How do RC car/plane "servo motors" act as servos with only three leads and two w

[Beamin]

As far as I can tell  the little black box servos in rc stuff is just a way geared down motor with three leads NOT a servo or stepper motor. How do you stop it from spinning at a certain point , make a phase down the third wire that limits the turn when the signals cancel out?

[beanflying]

The standard for these goes way back pre microcontrollers and traditionally relied on a Potentiometer for position sensing. The servo was sent a pulse that varies between 1 and 2mS which was 'normally' set at 1.5mS for center and the servo interpreted based on the feedback from the Pot and circuit. If the Pulse width doesn't agree with the Pot position then correct the angle by supplying positive or negative voltage to a DC motor to change it.

Some modern servos now use magnetic sensors for position and some run Micros internally to set end points, speed or rotation direction and a whole lot more.

There is plenty of info on controlling an R/C servo with an Arduino for example so grab a cheapy from evilbay or hobbyking and have a play 

[Nerull]

I'm curious what you think a servo is and why hobby servos don't count.

It's a motor with position sensing and a closed loop control IC that drives the motor to the position received on the data wire. That's...what a servo is.

Servomotors and stepper motors are completely different things and often competing technologies.

[Beamin]

I'm curious what you think a servo is and why hobby servos don't count.

It's a motor with position sensing and a closed loop control IC that drives the motor to the position received on the data wire. That's...what a servo is.

Servomotors and stepper motors are completely different things and often competing technologies.

I remember opening one up as a kid and just finding a motor inside no IC or positional sensor. I'm talking the 15.00$ motor that adjusts the steering or advances the throttle on the engine. Real simple device. Plus the receiver is a totally analog device that uses higher/lower tones to control it. Very simple too simple it seems to have a IC sending data back.

I shouldn't have said stepper motor, but I did read somewhere that stated the RC car servo motors weren't even servo motors too simple. The third wire measures resistance or something, cant find a good explanation of how they work.

[beanflying]

Until very recently there was no data sent back. The servo is a contained box that works as I described above.

The early ones to reverse the direction of travel for example was done by flipping the motor wires and each side of the Pot as the radio Transmitters didn't have that functionality. This all predates IC's common use in consumer electronics.

[Beamin]

So the position is mapped to the pot in there: IE center is 50k, 0 is full left and 100k is full right. the controller sends pulses until the pot reads what it should. Is this why it some time twitches when you try to get a precise angle as the coarse is too coarse and in-between values and it twitches between the two pot values. 

[beanflying]

Correct it is was then a crude PID of sorts to fix a final position with the jitter being caused by load, dirty pots or sometimes average engineering so they could swing a bit from the position. So Positive Negative and Signal wires from the Receiver.

The Receiver was a serial decoder after the Receiver bit. It took the AM or later FM and then split off the channels based on a fixed length pulse train from the Transmitter assigning them to each servo.

These days servos and radios run digital and two way communication in some cases so some of this doesn't apply but all servos sold for R/C still conform to the basics of 1-2mS pulse control.

[Psi]

So the position is mapped to the pot in there: IE center is 50k, 0 is full left and 100k is full right. the controller sends pulses until the pot reads what it should. Is this why it some time twitches when you try to get a precise angle as the coarse is too coarse and in-between values and it twitches between the two pot values.

Almost correct, 'until' is not quite right.

The controller sends a pulse which signifies a position (within a range) and it keeps sending it. If the controller wants a new position it changes the pulse to that. It's a never ending series of pulses.

The servo electronics forms a feedback loop that reads the pulse position and decides if it needs to move the motor one way or the other so that the pot position matches the pulse position. It's always trying to 'home in' on a match where it's in the correct position and the motor stops.

[Nerull]

I'm curious what you think a servo is and why hobby servos don't count.

It's a motor with position sensing and a closed loop control IC that drives the motor to the position received on the data wire. That's...what a servo is.

Servomotors and stepper motors are completely different things and often competing technologies.

