Hi, I suggest taking a look at ebay... I bought a linear "stage" as I=20
think these things are called that has approx 300mm of travel and had=20
ballscrew, aluminum enclosure, and (unfortunately) AC servomotor which=20
I'm replacing with a DC.  Around $250 I think.

I did a bunch of research while designing a (yet to be built, of course)=20
DIY pick&place.  $1700 worth of nearly everything needed waiting for=20
time to come around again.

For measuring position, it doesn't get better than a glass linear scale.

Some sources:

Ebay sellers:

yishi_xingtai
www_salecnc_com
linearmotionbearings2008
fa-parts
d_kwa2520
carolbrent
zsinstruments -> glass scales

Motors and drivers:
http://www.kelinginc.net/index.html
pololu.com has a variety of motors and drivers

I bought some things from several of the ebay sellers, the glass linear=20
scales came from a Singapore seller but I think the zsinstruments (in=20
the USA) sells the same things from the same china factory.

Have fun and good luck!
J



Dave Tweed wrote:
> I have a client who wants to do some experiments involving deliberate ant=
enna
> motion in a GPS-based system. He's done some simulations and likes what h=
e
> sees, so now it's my job to set up an actual prototype so we can see how =
well
> it works in practice.
>
> We have a small box of electronics that includes the GPS antenna and rece=
iver
> that masses a total of about 0.5 kg. We want to move this box along a
> horizontal linear track that's about 300 mm long, with a sinewave motion =
that
> has a period on the order of 1.0 seconds. This means that we need to be a=
ble
> to generate a peak velocity of about 0.95 m/s and a peak acceleration of
> about 6.0 m/s/s (and a force of about 3 Newtons).
>
> We also need to know with fairly good timing precision (on the order of 0=
..5
> to 1.0 ms and 0.5 to 1.0 mm) where the box is in its motion cycle. In cas=
e it
> matters, the motion will likely not be continuous -- instead, the box wil=
l sit
> most of the time at one end of the track, and on command, will make one t=
rip
> to the far end and back.
>
> I've never put together a motion control system before, but based on some
> preliminary Googling, I would guess that given these speeds and loads, th=
at
> a belt-driven stage powered by a stepper motor would be the type of thing=
 I
> should be looking for, along with a high-level motion controller that can=
 be
> programmed with the sinewave profile.
>
> Secondary considerations are that the unit should be relatively compact -=
-
> our box of electronics is about 50 x 50 x 125 mm and we don't want someth=
ing
> that completely dwarfs this. We don't want to spend more than about $1K t=
o $2K
> on this, but we want to buy as much of it off-the-shelf as possible. This=
 is
> just a one-off for now, but if it works out, we'll want a cost-optimized
> solution for a production design. The control interface would preferably =
be
> RS-232 (with a separate hardware sync signal?), but we can adapt as neede=
d.
>
> Does anyone have any recommendations for vendors for this sort of thing, =
or
> any other pertinent advice? Thanks in advance.
>
> -- Dave Tweed
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.