|
CCSRacer
Jul 28, 2010, 1:31 AM
Post #1 of 37
(15117 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Hey, just wanted to bump this thread to see if there are any updates. I'm a ME student with a lot of past fabrication and machine shop experience, and this sounds like a lot of fun.
The OP Link Cams are just gorgeous 
(This post was edited by CCSRacer on Jul 29, 2010, 1:17 AM)
|
|
|
|
|
CCSRacer
Jul 28, 2010, 2:19 AM
Post #2 of 37
(15098 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
I'm attempting to make a logarithmic curve directly in Solidworks using the 'Equation driven curve' command. I'm definitely having luck getting the rough idea with parametric equations:
x = e ^ (.244t) * cos (t)
y = e ^ (.244t) * sin (t)
It seems like Solidworks is using radians in this dialog box and this value would approximate a 14° camming angle, right? Then I can just add a coefficient to scale it to size.
Any tips?
|
|
|
|
|
adatesman
Jul 28, 2010, 3:04 AM
Post #3 of 37
(15087 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
CCSRacer
Jul 28, 2010, 3:13 AM
Post #4 of 37
(15083 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
That's a good point, kind of hit or miss to tweak the equations in SW. The nice part is that you can set up tangent center-lines, apply a dimension, then go back and tweak the curve equation.
What are the guidelines to locate the axle relative to the curve?
|
|
|
|
|
adatesman
Jul 28, 2010, 3:20 AM
Post #5 of 37
(15082 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
CCSRacer
Jul 28, 2010, 3:34 AM
Post #6 of 37
(15075 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Cool.
I had to think about that for a second with how I have my curve generated... I think I've got it right though. My curve starts when t=0, so my radius at that point is just 1 * my scaling factor. I added a tangent center line at the beginning of my curve, another at 90* from that point towards the axle's axis.
How does this look?
(This post was edited by CCSRacer on Jul 30, 2010, 6:50 PM)
|
|
Attachments:
|
Cam.JPG
(24.5 KB)
|
|
|
|
|
CCSRacer
Jul 28, 2010, 3:48 AM
Post #7 of 37
(15070 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Hmmm, in the assembly the pivot looks high relative to the contact surfaces on the lobes...
(This post was edited by CCSRacer on Jul 30, 2010, 6:51 PM)
|
|
Attachments:
|
Cam2.JPG
(26.8 KB)
|
|
|
|
|
kennoyce
Jul 28, 2010, 3:57 AM
Post #8 of 37
(15064 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
CCSRacer wrote: Hmmm, in the assembly the pivot looks high relative to the contact surfaces on the lobes...
I'd have to agree. I didn't measure or anything but eyeballing it, I think it does look like that is quite a bit greater than 14 degrees. When I'm designing cam lobes, I just use an excel spreadsheet and import the points into solidworks (probably like Aric does just in mathmatica or whatever he uses). Good work though, and enjoy building your first cam!
|
|
|
|
|
CCSRacer
Jul 28, 2010, 4:20 AM
Post #9 of 37
(15053 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Sorry to be a pain... Can you clue me in on importing a curve? Never done it. I've got a file with the curve data inside. Does it need a specific file extension? What is the command in SW?
Thanks for the help!
|
|
|
|
|
kennoyce
Jul 28, 2010, 4:31 AM
Post #10 of 37
(15051 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
Yeah, no problem, I don't have solidworks right here on my computer so i am just trying to go by memory, but it seems like you just save the file as a either a coma or space delimited text file (three columns one for x one for y and one for z). You then go to Insert, curve, from file, and you browse for the file you saved (make sure you use the drop down file selection to include text files and not just solidworks curve files). I know that it does matter weather it is comma or space delimited, and I think that for solidworks 2008 or earlier you need coma, and 2009 or 2010 needs space delimited. I could have this backwards though. Hope this helps and have fun!
|
|
|
|
|
CCSRacer
Jul 28, 2010, 4:51 AM
Post #11 of 37
(15044 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Awesome, you guys rock. Okay, so I'm running into the same hang up as with my other curve. How to plot the axle's center... By nature the curve doesn't have a center. Are you using the curve's start point?
(This post was edited by CCSRacer on Jul 28, 2010, 4:52 AM)
|
|
|
|
|
kennoyce
Jul 28, 2010, 4:57 AM
Post #12 of 37
(15038 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
When I make a cam lobe, I always have the axle at 0,0 and as long as you don't offset the curve somehow, it should be naturally centered around the origin. I'm out for the night, but I hope this is helpful, I can't think of how to describe it any better than that.
|
|
|
|
|
CCSRacer
Jul 28, 2010, 5:13 AM
Post #13 of 37
(15028 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Hahaha, that's when you know it's too late to be winging it!
I didn't notice that it had inserted that curve already oriented to the origin. Right before you posted that I had made an elaborate method with tangent centerlines and driven angle dimensions... as I'm dragging the axle circle around it snaps to the origin and the angles equalize, hahaha... Time for bed!
