[skippy]

Hi all,
I just spent the last day trying to draw a motorcycle gear shift drum which is basically a cylinder with 3 tracks (paths) for the selector fork pins to follow. I've been CAD drawing for some 15 years and have done some rather weird jobs but this one has me stumped and I'd like to hear the opinion of others who may have done something similar. The problem is that I don't know how to draw the tracks/paths (8mm wide x 6mm deep) to end up with parallel walls. I started off this job in Fusion 360 trying to use "Sweep" command for the track and then switched to SolidWorks (at my work) when I realised that Fusion can't do a wrap on a full cylinder. I then realised that wrap in SW produces NON parallel walls everywhere EXCEPT for the concentric paths. Right now it seems I'm left with 2 options which are:
(a) take the job as is and turn it from solid to surfaces and using offset surfaces and extend/trim/delete, etc. obtain what I want then convert back to a solid
OR
(b) waste no further time and leave the job as is and use the CAM (Fusion 360) to sort out the tool paths so that I get parallel walls.
I'm inclined to go for option (b) but just wanted others to weigh in on this. Maybe this kind of thing is only achievable (easily) with high end packages such as Catia or Siemens NX but I don't have access to those.

[deadlykitten]

hy, i guess the problem is that you have to keep the cutting section orientation constant, while the software is tilting it, in respect to path orientation

i could start my SW and give it a try, but anyhow, i would not aproach it like that, because i know that when a 4th axis is used, you have somehow to corelate the feed and everything, because the motion is interpolated, and not all cam's are doing it right, thus they may output the toolpath, but may fail in computing the correct feed

i would actually draw the cilinder in 2d, and output parametric code for the finishing, because i believe that you are looking for some tight tolerances, and like this i would obtain the level of control where i would be comfortable

otherwise, you may be lucky, if wall deviation is < required tolerance, and in such a case, you could machine it just like that ( for example, if you look for 0.2 tolerance and walls deviation is 0.1, then is all ok )

i am not saying that fixing SW is not possible ... i am just saying that i can do it faster in 2d + wrapping the final code ( using a macro eq, etc ), then the time required to dig for a sw fix / kindly

ps : i also hit into something similar a while ago, when trying to design reamers with tilted flutes ; truth is that a cut-sweep may output a toolpath <> then what a cutting tool would generate; in your case, the problem is keeping the section orientation constant, and my problem was that the section was much smaller in comparison to tool section ... maybe you should ask inside the sw forum

[deadlykitten]

me again : search "barrel cam mechanism" on youtube ... what you need is to obtain something some-how similar to that, but truth is that you can simulate the motion even if walls are not ||

you are looking to obtain kind of cam behaviour ( by cam i don't mean cam-software, but cam-mechanism )

when it comes to cam-mechanism, i simply compute the motion and code it directly, thus i don't go through a solid / kindly

[deadlykitten]

hy pls check attached images :

01) i sketched the path ( 2 circles dia8, conected with a tangent groove )
02) wrap-deboss ( 6mm depth ) over the od of the cilinder
03) to check the walls, i created a 3d sketch, and i have intersected the wrap-debos with a plane that contains the cilinder axis, and i have obtained a rectangle with height=6 and length>8; length would be 8 only if the groove was perpendicular on the cilinder axis

if you wish, send me the real required dimensions, and may sent a video back / kindly

[skippy]

Thanks deadlykitten. Damn, I've been around motorcycle engines for most of my life but never knew that the industrial version of a shift drum was called a "barrel cam" mechanism. No wonder I couldn't find anything in YouTube. I just looked under "Barrel cam" and there are plenty of videos. Thanks for your help and your offer re a video but could you give me a couple of days to look at and digest some of the ones on YouTube first?

Regards

[deadlykitten]

hi skippy

"barrel cam" is from the field of mechanisms with cams and articulations, which were heavy used to design machinery motion

if you know them, you will no longer look at a part as an individual, but you are actually capable of undertstanding it's place inside a motion chain, and you may figure it out what tolerance is needed in order to craft it thus, even on a wierd mechanic organ, is possible to imagine it's motion ...

today they are less used, because most motions are interpolated by cnc's

that is not a "true" barel cam mechanism ... if you wish to know, it is a "restricted" mechanism, because is not 360* ... that groove is a mechanical cam contour ... whatever

could you give me a couple of days to look at and digest some of the ones on YouTube first?

do whatever you wish

as for that part, one way to get the solid, is to open the 2d dxf ( considering that there is one ), check the contour, and paste it in sw, then wrap it on the cilinder

do you race ? kindly