Camera Remote Servo Based Tilt / Pan Head - Any experience? Suggestions?

Paul F said:
I don't think there is a solution with PWM and gearmotors. The issue is that they don't start moving until about a 15% pulse. There's nothing to be done about it.

I took a servo lens apart to see how they do it. Servo lenses have excellent range of control. The don't used PWM. They drive the motor with straight DC and vary the voltage. Contrary to the name servo-lens, they don't use servo motors, they are simple DC motors.

I don't know of a joystick motor controller that outputs a variable DC voltage. The ones I know of all output PWM.
Click to expand...

Servo motors don't use PWM either, they use PCM. That said, I don't see how a Servo Lens could do it with just a DC motor, there is no way it would know its current position.

Anyways, the best solution would probably be to go belt driven from a servo, not a gear motor (which is a DC motor w/ a gearbox). With a rubber belt, you will have a small amount of elasticity which should absorb the initial jerk. Just like the rubber band/tripod trick for smooth panning.
 
Paul F said:
Contrary to the name servo-lens, they don't use servo motors, they are simple DC motors.
Click to expand...

There is no such thing as a "servo motor". There is such a thing as a servomechanism. A servomechanism is a device that uses error sensing feedback (aka negative feedback) to correct the performance of the mechanism. Performance can be all kinds of things, from position to velocity to acceleration, to heat transfer, to SPL, to whatever. That definition doesn't specify what kind of device, nor what kind of a sensor is used. It can be just about anything, and it can be completely analog without any digital components at all. But it's darned difficult to make a servomechanism without any analog components.

You can build a servomechanism out of all kinds of things, from hydraulic cylinders to ribbon tweeters. The key characteristic is a feedback control system. You can make them be extremely accurate, extremely precise (and no, those are not the same things), or just minimally accurate and precise as in the case of your hobbyist "servo motor" which is typically a (really cheap) DC motor with a potentiometer built in, and a feedback controller in a single chip.

The best servo system I've ever dealt with used an eight inch 3000psi hydraulic cylinder, paired with an LVDT to measure it's position (there were five cylinders like this in the machine). Accurate within very small, and precision beyond measuring. And it cost like $400K back in the 1980s and ran part of a multi-million dollar machine tool. Man, I used to do the coolest engineering!

Back to the servo zoom lens, a DC motor is a good choice -- a DC motor exhibits its maximum torque at stall. It would be fairly easy to make one with a very predictable and controllable step-off, and an easily controllable ramp up/down. As it clearly has been done in thousands of servo lenses. But... there's a reason they cost more than your standard DSLR zoom lens. Smooth motion always costs more than a "bang-bang" control system does. Just sayin'.
 
atarijedi said:
... That said, I don't see how a Servo Lens could do it with just a DC motor, there is no way it would know its current position....
Click to expand...

The ENG lenses I've taken apart have no position sensing, tach, or any other servo function in them. They don't need to because they don't have any re-positioning function. But they are still commonly referred to as a 'servo' lens. But the big box cameras I've opened up do have a servo mechanism (thank you Bruce). So the terms are a bit abused.
 
Last edited:
I own the kessler head and notice it is pretty noisy. I'm wondering if their are heads available that are quiet, or is that how they all operate?
 
5 hydraulic cylinders controlling a machine tool. I remember seeing an article about that but forget what it was called. Hexapod? Anyway my experiance with servo motors is all CNC machine tool related. Cheaper DC motors typically use a tachometer for velocity control and a built in encoder for position loop. They can be programed to move at a creep speed to full output RPM, typ 3000RPM. Accel and decel can be programmed as well. Very good performance for the cost. Later machines used AC motors, typ 3 phase with phase monitoring feedback sensors for velocity control and again encoders for position loop. They can acheive very high performance out of a compact package. On the DC motors the noise is largely determined by the frequency of the pulses switching back and forth. As better drive devices came out the frequency went up till they were out of the audio range. Still there, you just can't here it. AC motors are typ very quiet.
I think the problems that you are dealing with is just very low tech motors and drives. Stepper motors disappeared decades ago in any serious machine tool. I remember them as being very noisey also.
Bill Wojo
 
You must log in or register to reply here.
Back
Top