would have agreed if we know that the size tolerance for a circle/cylinder is always formed by two coaxial circles/cylinders. But there's no guarantee of that.
See attached graphic -- in that example any cylindricity tolerance would have to be less than 0.6, not 0.3.

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

http://www.gdtseminars.com

 

[flash3780]

What you describe is an appropriate use of concentricity, but it would make for an exceptionally rare case. Most parts that require some degree of balance achieve it by a dynamic balance note.

Does the part incorporate some feature that can be ground away or drilled out to meet balance? If so, a note giving a balance requirement might be a better way to meet your design intent.

//signed//
Christopher K. Hubley
Mechanical Engineer
Sunpower Incorporated
Athens, Ohio
--

http://engineeringliberty.wordpress.com

 

[dingy2]

You are correct on this one J-P but I am sure that most people on the shop floor would not take rule #1 into consideration. In your example with minimum size, we can use up to the full tolerance range while if we made the cylindrical feature at its MM size, no cylindrical tolerance is allowed.

Dave D.

http://www.qmsi.ca

 

[pmarc]

The callout on OP's drawing is not that illegal as it seemingly looks to be. I am not saying it is the proper way to go (because in my opinion it will not grab the function of dynamic balance well), but technically I do not think it should fall under "not allowed" cathegory. It is weird and overcomplicated but there is a logic behind it.

That being said, I have to somehow disagree with Dave's statement that: "Concentricity is applied on a cylindrical feature relative to another cylindrical feature that is coaxial". This is how it is presented by some pictures in the standard and this is the most common situation (if we can even say "common" when talking about concenctricty applications), but the concentricity definition stated in 7.6.4 of '09 edition does not mention a single word about two cylindrical features that must be involved.

What I am trying to say is that datum axis is derived from unrelated actual mating envelope of the cylinder, but the median points for concentricity check are derived from actual surface of the cylinder, meaning that this is not exactly self-referencing.

So if the cylinder's actual geometry is similar to shown in fig. 5-1 b) or c), the concentricity tolerance of the cylinder's surface wrt to its axis derived from UAME may be met. Unfortunately in the same time each cross-section of such cylinder can look like bigger diameter of fig. 7-62. So the concentricity will still be met, but the function of dynamic balance will not.

Therefore in my opinion, assuming this cylinder is the one-and-only feature of the part, the only geometrical control that left in order to assure the function is tight cylindricity tolerance as a refinement of size tolerance if needed.

 

[Walterke]

Why not apply a (total) runout tolerance?

NX 7.5
Teamcenter 8