Suggested concentricity tolerances 2

[whotmewory]

Can anyone steer me toward a resource offering suggested concentricity tolerances for rotating elements at - say - various speeds?

Chris in NC NASCAR Country

 

[Barry1961]

I would think more in terms of balance or vibration if the measurement involves different speeds of rotation. Two objects can share a common center “perfectly” and be very unstable at any speed of rotation.

Barry1961

 

[TheTick]

If you are applying concentricity tolerances, think long and hard before you do so.

In most cases, when folks specify concentricity, they are actually thinking and measuring runout.

Concentricity in its true form is difficult to measure, and requires measuring size, roundness, and position of two surfaces.

Runout is what you measure by rotating a part clamped on one cylindrical surface with a dial indicator on another. Much easier, and in most cases more than adequate.

I could be the world's greatest underachiever, if I could just learn to apply myself.

http://www.EsoxRepublic.com

-SolidWorks API VB programming help

 

[alexit]

Try this, get permissible residual imbalance chart. It is special graph that you pick point equal to rotating part weight, design operating speed, and desired balance quality. Line through all points shows you maximum permissible imbalance. Now take 3D model and shift diameters off-axis until your imbalance equal data from chart. This gives you maximum run-out for that diameter. Of course this must be iterative for each diameter you move, so if you have 4V cam shaft with 16 steps plus journals...have fun.

 

[tbuelna]

chris in NC NASCAR country,

If your concern is strictly "concentricity tolerance", then you are concerned primarily with assembly fits. Any level of tolerance (concentricity or run-out) can be achieved in a given part or assembly. It's just a matter of what you are willing to pay to get it. And any geometric or dimensional tolerance you may specify can only be inspected under static conditions.

But if your concern is really "rotating elements at various speeds", then what you should be concerned about is dynamic imbalance. So I would recommend that you determine what is the maximum unbalance moment that you can accept for a given part/assembly, and develop a procedure to dynamically balance that part/assembly within that limit. It's a common practice with high performance race cranks, flywheels, clutches, etc. like those used in NASCAR.

Go Jimmie Johnson!

 

[Pressed]

http://www.engineersedge.com/general_tolerances.htm

http://www.me.metu.edu.tr/me114/tolerancing.htm

I have always used the Basic Hole Class of Fits chart in the back of my drafting book. I considered it took into account the part sizes, the runout of manufacturing and the assembly relationship etc.... The main thing is to learn what the classes are by trial & error and cautious reading. I have never had an error from using them. The 2 links above.

Other help:

Balancing act:

http://www.mpta.org/MPTABalancingPrimer.pdf

Reference, excellent:

http://www.efunda.com/DesignStandards/gdt/cylindricalspecs.cfm

Related:

http://www.mmsonline.com/articles/049602.html

Related:

http://adcats.et.byu.edu/

http://WWW/Publication/94-3/paper3-5_96.html

Related Services:

http://www.onsitemachining.com/tolerances.html

Related Specs: Concentricity tolerance (DIN ISO 1101)

Reference:

http://www.datamyte.com/PDF/Concentricity.pdf

Reference info, halfway down:

http://www.mitutoyo.com/pdf/4192-211.pdf

Just for sh*ts and giggles:

http://www.dstan.mod.uk/data/02/304/01000200.pdf

_______________________________________
Feeling frisky.........

http://www.tailofthedragon.com

 

[whotmewory]

Thank you all for the very helpful feedback.

I found an alternative to calling out concentricity in Lowell Foster's "Geometrics II" page 296 in which he suggests as an alternative to concentricity, calling out positional tolerance regardless of features size and referencing the same datum [you] would reference in the concentricity callout - with the datum too regardless of feature size.

cheers!

 

[MechNorth]

If you're worried about the balance, then the position route isn't the best way to go. In this case, use Total Runout with datum RFS(i.e. along the entire surface of the component, not at just at every discrete cross-section along its length as runout would allow) as it will cotrol the size and form of the surface and effectively center/balance it as well. Positional tolerance just controls the axis, not the surface. In general, don't use material modifiers (L&M) if you're concerned about rotational balance.