modeling a wave spring

[2ms1]

I think I've got a nice challenge here for some of you. Any tips on how to go about modeling a wave spring? The wave looking aspect really is a sinusoidal function. I know how to make a straight wavy strip by creating datum curve from equation, going in sketcher and "using edge" to create sketch by essentially tracing the curve so can protrude in z direction, and then using thicken to make it a solid with some thickness.


However, getting the "strip" to have helical shape, I have no idea.

 

[puppet]

1) make helix's
-----via helical sweet surface
-----or via equation

2) vss on ur curve edge or hss
---- draw a flat line horizontal with a offset to ur default references.

3) make that offset a equation like
sdXX = 10*cos(360*trajpar * 40)

10 = AMplitude
40 = how many full periods.

hope this helps.

 

[2ms1]

Oh gosh, I'm a little intimidated. I don't seem to be able to start becauseI don't know what options to chooseand, well basically,how to make a helical sweep surface.I don't even know what it is really. I've made helical sweep protrusions, so I know what they are visually and have some experience making them, but I don't even know what a helical sweep surface looks like. Could you go into a little more detail for a beginner in this area? Like what options to choose and stuff like that.
Edited by: 2ms1

 

[puppet]

hmm not in wf3. but right next to solid sweep protrution should be a surface option for it. Explore the menu a little u will find it.

 

[2ms1]

Ok I understand what I surface helical sweep is now. I'm not sure what section to use for the sweep and why you are recommending a surface rather than a protrusion, but I know how to make one.


However, I still don't know what you mean in (2) and (3). I don't know what you mean by making a vss with the offset equation. I understand that the sinusoidal equation is what would be giving the wave shape, but I don't know how to implement it. Could you elaborate a little more on the exact steps that need to be taken in (2) and (3)?
Edited by: 2ms1

 

[puppet]

hmm i think i over complicated it


ok starting again.

insert --> helical sweep -> protrusion
constant pitch || right hand || blah blah "DONE"

now select ur sketch plane.,
Draw a centerline which will act as ur center.
then draw a vertical line offset from ur centerline.
The length of this vertical line is the overall height of your sweep
tick tick

now ur in sketecher mode again. You will have 2 existing centerlines for you.
1)a horizontal one.
2)a vertical one which is produced from ur above vertical offset.


What ever shape u draw here. will be "swept" at what ever pitch u specified.

so ABOVE ur centerlines simply draw a box.
constrain symmetrically to your vertical Centerline and give it a offset value from your horizontal centerline.

then enter a equation like
sdxx = A * cos(trajpar * 360 *b)

(make sure u define A and B b4 starting)

This way u can change the values outside of your helical sweep(much faster)


values i used were

box = 20 horizontally, 2 vertically
centerline offset = 100 ( 200 if your using diameter)
vertical line height of 100
a = 3
b = 65
pitch 10


hope this helps.

 

[shankar_me]

Hi,


Here is one more solution. I defined two curves with equations


Regards,


Shankar

 

[swapnaja]

Use following equation
Cylindrical co.ordinate system



r=5
theta=t*3600
z=(sin(3.5*theta-90))+24*t
Edited by: swapnaja

 

[2ms1]

Thanks for all the help guys. I'm still not exactly sure how to implement some of these equations, or at least where to...


However, Puppet, I did follow your recipe and got the wave spring just like you made it and it's a beaut.


Now someone wants me to make basically the same thing except not helical. That is, I'd like to make something that's similar but stays in one circle, kind of like a wash washer.


I thought just doing things the same way with the trajpar in relation for offset of cross secton in a simple revolve would work the same way. But it doesn't.


Can anyone suggest how to do the same type of thing as the wave springs, except with all the waves being in one full circle (like wave washer) rather than a helix?

 

[puppet]

YAY you got it.

ok the next part is easy

draw a circle. 500 diameter
vss on circle

sketch this in your vss



is what i used this time

sdXX = 10 * cos(360*10*trajpar)

let me know how u go

 

[2ms1]

There's a problem. The model yielded by this technique changes cross section size based on position of curve because it will not sweep normal to the sinusoidal trajectory. Even if you specify for it to be normal to trajectory, in reality it isn't, but rather starts off that waybeing normal tothe sketch plane, but as trajpar changes the direction of the trajectory, the cross section stays at the same original angle (resulting in section size that is projection onto the original sketch plane).


Any suggestions on how to make the section truly constant? It's important I have the true geometry because I need to run Mechanica analysis.

 

[swapnaja]

try using datum curve> equation>select default coordinate system>Cylindrical co.ordinate system


then use the following equation
r=5
theta=t*360*10
z=(sin(3.5*theta-90))+24*t



vary the value of numbers to get the required dimensions
radius=5



then do a variable section sweep using that curve



.............swapnaja
Edited by: swapnaja

 

[swapnaja]

just for fun try the following equation to get a trangular spring





a=exp(sin(t*360*5))


b=exp(cos(t*360*5))


x=a+b
y=a-b
z=4*t


..........swapnaja