Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Choosing a Robot Drive Motor

Status
Not open for further replies.

cosmonavt

Member
When choosing a drive motor of a robot, which is the main thing to consider, the RPM or the Torque? Since RPM is reduced to a great extent when the load of the robot is applied over it, shouldn't the main thing be torque? The torque would be unchanged if the load is applied. And the acceleration of the robot depends on the force applied by the motor, which in turn is proportional to the torque.

If there are two motors with identical torque but different voltage and no load RPM, wouldn't the drive speed of the robot be the same if either of them is installed?
 
You need to consider both. The speed/torque characteristics are the most important thing to consider. I don't have much time to write now, but I can try to come back and give a briefing on that.


If there are two motors with identical torque but different voltage and no load RPM, wouldn't the drive speed of the robot be the same if either of them is installed?
No, and neither would the acceleration.
 
At constant speed, the motor torque has to equal the friction forces in the drive system. Those friction forces are practically constant and do not depend on the speed. You only need more torque to accelerate.

Maximum speed of a motor is limited by mechanical components and maximum allowed voltage across the motor. The motor speed (steady state) is directly proportional to the voltage applied to the motor. Different motors have different voltage/speed curves. When you select a motor, the maximum speed that occurs on the motor shaft should be below the maximum permissible speed of the motor.

Torque generated by the motor is practically directly proportional to the current through the motor. Different motors have different current/torque curves. The rated torque of a motor is derived from the rated current, which is selected to equal the maximum continuous current of the motor, i.e. the maximum permissible current from a thermal point of view.

The motor must be able to produce the maximum torque for the application. As a rule of thumb, during startup operation, the stall torque of the motor at the rated voltage should not be exceeded.

In order to make a good motor selection you must know:
- The maximum speed needed
- Friction forces in the drive system (gears etc.)
- Max. acceleration needed
- Maximum load (mass inertia of motor and load)
- Efficiency of the drive system (gears etc.)
 
Last edited:
If using **broken link removed**, won't the robot be very slow, since the no load RPM is only 60 and after adding load, it would be further reduced. Means less than 1 rotation per second? And yes, aren't the motors generally priced for their RPM?
 
If using **broken link removed**, won't the robot be very slow, since the no load RPM is only 60 and after adding load, it would be further reduced. Means less than 1 rotation per second? And yes, aren't the motors generally priced for their RPM?

If it can produce enough torque to overcome all the friction forces, it will be able to spin at the rated 60 RPM.. the robot speed depends on the wheel diameter. The motor is quite small, so it probably wont get a heavy robot moving. Difficult to say without the torque specs of the motor.

Edit: I found better specs for the motor: https://www.electro-tech-online.com/custompdfs/2011/12/2011414_83431_9296.pdf
 
Last edited:
Nice. I think these would be the specs of my robot:

Mass: around 1 - 1.5 KG
Drive type: Differential drive (two motors)
plus four wheels on each corner as a support (the body of the robot won't be hanging lose on the ground as seen in many differential drive robots). This would reduce friction too.
Speed: around 1 meter per second.

If I make the wheels around 10 cm in diameter and have a motor of torque say 1.41 Kg.cm = 14.1 N.cm.

Then the force exerted by the wheel will be 14.1/5 (radius of wheel) = 2.82 N right? There are two wheels so total force = 5.64N. The friction coefficient of wood against metal is 0.4 approx. The force (resistive) will be F = mu R = 0.4 * 15N (weight of robot) = 6N.

This means that I am very close. I just need to reduce the mass a little right?

I think I'm getting close......:) And I have taken the maximum weight in the calculations. It will be less than 15N.

If the initial force is less than the resistive forces, then the robot wont budge at all right?
And if the robot starts to move, then after a while the wheels will be spinning at max RPM right?
 
Last edited:
Actually, if I calculate the continuous power needed to keep your robot moving just by taking account the rolling friction:

0.4*1kg*9.81 ~ 4N

For a speed of 1 m/s it would take 1m/s * 4N = 4 Watts of continuous power to keep the robot moving.. and the motor you are thinking of is rated at 3.3 Watts. With two motors you would get 6.6 Watts of continuous power.

1) the rolling friction is the smallest component in the total friction forces. If you stay on hard flat surfaces with the robot, and use good ball bearings for supporting the wheels, the motors could do the job.
2) With 10cm wheel, you would get max speed of only 0.3 m/s. That speed takes 1.3 Watts of continuous power.
3) I don't know if the tables in the motor specs allready take into account the inefficiency and friction of the gearhead yms.. If they don't, that is another force eating your power. Gearhead can eat easily half the power in small motors.

