Hi,
Jinx wrote:
>Does anyone have suggestions (and hopefully practical
>experience) for methods of propelling a sled on a track ?

Well certainly not experience (at least not for a >9.8m/s^2),
but I have an idea, not having read the entire thread this is what I =
would
suggest.

As you only have 2 m to reach final speed and I assume at that time
you want to have no remaing acceleration and fairly good control of
the speed, I think this might work:

An 4 stage acceleration device consisting of:

* initial launcher, this is just to get the sled moving and
  enter the remaining three acceleration 'zones'. This could be
  just by pushing it by hand, using an rubber band, spring or something =
similar.

* Three acceleration zones, each with an motor running (pwm) and =
configured
with two interconnected flywheeels acting as a 'pinch' on the sled.

I think the best accuracy would be achived when the first =
acceleration-pincher
would be set to reach 0.4-0.5 final speed and the second at 0.9-1.0 of =
final speed.
Then the last one would be set to exact 1.0 of target speed and just do =
the=20
final correction.=20
I don't see an major problems with this configuration, getting hold of =
flywheels
with sufficient friction should be quite easy, linking thewn in pairs =
and getting
them to pinch the sled should also be quite easy (rubberband, spring, =
etc) the
only thing needed trimming and tinkering would be the pwm control.

Actually I sometimes help a friend with servicing and repairing tennis =
trainers,
they work almost exactly like this, getting the tennis ball in pinch =
between two fast moving
(aluminium) wheels gets it upto speed *fast*.
Speed is controlled by pwm, the accuracy is so-so as there is only one =
wheel acting upon
the ball. But is works ok for a trainer where the speed in wanted to =
wary to some degree.

/Tony

--
http://www.piclist.com hint: The PICList is archived three different
ways.  See http://www.piclist.com/#archives for details.