(This post was edited by CCSRacer on Jul 30, 2010, 6:51 PM)
|
|
Attachments:
|
Cam3.JPG
(26.0 KB)
|
|
|
|
|
adatesman
Jul 28, 2010, 1:52 PM
Post #14 of 37
(14992 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
CCSRacer
Jul 28, 2010, 2:03 PM
Post #15 of 37
(14984 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Cool link. Yeah, so it's not a 'center' in the traditional sense, more like an origin.
|
|
|
|
|
adatesman
Jul 28, 2010, 2:11 PM
Post #16 of 37
(14977 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
CCSRacer
Jul 28, 2010, 8:39 PM
Post #17 of 37
(14929 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Here is a render of a rough assembly. I got some cam lobes modeled with a 14° angle and mocked up some rough components to make the assembly. I did a quick simulation on the assembly too by fixing three of the teeth 'flats' and applying 40kN downwards on the stem. Using 6061-T6 for the lobes (6mm thick) and a steel alloy on the axle (8mm D), this assembly is very strong. 40kN barely stressed it. This was really just meant to get the camming action right and give a base to start refining the other components, surprisingly the mass comes out as 155g
(This post was edited by CCSRacer on Jul 30, 2010, 6:53 PM)
|
|
Attachments:
|
CamR1 (Large).JPG
(84.7 KB)
|
|
|
|
|
kennoyce
Jul 28, 2010, 8:51 PM
Post #18 of 37
(14925 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
I think that fixing three teeth like you are doing in the simulation is screwing up the results. By fixing the teeth you are not allowing for the cam lobes to rotate as the axel bends which is what will really be happening. I may just not understand what you are doing correctly, but the materials you listed shouldn't be able to hold 40 kN.
|
|
|
|
|
CCSRacer
Jul 28, 2010, 8:54 PM
Post #19 of 37
(14923 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
That's a good point. I am getting lobe rotation, but in order to rotate it's having to elongate and compress the teeth. I haven't tried an advanced restraint like trying to simulate the cams biting into a surface yet. A quick work around might be to put a frictionless pin through the lobe near the contact surface and fix the pin.
|
|
|
|
|
kennoyce
Jul 28, 2010, 9:00 PM
Post #20 of 37
(14920 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
You'll still have the same problem since the frictionless pin will remain fixed, but in a true scenario, the point of contact changes as the axle of the cam bends. I think you need to simulate a crack and put a friction boundery between the cam and the crack (I don't know how easy this will be in solidworks, you may need to export your assembly to ansys or comsol).
|
|
|
|
|
kennoyce
Jul 28, 2010, 9:08 PM
Post #21 of 37
(14914 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
One other thing I noticed, I can't tell if you left room in your assembly for the springs. If you didn't, make sure you do because moving the lobes farther apart to accomidate the springs will increase the bending stresses in the axle.
I hope I don't sound like I'm being a jerk, I'm just trying to point out some things that you may have missed as someone who has gone through this process in the past.
|
|
|
|
|
kennoyce
Jul 28, 2010, 9:11 PM
Post #22 of 37
(14916 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
Oh, one more thing, adding a PTFTW!!!!! to the assembly couldn't hurt either.
edited to hang my head in shame since Aric made this into a new thread thereby eliminating my PTFTW.
(This post was edited by kennoyce on Jul 29, 2010, 12:26 AM)
|
|
|
|
|
CCSRacer
Jul 28, 2010, 9:13 PM
Post #23 of 37
(14914 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Absolutely not! I really appreciate any and all informed feedback. I'm a noob! 
I hear ya on the shaft width / spacing. In reality, I'll have much less control over the axle, since it will simply be sourced. As long as it isn't poorly implemented, that shouldn't be an issue.
That first render has everything butting up to each other, but in the one I'm running a sim on now, there is a good bit of space in there.
|
|
|
|
|
CCSRacer
Jul 28, 2010, 9:22 PM
Post #24 of 37
(14911 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Hahaha, damn it, I even searched and haven't gotten my head around PTFTW! I don't troll around here much, so I'm out of the loop. 
|
|
|
|
|
adatesman
Jul 28, 2010, 11:16 PM
Post #26 of 37
(7782 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
adatesman
Jul 28, 2010, 11:21 PM
Post #27 of 37
(7776 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
kennoyce
Jul 29, 2010, 12:23 AM
Post #28 of 37
(7760 views)
Shortcut
Registered: Mar 6, 2001
Posts: 1338
|
adatesman wrote: kennoyce wrote: Oh, one more thing, adding a PTFTW!!!!! to the assembly couldn't hurt either. Not no more, sucka! 
Oh you are such a jerk Aric, It's not like this is the lab or anything..... oh wait, yes it is, forget about that first part.
Edited to add that I thought Aric just deleted my post, but I later realized he just broke this thread off of the original Dirtbag Cam Building thread, thereby eliminating my PTFTW.
