# Precise speed reversible motor

#### abstruse1

Joined Jan 18, 2021
13
Greetings. I want to build an unusual clock that will operate by pulling a weight up a 12' pole for 12 hrs. then letting it go back down for 12 hrs., etc. The weight will be in the range of 5-10 lbs. and it should move 1 ft. every hour, thereby telling time by the position of the weigh. The lifting mechanism will be a small steel cable that runs up the inside of the tube, around a top pulley to the outside, and down to the weight. There'd be a motor box at the base of the pole.

What kind of motor would work best for this? Stepper? Servo? Simple geared DC motor? I'd need to be able to vary the speed of the motor precisely to regulate the clock, and the speed should be consistent over several months. I'm assuming the cable would wind around a spool that wouldn't require stacking the cable, e.g. always having the same diameter.

The clock will be outdoors, so it will experience temperature changes of ~120 degrees F, but otherwise it'd be protected from the elements.

Joined Jan 15, 2015
7,517
If I were thinking about doing this I would likely consider a stepper motor with a reduction gear. Less fancy each step is 1.8 degrees so one revolution is 200 steps. Next figure in your spool diameter. The accuracy will be a function of your clock pulses but each pulse will be 1.8 degrees of rotation. Next is change direction which is pretty easy with a stepper motor. So 12 hours of up and 12 hours of down. The speed and travel will be as accurate as your clock pulses. Today pretty accurate clock oscillators can be had relatively inexpensive and their outputs divided down pretty easily. Things like the mechanics are up to you, Space limitations and environmental as well as the math for your spool. Just make sure your clock pulses are stable over a wide temperature range. A Google of TCXO Modules should get you started for the clock. This is how I would likely approach it but I am sure there are plenty of others ways to get to Rome.

Ron

#### BobTPH

Joined Jun 5, 2013
8,980
I don’t know how you expect to regulate the speed if winding a cable on a spool. A continuous loop belt or chain would be much easier.

Also, if the cable is inside the pole how does it connect to the weight? Is the pole slotted for the entire distance? This would weaken the pole considerably.

Edited: Ignore second paragraph, I re-read the original post and my question is answered.

Last edited:

#### panic mode

Joined Oct 10, 2011
2,752
the weight on a steel cable is inside clear tube?

Joined Jan 15, 2015
7,517
I don’t know how you expect to regulate the speed if winding a cable on a spool. A continuous loop belt or chain would be much easier.
Thought about that later and yes, I would run a continuous loop. I would skip the counter weight idea and use a spring in the loop for tension at the two ends Ns maybe add a few feet to the pole. The biggest problem to overcome would be exposure to the elements.

Ron

#### Alec_t

Joined Sep 17, 2013
14,317
I like your steam-punk approach to telling the time. Perhaps the 'clock' could be accompanied by a weather rock ?

#### BobTPH

Joined Jun 5, 2013
8,980
I’ve had he idea a digital sundial. it would sense the direction of a shadow using photosensors and use a microcontroller to adjust for the seasonal variation in the sun’s angle.

#### abstruse1

Joined Jan 18, 2021
13
Thanx, guys. I wasn't sufficiently clear in my description of what I want to do.

There is no "counterweight" and there is no slot in the pipe. The cable would be wrapped around a single level, motorized spool at the bottom, run up through the inside of the pipe to a redirectional pulley at the top, then it would run parallel to the pipe on the outside with some kind of an indicator on the end, e.g. a steel spider.

When 12' (or maybe 24') were spooled out, lowering the spider, the spool end of the cable would be at 12:00 on the spool, so every increment of spool revolving thereafter would raise the spider.

I can't picture how a continuous loop (e.g. chain) would work.

Joined Jan 15, 2015
7,517
I can't picture how a continuous loop (e.g. chain) would work.
The idea is the cable forms a loop, half outside and half inside. The cable is run up and down using a motor. I suggest a stepper motor. With a stepper the speed is controlled by the frequency in and also the direction of rotation can be changed. So using a stepper and a reduction gear box which is really pretty simple the pulley has a diameter so you calculate how much travel up or down for each full revolution of the pulley. You maintain cable tension by using a spring where the two cable ends are joined. With an average stepper motor each step is 1.8 degrees rotation so 360 degrees / 1.8 degrees = 200 steps per revolution or more simply 200 pulses to the stepper driver per revolution. The stepper driver also gives the option of rotational direction as Clockwise and counterclockwise. Think rubber band with an idler pully on one end and a drive pully on the other end. The merit to a gearbox is you can run the motor faster which results in smoother travel up and down. A Google of "stepper motor reduction gearbox" will bring up dozens of motors with gearbox already attached. Again, less seeing what you have this is how I would likely approach it.

Ron

#### LowQCab

Joined Nov 6, 2012
4,074
Incorporate a counter-balancing-weight if at all possible.
This will allow for a much smaller, ( and much less expensive ),
Motor, which will be under much less strain.
.
.
.

#### WBahn

Joined Mar 31, 2012
30,062
Greetings. I want to build an unusual clock that will operate by pulling a weight up a 12' pole for 12 hrs. then letting it go back down for 12 hrs., etc. The weight will be in the range of 5-10 lbs. and it should move 1 ft. every hour, thereby telling time by the position of the weigh. The lifting mechanism will be a small steel cable that runs up the inside of the tube, around a top pulley to the outside, and down to the weight. There'd be a motor box at the base of the pole.

What kind of motor would work best for this? Stepper? Servo? Simple geared DC motor? I'd need to be able to vary the speed of the motor precisely to regulate the clock, and the speed should be consistent over several months. I'm assuming the cable would wind around a spool that wouldn't require stacking the cable, e.g. always having the same diameter.

The clock will be outdoors, so it will experience temperature changes of ~120 degrees F, but otherwise it'd be protected from the elements.
How accurate do you need it to be? What is the time resolution that someone looking at this clock is expecting? Just to the nearest hour? Or to the nearest minute? As the weight is lowered, it will stretch cable more. Plus, both the pole and the cable will change dimensions over the wide range of temperatures.

How are you going to ensure that the cable winds nice and neat? Is the spool going to be grooved to guide it.

#### Ya’akov

Joined Jan 27, 2019
9,165
As @LowQCab suggests, a counterweight is a very good idea. It will greatly reduce the motor capacity required. If you put the counterweight inside the tube, it will not be visible.

Running this open loop is another difficulty. That is, depending on the assumption that feeding or collecting a certain amount of cable will correspond to the weight’s position may be a mistake. Some sort of empirical sensor, determining where the weight actually is would allow you to move it to where it should be at each update and compensate for the vagaries of the interaction of the many mechanical components.

It’s possible that using a rotary encoder or other position sensor on the driven pulley may be sufficient if there is no slip. The idea there may be slip is another consideration. The pulley will have to provide enough friction to hold and lift a substantial weight. It might be a good idea to use a capstan for the actual drive, out of sight and below the pulley.

A far as accuracy goes, instead of counting on timing to have the weight in the right place, if you have a closed loop (that is, you can get feedback about the weight’s position to sufficient accuracy) you can make empirical updates to the position at intervals. These could be very small, say one second, which shouldn’t be perceptible to an observer but will make the average position of the weight quite accurate.

One more thought. It might make sense to consider a brake on the cable that is released to update and applied afterwards to ensure the weight can‘t move unless it’s meant to.

#### Alec_t

Joined Sep 17, 2013
14,317
The weight will be in the range of 5-10 lbs.
Why so heavy? Personally, I'd make a dummy weight, say 1lb, with the appearance of a much heavier weight. The weight would have some guide mechanism on the pole so that it didn't flap about in the wind.