How to model a flashlight reflector?

[pigsmith]

Greetings,


I am trying to model a flashlight reflector and I have so far got only as far as producing half a parabolic curve. I'm getting so frustrated trying to figure it out that I'm afraid the keyboard marks on my forehead are going to become permanent.


Can someone please step me through the process to model a flashlight reflector?


Thanks very much for the help!

...

 

[jeff4136]

Revolve a conic arc?

 

[wsylvester]

Does a conic arc really give you the intent that you need to describe a parabola


A parabola is created from a cone granted but there are at least 3 factors involved


cone radius, cone height and offset distance and you need to find the focus pointbased on all this. I used to date a math major

she set me straight on all this and some other things too

So you need some start values and then assign them to the relations below it's not pretty as far as naming goes but it does work if the values are correct


HTH .. might just be more confusing .. it has been awhile since I have done any thing with this


------------------------------------------------------------ -------------


you have to create 3 parameters name .. cone_ht,cone_rad, and offset


then you can show those under model tree and edit the values.. be sure and
enter a start value , so you don't get a divide by zero error


then this goes in the relations table
------------------------------------------------------------ ---------------------


ratio=cone_ht/(2*cone_rad)
cone_ang=atan(cone_ht/cone_rad)
off_base=offset/sin(cone_ang)
max_offset=(2*cone_rad)/sin(cone_ang)
trig_leg=cone_rad-off_base
para_base=sqrt((cone_rad^2)-(trig_leg^2))
para_ht=((para_base^2)/offset)*ratio
focus_dist=(para_base^2)/(4*para_ht)


------------------------------------------------------------ -------
this is the datum curve equation
------------------------------------------------
x=para_base*t
y=((x^2)/offset)*ratio

 

[jeff4136]

Does a conic arc really give you the intent that you need to describe a parabola

I'm not sure I "know", but I've seen somewhere (Rhino documentation?) that a conic curve with rho:

= > 0 < 0.5 is elliptical,
= 0.5 is parabolic,
= > 0.5 < 1 is hyperbolic.

(???)

I've never seen a description similar to yours (I'm a mathematical dunce). Thanks. I'll definitily add it to my crib sheet.


Edited by: jeff4136

 

[Huug]

A conic with a value of sqrt(2)-1 is a true ellips. You can ad this as a parameter to get the value (0.41421356) correct.


Huug

 

[jrobi]

does the profile of the reflector need to be a simpleconcave or is a more elaborate profile required?

 

[pigsmith]

does the profile of the reflector need to be a
simpleconcave or is a more elaborate profile required?

I was stuck on the problem of what to do after I created the parabolic
curve -- how to make it into a solid model? -- and here's how I did it,
thanks to the kick-start from wsylvester and modified for WF2:



For the curve:

Datum curve

From equation

Done

Select CSYS

Select cartesian, then equation window should pop up.

Fiddle with the equation, save, exit.



Then for the solid:

Revolved protrusion

set up the orientation of the sketch planes

add c'line

sketch using offset edge

fiddle with that until you're happy,

OK out of there.



I'm pleased to say the keyboard marks in my forehead are fading nicely.

 

[jeff4136]

Good to know no permanent damage done.

Started a new thread

http://www.mcadcentral.com/proe/forum/forum_posts.asp?TID=27 098&PN=1&TPN=1

that may have something of interest in it.

(Ok, I'm too lame to figure out how to get the link to display properly. There's a space after "TID=27" that's not supposed to be there.

)
Edited by: jeff4136