TurboCAD Forums

The Ultimate Resource for TurboCAD Knowledge

Register
 
Be sure your post is relevant to the current discussion.  Create a new topic within the appropriate board if you are unsure.

curved wedge Pro 19
Read 6739 times
* June 01, 2014, 10:31:43 PM
Hi,  I want to 3d Add a curved wedge to a simple extruded bas, as in the drawing.  I can't think of a way to build the 3d wedge shape.  I've tried Lofting with no success yet.  Can you help me?

Thanks, Leigh

Logged


* June 01, 2014, 11:09:55 PM
#1
How about something like this for a start?

Logged
Gary Wooding
Win10 64-bit,
TC21.2 x64 Plat, Bld59
TC16.2 Plat, Bld54.0
TCC 3.5


* June 01, 2014, 11:22:51 PM
#2
You can probably skip the Slice part. Select the revolved part and change the "Angle of Rotation" to something less than 360° in Revolution Shape Properties; provided it's still a 'Spin' object.

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1903), 64-bit


* June 02, 2014, 12:28:02 AM
#3
Thanks, but sorry I didn't make my question clearer.  The raised part is a wedge shape, starting at the level of the base, and rising up as it curves around.  I hope this picture shows it clearer.

Thanks.

Logged


* June 02, 2014, 01:14:38 AM
#4
I've got to go out, so I haven't time to document the steps, but you can draw it by combining two shapes.

1. Create the section of flattened torus by rotating a rectangle as shown.
2. Draw a smaller rectangle that will be the largest end of the actual wedge. Let its hight be h
3. Draw a helix with centre at the centre of the torus, top and bottom radii equal to the innermost corner of the smaller rectangle, one turn, and hight equal to H.
H is equal to 360/angle of torus section times h.
Sweep the smaller rectangle along the helix. Cut a section of the resultant sweep to fit on the torus section, position it carefully, then add the two pieces together.
I'll try to document it properly when I return later this afternoon.

Logged
Gary Wooding
Win10 64-bit,
TC21.2 x64 Plat, Bld59
TC16.2 Plat, Bld54.0
TCC 3.5


* June 02, 2014, 01:33:15 AM
#5
After you've created the "Spin' object and cut it down, try the Facet Editing tool to create the ramp.
  • Select the top facet with the tool
  • Move the Reference Point to one corner ("D" key)
  • Move one of the rotation nodes to the other corner ("Y" in the picture)
  • Hold the Ctrl key and mouseover the node. When the cursor becomes a hand, you can click on it to pick it up, then snap at the new location.
  • select that rotation node and move it down.
  • Once I started moving it, I used the up/down arrows to get a rough idea of the degrees, then fine tuned if necessary.

My 2¢

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1903), 64-bit


* June 02, 2014, 04:54:53 AM
#6
Thank you people,  I shall study your solutions and let you know how I go, but at first glance I feel you have given me the answer.  I haven't dabbled with facet editing, so something new.. great!

Much appreciated.

Logged


* June 03, 2014, 01:26:55 AM
#7
I've realised that this computer has Pro Std 15, not 19.  I haven't been able to find a Helix tool to follow lemel_man's suggestion.  I have almost succeeded with John R's facet editing solution, but not quite yet.  Still experimenting.  It's nearly there, but matching the thin end to meet the edge of the base is challenging.  Is it time to upgrade?

Thanks for your help.

Logged


* June 03, 2014, 02:56:35 AM
#8
Sweep will do the helix for you with a >0 twist figure.  Draw a line as path the height of the wedge, sweep the wedge's profile and specify the degrees of twist that the wedge needs.  You can also do it with a revolve using spiral parameters.  Do it using either with compound profile checked and you can edit the parameters after creation.
« Last Edit: June 03, 2014, 04:47:40 PM by murray dickinson »

Logged


* June 03, 2014, 10:25:09 AM
#9
re: …Pro Std 15, not 19……I haven't been able to find a Helix tool…

The Helix tool was introduced in V16. You may stop looking now.