I remember opening one up as a kid and just finding a motor inside no IC or positional sensor. I'm talking the 15.00$ motor that adjusts the steering or advances the throttle on the engine. Real simple device. Plus the receiver is a totally analog device that uses higher/lower tones to control it. Very simple too simple it seems to have a IC sending data back.

I shouldn't have said stepper motor, but I did read somewhere that stated the RC car servo motors weren't even servo motors too simple. The third wire measures resistance or something, cant find a good explanation of how they work.

The control loop is contained inside the servo. The third wire is a typically data wire, carrying a position encoded in a pulse train. More modern servos can use digital single-wire communication busses.

The existence of a position feedback control loop makes it a servo. It doesn't matter how its implemented.

Steppers have no position feedback. This is why missing a step can can be so disruptive - the stepper has no idea where it is and when it isn't where the controller thinks it is.

[janoc]

Why don't you google up how RC servos work?

E.g. this video explains it very nicely, including some animations on how the feedback loop works:

[brucehoult]

I'm curious what you think a servo is and why hobby servos don't count.

It's a motor with position sensing and a closed loop control IC that drives the motor to the position received on the data wire. That's...what a servo is.

Servomotors and stepper motors are completely different things and often competing technologies.

I remember opening one up as a kid and just finding a motor inside no IC or positional sensor. I'm talking the 15.00$ motor that adjusts the steering or advances the throttle on the engine. Real simple device. Plus the receiver is a totally analog device that uses higher/lower tones to control it. Very simple too simple it seems to have a IC sending data back.

I shouldn't have said stepper motor, but I did read somewhere that stated the RC car servo motors weren't even servo motors too simple. The third wire measures resistance or something, cant find a good explanation of how they work.

No, that's exactly how even the $1.70 SG90 RC servos work. e.g.

https://www.amazon.com/Cenrkay-Micro-Helicopter-Aircraft-Control/dp/B07N1C8D2F

You send an analogue 1.0 to 2.0 ms pulse for the desired position on the 3rd wire at regular intervals. It doesn't really matter what the interval is, but a single pulse isn't enough to move the servo from one extreme to the other. I think 100 or 200 Hz is the recommended repetition speed for the pulse but I've found 20 Hz to be enough at least with these ones. I've used a bunch of them.

The slightly more expensive MG90S -- $2 instead of $1.70 -- uses metal gears instead of plastic ones and can produce (and take without stripping gears!) more torque.

Unless you have a pretty big model there's little point in paying $15 or $30 for a servo these days.

[rstofer]

The repetition rate for the pulses is around 20 ms or 50 Hz.  A 1 to 2 ms pulse every 20 ms.  If the pulse train fails, the servo will shut down electronically and there won't be any holding torque.

[brucehoult]

It depends how much torque you need to resist. These things are pretty hard to force to turn even turned off, due to the high gearing of the motor. But of course actively holding position is stronger.

[beanflying]

While this is heading off the topic this is why those of us playing with higher performance R/C spend way more than $30 a servo 

Proper high quality so called Digital Servos have very high holding torques partly due to the motors and also the electronics they will hold position for a period of time without a signal. The first of these I brought in the early 90's were about $150USD each and hand assembled in Germany. By comparison an equivalent sized servo now would have 4-5 times the power and cost under half of that. These are still nothing like the Hobbyking junk end of the market which I also use a bunch of in appropriate aircraft.

[brucehoult]

Exactly. It depends on the application.

Over on the stepper motor side, the 28byj costs around $2.50 WITH a driver board (in qty 5 on Amazon).

Those things are used for example to drive the louvers on hundreds of millions if not billions of air conditioning units, which they appear to do just fine.

[beanflying]

Just for a little visual of the guts of a fairly recent JR507 servo. From the geartrain side you can see the motor at one end and then the pin on the near side gear goes through the case to the pot on the board to sense direction on the output shaft.

I still have a couple of pre IC servos and if I can find them so I will add them here too.

EDIT No Servo found but here is probably my first Receiver with an actual 'IC' in it. I don't think I have any of the older ones left.