kinetix 6000 question

If a single servo motor raised and lowered some component that would fall freely by gravity, how does the Kinetix drive determine (using motor encoders and whatever else it uses) that it needs to apply some sort of force in the opposite direction of the desired move (MAM) in order to control the rate of descent?

Can this be adjusted in the axis properties?

If I remember right, during tuning you can select that it is a vertical axis. Then during tuning, it will determine a torque offset

Thank you... I intentionally simplified the setup because I would like to know how a "braking" effect is accomplished using the encoder.

It seems to me that there would have to be a period of time elapse before the drive would "know" that it needed to exert a force in opposition to the direction of the requested move.

I would like to know if there is some setting that determines a period of time like that... I am at home now, but there was some setting in the axis properties that sounded like it may be what I am trying to talk about without knowing what I am talking about.. it was set at .25 sec.

If it was, then it seems to me that four times a second the drive would check to see if the encoders indicated that the axis was where it was supposed to be.

When the servo is enabled, it will maintain position continually.
Of course, when the servo is not enabled anymore, a vertical axis may drop. If it can drop but it should not, a brake on the motor should be used. For a servo with a brake, there are two values that can be set to control when the brake activates and de-activates

realolman said:

Thank you... I intentionally simplified the setup because I would like to know how a "braking" effect is accomplished using the encoder.

It seems to me that there would have to be a period of time elapse before the drive would "know" that it needed to exert a force in opposition to the direction of the requested move.

I would like to know if there is some setting that determines a period of time like that... I am at home now, but there was some setting in the axis properties that sounded like it may be what I am trying to talk about without knowing what I am talking about.. it was set at .25 sec.

If it was, then it seems to me that four times a second the drive would check to see if the encoders indicated that the axis was where it was supposed to be.

Click to expand...

The servo has closed the loop on position - it will apply as much force as required to hold that position (up to the limits of the servo)
when moving the servo has continuously changing position reference that it is following (how closely depends on tuning) - this means that for a hoist when it is lowering it uses less torque and the axis lowers.
Conversely for raising More torque and the axis raises (assuming the the up direction is Positive)

The Torque loop in a kinetix is running at sub ms rates (Specification is a bandwidth of 1300Hz depending on tuning)
the velocity loop 500 Hz and the position loop is similar
Adding this all together causes the servo to "monitor and correct" the position of the servo every ms or so - this correction is what causes the servo to know what torque is required to hold position.

I think that the 0.25 sec that you quoted is for the "human update rate" and averaging of the velocity and acceleration rates NOT the control rate

MichaelG said:

The servo has closed the loop on position - it will apply as much force as required to hold that position (up to the limits of the servo)
when moving the servo has continuously changing position reference that it is following (how closely depends on tuning) - this means that for a hoist when it is lowering it uses less torque and the axis lowers.
Conversely for raising More torque and the axis raises (assuming the the up direction is Positive)

The Torque loop in a kinetix is running at sub ms rates (Specification is a bandwidth of 1300Hz depending on tuning)
the velocity loop 500 Hz and the position loop is similar
Adding this all together causes the servo to "monitor and correct" the position of the servo every ms or so - this correction is what causes the servo to know what torque is required to hold position.

I think that the 0.25 sec that you quoted is for the "human update rate" and averaging of the velocity and acceleration rates NOT the control rate

Click to expand...

Hello... it's me again... thank you very much for your information.... IIRC I think the .25 had something to do with a velocity averaging but I'm not sure.

So what you posted here sounds like the monitoring and correcting of the position and torque is not causing my problem as I thought it might be if something was not set correctly. hmmm ...durn

I was on a chat line with a guy from Rockwell who put me in the direction of the gains. I have experimented with that a bit, but not much.. I thought I would try on Monday.

The gains were displayed to about 4 or 5 places beyond the decimal point so I'm thinking that this was not arrived at manually. When I first tried to modify anything, a message box appeared complaining about, I believe, braking torque ...supposed to be set between 0 and 233 and it was set at 283 ... when I finally figured out how to change that, it quit squawking about it and allowed me to change other things

I have been looking in the manual on line and it talks about tuning and it says to remove the load. And then it says to do it with the load... It seemed to me to leave some pretty big gaps in explanation there, and I'm not sure what's going to happen if I try to tune it.

It seems to me this problem we are having may have something to do with the fact that we modified the pick head and it is considerably lighter than it used to be. Thing is we did it to 3 machine and are only having the problem with one.

Is there something that explains the tuning procedure better?

thank you