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1903), 64-bit


* June 04, 2014, 04:30:31 PM
#10
I fired up V15 and created the file attached, one created with sweep, one with revolve, as I'd suggested earlier.  It's interesting to note how far TC's come.  I could get away with a lot more in later versions than in V15, V15 peppered this exercise with selfintersection warnings, and limited the revolved example to no lower a figure than .21whatever-of-a-coil, presumably for some variation of that.  With sweep, an open profile for the ramp was much more tolerant than closed, I assume because a closed profile will produce a selfintersecting solid.  I remember making the decision to upgrade from V15 to V17 because V17 introduced more sophisticated beziers, but I think it's easy to overlook the other improvements that've happened in less obvious areas.  (Maybe, just maybe, I'm better than I used to be, too ;)).  Aside from this, which I assume is ACIS kernel behavior, the other thing that struck me is how limited the inspector bar is compared to later versions.  No deltas, no position, no size.  Kudos IMSI.   Still don't like the dark colour scheme, though.  I'm older, my eyesight's worse, my PC's are smaller.  Dark is detrimental, not an improvement.

Logged


* June 04, 2014, 04:39:22 PM
#11
Thanks Murray, and others who have offered help.

I might have to bite the bullet and upgrade.  Cad is only part of what I do, and mostly I can easily do what I need to with v15.  But I am sure that there have been vast improvements (apart from the colour scheme, one can't change the skin?).

I have a few projects on the go at present, so haven't been able to spend the time to crack this challenge, though making progress.  I shall study this drawing, and no doubt will be successful.  If not, I'll be back.

Thanks again.

Logged


* June 04, 2014, 05:20:11 PM
#12
re: …V15 Inspector Bar…

Have you checked "Selector 3D Properties"? I remember they used to leave some of that stuff turned off by default.

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1903), 64-bit


* June 04, 2014, 06:16:00 PM
#13
re: …V15 Inspector Bar…

Have you checked "Selector 3D Properties"? I remember they used to leave some of that stuff turned off by default.

You're right, John.  I hadn't.

Logged


* June 04, 2014, 09:18:55 PM
#14
After you've created the "Spin' object and cut it down, try the Facet Editing tool to create the ramp.
Facet editing will not work.  The wedge surface has a radial twist.  Even the method posted by lemel_man using a helix as a path for a sweep will require a twist.

Logged


* June 04, 2014, 10:50:34 PM
#15
I didn't experience any problems with the Facet Editing tool in V15. As I posted above, I selected just the top facet to the wedge section. Moved its reference point and then the "Y" rotation node to the top corners. Once they were set, I zoomed in and moved the "Y" rotation by hand at first, then entered various values. I finally ended up with 355.6°.

Logged
John R.

V17—V21, 2015—2019
Designer, Deluxe, (Professional, Expert, Basic), Platinum
RedSDK enabled
Windows 10 Pro (1903), 64-bit


* June 05, 2014, 05:58:13 PM
#16
Measure the angle of the front upper edge to the bottom edge.  I get 1.6866°, so it's not technically accurate.

That said, the method will suffice for visual presentation... i.e. provided there are no annotations indicating the inaccuracy  ;)

Logged


* June 05, 2014, 06:02:06 PM
#17
Can anyone describe what the upper face of the wedge is supposed to be? To my mind, the attached looks reasonable, but that doesn't mean it's right.

Henry H
« Last Edit: June 05, 2014, 06:15:44 PM by Henry Hubich »

Logged


* June 05, 2014, 06:18:06 PM
#18
I don't know of any technical word or word combination.

I assume the desired upper surface could be described as one following the radials of a helix, similar to the sides of a square thread.

Logged


* June 05, 2014, 06:38:17 PM
#19
I don't know of any technical word or word combination.

I assume the desired upper surface could be described as one following the radials of a helix, similar to the sides of a square thread.

That's how the object in my attachment is shaped. The wedge was lofted, then added to a sector of a hollow cylinder. Can't fillet the reentrant corners in v15, but I can do it in v21.

Henry H

Logged


* June 05, 2014, 06:57:20 PM
#20
Solved without using any of the previous mentioned methods.

Use Revolve with spiral pitch.  Easiest to create an intersecting part, then subtract from the flat top base model.

Logged


* June 05, 2014, 07:02:28 PM
#21
That's how the object in my attachment is shaped. The wedge was lofted, then added to a sector of a hollow cylinder.  ...
I have to wonder if lofting creates a technically accurate part.  The upper surface at loft profiles should be accurate.  Its the surface in between the profiles that I'm not too sure of.

