Drafting of Complex Pipes 3

[boisebiker]

I am trying to dimention pipes that are to be bent on a CNC bender and have troubles getting the true angle of the bend. To get the true angle you need both legs of the bend on the plane of the paper. If the pipe has multiple complex bends then is is very dificult to get both legs on the plane. From the 3d software I can create a plane using the centerline vectors of the bend and then place a view using the created plane. But I am finding this method to have a lot of resistance from the checker because I cannot explane the orientation of the view from the standard views. Is there a drafting standard for this type of problem, can I place a view that has no easy tracability to the standard 3 views? What is the preferd method of drafing compex pipe/tube shapes?

 

[ctopher]

Add a bend table to the drawing with point coordinates of each bend.

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[boisebiker]

There is already an XYZ coordinate system on the draft for the CNC bender, but this doesn't help the guy who has to inspect the pipe. I want to put down the dimension that he can measure with out getting out his calculator.

 

[ewh]

???
While it may take much thought, properly defining view orientation is important if the part is to be defined on paper. It may take three or four projections to get a single view to dimension, and many drawing sheets to completely define the part.
Think what drafters went through when descriptive geometry was the only way to accomplish this and count your blessings.
I realize that this isn't what you wanted, but have patience and think it out.

Believe it if you need it or leave it if you dare. - [small]Robert Hunter[/small]

 

[MikeHalloran]

I used to work in a shop that did complex marine exhaust pipes with short radii and double walls. We dimensioned the location of both ends of each elbow or tube in regular coordinates, and put the included angle and radius of the elbows in the bill of materials.

This would not be of use to anyone programming a tube bender, but our radii were too short to bend anyway, and it worked okay for inspection.



Mike Halloran
Pembroke Pines, FL, USA

 

[ctopher]

Look at the example here:

http://www.tube-bending-service.com/pdf/tube-design.pdf

You need a table to indicate the bends.

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog

 

[MadMango]

Check out this book, specifically page 42.

Tube Forming Processes
Gregory Miller
ISBN 087263633X

Its hard to use strictly XYZ coordinates with working with compound bends in a tube. You will need a table that defines starting angles, ending angles, CW/CCW rotation in degrees, etc. You drawing will have many auxiliary views to get the straight tube length for inspection. You finished drawing should look similar to fig 1-7 of the above book.

The work around consists of multiple Go/No Go checking fixtures.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[boisebiker]

Thank you for the comments so far, they have helped a lot. Is there any way to place a view on a drawing that is not a projection or auxillary view of an exsisting view? Can I defign or name the view so the reader knows where it comes from, or do all views have to have visual orientation to a standard 3-view?

 

[MikeHalloran]

At this point, I think you and the checker have to agree on just exactly who is going to be reading the drawing and for what reason.



Mike Halloran
Pembroke Pines, FL, USA

 

[MadMango]

What you can and can't do with your CAD program should be asked in the appropriate forum, since its not related to drafting in general.

"Art without engineering is dreaming; Engineering without art is calculating."

Have you read faq731-376 to make the best use of Eng-Tips Forums?

 

[boisebiker]

This is not a CAD capability question but rather a Drafting Standard question.

 

[Xplicator]

I believe that ewh has given you the solution. It seems that it just doesn't want to be accepted, because of the work involved.

If you care to look at the the standard (ANSI/ASME Y14.3-2003) for Multiview and Sectional View Drawings, I only have a copy of Y14.3-1975[affirmed 1980] and unless it has changed, which I doubt, use of auxiliary views from standard view to first aux then to second aux is defined to show true shape and relationship of features that are not parallel to any of the principal planes of projection is defined. In the days before CAD, I have even prepared (board) and witnessed on roll size formats (H,J & K) chained auxiliary views. Where starting from the standard projection plane of the most important view (defining datum)there was a first aux then second then third. I have even witnessed fourth. I do realize, as I'm sure all here do, this corresponds to lots of work.

However, I believe that for this type of product/industry, the best way to depict and define is this method as opposed to the ambiguous non-standard "shotcuts" evolving, mostly because of the time required to do it correctly and unambiguously in a standard that all understand.

My suggestion is select the most important section/interface end then be consistent in depicting and specifying with respect to (w.r.t.) it. There is also the option of using partial relevant only views in the auxiliary view to save real estate if you choose.

 

[CheckerRon]

I too think that ewh (and Xplicator) is right. Fig. A3, "True Lengths and Point Views of lines" is in ASME Y14.3- 2003, and defines such views, but doesn't show angles, radii, or straight lengths.
I my piping detailing days (spools and ISO's on the board) we did single line representation, and showed compound bends by two adjacent views, 90° apart and dimensioned the bend (say) 42°H in one view and 17°V in the other view.

 

[CheckerRon]

At my current job, we detail bent tubing for fuel and oxidizer lines using a table showing X,Y,Z bends and lengths center to center for an NC bender. All dimensions are Basic, and an all around profile tolerance without datum is applied. The drawing pictorials are just that---pictorials, defining the X,Y,&Z directions but no dimensions. The bending table is fine, but I would prefer seeing length, angle and radii imensions on the pictures also, as I think boisebiker's checker expects to see.

 

[11echo]

You don't need to show any "views" of this fabricated bend! All you need to do is call it out. Just show how it would look in a normal plan and or section drawing, then call out the fabricated elbow as follows: 12" 29Deg. 30min 10R. Elbow (my example only)
With this call out a fabricator can manufacture any elbow required. The only other thing is I would try to take the accuracy any tighter then 0.5 Deg.s, that is about how accurate a fabricator can get. The only coordinates or dimension you will need is for the contractor to use this fabricated piece. Good Luck!

 

[ewh]

What determines a compound bend with that callout?

Believe it if you need it or leave it if you dare. - [small]Robert Hunter[/small]

 

[11echo]

When piping is laid out you call out the components, then dimensioned as to how they are laid out. Case in point, if this was a “standard” type layout with “common” type fittings like a 45 Deg. elbow. You would call out the component a 45 Deg. Ell and its position (Vert. or Horiz.) and then you would dimension how these fitting were assembled. Same applies here, the elbow may be a non-standard angle …i.e. a fabricated fitting, but a single call out would identify it, as I have indicated. If this was a double rolling offset, it would normally be fully documented in an isometric dwg. or detail, where you can "see" the true relationship of the piping.

 

[ewh]

For simple angles that may work. The OP was referring to compound angles, which, as you note, need to be somehow be fully documented.

Believe it if you need it or leave it if you dare. - [small]Robert Hunter[/small]

 

[boisebiker]

Thanks to all of the help from here we have come to a resolution. We have added a table of segments and bends to the drawing to defign the pipe. The drawign views are just pretty pictures for the reader to help visualize the finished pipe, with a few dimensions to give orientation with the table.
Thanks for all of your help and sugestions, it has been very insightfull.

 

[ctopher]

boisebiker,
It is the only way I know of how to do it.
Thanks!

Chris
SolidWorks/PDMWorks 08 1.1
AutoCAD 06
ctopher's home (updated 10-07-07)
ctopher's blog