AC vs DC servo motors in large scale cable robot design

[cpcoe]

hi, I'm designing a 50ftx50ft, 8-cable Cable Driven Parallel Robot and am thinking this is a pivotal decision. the corner posts are 10ft tall and the machine should be able to lift 100lbs at least 5ft off the ground. traveling speed of the end effector is less important than accuracy but I'm not trying to build a slug either! I am not attached to AC or DC, and the prices aren't that different in the 1/2hp range, will the controls and drivers present any hidden complications? what are the pros and cons of each system?
thanks for your help

 

[Baluncore]

Welcome to PF.
The motor selection will be dependent on the controllers available to you, and what will be controlling the system.
It is important that you make it work ASAP. The details can be changed later if you need to manufacture in quantity.
I would use PWM, DC servo motors, controlled by off the shelf H-bridges, with shaft encoders on the wire drums.
That will give a continuous estimate of speed, position, torque and load.

 

[cpcoe]

Thanks, I'm thinking the rotary encoders should be independent of the winches because the cable winding on the drum creates a variable axle size. I imagine a pulley wheel/ encoder mounted between the winch and end-effector for cable length and velocity measurement. In regards to controllers and controlling systems, I would like to be able to manually 'drive' the machine with a joystick (or something) and record the drive for repetitive playback. Any insight as to how this might look in hardware and software?

 

[Tom.G]

rotary encoders should be independent of the winches because the cable winding on the drum creates a variable axle size.

Taking into account Murphy's Law^*, there will be slippage between the cable and the idler driving the encoder. I suggest a follower arm riding on the cable as it winds on the winch, and putting the encoder somewhere on the winch drive. Servo motors typically have an encoder mounted on their back end (it makes the gearing to the driven element simpler).

* Anything that can go wrong, will.

Cheers,
Tom

 

[Baluncore]

Thanks, I'm thinking the rotary encoders should be independent of the winches because the cable winding on the drum creates a variable axle size.

Correct. The rotary encoders are needed on the motor / drum to control the motor direction and velocity. The position of the load needs to be measured quite independently of any loaded positioning cable. A slipping or broken positioning cable will cause a detectable difference between the drum estimate and the real position. How else can it detect the next fault to occur?

Do not use the energy transfer systems to obtain critical positioning data. It will crush or strangle the operator.

If you must use cables to measure position, use a totally independent and parallel system that maintains the loop length without a spool. Make sure that the tension in the sensor cable is within narrow specified limits by using a spring or mass with switches to detect any length error. Kill the power and sound the alarm if it goes outside range.

Take a look at the modern equivalent to “ball chain”, as used on vertical window blinds. The string of spaced plastic balls moulded onto the fiber will not slip on a position index encoder wheel made with holes to accommodate the beads and a Vee groove to accommodate the fibre.