Forum Navigation
You need to log in to create posts and topics.

No position feedback to Masso?

12

Hello All,

I have been following the forum for a long time, but this is my first post.

We have previously built some hobby CNC plasma pipe cutting machines (4 axis, x,y,z and rotary). Now, we are building a commercial machine. We plan to test Masso controller. Upon success, Masso looks like the first candidate for our machine.

Our concern about Masso is the speed control ability. As far as I see there is no encoder input from axis drivers, https://masso.com.au/masso-documentation/?section=section-title-1.  Sample servo connections in the documentation conforms that issue, only step and dir connections.

To best of our knowledge, without position feedback from servo encoder of each axis, speed control is not possible. Because, interpolation of the motion is not possible without real time position data. We will use servo motors. In this case, servos are used like closed loop steppers. The position control is ensured by communication between driver and the motor. But the position info on the controller is not real time it is just calculated from sent pulse signal. In our design the X axis is 8000 mm long, this even increases our concern.

My understanding is summed above. Any explanation or advice is appreciated.

Fred

Hi @fred

It sounds to me like the question is more about the mechanical design of the equipment then the capability of the Masso control. I'm thinking that the Masso calculates the interpolated moves and sends out the appropriate pulse and signals to execute the interpolated move. It is implied that the servo goes where it's told at the appropriate speed in relation to the other servos in the system.

System resolution can be addressed through mechanical design. Feed speed and rapid speeds can be addressed at the mechanical design stage also. The following error on each individual drive can be set to reflect the result you are looking for. And of course feeds rates, rapid rates, and positioning can be controlled with g code.

If you are able to design a system that mechanically can do what you are trying to do then the Masso control will be able to do what you need.

Guy

@fred thanks for your message

closed and open-loop topic does get a bit confusing. In a closed servo based system, the encoder feedback goes to the servo drive because without it it will keep spinning and not hold any position. Controllers calculate and provide the movement data to motor drives (these can be open or closed loop) and the drive positions the motor shaft, so things like feedrate and position are all calculated on the controller and does not make any difference in the machine running if its a closed or open loop system.

In the closed-loop system, there is one benefit that if for some reason the motor could not execute the requested position then the encoder feedback from motor shaft will issue an alarm on the drive when can then send an alarm signal to the MASSO and stop motion.

 

 

Quote from evermech on July 31, 2019, 1:24 pm

Hi @fred

It sounds to me like the question is more about the mechanical design of the equipment then the capability of the Masso control. I'm thinking that the Masso calculates the interpolated moves and sends out the appropriate pulse and signals to execute the interpolated move. It is implied that the servo goes where it's told at the appropriate speed in relation to the other servos in the system.

System resolution can be addressed through mechanical design. Feed speed and rapid speeds can be addressed at the mechanical design stage also. The following error on each individual drive can be set to reflect the result you are looking for. And of course feeds rates, rapid rates, and positioning can be controlled with g code.

If you are able to design a system that mechanically can do what you are trying to do then the Masso control will be able to do what you need.

Guy

Thank you Guy,

In fact the resolution of the drives will be adjusted to maximum pulse generating capability of the Masso. This means we have plenty of room there.

For the mechanical design, torque and power requirements of the each axis is calculated carefully. Yes, the communication between servo motor and the driver promises no position lose. But this may not be enough for speed control, or momentary position control. I am pretty sure that high end industrial machines are designed considering mechanical issues. Nevertheless, they have encoder feedback to controller.

Quote from MASSO Support on July 31, 2019, 8:31 pm

@fred thanks for your message

closed and open-loop topic does get a bit confusing. In a closed servo based system, the encoder feedback goes to the servo drive because without it it will keep spinning and not hold any position. Controllers calculate and provide the movement data to motor drives (these can be open or closed loop) and the drive positions the motor shaft, so things like feedrate and position are all calculated on the controller and does not make any difference in the machine running if its a closed or open loop system.

In the closed-loop system, there is one benefit that if for some reason the motor could not execute the requested position then the encoder feedback from motor shaft will issue an alarm on the drive when can then send an alarm signal to the MASSO and stop motion.

 

 

Thank you Masso,

Yes. I am not an expert in electronics. But, to best of my knowledge this is the point servo motors differ from stepper motors. The stepper motors rotates as much as sent pulse count. However, a servo keeps spinning until the pulse signal from the encoder match the sent signal. This means that theoretically there is always, even a very small amount, a deviation from actual sent signal.

In our previous projects, we have used some arm based Chinese controllers. Even they had pins for encoder feedback (Phase A,B and Z). But I know that this feature do not make them very precise controllers.

Hi @fred

when you are saying speed control do you mean the actual speed that your axis is traveling, or do you mean speed control setting in your servo drive? You should be able to control axis speeds through g code. I even use it to change the speed of my X axis motion half way through a cut because in my case the load lessens as the axis moves.

speed control mode on the servo drive requires PWM signal from motion controller, and I'm nearly certain that the Masso does not provide that type of signal. I do believe Masso  provides step ( pulse ) and direction signals only.

Guy

Thank you @evermech

I am talking about speed control mode between servo drive and the controller. As far as I know, step signals are not enough for speed control. Encoder feedback needs to be evaluated continuously to adjust the amount of sent pulse signal, this is un-linear control.

Let me suggest an example. Lets assume that we are drawing a line using X and Y axes. Thanks to servo algorithms, the start and end points will be very accurate without encoder feedback to controller (if mechanics and servos are ok). But, due to some reasons (mechanical or other), there can be momentary speed variations in the axes. In this case this motion is not a perfect line. I hope it is clear.

As I am not an expert in electronics, I am not sure how this variation effects the accuracy.

Fred

"Edited"

 

Hi @fred

so I'm thinking that you are considering brushed dc servos for your project?

I don't know too much about PWM and speed control in the servo drive but I have a milling machine that is aging that is controlled that way. I am considering putting a Masso controller on but I would need to replace all four servo drives also and maybe even the servos ( I'm not sure if I can drive a brushed dc servo with step and direction )

i am wondering if there is some sort of interface between pulse signal, change to PWM signal available? Have you heard of anything like that?

im wondering what your hesitation about pulse signal system is based on. There are many machines using the same sort of control environment as the Masso that are very accurate. You should ask yourself just how accurate do you need to be. One thousandth of an inch repeatability should be no problem, providing you have a good mechanical design.

Regards. Guy

Hi @evermech

You are wright, PWM and step/dir signals are not the same. I edited my previous post. Our servos are step/dir controlled industrial AC Servos.

I am not sure about it. But as far as can remember I have read something like PWM to step signal conversion in another forum.

My hesitation is based on closing the loop between the servo motor and the driver. I believe that closing the loop at the controller level ensures a more accurate interpolation. I made a research about this topic. I saw that with the modern servo motor/driver technology, closing the loop at the driver-motor level is able to yield satisfying results. Like you said, in this configuration, 0.01 mm precision looks possible with a good mechanical design and a capable controller. This precision is acceptable for our design.

@fred

after a quick Google search I couldn't find a pulse and step to PWM converter that's looks like it would be usable for an application like what we would be looking for. There is an opportunity for development of a device like that as far as I can see. Unfortunately that's not my field of expertise.

Other than that are there any other issues that you might see with the Masso control? Like you I am an Oem of custom equipment and am always looking for control solutions that match the project I am working on.

Check out this machine that I used the Masso controller on for a customer.

Guy

12