Logged


* June 05, 2014, 07:08:23 PM
#22
That's how the object in my attachment is shaped. The wedge was lofted, then added to a sector of a hollow cylinder.  ...
I have to wonder if lofting creates a technically accurate part.  The upper surface at loft profiles should be accurate.  Its the surface in between the profiles that I'm not too sure of.

Yabbut the actual part is a casting. How precise is that? :-)

Henry H

Logged


June 05, 2014, 07:11:30 PM
#23
I had a play with it [while Henry was Posting Reply #23], and it seems to me that the top of the curved-wedge has to either be accepted to have a "twist" incorporated, or one of the ends not 0°... as shown.

If I think about it, if I were actually machining this with something flat, it would turn out as I've drawn.
But if I were machining it with a grinder or something-- and I had to have the end not have an angle-- I'd have to incorporate a little twist on the that top wedge "plane" as I was grinding (right at the very higher end, just after I get to one end-point... "twisting" to the other end-point).

Where's our geometrician Murray when I need him :P

[Not so sure about my "twist" theory:  If I drew a helical ramp-- like at football-stadiums say, all walkways level-- and then just sliced it to obtain one ±45° portion, I think that portion-- and the ends of it-- would be level.  I'll let the big-boys figure it out. ???  It was still an enjoyable experiment.]

Yeah, I was wrong in my theory.  I see that easily now.
« Last Edit: June 05, 2014, 08:14:19 PM by Alvin Gregorio »

Logged
Alvin Gregorio
Intermittent TurboCAD user since yr. 2000 (ver6.5).  No formal CAD Training.
---TurboCAD: V21PP; V20.2PP; V19DL; V11.2Pro; Windows-7-Pro/64-bit; Intel-Core-i3 CPU; 2.27ghz; 4GB RAM; Intel HD Graphics (CPU based)


* June 05, 2014, 07:13:31 PM
#24
Yabbut the actual part is a casting. How precise is that? :-)
1) Depends on how precise the mold is   ;)

2) Surface could be machined after casting.

Logged


* June 05, 2014, 07:15:49 PM
#25
Can anyone describe what the upper face of the wedge is supposed to be?
Henry H

It's a helical ramp, a square thread, as Joe says.   The radial profile is normal to the axis.   A Guggenheim floor....    A degree 3 loft will be accurate if the profile intervals are at 30 degree or closer.  It can be ground with a cylindrical stone normal to the axis, that's the path a bearing follower would traverse, and strictly speaking, there's only a singular point of contact, like a worm and wheel.  Lotsa lube....
« Last Edit: June 05, 2014, 07:37:54 PM by murray dickinson »

Logged


* June 05, 2014, 07:16:34 PM
#26
...
But if I were machining it with a grinder or something-- and I had to have the end not have an angle-- I'd have to incorporate a little twist on the that top wedge "plane" as I was grinding (right at the very higher end, just after I get to one end-point... "twisting" to the other end-point).
Machining method presents another challenge all by itself  ;D

Logged


* June 05, 2014, 07:38:12 PM
#27


It's a helical ramp, a square thread, as Joe says.   The radial profile is normal to the axis.   A Guggenheim floor....    A degree 3 loft will be accurate if the profile intervals are at 30 degree or closer.

What's that, Murray?

Henry H

Logged


* June 05, 2014, 08:07:20 PM
#28
Degree is the polynomial degree of a curve or surface, 30 degrees angular separation will maintain the continuity of a degree 3 surface. Rhino .3dm can exchange higher degree surfaces, maybe STEP can too, but IGES not.  TC's curve properties can go from degrees 3 to 9, but that's the U direction of a surface, I think that the V direction, through the profiles, is fixed at 3.  Maybe ACIS has been improved, but it used to be that way.
« Last Edit: June 05, 2014, 08:12:54 PM by murray dickinson »

Logged


* June 07, 2014, 01:01:24 AM
#29
Thank you everyone.  It's been an interesting conversation.  I couldn't quite get the facet editing method to meet the lower edge accurately, but might with a bit more playing.  Also the Sweep and Revolve were not entirely successful.  I had good success with Henry's lofting approach, so I stuck with that for now.  This job is to produce a 3D print version of this cast part to give a temporary fix for a machine with this part broken.  If it works, I could print a PLA version for use in a lost wax method to cast a metal part.

Of course it would be lovely to upgrade to the latest version of TC, but just at present I'm not doing quite enough cad to warrant it.  $700 to upgrade in Australia. :(

All the best,  Leigh

Logged