I would go for bigger motors. Something like this: https://www.robot-electronics.co.uk/htm/emg30.htm

The motor you are considering is quite cheap, so I think there is not much risk trying it out. The motor I suggested cost 4 times more, because it comes with sensors attached.
If you wan't to find a cheaper, but bigger motor. Look at the specs of the motor I suggested for advice. I have used those motors in a 2 Kg robot which had 10cm wheels. The motors performed well, but they didn't have any extra power.
 
Last edited:
Actually, if I calculate the continuous power needed to keep your robot moving just by taking account the rolling friction:

0.4*1kg*9.81 ~ 4N

LOL its interesting how taking g to be 10 instead of 9.8 can make a difference.

For a speed of 1 m/s it would take 1m/s * 4N = 4 Watts of continuous power to keep the robot moving.. and the motor you are thinking of is rated at 3.3 Watts. With two motors you would get 6.6 Watts of continuous power.
Are you talking about **broken link removed**? It was rated 0.16 A and 12 V so that makes 0.16*12 = 1.92WATT. They didn't mention the power rating anywhere. I just wrote to them to ask the current rating and they said 0.16. That can be a problem

1) the rolling friction is the smallest component in the total friction forces. If you stay on hard flat surfaces with the robot, and use good ball bearings for supporting the wheels, the motors could do the job.
Means that the friction coefficient would be way less than 0.4 for a wheel?

2) With 10cm wheel, you would get max speed of only 0.3 m/s. That speed takes 1.3 Watts of continuous power.
I could always increase the dia of the wheel, given that the torque doesn't fall below the required value.

3) I don't know if the tables in the motor specs allready take into account the inefficiency and friction of the gearhead yms.. If they don't, that is another force eating your power. Gearhead can eat easily half the power in small motors.

I would go for bigger motors. Something like this: https://www.robot-electronics.co.uk/htm/emg30.htm

The motor you are considering is quite cheap, so I think there is not much risk trying it out. The motor I suggested cost 4 times more, because it comes with sensors attached.
If you wan't to find a cheaper, but bigger motor. Look at the specs of the motor I suggested for advice. I have used those motors in a 2 Kg robot which had 10cm wheels. The motors performed well, but they didn't have any extra power.

I have to stay within the budget for this thing so no.
 
Are you talking about **broken link removed**? It was rated 0.16 A and 12 V so that makes 0.16*12 = 1.92WATT. They didn't mention the power rating anywhere. I just wrote to them to ask the current rating and they said 0.16. That can be a problem

One other document said 0.227 A. I don't know witch one is correct.

Means that the friction coefficient would be way less than 0.4 for a wheel?
The frictions could be anything. Quite useless to be guessing actually.

I could always increase the dia of the wheel, given that the torque doesn't fall below the required value.
Yes, but I think the motor you are planning to buy is too weak already for 10cm wheel

I have to stay within the budget for this thing so no.

You are not planning to use any encoders? Or do you already have encoders?

anyway.. they are not bad motors. If your robot becomes too heavy, you have to figure out how to make them work. And I believe you can overdrive the motors slightly.
 
Last edited:
Encoder you mean driver? Hell yes, I am using H bridge. Can I really throw 0.227 A at a motor rated 0.16 A? Would it really add to the torque? Would it come up to 1.4 or something? Damn the high torque, high RPM motors are expensive. I don't get, there are some Chinese toys priced around 23$ which contain powerful motors. e.g. RC Helicopters. Why in heavens name do motors separately cost around 50-60$? That's way too much.
 
By encoder I mean a sensor. Some kind of sensor that tells you the speed or travel distance of the motor. That is a pretty useful component in a moving robot.

Rated current is what you can put through the motor continuously. You can easily put twice that for a few seconds. Torque will double if you double the current.

The motor I suggested costs around 30 USD and they come with sensors. The sensor is probably more expensive than the motor itself so 30$ for both is cheap.
 
Last edited:
I've bought the sensor. But nice find man! 170RPM + 0.56A + 1.5kgcm torque in just 30$ thats awesome. Gives me some hope that I'll find the motor I need for cheap.
 
One question. Why is torque shown in kgcm and not Ncm?

I don't know.. mNm or Ncm are usually used for small motors, but some like to be different :)
 
How goes this motor: **broken link removed** :p

Also, the same seller is selling this motor: **broken link removed**. The details say that it has a torque of 90 N*cm. How is that possible?
 
How goes this motor: **broken link removed** :p

Also, the same seller is selling this motor: **broken link removed**. The details say that it has a torque of 90 N*cm. How is that possible?

Difficult to say. The specs are so limited, and 100RPM 90Ncm really sounds suspicious. If that is true it might be a good choice.

High torques are possible even with small motors. High torque is achieved with high gear ratio. That is why 100RPM and 90Ncm sounds suspicious. 10RPM and 90Ncm would be more realistic.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top