Runout includes profile?

[Belanger]

This is a two-pronged question for all of you regarding total runout...

First, I claim that total runout may be applied to a cone, IF the angle of the cone from the datum axis is given as a basic angle.

A colleague insists that it can't; his main evidence is that Y14.5 doesn't give an example of one (flimsy, eh?). But I point to paragraph 9.4.2.1 (of 2009 std), which glaringly omits "cylindricity" from the list of other characteristics covered by total runout. Thoughts?

Second, that same paragraph mentions "profile of a surface" as one characteristic that total runout may control. We know that profile of a surface must be applied to a true profile (meaning a basic diameter), but I maintain that runout tolerancing is not meant to control size -- the diameter of the part should not be basic. Though the standard doesn't say this latter item, I infer it because they make no mention of the dial indicator being zeroed at a specific distance from the datum. Thoughts?

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

 

[TheTick]

Runout does not control size. Simply picture a cylinder twice the diameter that it should be but perfectly straight. Runout is in, size (and profile) are out.

 

[fsincox]

Maybe it is too late in the day, but, if I have a diameter tolerance of and .005 a runout of .001, once you establish a number, it can only vary by .001, right? that is what I mean.

 

[mfgenggear]

yes I agree my last staement should read can not instaead of should not control runout.

fsincox

that is correct diameter & runout seperate attributes.

however total runout can incorporate perpendicularity & taper depending on the feature.

 

[mfgenggear]

sorry for the bad typing
bad on me :>/

 

[Belanger]

fsincox -- No, they are two different things. You asked: "If I have a diameter tolerance of and .005 a runout of .001, once you establish a number, it can only vary by .001."

It's not problem if the diameter over a run of 50 parts varies by .005. But runout is asking if an individual part is consistent around its circumference. So imagine making two shafts each with perfect runout (i.e., no wobble as it's checked with a dial indicator), yet their sizes (the distance across the center) might differ significantly.

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

 

[Belanger]

One other point -- I think when they say that total runout can control taper, that doesn't imply that it must be a straight cylinder.

If I call out a cone with a basic angle of 10º we can say taper is properly controlled if the actual part doesn't deviate from 10º.

What I mean is that I don't read the phrase "controls taper" to be equivalent with the phrase "prevents taper."


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

 

[mfgenggear]

Belanger

if a .001 total runout is specified on a flange face.
the face can not deviate more than .001 across the face.
is it not controlling perpendicular?

if a .001 total runout is specified on a cylindrical diameter
it can not deviate more than .001 across the length regardless of the diameter size.
is it not controlling taper?


Best Holidays to every one

Take Care

 

[fsincox]

I am sorry, John, I just want to check one thing, if I have a part with a .500-.505 diameter size callout, and the surface is indicated total runout of .001 to a datum axis. This same diameter can measure .500 in one place and .505 in another? I think, I misunderstood runout.

 

[Belanger]

mfgenggear -- I would say yes it controls perpendicularity, and yes it controls taper. My original post went further, and speculated that if a basic angle were given to describe an intended taper, then total runout could still be used to maintain the taper and keep it from deviating too far from that prescribed angle (just tilt the rail that the dial indicator rides on).

Fsincox -- No, the part you propose would not be legal, because of the runout tolerance. But I think we were saying this: Just because a part passes a runout check (it might register zero deviation) that does not mean that the part is the correct size. Thus, runout does not control size.

You kind of said that already when you stated earlier (time=16:33) that "once you establish a number" in your scenario, but I goofed and missed that. You say that we will have "established a number," meaning that we verified the diameter some other way. The runout tolerance will then ensure that the rest of the part maintains that size, so I guess in that respect in has some influence on size.

Merry Xmas, all!

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

 

[mfgenggear]

Fsincox

look at it this way.

The diameter would be inspected by means of a micrometer.
measurements would be taken 120 degrees apart or more.
the maximum & minimum are the variation in size.

The run out would be inspected by means of a dial indicator.
the part would be rotated on it's axis point , centers or diameters. the movement on the dial would be it's total run out. now verifying all the way across the length would be total run out.
if the diameter is at the low limit & it passes run out.
it's good.
if it's on the high limit of the diameter & it passes run out
it's good.

It is two independent verifications.

 

[axym]

I agree with John-Paul, but I think I see what fsincox is trying to say.

Total runout does not control the size of the feature. But one might say that, for a cylinder, total runout controls the range of local sizes that the feature can have.

Another way of looking at total runout is in terms of deviations from the basic profile. The deviations do not have to have a certain value, but they must be consistent.

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[fsincox]

Thanks, guys, I have started to respect your opinions enough, I was beginning to think I missed something. I realize the point I was trying to make is obscure, but, that in it self kind of helps make the point I was going for, namely a listing of the thing controlled by runout would not contain size yet it can in this case. Runout is not a size control tolerance, as many here have strove to point out to me, but it can have an effect on size tolerance variation in a part. I think the statement in the standard as referred to earlier is a general statement not an absolute.

 

[flash3780]

Legality issues aside, why not just use runout and (if absolutely necessary) a straightness callout? Checking straightness can't be any more difficult than trying to move a dial indicator along a part at some oddball angle. I see no benefit to using total runout on a taper.

 

[Belanger]

I think because straightness can't be referenced to a datum, and thus it would not verify that the taper is at the correct angle. Total runount -- with the mention of a basic angle -- does control the degree of taper.

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

 

[flash3780]

I think because straightness can't be referenced to a datum, and thus it would not verify that the taper is at the correct angle. Total runount -- with the mention of a basic angle -- does control the degree of taper.

Good point. Circular runout also provides such a control by constraining the position of each circular element of the taper to the datum within the allowable tolerance. A refinement could be made with concentricity if necessary (maybe the only good use of it).

Thing is, shops are set up to do TIR on a cylinder or a flat. I think you'd get some funny looks if you put it on a taper.

 

[fsincox]

So in the final analysis can we now agree here that while total runout is not advertised as a size control, it will have that effect given the conditions I have stated above, the only thing restricting the size variation is the runout tolerance, what do you want to call it? The point being, that the standard in making those kinds of statements is generalizing and you must be actually looking at it from their perspective because it is not “literally” true.
Frank

 

[fsincox]

To the OP, I do not see why cylindricity is not covered, bottom line.
Frank

 

[Belanger]

Frank -- total runout certainly CAN control cylindricity, but my point was that it is not a "must."

If the standard stated that total runout always "must" control cylindricity, then that symbol could never be used on a cone, which was the thrust of my OP.

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

 

[fsincox]

John-Paul,
Thanks,
Frank

 

[fsincox]

I am not an inspector nor do I play one on TV, but it does seem that if you could set it up at the basic angle you could, did you ever really get an agreement/consensus?
Frank