×
INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!
  • Students Click Here

*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Students Click Here

Jobs

Is anyone really good with curves functions ?

Is anyone really good with curves functions ?

Is anyone really good with curves functions ?

(OP)
I'm trying to some angled curved cut outs on the cylindrical body.  I have tried just about everything to make the cut work but I cannot get it, so I was wondering if there is any one who's really good with curves that might wanna help me out. It's kinda hard to explain in writting, if someone would like I can send a particle model to help describe what I am trying to do.

RE: Is anyone really good with curves functions ?

Can you show a pic of your part? (see FAQ)

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

RE: Is anyone really good with curves functions ?

I'm still not clear what you want to do.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

RE: Is anyone really good with curves functions ?

(OP)
If you look at the one photo which shows the top view you'll notice that the cut is on a slight angle.. this was my process.... i drew my cylinder .. then i drew my path of the cut on the top plane.. i think used the project sketch from the curve tool bar to project it up to the curve surface... then i created a plane "normal to curve"  and created a profile to cut with .. then used the sweep cut tool.. this works but leaves the slight angle shown in the top view (1st link)  i cannot seem to get rid of this angle and i cannot find what's causing it.

RE: Is anyone really good with curves functions ?

It's projecting the cut towards the center. Make a plane first, create the sketch on the plane, then cut thru the part.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

RE: Is anyone really good with curves functions ?

(OP)
hmmm i don't follow you chris ?  I want the cut to be projected about the centre because it will be guiding a cam follower which has a common axis with the centre

RE: Is anyone really good with curves functions ?

Project the curve twice and loft cut between the two.

RE: Is anyone really good with curves functions ?

Oh, I see. A loft cut as kmarencswp suggested.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

RE: Is anyone really good with curves functions ?

(OP)
Alright I'm gonna give the loft a try I didn't think of that .. .thanks guys.

RE: Is anyone really good with curves functions ?

(OP)
kmarenscwp I'm having trouble using the loft cut feature to do that cut ? I might be lost in the part where you say to project the curve twice? I have the one line projected up on the surface but then i'm not sure what to use to make the cut?

RE: Is anyone really good with curves functions ?

Looks like what you are wanting is a barrel cam profile.  This can be easily made by a machinist using a cutter as close as possible in diameter to the follower, but recreating a true barrel cam profile is not easy in a solid modeler.  The reason is that it requires a sweep cut of a cylinder along a helical path, which creates a groove that is actually slightly wider at the bottom than at the top. You can approximate this with various techniqes, including a sweep of a rectangle along the path, but its accuracy to the true profile is dependent on the cam diameter, cam slope, follower diameter, and follower length.  Accuracy increases as cam OD increases, slope decreases, follower diameter decreases, and follower length decreases.  If at all possible, the best way to specify a cam like this is by making the model close enough to give an idea of the shape, then specifying follower travel vs. degrees of rotation.  You won't be able to dimension the slot on a 2D drawing anyway.

RE: Is anyone really good with curves functions ?

(OP)
YES this is exaclty what I am trying to do!  I didn't know the term for it but barrel cam makes sense because this groove is for cam followers to ride in.  It's bin driving me crazy why I cannot make the groove as it should be.  
But are you saying that it is actually not possible to make this cut accurately in solidworks?

RE: Is anyone really good with curves functions ?

jamaz, I don't think it is possible to make the barrel cam exactly as it is supposed to be in SW as you simply cannot describe it correctly. I'm not even sure what "correctly" is haha. My level of expertise is not cams at least not barrel cams.  
I did a machine with a full 180 degree barrel cam. actually 2 of them. It was a bitch. SW support essentially said I was on my own. I found that I could get 90 degrees around the barrel but the sweep would roll over itself if it would work at all. I finally projected points on the surface of the cam and created spline lines for a path that was adjustable. I then made the roller and in an assembly, rolled the roller along the created path and adjusted as required. It was another trick to get the roller to be tangent or to just stay in contact. I really used a narrow roller and then redid the path untill I got what I needed. It was a nightmare.  As it turned out we couldn't make this thing in our CNC as we don't have a 4th axis so I then cut the cam into sections to fit material thickness available. This was another nightmare as for some reason I couldn't break the link between the original model and the save as copy new models I made for each section. I now have to be very carefull when making changes to the cams, as they can cause unexpected results in the other models.  One of the sections was modified at assembly and I had to recreate this one. I used a digitized point cloud, again it was said it couldn't be done, and created a surface which I used to cut the blank so to speak. This was machined nicely by our cnc and it worked even better than the hand ground model. Fortunately the whole thing was in HDPE so machining was relatively easy.  The machine is well past the 5 million cycle mark so it is working well.
There are cam software programs that do just this type of thing but we couldn't justify the cost since this was a 1 off machine.  Camco and Stelron are two  big users of barrel cams.

99 Dodge CTD dually.

RE: Is anyone really good with curves functions ?

