Applying an accurate tension on a line being reeled in

In message , dated Sun, 3 Sep 2006, Dave writes

Coil-winding machines have a spring-loaded arm with a pulley on the end, over which the wire runs. You adjust the tension by stretching or loosening the spring with a threaded stud and a handwheel.

Another way is to have two vertical pulleys separated horizontally. Between them is a pulley block loose on the wire, with a weight hanging from it. For electronic control you could use a linear motion transducer and use its signal to control a brake.

I guess I was hoping someone would say that some XYZ type motor naturally generates a constant torque. I'd be reeling in line at up to

You could use a mini semiconductor pressure sensor. You would pull the line up 90 or 45 degrees away from the sensor, this will give you a reading, you will then need to compensate for the friction that the line will create. Maybe put a small foam cushion on the sensor and lay the line on this, and then pull it up 45 - 90 degrees. This will limit the friction that reaches the sensor, while still being able to give it some reading. Your gonna need to amplify it alot.

Ok. Check out these two pages,

Tp a pretty good approximation, the torque generated by a DC motor (brushed or brushless) is a single-valued function of the current through the coils, while the voltage you need to provide to drive that current through the coils is the sum of the resistive drop through the coil and the back-EMF of the motor. The back-EMF is proportional to the speed of rotation of the coils and the magnetic field through which they are rotating.

Philips had a patent - long since expired - on driving DC motors at constant speed from a circuit with a negative resistance equal to the winding resistance. I've puilt a couple and they work pretty well. The driving impedance is only negative at DC and up to around 100Hz or so, well below the self-resonant frequency of the motor coils.

Your problem is easier - you simply need to set up a constant current drive for the motor.

The gotcha's are in the maximum speed that you want the motor to run at

- which determines the maximum voltage that your constant current source has to deliver - and the problem of heat dissipation in the motor coils.

The torque (in Newton-metres, or ounce-inches) that your motor delivers per amp of coil current depends on the magnetic field generated by the permanent magnets involved (at least they do in the small DC motors you seem to be talking about). If you generated enough heat in the motor coils for any length of time, the magets will get hot and the magnetic field they produce will drop, and with it the torque per amp.

You can sense the motor temperature and increase the coil current to compensate for the lower magnetic field, but you have to make sure that you don't get to the Curie temperature of the magnets, because that is the point where they stop being magnets, and end up demagnetised after they cool down.

You can also generate a constant torque with a microstepped stepper motor - equally well described as a synchronous AC motor - but for that you need a shaft encoder of the sort that you find in brushless DC motors or in the commutator plus brushes you find in a simple DC motor.

A DCPM (D C Permanent Magnet) motor naturally produces constant torque when supplied with constant current drive. A 100 watt motor rated at 1800 RPM produces about 4.7 lbf-in of torque at rated current, regardless of speed. A DCPM servomotor would be ideal here if you can find one. Most small DC motors deliver higher speed and less torque. This could be reduced with gears or belt but the speed reducer will soak up some torque in proportion to speed. Still, you could probably "dial in" your current to produce the torque you want.

mechanically ? use a catenary arm that actuates a scrub brake on the payoff side if there is one, if its a loose line your pulling and your trying maintain tension on the take up end then you need a drag pulley which is control by the catenary arm.. you simply have an adjustable spring on the arm, the arm has a roller on it where the line runs on to monitor the tension.. this arm simply releases the scrub brakes as it gets pushed down..

Can you explain the application more? Is it a do-not-exceed situation?

Dave,

You haven't given enough information about the problem you are trying to solve.

Do you have an active load on the line? e.g. If you were talking about reeling in a fish, then you'd have to be able to let line out if the fish pulled too hard, and take line in if the fish swam toward you. And that's just scratching the surface, for an application like that one.

I also saw your subsequent post about motors with constant torque vs current. I don't see how that would help you, unless you have a very well defined, repetitive application, for which the mechanical parameters are all already known and don't change much. But if that were the case, your problem would already be almost solved.

So what are you trying to do? What are the dynamics, variables, constraints, et al?

- Tom Gootee

I used to use some RCA 16 mm TV film chain projectors that used a tension arm mounted on a variac. The variac adjusted the voltage to the reel drive motor. They used the same setup on the supply reel so the film was handled as gently as possible, as long as everything was powered by a Sola Adjust-A-Volt stepping AC Voltage regulator which adjusted the AC line in steps under one volt. On the other hand, if the AC supply varied more than about +/- 5%, all bets were off.

No, this is a very smooth situation. No fighting fish pulling on the line. The concept is to conduct hull characterizations (in some cases verification of calculated) of towed kayaks.

There are compact controllers which can control the torque applied and speed according to your settings. Most are designed for a 90 VDC motor.

maybe run it round a capstain that's driven at a constant torque

some idea of scale would help, is this 10 pound fishing line or a 100 ton hawser?

Bye. Jasen

ideas

In an electric motor, current is proportional to the torque. Control the current and the tension is controlled.

AH!

pretty-much any DC motor fed a constant current will give reasonably constant torque.

derate heavily provide cooling while the motor is stalled or running slowly,

be aware that as the take-up spool fills more torque will be needed for the same tension.

Bye. Jasen

Or use one motor to only tension, and another motor, loop-controlled, for the take-up spool.

...Jim Thompson

A few years back, there were many surplus magtape drive motors available. You can probably find one on ebay if you know what to look for. These were designed precisely to provide constant torque at relatively low speeds. Many were manufactured by Indiana General. They are about 4 inches diameter, with length varying according to the size of the tape drive. Whether you find a tape drive motor or not, look for a PM type with a relatively low RPM/volt rating, say 10 to 100 rpm per volt, and a low armature resistance, in the low ohms. This is regardless of the nameplate voltage of the motor. For example, either a '90 volt' or a '12 volt' motor will work, provided that neither the rpm/volt nor the armature resistance are too high. PM motors with low rpm/volt tend to have lots of commutator segments and to be designed for torque. Low armature resistance insures that the motore won't dissipate too much power. This type of motor is popular with builders of small wind generators, also. For en example, see ebay item