Geometric Tolerance question

[beachcomber]

Probably not the right place to ask this, but there are so many brilliant people on here -

I have a stepped cylindrical part and want to apply a perpendicular tolerance relative to a plate that it is welded to.

If I attach the geometric tolerance symbol to just one section of the cylinder, does that imply the remaining sections are included ?

Or should I attach it to the axis centreline of the cylindrical part - which makes more sense to me but I'm not sure it is correct.

Every example I can find shows only a plain cylinder, with the symbol attached to the cylindrical surface.

[KennyG]

I would expect it to need to be attached to each feature , so if stepped, each cylcindrical step would need it's on FCF.

[hawcad]

Hi @beachcomber

without having a draft or picture it isn't that easy to give the right answer, but IMHO and depending on the length of those cylindars, I would like to say that either You take the longest cylinder as datum or You put all (or some) of them into a common datum base A-B

The only wrong method AFAIK is to use the center axis itself.

This is something what isn't defined any more (this was possible before 1998 or 2002)

[bshand]

I tend to agree with @KennyG. Since a centerline is virtual it's not really appropriate since you're trying to control the cylindrical faces. I suppose it might be legal to apply it to one face and attach a note saying it applies to all cylindrical features.

Just my .02

[hawcad]

HI @beachcomber

sorry, it seems as I have confused the situation.

You want to have the plate plane as datum and the cylindars as target?

Then You have either to put a geometrical tolerance to each of those cylindars, or - what I suppose as to be better - give them a common zone, that all of them have to be in one and the same tolerance area.

[beachcomber]

Thanks for all your input - you have pretty much confirmed what I thought. I was sure that in the past you could apply the tolerance to the axis - as @hawcad indicated.

The image below shows the assembly in question.

The problem for me is that the important bit is the end indicated - this is where a mating part locates.

It is only a short length of the part (dimensions only shown for your reference) and I have been given the tolerance value of 0.4mm. The 19 & 56 long sections do not really matter.

I think the 0.4 is a bit loose for the short length, but as time is tight on this job its going to be a case of JFDI here.

[bshand]

Not sure if it applies here for you but a projected tolerance zone can help to keep a short feature's effective nominal tolerance tighter.

[beachcomber]

@bshand - just looking on the 'net and I don't think the projected tolerance zone can be appled here.

It seems to be applicable to holes into which a stud is inserted. We don't use GT all that often, other than positioning of multiple holes, so I'm a bit behind the times on the subject and hadn't even heard of this.

It could, I think, be applied to the hole in the plate to maintain that perpendicular to the datum face.

[bshand]

It can be applied to male features as well. Example, a dowel pin. Or even a stud.

It's a theoretical zone so agnostic regarding feature type.

(I think. I was wrong once.)

You have a possible point about using it for holes only. Though I can't find any prohibitions about using it for male features I can't find any examples of that use online.

[hawcad]

Hi @beachcomber

is it possible to indicate the mating cylinder faces on that image?

Will the mating part fit on the back only, the front only or on both cylindars ?

If both then You should use a common tolerance zone for that indicating the two partly cylindars to clarify it clearly for every manufacturer.