Electronic motor speed control.

I am interested in buying or modifying an electronic speed control with two types of controls. The motor will be used on a gantry (for the tool motor)
and should be selectable by the operator, but yet be changed by an encoder mounted to keep track of how fast the gantry is moving. The idea is to keep the tool motor turning at a constant rpm-to-linear travel. For example I should be able to set the speed of a sewing machine to get a specific number of stitches per inch regardless if the gantry it is resting on is being moved manually at 2 inches per second, or 8 inches per minute. This is necessary to keep the stitch the same size regardless of speed. Does anyone know of such a control for less than 200 dollars? Sewing Robot
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Don Davison wrote:

No.
For around $1000, most of the Galil controllers can do it, though.
                    John Nagle
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
On Tue, 20 Sep 2005 22:33:00 -0700, Don Davison wrote:

Really depends on required speed accuracy - if the sewwing machine already has a electronic speed control and you can tolerate a 10-20 % tolerance, you may need nothing more than a simple tachometer circuit for the gantry that drives the sewing machine controller in an open loop fashion...
Peter Wallace
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Thanks Peter: That is what I had in mind but was thinking of an encoder (tach?) Actually, the machine does not have a speed control on it because I haven't built it (I'm a real dreamer) I know there are machine that do this but extremely expensive. I think I can build the simple electronic speed control, but don't know how to retrofit it with a closed loop tach or encoder. In theory if the control is linear the needle will stop when the gantry stops, requiring the operator to pulse the circuit manually even if the gantry is not moving. With that in mind I think it would require a comparitor circuit ??? Do you have something in mind that would provide accuracy within 10-20%?? For this application, that would be a very small amount since the stitch may only be 1/16 inch. Ideally, this should be for an AC motor, maybe 0.125 hp or 0.0625 hp. I have experience with milling and welding and knowledge of aluminum foundry operation but need help with the control circuit.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Don Davison wrote:

$200 is close to the price-point for low-end (e.g. Disney brand) quilting machines with embroidary functionality built in. However, these are completely computer-controlled x-y tables covering only 6" or so. You want something the operator moves over wider distances.
As Peter suggested, you might be able to get by with an open-loop control. Thus some part (tach mounted to a table axis? optical mouse? roller bars near the point of sewing?) will detect how fast the gantry/fabric is moving. Based on the measured velocity, determine a voltage (or PWM %) that drives the sewing machine at the proper speed. The required motor velocity is easy enough: stitches/inch * inches/second = stitches/second
If you're lucky, a formula like voltage=a*speed+b will work, where B is the cut-off voltage where the motor doesn't turn. Different slopes A would then give you different stitches/second ratios.
If you're not lucky, you'll have to find some nonlinear formula. A look-up table would probably be good here. Apply a known voltage to the motor. Measure how many stitches/minute you get. Do this for several voltages between no stitches and maximum speed and compile a table of (stitches/minute,voltage) pairs. Then for a desired rate of stitching, you can find the closest two entries in the table and use linear interpolation to find the required motor speed. Most cheap microcontrollers (PIC,AVR,...) could do this math fast enough.
The addition of a simple rotary encoder (servo pot or even just a cam-drive microswitch) to the motor's output shaft (or the cam which drives the needle) would allow you to do some feedback and get better results than the open-loop control outlined above. This may be needed as you sew through different thickness fabrics.
For me, the hardest part would be detecting the fabric motion. That, and building the required circuits.
Q: Is this a solution in search of a problem? I thought the foot-pedal worked quite well for adjusting machine speed while sewing. Yes, the pedal takes training; but a simple stitches/inch system won't work at unusual spots like the beginning/end of a seam, will it? Maybe a toggle between stitches/inch and fixed rate would address this problem.
Later, Daniel
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.