Lofting from cylinders/circular cross-section sweep

[Chuffmechanicalengineer]

I'm rather puzzled at how I can loft from a circular-section sweep, to a flat surface (or to anything else, but I'm not worrying about that).

I have attached a screenshot and file below of my specific problem in question; I would like to 'wrap around' a surface from the curved sweep sweep in-front to each edge of the flat surface behind (this then needs mirroring for the other sweep at the back). The intended result is suggested by the sketch at the top.

I've tried all the ways of splitting it up I can think of, but the loft function (either surface or solid) just doesn't seem to be able to cope with cylindrical surfaces, or circular-section sweeps.

Could anyone please help?

[jhackney1972]

It is unclear, at least for me, what you desire to do.  You can extrude the sketch at the top using a surface extrude and turn it into a solid to join the pipes, that is what the sketch suggests to me.  A little more explanation is warranted.

[Chuffmechanicalengineer]

Sorry for the inclarity.

What I wish to do is as stated for the upper portion, but then with a similar shape along the entire of the fork, and the middle-section (planar surface modelled) is not of constant width, so I can't use simpler commands like extrude or sweep with guide surface.

[Chuffmechanicalengineer]

placeholder - duplicate message posted

[davebYYPCU]

That would be Sweep with Guide rail operation with 2 mirrors.

Taking the closest pipe to the camera in that pic, find the quadrant edge as if that pipe was sliced by the flange centre line. (9 o’clock position of the pipe)

Take your flange sketch and use one quarter as the profile, (half arc and tangent line)

Path is the edge of the vertical surface.

Guide rail will be the corresponding quadrant edge.

Can Demo later in the day if you get stuck.

Might help....

[jhackney1972]

---

@davebYYPCU wrote:

Can Demo later in the day if you get stuck.

---

I am "really" interested in seeing the actual solution model to this one just to understand what the original poster wants.  Please demo!

[davebYYPCU]

Reasonably sure he wanted this or close to it.

Geometry was already there.

3d edge from this centre line surface, 

1/4 of the flange as profile, 

Mirrored twice and capped.

Might help....

[jhackney1972]

Thanks, for the interpretation and the model, I hope you are correct.  I read his requirement any number of time and I could not see this. 

[Chuffmechanicalengineer]

That's the vague idea, but the 'pipe' (for want of a better term) in the front was meant to have a constant diameter, not to get scaled along with the new 'connecting' surface.

Images below should hopefully clarify this issue

proposed xz sectionvs. intended xz section (minus ends of pipe) — this has constant 'pipe' diameter and but still varying width of new surface

[Chuffmechanicalengineer]

Any ideas about how to do this? Has been quiet for a while...