Absolute Beginner - Need help with selecting motor for winch system

Using Arduino Motors, Mechanics, Power and CNC
#1

I have a contraption designed, and one of the steps requires a weight of about 1kg to be lifted at a controllable speed of 1-5 m/s, max height is 5m.

The initial plan was to glue a spool to a cheap Hi Speed DC Motor that I found in a robotics kit. The following are its specifications:

Voltage Range 3.0-12V Rated voltage 9V No load speed 20160RPM No load current 680MA Maximum current 3.8A Rated torque 600G.CM Rated Power 20W Starting torque 1000G.CM Size Diameter 27.7MM, length 38MM, shaft diameter 2.3MM, axial length 12.7MM

I also have a NEMA 17 stepper motor in the kit.

FYI I am a total beginner and don't have a good understanding of what these mean.

I require the position of the weight must be accurate to the cm scale. So I thought of using the stepper motor, but I read that they are inefficient and cannot hold onto the weight in a no-power state. And I read that DC motors are inaccurate. So I'm really confused right now on what I must do.

I'd really appreciate if anyone could help me out in a detailed solution. Thanks in advance.

#2

I don't think a cheap high speed DC motor is the way to go. You can't get the positional accuracy you need for your project without an encoder or some other way to measure the position of the weight. Integrating those things together will require lots of programming and probably implementing some kind of PID control scheme. That would be a lot of hassle.

Using the NEMA 17 is your best option, this is the kind of application they are built for. It has built in positional sensing so you know where your position is at all times. This will give you the required accuracy you need. No motor can hold onto any weight in a no power state without external help. However NEMA motors are capable of producing a holding force as long as they are powered. I can't speak for their efficiency but isn't getting something built and working a higher priority than efficiency? There are also lots of resources on connecting NEMA motors to arduinos.

Whats the bigger picture and what other options can you explore? Instead of using a spool could you use a belt drive or chain drive instead? As a spool winds the diameter increases because the cable is laid on top of each other. This decreases accuracy. Chain and belt drives don't have that problem and are commonly used. It is also possible to add counterweights to belts and chains to reduce needed torque and hold position with power off.

You could even look at GBRL and g code senders made using open source components like arduinos. G code and CNC was developed to control position and velocity using text. It is very easy to use and very common on CNC machines. It is also one of the most highly developed methods of controlling servos.

#3

Thank you very much for your answer. I really appreciate it :)

I have a few more questions, if you don't mind

Your input on using a belt drive mechanism is very nice. That's the way we will implement it now. I'm thinking of making something like this:

Actually this is an art installation with many such "movable" pieces, and hence the concern for power efficiency. I don't want 10-20 odd motors spinning all time, just to keep the parts hanging.

Could you suggest a work-around for this?

Also, I'm not sure how to do the math, if my motor can lift the 1kg weight at the desired speed. Any inputs on this is also deeply appreciated.

It is also possible to add counterweights to belts and chains to reduce needed torque and hold position with power off.

Didn't really get this. Could you please elaborate?

#4

1kg at 5m/s is 50W mechanical power not allowing for friction nor the weight of the rope/cable.

Assume 70--100W mechanical is needed. A NEMA17 stepper is completely unworkable. Without power do you want the load to hold position?

Are there any safety considerations if the motor/power fails?

#5

For a system to be fail-safe with no power you need a brake device. Normally a brake is fitted to the motor shaft which, when energised, releases the brake mechanism and permits the motor to rotate. If motor power fails or is intentionally de-energised the brake instantly applies and the motor shaft becomes locked.

You could do worse than getting hold of a powered wheelchair motor unit. These come fitted with brakes and normally operate off 24 volts. To determine load position you could fit a sensor disk to the motor shaft and count pulses.

This type of motor also comes with a worm gearbox fitted and typical output RPM would be suitable for your purposes. Once you know motor RPM you can determine capstan drum diameter to suit your desired hoist speed.

#6

The power required to move an object at 5 meters per second against a force of 9.81 Newtons (the downward force of a 1 kg mass) is 49 Watts, your 20 Watt motor won't hack it.

#7

Never mind the lack of power, the 20,000 rpm will be a problem ! :o

#8

If space is not an issue, you may want to consider an automotive wiper motor. The drive used in them (not sure if all of them but any that I have seen) is worm gear which equals a lot of torque and position held when un-powered. - Scotty

#9

Here is what I meant by counterweight part.

Counterweights are weights added to a pulley system, belt drive, or lever system to cancel out the force of gravity. In your case I envisioned a pulley system similar to an elevator. A very heavy elevator's weight is cancelled out by a large counterweight mounted on the other side of the pulley. It allows the motor to be smaller than otherwise allowed. And since the force of gravity is negated the energy needed to hold the motor in place (holding torque) is much smaller.

Motors only spin when they are moving something. They are not car engines that spin when idle. I think you got power efficiency confused with low noise levels.

Here are some comments about the math.

Before you do the math you need to figure out what's really important. Is moving at a precise speed important or did you just give that number because you had to fill in a blank space? Is getting position of weight within 1 cm of a target what you need or does it just need to look good? The more inaccurate you can be the easier it will be to do your project. In your case it just might be easier to tinker and go big. The basic mathematics are laid out in books on engineering statics and dynamics, check one out from your library.

I'm a mechanical engineering student and have always been interested in projects like this. If you post some pics or sketches I might be able to help optimize the mechanical aspects of your project.