read servo-moteur position microcontroller

[FvM]

The post explains the method.

It gives some hints, at best. But I think it's almost obvious. The only way to control a servo is to have a controller generating a pulse width. The controller uses the feedback potentiometer of the "master" servo as input.

But I don't see the other thread related to the present discussion.

 

[jpanhalt]

But I don't see the other thread related to the present discussion.

I think the method described in the link I provided goes directly to this question of the originator of this thread:

i know that the servo is controlled by a PWM from the MCU, what i want is to block the position of the servo if someone try manually to do this, so i need to read the position of the servo in order to block the servo in the position in order to fix the servo in that position

---------- Post added at 16:55 ---------- Previous post was at 16:51 ----------

i need just to read the position and save it and after i will send this position to the servo

Aside from hacking the servo, that is the only way I have read where someone has "read" the position of a model servo -- not the pwm pulse driving it.

John

 

[FvM]

Perhaps I didn't get it, can you please explain in a few words how you think the servo position is read in the video?

Do you mean, it is done by monitoring the current consumption?

P.S.: Perhaps you understood the intention of the original poster better than I did. If so, it has been stated very unclear. But I agree, that it should be feasible to make the servo controller track a manual override action by sensing the supply current.

 

[jpanhalt]

This seems clear to me: https://www.edaboard.com/threads/215058/#post911153

I have not actually done it, but like you and several others here, I was not aware of it until bigdogguru pointed it out.

John

 

[FvM]

Yes, but I think the method is limited in so far as it can't exactly detect the position of a servo, only try to get out of the way of an external force. To detect it's direction, it needs to make short moves to both sides. I would prefer a simple potentiometer (a modified servo) in most cases.

But thanks for insisting.

 

[jpanhalt]

Yes, but I think the method is limited in so far as it can't exactly detect the position of a servo, only try to get out of the way of an external force. To detect it's direction, it needs to make short moves to both sides. I would prefer a simple potentiometer (a modified servo) in most cases.

But thanks for insisting.

I agree. That is what I would probably do too, particularly with the Futaba 3003, which is huge in dimensions by current standards.

John