Update on constant-torque/tension

Jun 17, 2010 4 Replies

I've made a decision based upon disassembling and evaluating the clutch that was on my machine.



The existing clutch is a single friction disk, two pressure plate spring- loaded clutch that releases from stiction at about 21 ft-lbs, and maintains about 11 ft-lbs in slip. Lousy hysteresis.



I have evaluated hysteresis clutches, magnetic particle clutches, and multi-plate friction varieties. The hysteresis types don't reach the desired torque, the mag-particle clutches in the 20 ft-lb maximum I need are thousands of dollars each, and the mechanical ones have too much hysteresis, even in multi-plate, wet versions.



So... A small (1/6 hp) split-phase gear motor running at 67rpm with



55in-lb torque will be pulley'd down 4:1, giving roughly 220in-lb torque (around 20 ft-lb max) -- about the max the original clutch would deliver.

That will spool up the roving via a dancer arm sprung with a LONG, strong spring that will vary only about 4% in tension over the throw I need. That dancer arm will have two limit switches. One to turn the motor on or off as the arm moves through the acceptable tension range, and another to turn the whole machine off if too low a tension is detected. A momentary "override" switch will allow me to run up the tension when first spooling the roving.



I think this electro-mechanical "bang-bang" servo will work as well as a higher-tech solution.



LLoyd


What's that Lassie? You say that Lloyd E. Sponenburgh fell down the old rec.crafts.metalworking mine and will die if we don't mount a rescue by Thu, 17 Jun 2010 17:41:19 -0500:

you could hang a weighted roller between two other rollers to have real constant tension over a long length, rather than your spring. (if you have the space.)

snipped-for-privacy@privacy.net (dan) fired this volley in news:4c1be96e.1936139 @news20.forteinc.com:

Another good idea. Actually a LONG spring being extended a small percentage of its length operates approximately linearly. But you need room for the long spring. There's not much space in this machine to play with, but a weighted roller could fit. Thanks!

LLoyd

What's that Lassie? You say that Lloyd E. Sponenburgh fell down the old rec.crafts.metalworking mine and will die if we don't mount a rescue by Fri, 18 Jun 2010 16:57:55 -0500:

Don't forget that a weight also has mass. If your motor starts, um 'briskly' it may tug a little on your media, as the weight accelerates.

snipped-for-privacy@privacy.net (dan) fired this volley in news:4c21f842.5732655 @news20.forteinc.com:

Yup, and that would create an over-compensated oscillation that might continue ad-infinitum (PIO -- Pilot-induced oscillations). But if there's room for a weight, there's room for some viscous or hydraulic damping, along with a PID loop in the controller that expects the mass to be there.

LLoyd

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required