Flatness or Profile of a Surface? 1

[powerhound]

Recently I've been pondering something; I create drawings of tubes of all different sizes, the largest being about 13 1/2" in diameter with about a 4-6mm wall thickness. The way these tubes have been specified to keep them from rocking when set on end is to put a flatness callout on the end. Since this is essentially a ring and there isn't much area to establish a flat plane, should profile of a surface be used instead and just don't use the option of dimensioning it from a datum? I've had some good input from someone I highly respect for his knowledge of GD&T and I'd like to pose the question to the community and see what the general consensus is. Thanks.

Powerhound
Production Supervisor
Inventor 11
Mastercam X
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[MechNorth]

Profile of a Surface without datum references results in a form control only. In this case, it would be a flatness control. In other words, it would be the same either way.

If you make either the inner or outer cylindrical a datum feature, you can relate the base to it using the Surface Profile, which would give you perpendicularity to the central axis, if that's of value. If you used a composite surface profile in this same example, with a datum reference on the first level but no datum ref on the second level, you'd have the perpendicularity from the first level and flatness from the second level.
Hope that helps.

Jim Sykes, P.Eng, GDTP-S
Profile Services
CAD-Documentation-GD&T-Product Development

http://www.profileservices.ca

 

[CheckerRon]

I like MechNorth's suggestion of composite profile. It may seem like overkill, but it controls both the rocking problem and perpendicularity (tilt) of the tubes when standing on end (assuming they are freestanding). I guess a tight perpendicularity or runout tolerance datumed to the OD would do the same.

 

[dingy2]

If you are interested in flatness of this surface, place flatness and it can be checked or confirmed relatively easily.

I do agree with MechN that profile of a surface relative to a datum at the ID or OD will give one perpendicularity to the axis but it is difficult to confirm. I probably would suggest a circular runout which would be a bit easier.

 

[gpavlab]

The most accurate GDT method you could use for a pipe would be cylindricity, that would provide the most precise control. Cylindricity is probably not being applied in your case for the same reason it is rarely in most cases-that is it is the most difficult feature control to manage. That would also mean one of the most expensive to work with. Straightness however is one of the easyist to apply,control and manage (hence good costwise). So even though it is not the perfect feature control for your case it will work, especially if you use it along with concentricity of the I.D. and O.D. of the pipe.

 

[MechNorth]

gpavlab, the original post was about the end of the cylinder rather than the sides. Cylindricity would provide a form control for the outside of the cylinder, but nothing for the flatness of the end or for the perpendicularity of the cylinder end wrt the axis of the cylinder.

Jim Sykes, P.Eng, GDTP-S
Profile Services
CAD-Documentation-GD&T-Product Development

http://www.profileservices.ca

 

[dingy2]

Cylindricity is not relevant here.

MechNorth - Profile of a surface without a reference datum usually reflects flat surfaces that share the same plane rather than a single surface or feature. One would set the features up as one plane and confirm the profile similar to confirming flatness.

Flatness symbol is used on a single flat surface so it may be more appropriate in the second level of the composite feature control frame.

What do you think?

 

[powerhound]

Thanks for all the input guys, I don't guess I was too far off. I'll stick with flatness for now. Perpendicularity is always specified seperately because we generally have a lot of leeway in that regard but not as much in the flatness department. You guy definitely rock.

Powerhound
Production Supervisor
Inventor 11
Mastercam X
Smartcam 11.1
SSG, U.S. Army
Taji, Iraq OIF II

 

[MechNorth]

Hi Dingy2,
I believe you're referring to Fig. 6-20, which is a specific case where no datums are specified and two surfaces want to be related (i.e. within the same flatness and therefore loction zone), however this only works on nominally co-planar features. Taking a step back from that, if you have only one surface, the surface profile without any datum reference only controls the form; which form it controls is based on the type of feature (line element = straightness, planar surface = flatness, circular line element = circularity, and cylindrical surface = cylindricity).

Surface Profile is an inclusive control, and amazingly robust once you get used to it. Learning now to read it in the different situations seems to be a challenge for many of the people I've worked with, but once they understand it, it becomes a preferred control.

Jim Sykes, P.Eng, GDTP-S
Profile Services
CAD-Documentation-GD&T-Product Development

http://www.profileservices.ca

 

[jakfishjak]

Are these tube ends faced in a lathe? would runout confuse the issue?
regards
Jack Abram

 

[ctopher]

In addition to above suggestions:
"...13 1/2" in diameter with about a 4-6mm wall thickness..."
The inches and mm should not be mixed.
Just my 2 cents.

Chris
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 10-27-06)

 

[ringman]

Wat if they purchased metric stock sheets and formed, then welded? Hmmm. Just a thought.

 

[ctopher]

... then you would have a metric welded part.

Chris
SolidWorks 06 4.1/PDMWorks 06
AutoCAD 06
ctopher's home (updated 10-27-06)