I wouldn't say it's impossible, but it would be really hard to get correct.  As I mentioned in my first post, the bottom of the groove is going to be wider than the top if you're using a cylindrical follower.  The sides of the groove will not be parallel.  This is due to the slope of the cam.  If you had a cam with no slope (just a circumferential groove) the sides would be parallel.  But once you give it a slope you run into issues.  The surface of your follower is cylindrical.  With no slope the contact line of the follower to the side of the groove is parallel with the follower axis and perpendicular as well as coincident to the barrel axis.  As the profile slope increases, the contact line moves away from being coincident with the barrel axis.  It's just like rolling a wheel up a hill.  The contact patch of the wheel is no longer under the wheel's center.  Therefore the portion of the cam grooves that's parallel to the follower axis needs to be skewed away from perfectly radial.

As I said previously, if you can get the same size cutter as follower to follow the correct path at the cam outside surface then the cutter will get the sides right for you.  Also, you may be able through trig to show that the difference from parallel on the cam sides is insignificant.  It looks like your barrel is rather large, your groove rather small, your motion profile fairly small in slope, and the line of contact fairly short.  All of those tend to minimize the "dovetail" effect on the groove.  

RE: Is anyone really good with curves functions ?

Hi all,

1. handleman is making a lot of sense here. Understanding helix geometry is the key to this problem.

2. I would discourage anyone from using the loft method. It is a difficult way to get even a pretty model and it is not useful if you want an exact representation of this barrel cam.

3. Swept solid capability would make this easy to do exactly.  Until then, I think that using surface modelling techniques is the easiest way to get an exact representation as follows:
  
a) make your base ring, cut two holes for your start and end positions, and then do a simple undersized helical sweep to join the holes
http://img51.imageshack.us/my.php?image=akengcam13ux.jpg

b) Create a "surface-sweep" by sweeping a line that represents your cam follower axis along your helix path. Then create two "surface-offset" features by offsetting by the follower radius plus clearance.
http://img72.imageshack.us/my.php?image=akengcam2a1st.jpg

c) Now use these two surfaces to do two separate "Cut-with Surface" operations. Done.
http://img72.imageshack.us/my.php?image=akengcam3a8kb.jpg

The key point is that the surface offset creates the cam walls exactly.  Theoretical mathematical types can prove this using "envelope theory".

This is actually a relatively simple problem.  It gets a lot more complicated if you have a track bottom where there is a complex fillet generated at the corners.

I hope this helps. Please share if anyone has a better way of doing this.

ERT
http://www.akeng.com

RE: Is anyone really good with curves functions ?

(OP)

Ok, I've tried some of your suggestions and have had some
luck but i'm not completely convinced that this is the path i am looking for.  In the link below I drew a stetch showing the mechanism, could anyone look at this and give there opinion on what the cam slot would look like ?
ERT in the model you created does that cam slot rotate away from being normal the surface of the barrel ?



http://img308.imageshack.us/img308/1928/camlayout4ji.jpg  

RE: Is anyone really good with curves functions ?

Maybe create the part, then a separate part of the path shape, then 'subtract' the path-shaped part. Just an idea.

Chris
Systems Analyst, I.S.
SolidWorks/PDMWorks 05
AutoCAD 06
ctopher's home site (updated 06-21-05)
FAQ559-1100
FAQ559-716

RE: Is anyone really good with curves functions ?

(OP)
Is this done thru the combining bodies feature ?

RE: Is anyone really good with curves functions ?

If you have the appropriate section data for the slot this is very very easy to do using the wrap-deboss feature.

Simply sketch the geometry of the outside of the cut (on the surface)--you need the arclength. Dimension the vertical (axial) distance relative the cylinder appropriately. Use wrap to wrap the sketch onto the outside face (this is why you used the outside[final] dimensions) and select deboss. Pick an appropriate depth and voila. Keep in mind that it literally wraps the sketch not projects so the dimensions of the sketch are unchanged when put on the face of the cylinder.

Here's a picture of the result with the measurement to explain what I mean: http://img51.imageshack.us/my.php?image=deboss0ji.jpg

RE: Is anyone really good with curves functions ?

Kudos to ERT for putting much more eloquently what I was trying to say.  Yes, Jamaz, the sides of the cam slot are not normal to the barrel surface.  The wrap sketch method will create sides that are normal to the barrel surface, which is close but not quite the correct profile as generated by ERT.  You can verify this by creating sketch wrapped cam (with some slope, remember, not just a circumferential groove) and putting the follower in.  You'll have tangency at some point, but also interference.

RE: Is anyone really good with curves functions ?

Oh, they aren't normal? That's why I suggested it in the first place, oops.

RE: Is anyone really good with curves functions ?

Jamaz17,

My example only makes sense if the follower axis and cam axis always intersect.  I assumed that the follower translates parallel to the cam axis.  A follower that swings on a lever would require a different approach.

I find your mechanism sketch confusing.  Can you explain a bit further?  Does the cam translate and rotate?  What is the objective?  I think you may have very poor pressure angles, but it is difficult to be sure until I understand the mechanism.

ERT    

RE: Is anyone really good with curves functions ?

(OP)
ERT, the idea of the mechanism is that as you push the barrel cam shaped piece inward, the fixed legth rod angle can be changed.  So when the follower is at one end of the cam slot the rod is at 0 degrees, while at the other end it's at about 25 degrees.  I agree that this will not yeild the best pressure angles, as the slope is very shallow.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!


Resources