I am modeling a screw and nut that are supposed to fit into each other
without collision and stay without collision throughout the process of
moving nut on screw  from top to end.

However, I have used approximation formula that might not be exact:

isosurface {
  function { f_helix1(x,y,z, 1, 20/3*omega1, dy, inner_r+dy+dy*2/3, 1, 1.5,
0) }
         // distance_wind, thicknessm, r, shape, 0, cross-section type
         // 20/3 is an empiric factor connecting number of windings of helix
         // per unit of length with parameter P2 of f_helix1()
         // ***********************************
         // *** WHAT IS THE EXACT VALUE???? ***
         // ***********************************
  contained_by {
     box { <-2, -l/2, -2>, <2, l/2, 2> }
  }
  max_gradient 1.5
}

Later I use this screw from macro:

    object {
       Screw_Approx_Type_01(r_rod, l_rod, omega1, 60+tt)
       rotate y*(-l/l_rod*omega1*360)
       texture {
          //Silver_Texture
          T_Chrome_4D
       }
       translate <0, 0.27-l, 0.0>
    }

..... but it is not exact fit - after 5 turns of screw inside nut, there
seems
to be like if in 25 turns the chrome would meld with chrome, if I am making
myself clear.

So, I'd need a more correct value of 20/3 factor multiplying omega1, but it
is very hard to determine experimentally, if it is linear at all?

Many thanks in forward,
Marvin

Post a reply to this message

Wasn't it Marvin who wrote:
>
>I am modeling a screw and nut that are supposed to fit into each other
>without collision and stay without collision throughout the process of
>moving nut on screw  from top to end.
>
>However, I have used approximation formula that might not be exact:
>
>isosurface {
>  function { f_helix1(x,y,z, 1, 20/3*omega1, dy, inner_r+dy+dy*2/3, 1, 1.5,
>0) }
>         // distance_wind, thicknessm, r, shape, 0, cross-section type
>         // 20/3 is an empiric factor connecting number of windings of helix
>         // per unit of length with parameter P2 of f_helix1()
>         // ***********************************
>         // *** WHAT IS THE EXACT VALUE???? ***
>         // ***********************************
>  contained_by {
>     box { <-2, -l/2, -2>, <2, l/2, 2> }
>  }
>  max_gradient 1.5
>}

If you use the value "2*pi" where you've got "20/3*omega1" then you get
exactly one complete turn as the y value increases by 1 unit (e.g. from
-1/2 to +1/2)

-- 
Mike Williams
Gentleman of Leisure

Post a reply to this message

I2w### [at] econymdemoncouk...
> If you use the value "2*pi" where you've got "20/3*omega1"

One have to say it omega3 is better for your health

Sorry couldn't resist

Marc

Post a reply to this message

"Marvin" <mto### [at] grfhr> wrote:
> I am modeling a screw and nut that are supposed to fit into each other
> without collision and stay without collision throughout the process of
> moving nut on screw  from top to end.

Bizarrely I've just been modelling something very similar although without
that degree of accuracy ;)  See my post in image files...

Pete

Post a reply to this message

"Marc" <jac### [at] wanadoofr> wrote:

> I2w### [at] econymdemoncouk...
> > If you use the value "2*pi" where you've got "20/3*omega1"
>
> One have to say it omega3 is better for your health
>
> Sorry couldn't resist
>
> Marc

This was funny, Marc.

You cheered me up.

Marvin

Post a reply to this message

Mike Williams <nos### [at] econymdemoncouk> wrote:
> Wasn't it Marvin who wrote:
> >
> >I am modeling a screw and nut that are supposed to fit into each other
> >without collision and stay without collision throughout the process of
> >moving nut on screw  from top to end.
> >
> >However, I have used approximation formula that might not be exact:
> >
> >isosurface {
> >  function { f_helix1(x,y,z, 1, 20/3*omega1, dy, inner_r+dy+dy*2/3, 1, 1.5,
> >0) }
> >         // distance_wind, thicknessm, r, shape, 0, cross-section type
> >         // 20/3 is an empiric factor connecting number of windings of helix
> >         // per unit of length with parameter P2 of f_helix1()
> >         // ***********************************
> >         // *** WHAT IS THE EXACT VALUE???? ***
> >         // ***********************************
> >  contained_by {
> >     box { <-2, -l/2, -2>, <2, l/2, 2> }
> >  }
> >  max_gradient 1.5
> >}
>
> If you use the value "2*pi" where you've got "20/3*omega1" then you get
> exactly one complete turn as the y value increases by 1 unit (e.g. from
> -1/2 to +1/2)

Thank you, Mike, this worked exactly, and my screw and nut turn w/o internal
collision now.

I wish there was a better open source screw and nut model already on net,
because I don't have exactly the time to model screw, I am actually doing
the transformer ...

Sorry for digression.

Marvin

Post a reply to this message