Controlling coaxial features

[cwdaniel]

I stumbled across this dusty thread: thread1103-200480
I found it quite educational. Basically I've been assuming implied coaxial features on the same center is acceptable.
I'll be correcting my practices regarding this in the future.

My question is: What is an acceptable practical way to govern the coaxiality of multiple cylindrical features (on a turned part for example) without using GD&T?

A general title block or drawing note?

Adding a datum and a FCF to something like a round knob seems to be overkill. I understand that doing that with a appropriately loose tolerance shouldn't be a problem.

The reality with a majority of our vendors may cause a cost increase when they see that.

 

[ewh]

You could dimension the surfaces of the cylindrical features from each other (difference in radius) and control the tolerances that way.

Believe it if you need it or leave it if you dare. - [small]Robert Hunter[/small]

 

[drawoh]

cwdaniel,

Adding a datum and a FCF to something like a round knob seems to be overkill.

Apply a sloppy tolerance.

Sometimes, I apply an FCF because the part looks precise, and I want to communicate to the fabricator that I do not care. Do not take the time and effort...

JHG

 

[ringman]

Try a drawing note, such as 'Diameters indicated to be concentric within .oox TIR.' As memory serves me it used to work for most applications without the ANSI callouts.

As FOOTNOTE and for thought, I recently ran across a ASME instructor who had it 'right'. "One of the reasons for using Geometric Dimensioning is that it makes the drawing more clear and AMBIGUOUS."

Think on that for a while.

 

[KENAT]

Look at 14.5, 2.7.3 (c), it gives a suggested note. However this note dictates perfect at MMC so may be tighter than you need, although if the size tolerance is relatively loose then maybe it works even for your knob example.

KENAT, probably the least qualified checker you'll ever meet...

 

[axym]

Ringman,

A couple of comments and questions about your last post.

The note "Diameters indicated to be concentric within .00X TIR" probably covers the coaxiality requirement, but as a former QA person I'm compelled to point out the inspection issues.

The TIR suggests something similar to a runout tolerance, but some important details are left out. For one, what are the diameters to be concentric to? Each other? Some other datum axis of the inspector's choosing? Also, there is no indication whether the TIR applies to the entire feature (like total runout) or to individual cross sections (like circular runout).

Regarding the ASME instructor's comment, did he/she really say that GD&T makes a drawing more ambiguous? If so, I would have to disagree. ASME Y14.5 has its share of ambiguities and gray areas, but they're nothing compared to plus/minus tolerancing. Did the instructor give any reasons for that opinion?

Evan Janeshewski

Axymetrix Quality Engineering Inc.

http://www.axymetrix.ca

 

[cwdaniel]

axym,

T.I.R. - Total Indicator Runout. Yes?

As to what is the datum, Could you pick a convenient round feature for a datum, make sure the other coaxial features meet the requirement and satisfy the callout?

After seeing the light from this thread. I've added a general note addressing this to our format. Do wee need to do more?

 

[KENAT]

I'd be sceptical of the old TIR note too, axym raised some good points.

ASME Y14.5M-1994 2.7.3(c)says:

Indicate this control for the features involved by a note such as PERFECT COAXIALITY AT MMC REQUIRED FOR RELATED FEATURES

cwdaniel, have you got 14.5 and if so have you read 2.7.3?

KENAT, probably the least qualified checker you'll ever meet...

 

[TheTick]

re: Runout GD&T frame

The reality with a majority of our vendors may cause a cost increase when they see that.

Send them a draft of each and see what happens. If they raise the price, start asking questions. Maybe teach them a thing or three.

Obviously, you need runout control. If the only thing driving the price change is the fact that the spec is GDT and not some convoluted nearly-adequate scheme, you need to have a conversation with your vendors. You also need a conversation with new vendors.

This is the price one pays for low grade vendors.

 

[cwdaniel]

I have 14.5 and I just read 2.7.3.

Fine, this works for me too:

Indicate this control for the features involved by a note such as PERFECT COAXIALITY AT MMC REQUIRED FOR RELATED FEATURES

Still, how does one verify perfect coaxiality without a datum? Same question applied to the T.I.R.

 

[KENAT]

I was thinking the same, I suppose it doesn't matter which you pick as datum. At MMC they must be perfectly aligned.

KENAT, probably the least qualified checker you'll ever meet...