Helical Sweep Cut 1

[hugou]

I am trying to model a part in Pro E and I’m really struggling. Imagine something like the gear selector drum in a motorcycle gearbox. So in other words if I were to plunge an slot drill into a cylinder perpendicular to the axis and then rotate the cylinder an eighth of a turn while moving the cylinder an inch along its axis, then continue to rotate it by a further eighth of a turn while moving the cylinder back by the same distance in the direction it came from, I will have created a V-shaped slot around the circumference with a full radius to each end of the slot and a radius in the bottom of the V.
I have tried doing this using a Helical Sweep Cut both Through Axis and Normal to Trajectory, neither has given satisfactory results.
The trouble with doing it through axis is that the slot becomes too narrow so I have try and work out a width which will work out correct when projected at the desired pitch plus the ends of the slot don’t end up square to the walls of the slot.
The trouble with doing it normal to trajectory is that I end up with a ridge in the base of the slot at the V and the walls of the slot do not appear to be square to the base of the slot for some reason.
The trouble with both is that when I try to put a round in the V of the slot it does not come out normal to the spine of the original intersection (despite me selecting normal to spine).
I hope all of that makes some sense to somebody as it has me beat. I will try attaching the model.

Thanks.

 

[hydromech]

You can try doing it this way, it is done by offsetting the surface. But this creates a problem at the ends of the feature.

You might be able to work around that though. I am not entirely sure what you want to do.

Regards

Adrian

 

[hydromech]

It looks like this if the offset is normal to the sketch and not normal to the surface.

It is done by using the offset command and then offsetting a sketched profile.

Let me know what you think.

Regards

Adrian

 

[hugou]

Unfortunately Adrian I can't open those files, are you on WF4? I'm on WF3.
Attached shows the issue with rounds, they don't remain perpendicular to the spine. I just can't get it to make the shape that the milling cutter would.
This pic shows the sort of slot I am trying to achieve

http://www.dansmc.com/shaft_shifter2.jpg

I'll try doing an offset surface, but I'm not really sure how.

 

[hugou]

Was it the offset with draft feature you meant, I have just tried that with some degree of success, but that involved sketching on a flat plen which is going to make things very complicated to get the correct geometry.

 

[hydromech]

I'll explain how I did it, I think it is what you are after.

Firstly, sketch the profile on a plane that is outside of the cylinder OD.

Then select the surfaces on which the feature will be cut. With the surfaces still selected clich edit...then offset.

Select the sketch and then click to flip the feature to make it into a cut.

Click options to make the cut normal to the sketch instead of the surface.

I have done it before in WF2 so it must work in WF3.

I will try again in the morning in WF3.

Adrian

 

[hugou]

I'm pretty sure I've done what you suggested but it seems you do have to use the 'offset with draft' option to be able to select the sketch (just left draft as 0).
I also made sure I selected 'side surface normal to SURFACE' (as that is what I am trying to achieve). But it's still no good as it's projecting the sketch onto the surface rather than wrapping it (hence the distortion at the ends of the feature which I think you mentioned).
So this got me playing with the wrap command, and it does exactly what it says on tin. I might be able to get the wrap command to work for me if I can find a way of turning it into a cut. So far I have tried offsetting the suface of the cylinder and then wrapping the sketch onto that new diameter surface then doing a boundry blend between the two 'wraps'. Unfortunately this again creates a surface which is normal to the sketch not the surface of the cylinder.
If you do a google image search for gear shift drum you will find some pics of the type of thing I am trying to model.
There has to be a way of doing this, I just can't believe that something so simple can be so dificult to model.
Thanks.

 

[jeff4136]

http://www.mcadcentral.com/proe/forum/forum_posts.asp?TID=36813

may help you. There are other examples laying around there, as
well as other forums if you'll search for
barrel | cylinder | drum + cam

Groove sides are no problem. Sweep a surface representing the
cutter axis trace and thicken cut. I wouldn't worry about groove
bottom as I don't believe it can be correctly modeled in Pro/E.


-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.

 

[hydromech]

I see what you mean, there are more options in WF4.

See attached...actually dont see attached, the image won't upload. You'll have to trust me WF4 has more options.

Good Luck!

Adrian

 

[hugou]

Thanks Jeff, v.helful. Took me a while to figure out how you did that veriable section sweep as I have never come across Trajpar before! Done a bit of rading up on it and also found datum graph feature which may help me define the acutal path.
Reading your posts in that other thread I see what you mean about the base of the groove, more complex than you would initally expect.
I'll have to have a play with it as a lot of this is new to me but I'll keep you posted.
Cheers.

 

[jeff4136]

> but I'll keep you posted

That would be cool. I'd also communicate with the machine shop re their requirements. It's possible they really want a planar displacement curve or a curve wrapped on cylindrical surface, in which case the modeled surfaces are just for show and you can use whatever methods seem appropriate for that purpose.


-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.

 

[hugou]

I think I have finally got one of the grooves done, just two more to go at different radial offsets around the cylinder.
Jeff does this look 'right' to you? In so much as an 8mm peg will follow it around moving 10mm one way then 10mm the other. I'm aiming to go from the 'neutral' position to 10mm one way in a 1/16 of a turn, then to 10mm the other way in in 1/8 and back to neutral in another 1/16.
Cheers.

 

[jeff4136]

There are a couple of things you can do as a courtesy when posting models.
1) Make known the generating software version.
2) Suppress all features and save as *.zip.

I can't read your part with WF2, assume it is from 3 or 4.
(Oops, I see you've stated WF3 back a ways.)

If you want to post a neutral (.neu, .stp, .igs) I'm happy to take a gander.
Include any pertinent 'construction' geometry (cutter axis trace surface?), displacement curves, 2D layouts, etc. you might have to help clarify intent.


-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.

 

[hugou]

Sorry Jeff, didn't think to .zip it.
In attached .zip file you should find an iges file and some screen shots from pro e which should help you understand what I did.
Basically I created graphs (each of the vertical construction lines in the graph represent 1/8th of a turn along the circumference of the cylinder), one graph for each of the veriable section sweeps driven by relations refrencing the graphs.
I then created a boundry blend between the tops of each of the variable section sweeps and thickened that to cut the groove. Hide the veriable section sweeps and other refrence geometry and hey presto. The final thing I did was to extrude a smaller cylinder inside just to 'clean up' the suface forming the base of the groove.
I don't have mechanisms so can't check it with that, but it looks like what I was aiming for. Whether or not I have gone about it the right way I don't know.
Cheers again.

 

[jeff4136]

The attached is how I'd go about that particular (with relatively tight min rads) groove. Give it a look over and if there's anything you'd like to thrash out some more, holler back.

(Attachment: groove.zip WF2 ~200 KB)

-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.

 

[jeff4136]

Hugou,

My apologies. I was looking at that model again
(there are nuances to that geometry that still
baffle me, but anyway) and noticed I'd screwed up
some of the relations. If you're trying to get a
handle on it those mistakes won't help you any
so here's a reworked model.

Again, if there's anything that seems inconsistant
with what you're trying to do, holler back.

(Attachment: new_part2.prt.zip WF2 ~ 80 KB)

-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.

 

[jeff4136]

... try again :-|

-Jeff Howard (wf2)
Sure it's true. I saw it on the internet.