(This post was edited by kennoyce on Jul 29, 2010, 12:30 AM)
|
|
|
|
|
CCSRacer
Jul 29, 2010, 1:15 AM
Post #29 of 37
(7750 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Yeah, the sim stuff is really cool because it lets you see where the stress, strain and deflection are distributed. You can go back to the model and make tweaks to trouble shoot spots, then retest in see how it looks.
Thanks for splitting the thread off for me. I'll post an animation of the loading sim for you guys to check out.
(This post was edited by CCSRacer on Jul 29, 2010, 1:19 AM)
|
|
|
|
|
adatesman
Jul 29, 2010, 1:43 AM
Post #30 of 37
(7739 views)
Shortcut
Registered: Jul 13, 2005
Posts: 3479
|
|
|
|
|
|
CCSRacer
Jul 30, 2010, 6:57 PM
Post #31 of 37
(7665 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
I went ahead and embedded the earlier pics.
I'm still screwing around in Solidworks simulation to get a more accurate analysis (and to better learn the software). The problem is that these more complex sims with multiple surfaces contacting, friction etc, take forever to run, and more ofter than not fail at some point.
Here is one where I just fixed one small surface near the contact area of each cam. I scaled the stress coloring to better show the load. Working on one right now with actual flat walls tangent to the cams using a given CoF... we'll see if it works....
http://s131.photobucket.com/...¤t=PT1.mp4
(This post was edited by CCSRacer on Jul 30, 2010, 7:12 PM)
|
|
|
|
|
jeremy11
Jul 31, 2010, 2:01 AM
Post #32 of 37
(7635 views)
Shortcut
Registered: May 28, 2004
Posts: 597
|
Excellent thread! I'm going back to school for Mech Eng. this fall. Building my monster cam project last year I designed it all on paper (along with Paintbrush) and was wishing for some Solidworks Skillz to answer some stress questions and to minimize weight in the Al axle without guessing (leading to overkill on the weight)
Keep it coming!
|
|
|
|
|
CCSRacer
Jul 31, 2010, 4:30 PM
Post #33 of 37
(7621 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
Aluminum axle huh? Did that become the point of failure in testing?
Feel free to post up on any stuff you are doing too. I don't have any design projects right now, but hope to use this build down the road in something like that. I am going to work part time at the UT machine shop this fall which will allow me access to the sweet CNC mills and lathe.
I am still struggling big time with getting simulations to solve with dummy walls and complicated (and changing) contacts between the lobe and wall.
|
|
|
|
|
jeremy11
Jul 31, 2010, 10:45 PM
Post #34 of 37
(7594 views)
Shortcut
Registered: May 28, 2004
Posts: 597
|
CCSRacer wrote: Aluminum axle huh? Did that become the point of failure in testing?
Nope, the 3/4" 7068 T6 axle was unmarred with a 14-16kN force that was enough to bend the lobes and pull the cam out.
Details here:
http://www.rockclimbing.com/...8;page=unread#unread
|
|
|
|
|
marde
Aug 1, 2010, 9:51 AM
Post #35 of 37
(7562 views)
Shortcut
Registered: Sep 3, 2006
Posts: 169
|
CCSRacer wrote: Aluminum axle huh? Did that become the point of failure in testing?
The friction between the axle and the lobes would be less if you used different materials.
Same materials moving over each other usually create more friction and weird corrosion problems.
|
|
|
|
|
CCSRacer
Aug 1, 2010, 5:10 PM
Post #36 of 37
(7543 views)
Shortcut
Registered: Oct 2, 2009
Posts: 48
|
I would really love any insight I can get into running simplified, but somewhat accurate simulations of the cam operation in Solidworks.
I have rolled back the tooth cut operation on the cam lobes to simplify surface contact and butted the lobes up against two sections of flat wall (fixed in space). I have an assigned CoF between the lobes and walls, a small (~40 N) force preloading the lobes into the wall and a larger (~4000-10000 N) force pulling down on the shank bottom surface. I still crash the sim most times. Have tried with and without the large displacement flag.
The error I'm getting lately is that the incremental elastic strain is too big >25%. I'm thinking that means the cam lobe is getting ripped near the point of contact with the wall, but am not sure.
|
|
|
|
|
jeremy11
Aug 1, 2010, 5:45 PM
Post #37 of 37
(7538 views)
Shortcut
Registered: May 28, 2004
Posts: 597
|
marde wrote: CCSRacer wrote: Aluminum axle huh? Did that become the point of failure in testing? I don't think failure is the major concern here. The friction between the axle and the lobes would be less if you used different materials. Same materials moving over each other usually create more friction and weird corrosion problems.
That was discussed in my Monster Cam thread. Any more comments on my cam, please add them there. Al on Al is an issue, but the bigger issue was my 3/4" drill bit was smaller than my 3/4"axle so I had to file and sand the holes bigger, so they aren't perfectly round.
Now, back to the original topic.....
|
|
|
|
|
|
