Circular runout measuring argument

[sbozy25]

OK, so I am in the middle of a huge argument with my head quality tech. He has no college degree and no formal training with GD&T, however he thinks he knows everything there is to know about everything in the world...

I on the other hand have a formal degree, have had formal training in GD&T, and have dealt with it for a while now...

So any way, here is a picture of what we are arguing about..

http://i77.photobucket.com/albums/j66/sbozy25/GDT.jpg

Sorry for the crude drawing, but I am in a hurry. Ok, this is a bellview washer shown as a section view. We are trying to check the circular run out of the inside diameter's surface in a couple places. I personally think we should be checking total run-out, but that is another argument for another time. Well any way, our quality tech. approved the gauge for this part without speaking to me, and I think he goofed big time. The gauge he approved, comes up from under the part and 3 prongs extend to grab the ID and hold the part steady. Then a wheel hooked to a dial indicator comes in and touches the entire OD. The part is then rotated 360° and the high point is then displayed on a computer screen. He seems to think this is an ok practice for checking the run-out of the ID.

Well, here are my problems...

#1, this is a stamped part so there is shear wipe to deal with near the edges.

#2, he is checking the entire surface @ once rather than the 3 places he is asked for.

#3, he is an idiot...

Personally, I think the part should be grabbed on the OD, and an indicator should come down and touch off on 1 spot on the ID and then the part should be rotated. Am I way off base here, or am I right in thinking this guy has no clue what he is saying...

 

[ctopher]

I think you are correct. Especially for a stamped part. The part should be inspected per the print.
BTW, there should not be a leading zero. It may be confused for metric.

Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 04-21-07)

 

[ewh]

I agree with you. With his setup, he is inspecting total runout on the OD, nothing to do with the requirement listed.

 

[sbozy25]

Thank you both!

Oh, and the actual print says that all dimensions are in inches. So there should be no confusion.

 

[btrueblood]

1. Agree your QC tech is warped.
2. The all-around circle on the leader is redundant.
3. Runout relative to what datum? The way it's shown in your drawing, you are implying runout relative to the center hole, since no datum reference to the o.d. is shown.

 

[TheTick]

No such thing as an implied datum. Your runout needs a datum reference.

Otherwise, consider a profile tolerance.

 

[sbozy25]

Sorry...

I was in a hurry and used microsoft paint to produce the picture. Yes there are datums, and no the all around circle is not on the real print. I quite literaly took 60 seconds to sketch that out. I was just looking to see if I was thinking correctly or not. I believe there is one datum on this part and it is the center axis of the part... I am working off of memory here, seing as how my quality tech is gone for the day and has the print somewhere in his desk. Unfortunately, our customer maintains design control and we can not change how things are called out. So I am stuck with what they are asking for... I just need to find a way to get it in quality's head that they are wrong...

Keep in mind, this is the same guy I spend 45 mins arguing with one day because he swore up and down that he had the ability to check to 6 decimal places with his measuring equipment. I finally convinced him of his wrongness (is that a word?) when I was able to fudge his last few decimals by flipping the light switch on and off....

 

[ctopher]

Maybe you can show us a pic of your CAD dwg? We may help you more if we see the whole picture. Showing us a quick sketch and telling us this guy is an idiot will attract more questions.

Chris
SolidWorks 07 4.0/PDMWorks 07
AutoCAD 06
ctopher's home (updated 04-21-07)

 

[ewh]

While I still agree that your part is not being inspected correctly, there may be other problems associated with its definition.
Your datum should not be "the centerline of the part", but the centerline of a feature, in this case perhaps the conical surface of the washer. Para 4.3.2 of ASME Y14.5-1994 states "The datum feature symbol identifies physical features and shall not be applied to center lines, center planes, or axes...". The exceptions include equalizing datums and datums established from complex or irregular surfaces, neither of which is the case here.

 

[sbozy25]

Trust me I would love to post up a picture of the CAD drawing. However, we were not and will not be given one. This customer does not believe in handing out electronic versions of it's prints. you only get hard copy.... Sorry...

But thank you all for your responses. I believe I have enough fire power now to go after this guy and his boss and get something done for a change.

 

[dingy2]

The circular runout (wouldn't use total runout) doesn't reference a datum but I assume that OD but one would need a another datum since the feature (OD) is very thin. I would suggest the mounting surface. I would further suggest that the mounting surface is datum A and the OD is datum B.

Yes the checking fixture should locate on the OD and the feeler contact should be at least contacting close to the "cut" and then rotate the part 360 degrees. If the requirement is 3 places (I don't know why) and the locations are not specified on the drawing, then I would take readings near the "cut", centre depth and then near the "break". You will find the readings (TIR or FIM) will be similar in the 3 locations.

The checking fixture is not made correctly.

Hope this helps.

Dave D.

http://www.qmsi.ca