concentricity

[Duckhunter]

I have a cylinder with 3 bores all in line. the first diameter is dat. k the next diameter is not a datum, and the third is dat.f........The print calls for the concentricity of the second diameter to k-f. Do I create an alignment to k-f then dimension the conc. to either of the diameters?

[kbotta]

could you post a pic of the drawing please (or a portion thereof)?
Thanks,
Kevin

[Wes Cisco]

How I would do it

I see this kind of call out a lot. If possible measure one autocylinder that takes hits on both datums. If this is not possbile to due to different diameters or location on part, then measure a circle on each datum with at least 7 hits for each circle. (Use an odd number of hits.) If the gods are on your side you will be able to create a cylinder from the two circles. Be careful, pc-dmis may use you data to create somekind of monster cylinder with wrong diameter and vectors. If it does make a good cylinder, then treat it as one datum and make an alignment that levels to it. If it does not make a good cylinder you will have to use the two circles to create a 3D line. Create an alignment by leveling to the line, then use one of the datum circles for origin. ( Use your best judgement as to which one.) Make sure you check the location of both datum circles after the alignment to make sure it is correct. HTH

[Duckhunter]

Quick drawing

See if this helps

[Duckhunter]

I kinda thought that I needed to create a 3d line, level, and origin to one of the cylinders...wanted a little input

[Wes Cisco]

yes 3D line

3d line is best way I know of .

[craiger_ny]

I hate that datum deal with the hyphen (A-B). I'm not sure it is even legal. One of our suppliers asked us to add it to one of our drawings for a part they supply during a design review and against my protests our engineers added it anyway. Not only did they add it but they picked it up as a convention and use it on other drawings. Best part is I can't get them to interpret it. The best answer I get it "It means it has to run true to both datums". I say to them "why don't you control the two features back to one datum instead of controlling one to some made up datum structure like this A-B crap?"

Funny how it works, if one of our suppliers wants a drawing change they get what ever they ask for because it might effect delivery. If one of us manufacturing guys asks for a change engineering acts like we are telling them how to do their jobs and doesn't give us squat. They know good and d@mn well we aren't going to hold up delivery even if the drawing can not be interpereted. If we do it is our a$$ and that is not engineerings problem.

Ayway I am done ranting, I hate that callout.

[kbotta]

I think Wes gave some good advice.
I was hoping to see the FCF in question. This looks like more of a total run out application...
Kev

[Paul Sarrach]

I'd go with the 2 circles

[Underspec]

You don't have to set origins to either circle. It would be easier to analyze because you zero out the X/Y of the datum circles.

You can set datum to k then take middle diameter then insert concentricity. Set the datum to f then do the same. It really doesn't matter, PCDMIS just calculates the difference between the two anyway.

Align the Plane and use circles.

[craiger_ny]

I think Wes gave some good advice.
I was hoping to see the FCF in question. This looks like more of a total run out application...
Kev

Here is the thing I was ranting about. It can't be legal, our engineers can't tell me what it means yet they are putting it on other drawings as well. Stupid, just plain stupid. I circled in red the lunacy.

[Matthew D. Hoedeman]

One diameter is the C datum, one is the D datum, then each one has to be within runout 0.002" to the line from C to D. What it boils down to is they want a 0.002" roundness held on each diameter, that is all that can be 'pulled out' from those callouts.

[george frick]

i would measure the two dias k&f and create a 3d line between the two. then level & origin to the line. then measure the middle dia and dimension the concentricity. this will be evaluated as concentricity between the axis which i believe the print implies.

[kbotta]

Yes it is 'leagal'. The 14.5 standard even gives a sample of this.
There is Figs on
pg 173 FIG 6-21
pg 68 FIG 4-21

And an explanation on pg 67 4.5.7.2

This shows it 'legal', and what it means. Absolutly correct.
Kevin

[kbotta]

Although, nose pickers application seems a little funny...