Circularity of a medium quality extrusion

[lsmaclea]

The part in question is a simple tube.

I've attached a picture of the dimensioning scheme I have going so far. I want to keep this part simple and cheap. The problem that the part is made from an extrusion of nylon 6. The machine shops that manufacturing has talked to so far want to shear the part to length, rather than cut it on the lathe.

I don't want to "ovalized" the tube during manufacturing, but the tube does not need to be a perfect circle either.

I want to put some sort of circularity control on the part. Can I spec a circularity control that is actually more generous than the envelope 1 diameter FOS controls already put in place? I don't think that the shops will feel compeled to stay within the envelope 1 diameter FOS controls.

Thanks!

 

[KENAT]

What drawing standards do you work to as this may be a case where there is a default difference between ISO & ASME?

You could look at specifying in the constrained condition if this suits your function.

KENAT,

Have you reminded yourself of faq731-376 recently, or taken a look at

http://eng-tips.com/market.cfm?

 

[dingy2]

I noticed that you have total runout to itself on this drawing which is incorrect.

If you want to control the roundness and straightness of this part, I would suggest that you call out a straightness diametrical tolerance zone tolerance of a value beyond MMC. You could also use a zero tolerance at MMC which is quite tight but legal.

One would then make tube for checking this part with the ID straight and its virtual condition size.

I am a bit confused on the size of the OD. Is it .38 or .31? There must be a tolerance of size somewhere but it was not on the drawing. Let's say the OD was .380 rather than .310 and the tolerance is +/-.003. If we had a straightness of 0 beyond the MMC, we would get a virtual condition size of .383. That would be the ID size of the gauge.

The part must fit through the gauge after processing which is pretty simple.

Dave D.

http://www.qmsi.ca

 

[powerhound]

...not to mention that the gauge must be at least the length of the part.

Powerhound, GDTP T-0419
Production Supervisor
Inventor 2008
Mastercam X2
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[dingy2]

Good point powerhound

Dave D.

http://www.qmsi.ca

 

[lsmaclea]

Ok, great stuff so far.

I'm not sure I understand why my Runout is improperly located. I've been working from this site for my runout specifications.

http://www.tec-ease.com/tips/march-00.htm

Do I have to actually put my Runout tolerance near the part diameter callout, even though they are from different views?

To clarify a few questions, the part's OD is 0.38 plus or minus 0.010.

The implied zero tolerance at MMC is too tight for this part. It is meant to be a very cheap, simple part. I've gotten rough quotes of about a dollar fifty each. I do not want to put a heavy inspection in place.

Is inspecting a straightness tolerance time consuming?

 

[TheTick]

You can apply the runout if you specify datum targets. Otherwise, you have a circular reference, where a datum controls itself for the entire surface.

You may want to consider single runout in this case.

 

[dingy2]

Both circular and total runout on an OD should be relative to another diameter usually the ID. In your pic. you had a total runout to itself which is incorrect. The tec-ease example is an OD to and ID.

Straightness at MMC - how long does it take to push your part through a tube once the gauge is made - 10 seconds maybe. That is it.

Dave D.

http://www.qmsi.ca

 

[TheTick]

Let me clarify...

If you specify a datum target zone where the OD is to be clamped, the unspecified parts of the OD are valid for the runout control.

 

[dingy2]

TheTick:

Sorry but that is not quite correct. Please go to page 189 of the standard. It states "runout is a composite tolerance used to control the functional relationship of one or more features of a part to datum axis."

Please see examples on pages 190 - 193.

Datum targets only show where on the datum that one is to contact the part. Datum targets (in most cases) do not create the datum but only set up on it.

Dave D.

http://www.qmsi.ca

 

[TheTick]

"runout is a composite tolerance used to control the functional relationship of one or more features of a part to datum axis."

If the datum axis is determined by datum targets, how is this approach invalid?

If there is a .0005 deep line scribed on the OD defining the datum target, then somehow magically we have two separate features.

 

[TheTick]

<

http://www.tec-ease.com/tips/november-99.htm>

 

[lsmaclea]

Interesting. So, if I move my datum target to the ID of my part, then my total runout is valid.

Does that make the inspection any more complicated than putting the part in a lathe and figuring out what the total swing of an indicator is as the part turns? I don't have to do anything like calculate a virtual axis, etc?

And around here, getting the documentation through to create an inspection gauge is a PITA. My goal is to create a dimension that the machine shop can very easily hit provided that they start with a relatively straight piece of material. However, some of the shops want to buy the raw material in rolls. The parts made from the roll are quite bowed. I need to make it clear that the parts should start from straight material stock.

 

[dingy2]

You do not need a datum targets shown on the part but you do need a different datum shown. If you datum is now the ID, then one could have either a circular or total runout of the OD. Now it is valid but both runouts are time consuming to check.

Anyway, the straightness of the OD at MMC will do the trick and making a gauge longer than the part of course should not be a problem.

Dave D.

http://www.qmsi.ca