Choosing a high torque, precise motor for a light-weight device

Hi,

I am currently trying to make a device controlled by an Arduino which would be able to travel on a blackboard and use the chalk on it to draw out functions(f(x)).
My idea is to use the principle of a vacuum cleaner to stick my device to a blackboard by using a brushless motor, so it would be like a small car traveling on a blackboard.
But here's my problem, the motor I use to make the car go needs to be precise for me to draw a good picture, it also has to be light-weight so the car won't come off, AND it has to have enough torque cuz it may have to travel vertically.
My device would contain an Arduino uno, a LiPO battery(for brushless) another battery for Arduino and maybe the motor, a fan(about 5grams) , two wheels and a A4 sized acrylic sheet. I'm not sure how much it would be weigh but I don't think it would weigh over 1 kg.
I don't know which kind of motor I should choose, please give me some advise, thanks! :~

Could I suggest that you might also find it useful to duct the air you're pushing out,downwards to help with the issue of gravity. Theoretically the tyres could provide the necessary friction but it would probably help.

For the driving motor I'd suggest you go for a stepper motor. It has more torque at low speeds and is much easier to control precisely. If you have an old printer hanging around you'd probably find something suitable within.

KenF:
Could I suggest that you might also find it useful to duct the air you're pushing out,downwards to help with the issue of gravity. Theoretically the tyres could provide the necessary friction but it would probably help.

I'd like to, but since the car would turn in all directions, I don't know how to keep the air flow pointing to the ground.

KenF:
For the driving motor I'd suggest you go for a stepper motor. It has more torque at low speeds and is much easier to control precisely. If you have an old printer hanging around you'd probably find something suitable within.

Unfortunately I don't have an old printer hanging around, so are there any specific steppers that I can use? Cuz I am confused by all the different ones out there.

There's no way you are going to get that to work without an x/y table frame with the chalk held in the x axis (moving left & right)
while the x axis moves up and down. Each axis is a strip just wide enough to hold the chalk and secured at both ends.
If you have a mechanical model that is better I would like to hear it. There is a system designed for writing on store front windows with a white pen that draws intricate patterns. I don't recall what it is called but I saw a video and it worked well. (Wooden structure by the way. Little or no metal parts other than the motor.

raschemmel:
There's no way you are going to get that to work without an x/y table frame with the chalk held in the x axis (moving left & right)
while the x axis moves up and down. Each axis is a strip just wide enough to hold the chalk and secured at both ends.
If you have a mechanical model that is better I would like to hear it.

Actually my idea is quite different. If we put the blackboard on the ground, the device would just be a car, by controlling the angle each wheel goes precisely, we can make the car go along the function by always facing the direction of the tangent line, and with a chalk fixed on the car, it could draw out the picture.

I'll have to see it to believe it. Good luck.
By function, what are you talking about ? Can you post an illustration of the hypothetical finished function (drawing ) ?

I imagine you're actually going to need 3 motors, (one plain vanilla DC motor for the fan + 2 steppers for locomotion). By having one motor on each rear wheel you could use their respective movements to do the steering. (having a simple caster arrangement for the front). Assuming a total mass of 1Kg you'd need 10 Newtons of force just to overcome gravity.

Stepper motors performance are usually quoted in Newton/metres so; Assuming the motors are fitted direct to the drive shafts with wheel diameter of about 5 cm you'd need to have motors rated at around 50 mNm (milliNewtonMetres)

Having a quick look on ebay I found these. http://www.ebay.co.uk/itm/Bipolar-Hybrid-Stepper-Motor-1-8-Degree-200-Step-NEMA14-Frame-Size-7-4V-0-3A-/271625594201?pt=UK_BOI_Electrical_Components_Supplies_ET&hash=item3f3e25a559

At first glance they look up to the job (only just, and assuming that the figures quoted are realistic) Even so if you can get the weight down, so much the better,It gives a hint of what's available out there.

Have you had a look at the old etch a sketch toy ?

You certainly don't want steppers for locomotion, far too heavy for any
reasonably torque (as would be any motor without gears.

You need reduction gears to get decent torque out of a small light motor,
but there are lots of good little gear motors on eBay like this:

http://www.ebay.co.uk/itm/Micro-Metal-Gearmotor-100-1-/220915201613?pt=LH_DefaultDomain_0&hash=item336f92764d

However unless you use a geared stepper you'll also need encoders to
measure position and used closed loop position control.

An accelerometer would be able to give orientation BTW, which would be
useful.

Some sort of LiPo battery pack to handle the BLDC suction motor would be
the heaviest item I suspect. Good attention to air leakage would allow using a smaller
suction fan - probably best to use a ducted fan (you have an inverse hovercraft!).

