Servo motors on rotary table, no slip rings!

[BigSteve]

Hopefully I have the right forum!

I have an application where I need to have a servo motor and some I/O mounted on a rotary weld table, but we don't want to use slip rings. I realize I will need to for motor power and for the welding ground cable, but I don't want to run communication signals through a slip ring. This would include the servo encoder/resolver, and the external I/O.

Many companies make remote sensor blocks which can send power through an air gap, but they all need to be centered on the axis of rotation, as do the slip rings.

So, does anyone have any experience with these kind of problems? The last similar job we manufactured with cable drums on the rotary axis and set up positive stops to prevent cable stretching. We would like to avoid this if possible, as even flex cable has a finite lifespan, and it is very difficult to replace.

 

[ve7brz]

Hi Steve, Although I've not dealt with this particular problem before I'm wondering if Fibre-optic cable would stand up to the fatigue you are dealing with here. As I see it only two (or perhaps four) strands are required. If the encoder logic can be handled at the table then all you require is a serial data link between your on-table I/O board and your computer. These serial I/O interfaces are an industry standard as you probably know. If the response of this system is inadequate to accomodate your encoder (I'm assuming a quadrature encoder) then your two encoder signals can be brought out separately.

An alternative might be to achieve signal coupling via wire coils centered on the table axis with Hall Effect sensors positioned in proximity to each coil in order to sense the magnetic field generated by each coil. This would of course require coils fixed to both the table and a stationary frame. Although this might be done using pure inductive coupling (coil to coil) the coil to Hall Effect sensor method would be interference free and far more reliable.

Hope this gives you food for thought.

Bruce

 

[electricuwe]

Some years a robot has been built by several departments of Aachen Univeristy of Technology which used inductive coupling for energy transfer and inductive and/or capacitive coupling for signal transfer. This work had been funden by the DFG under ref. no. SFB208

You may find some information here:

http://www.isea.rwth-aachen.de/Ww/english/

Search for Publications on that topic by T. Kalker, A. Eßer and A. Nagel

Later some Japanese servo-drive manufacturer, unfortunately I do not remember the name, continued to work on that topic.

 

[jbartos]

Suggestion: Wireless communication could be considered.

 

[BigSteve]

Yes, wireless was and still is being considered, but I don't think it will work. We have to run servo power, which cannot be wireless or inductive. We have to run servo communication, which can be done wireless. To run a wireless network, I need to have power on the rotary unit, hence my dilemma. I also have a 300Amp conductor that has to be mounted as well.

The deeper I get into this project, the more I am swayed to using slip rings. I have contacted a company that will custom-make a non-metalic slip ring that will combine all the signals I need in this application.

Any other hints?

 

[jbartos]

Suggestion: How about making the rotary weld table a little bigger and locating there a diesel generator?

 

[SERVOCAM]

I have ran into these cases before on applications. I just used a rotary tack & pinion where the center was open to deliver cables to the system. I have also used direct drive rotary motors to drive the table that have an open center to feed wires through. Also Camco's Flex-i-dex.

On these, the motion was only 180 degrees fwd and back. So the cables were just high-flex and came through the center to the motors and stages mounted on the system.

Maybe do an 3-ph AC slip ring to put power to the table top then mount the control pannel there for the drives & motors.

Cameron Anderson
Sales & Applications Engineer

http://www.aerotech.com

"Dedicated to the Science of Motion"

 

[BigSteve]

SERVOCAM: We have done similar things on past projects. We ran our cabling through a drum that was mounted to the axis. The drum contained enough cable to accomodate the rotation of the axis, and then was passed through the middle of the table. This worked fine until the cables wore out, even with the high flex cable. My boss wants me to try and find other ways to solve this problem.

Does anyone know any places to get slip rings?

 

[jbartos]

Suggestion to BigSteve: What a pre-conception.