Datum(s) on formed tube 5

[cpdpeckh]

Is there a recommended method for specifying datums on formed tubing, such as formed hydraulic tubing or formed exhaust tubing. I care about the profile along the length of the tube and I care about the location of the ends with respect to each other.

 

[ctopher]

I would use total-runout or cylindricity with profile tol.
Look up ASME Y14.5M 1994.

Chris
Sr. Mechanical Designer, CAD
SolidWorks 05 SP1.1 / PDMWorks 05
ctopher's home site

 

[Heckler]

Datum Targets....three points define a plane

Best Regards,

Heckler

Do you trust you intuition or go with the flow?

 

[MadMango]

There is/was a MilSpec for tube and bar bending, try researching MIL-D-9898C.

It delt with assigning points on the tube, like "Position 0" for starting end, Position 1 for first bend, etc. It also mentioned Bend Angles, straight length dimensions, direction of rotation (CW or CCW) before next bend, etc.

[green]"I think there is a world market for maybe five computers."[/green]
Thomas Watson, chairman of IBM, 1943.
Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[MikeHalloran]

A tube _drawing_ is easier to understand if you orient the longest straight piece with the drawing's coordinate system. Then you can dimension the ends relative to the coordinate system and any tertiary datum along that length of tube.

There are also good reasons for orienting the part relative to one end, or the extreme tangent of one end, which is what I think you need in order to program a CNC bender.

Third choice, sometimes the least awful, is to use the coordinate system of the assembly into which the tube fits as your datum coordinate system. This can get messy fast because in the general case every important point requires three linear dimensions and three angles.

The best datums, IMHO, are not imaginary like centerlines, and not arbitrary like tube walls at particular planes, but are the actual features at which the tube touches the remainder of the world, e.g. flange faces and holes in hanging features.





Mike Halloran
NOT speaking for
DeAngelo Marine Exhaust Inc.
Ft. Lauderdale, FL, USA

 

[drawoh]

cpdpeckh,

If it were me, I would use one end diameter as my primary datum, the end face as the secondary datum, and a point out on the tube somewhere to control rotation.

How are you going to inspect this thing? A sketch of your inspection fixture might tell you how your datums should work.

I do not think there is a recommended procedure anywhere here beyond making sure your part is constrained in six degrees of freedom, and that your datums are some sort of functional feature if at all possible.

JHG

 

[cpdpeckh]

Thank you for all the comments and suggestions.

Does anyone use true position to control tolerance between the two ends of a formed tube?

Don Peckham

 

[ctopher]

no

Chris
Sr. Mechanical Designer, CAD
SolidWorks 05 SP1.1 / PDMWorks 05
ctopher's home site

 

[ewh]

An inspection fixture should be easy to set up to inspect true position between the ends, so I would think that this would be acceptable.

 

[Heckler]

The datums used to define the fixture should also represent the functional datums for the finished part.

I wouldn't use total runnout you would never get good parts. I would use a profile tolerance providing your datum scheme is thoughtout.

Just remember the part has to be manufacturable and inspectable.

Best Regards,

Heckler

Do you trust your intuition or go with the flow?

 

[drawoh]

cpdpeckh,

A positional tolerance sounds good to me. You want the other end of the tube located somewhere, and the positional tolerance defines that. I am sure there are other ways to do this, but I would not spend any time worrying about it.

If I were inspecting your drawing, the primary thing I would look for is your tolerances. If they are not loose enough, the fabricator will assume you do not know what you are doing. Bending tubes cannot be a very accurate process. This thing will be somewhat flexible, right?

JHG

 

[cpdpeckh]

drawoh

We are using SAE J2551 (Recommended Practices for Fluid Conductor Metallic Tubing Applications)recommended total true position tolerance for hydraulic tubes so tolerances should be loose enough.

In the past, we specified true position tolerance at the "B" end of the tube and IMPLIED that the tolerance was with respect to the "A" end of the tube.

It is my understanding of ASME Y14.5M that "Location" tolerances must specify the datum. Therefore, I need to identify end "A" as a datum, and reference the datum in the position tolerance.

I think this will work.

Don Peckham

 

[type26owner]

Position is used to locate features_of_size by their centerlines or centerplanes not to locate surfaces.

Runout is used to define parts that spin on an axis. Either in usage or when being manufactured.

If you would describe how the part functions I can suggest the best datums to define it. The most important features which mate or aling or both are the best datums according to the standard. Not adhering to this basic principle causes no end of confusion.

 

[cpdpeckh]

type26owner

Thanks for your comments.

The tube is 1/2" diameter and has ORFS fittings on both ends. The tube is "U" shaped with the "A" end connected to a straight fitting on the side of a transmission. The tube runs horizontal to the left, vertical, and then horizontal to the right and about 5 degrees toward the rear in a horizontal plane. The "B" end connects to a 90 degree fitting on top of the transmission.

For this particular tube there are no clearance problems along the length of the tube. However, we need good control of the location of the ends.

I welcome your suggestions.

Don Peckham

 

[type26owner]

"However, we need good control of the location of the ends."
Those ends are the datums. This is a rather unusual part but the logic still applies. Since this part is probably flexible then it's the usual practice to simulate how it's restrained in use and replicate that in the fixturing to check this part for acceptance or not to the tolerance.

I still can't clearly comprehend this part. Have not a clue what those fittings look like. Are they brazed onto the tubes for instance? If they are then I suggest the detailed drawing consist of the finished assembly and not the tubing alone. This way the fixturing is simple just simulate the mating fittings in the gauge to a gauge tolerance.

 

[cpdpeckh]

type26owner

Thank you for your comments.

The tube ends are flared and the contact surface is on the flare. The "ORFS FLANGE" figure on page 1 at the following link shows an example.

http://www.parker.com/tfd/fittingsolutions/TUBEPR.pdf

The three straight lengths are 9" to 12" long and this is a 1/2" diameter tube.

Don Peckham

 

[type26owner]

Suggest the mating gasket surface be the primary datum and the female thread of the aligning and mating nut be the secondary datum. Those two datums of one end's fittings would surfice to constrain 5 of the 6 degrees of freedom. The other fitting's similiar features would constrain the sixth. See the logic of detailing the flared finished assembly with captured fitting pieces included versus just a weird shaped piece of tubing?