In another thread someone brought up brushless "gimbal" motors. Unlike typical hobby brushless motors that spin 1000s of RPM, these are designed to only rotate a few degrees per second with a decent amount of torque. But since they are brushless and require no gear reduction, they are relatively light weight. I believe they can rotate continuously.

Might be worth looking in to.

https://www.google.com/search?q="brushless+gimbal+motor"

https://www.google.com/search?q="brushless+gimbal+motor"

https://www.google.com/search?q=brushless+gimbal+motor+controller&biw=1113&bih=690&tbm=isch&tbo=u&source=univ&sa=X&ei=PPMyVMCeD8ngoATvrIGIDg&ved=0CFgQsAQ#facrc=_&imgdii=_&imgrc=GRXTvr9ETF04sM%253A%3Bls65qSbieFZw0M%3Bhttp%253A%252F%252Fwww.foxtechfpv.com%252Fimages%252F766TG8Y80.jpg%3Bhttp%253A%252F%252Fwww.foxtechfpv.com%252Fbrushless-gimbalmotorkit-brushless-gimbal-accessories-c-191_201.html%3B800%3B600

tylernt:
In another thread someone brought up brushless "gimbal" motors. Unlike typical hobby brushless motors that spin 1000s of RPM, these are designed to only rotate a few degrees per second with a decent amount of torque. But since they are brushless and require no gear reduction, they are relatively light weight. I believe they can rotate continuously.

Might be worth looking in to.

"brushless gimbal motor" - Google Search

To get decent torque out of a small motor you need reduction gearing, its the
laws of physics. Torque is roughly proportional to motor volume for standard designs.
[nice little exercise for the reader, or read a text on motors]
You can use superconductors, or liquid-cool windings to overcome this limitation,
but in practice reduction gears are a better bet.

If the finished item is going to be driving over a vertical surface and will, at times, have to drive towards the ground, he will have the problem that gravity will take over. Under these conditions a standard brushed motor would be a nightmare to control. Perhaps having to reverse, slightly to reduce forward motion. Hence, I still think steppers are the way to go. They are the only way he's going to get precise control over the varied angles relative to the horizontal.

It's never going to be a fast machine. Just precise and controllable.

MarkT:
To get decent torque out of a small motor you need reduction gearing, its the
laws of physics.
...
in practice reduction gears are a better bet.

I'm sure the OP can determine the torque needed for their application and calculate if a brushless gimbal or a traditional brushed+geared motor (or, for that matter, a stepper) has the better power-to-weight ratio. Or price. :wink:

Volume doesn't seem to be an over-riding factor here.

If we put the blackboard on the ground, the device would just be a car, by controlling the angle each wheel goes precisely, we can make the car go along the function by always facing the direction of the tangent line, and with a chalk fixed on the car, it could draw out the picture.

raschemmel:

If we put the blackboard on the ground, the device would just be a car, by controlling the angle each wheel goes precisely, we can make the car go along the function by always facing the direction of the tangent line, and with a chalk fixed on the car, it could draw out the picture.

I gather that's for testing. Otherwise he wouldn't need the vaccum fan idea to get it to hang on.

It's not going to work in the vertical position because in order for the car to stay in place and not slide down immediately, the vacuum has to be strong enough to prevent movement, thereby making it impossible to move from that spot. If the vacuum is decreased enough to permit movement, the car will immediately slide to the floor. If short, the design concept is flawed and the
project will never work vertically. Placing the blackboard in the horizontal position , eliminates the previous problem but doesn't necessarily make the concept doable. In order for the chalk to draw, it has to press against the blackboard. It is easy to underestimate the amount of force a human exerts on a piece of chalk.. The project is plagued with numerous issues, not the least of which is the software. If this has been done before, please post a link so we can all see it working.

raschemmel:
in order for the car to stay in place and not slide down immediately, the vacuum has to be strong enough to prevent movement, thereby making it impossible to move from that spot. If the vacuum is decreased enough to permit movement, the car will immediately slide to the floor.

Surely there's a window of opportunity where the force of suction is, say, 1.1G but the torque of the wheel traction motors can transmit 1.2G of force to allow movement?

Inefficient as heck, yes, but possible...

If you are still referring to the vertical model, no there isn't because gravity will win.
If you are referring to the horizontal model, there is some hope of success, but it will be painful and fraught with trials and tribulations...definitely not a walk in the park. I daresare the end result will look like a toddler's doodling.

raschemmel:
If you are still referring to the vertical model, no there isn't because gravity will win.

I guess these guys faked their robot. Including a video of it in motion.

http://www.robotic.diees.unict.it/robots/alicia3/alicia3.htm

:roll_eyes: