[Frogblender]

The goal is a tube: with round cross-section at one end, and square at the other. (see picture).

The tube follows a given squiggly line. I tried bobcad's Surfaces->Sweep, but it only allows a one "profile curve" (in the picture, I've chosen the circle as the profile curve).


Anybody have any ideas?


FYI: this question comes from a Rhino vs. Solidworks vs. OnShape shootout, which I tried to replicate in bobcad, but got stuck

Attachment 395212

https://www.youtube.com/watch?v=xTZpGM2P9u4

[magicniner]

What are you making?
I use BCC to make things, I don't think it's the right tool for "Cad-Kiddie" modelling of things which will never be made as real world products ;-)

[Bazzer]

I tend to agree with Magicniner about using CAD for some real life purpose.

However I would attack that task with MOI (very similar to Rhino), I done it in about 2 minutes, circle at one end, square at the other and sweep between the two, but why?

Barrie

[Podunk]

One reason would be intake runners for a plenum style race car intake manifold. I'm doing one now but the centerline is straight so it is not a problem. The next one could be curved. I would be great if some guidelines from someone who's been there were available.
Terry

[Bazzer]

Terry

Yes I see your point.

I will try something today.

Regards

Barrie

[BurrMan]

The goal is a tube: with round cross-section at one end, and square at the other. (see picture).

The tube follows a given squiggly line. I tried bobcad's Surfaces->Sweep, but it only allows a one "profile curve" (in the picture, I've chosen the circle as the profile curve).


Anybody have any ideas?


FYI: this question comes from a Rhino vs. Solidworks vs. OnShape shootout, which I tried to replicate in bobcad, but got stuck

Attachment 395212

BobCad wont have tools to do this properly. Rhino will. Not sure about the Solidworks or Onshape...

Barrie mentioned MoI, but he also mention an issue which would be with all of these... The WRONG METHOD has been chosen.

Sweep along the path wont be it. As the sweep moves the profiles along the path, it will morph the shape... Super high end packages can give "More options" to hold fast, but there is a better option. Rhino has this.

You would do a "Flow", using either curves or surfaces to control the path... The object would then hold it's dimensions through the operation.

Here is a result! bbcd file zipped

[RAF.]

BobCad wont have tools to do this properly. Rhino will. Not sure about the Solidworks or Onshape...

Barrie mentioned MoI, but he also mention an issue which would be with all of these... The WRONG METHOD has been chosen.

Sweep along the path wont be it. As the sweep moves the profiles along the path, it will morph the shape... Super high end packages can give "More options" to hold fast, but there is a better option. Rhino has this.

You would do a "Flow", using either curves or surfaces to control the path... The object would then hold it's dimensions through the operation.

Here is a result! bbcd file zipped

BurrMan

What do you think about using Variable Radius?

RAF

[BurrMan]

BurrMan

What do you think about using Variable Radius?

RAF

Wow. Thats an awesome result RAF. I like to use fillets whenever possible, because the are calculating arcs to determine trim boundries.

They are super complex calculations though and can get problematic on mor complex geometries.

It's funny though. As weak as the filleter in BobCad is, i remember it handles calcs down to a zero point well, where others cant. Did you do tye final circle down to "0", or leave a small flat?

I'll have to test that on a curve that moves doen in a 3rd axis also...

Good thinking!

[RAF.]

BurrMan

I used 0 for the end rad Seems to work with some Z action maybe you can try one with a more complex shape.
My old computer started smokin and shaken and rockin and then went into a deep think mode and I was afraid to wake it up.

RAF

[BurrMan]

BurrMan

I used 0 for the end rad Seems to work with some Z action maybe you can try one with a more complex shape.
My old computer started smokin and shaken and rockin and then went into a deep think mode and I was afraid to wake it up.

RAF

That's Rad RAF!

BTW: the "Zero" was referring to the point to where the fillets meet. But it looks like you are using a 1" as the base. So all the ".25's" meet at a full arc!!!
A lot of times, that has to be "Faked" with a small space between where the fillets need to meet. Again, ole BobCad has always seemed to handle that!

Will have to fool around with this suggestion a bit more!
Thanks.

[BurrMan]

Well, looky there!

[The Engine Guy]

RAF
Burrman

Absolutely brilliant , I never would have thought of using the variable rad fillet, seems to work beautifully on any shape I have tried so far.

Keep up the good work, good old BobCAD-CAM ain`t as dumb as some folks would like to make out

Regards
Rob

[RAF.]

RAF
Burrman

Absolutely brilliant , I never would have thought of using the variable rad fillet, seems to work beautifully on any shape I have tried so far.

Keep up the good work, good old BobCAD-CAM ain`t as dumb as some folks would like to make out

Regards
Rob

Thanks Rob
RAF