Z axis runaway

[camaro_dan67]

The encoder connector on the motor has 17 pins.

 

[memphisjed]

If your machine has a dro on the monitor you can tell if it is motor or encoder. Look where the machine thinks it is at before reset, then watch the z value. If it is going down when the z goes down the encoder is sending correct signal / direction, the motor is spinning backwards throwing itself into crash - servos normally follow error. Reversing motor might be as simple as switching a setting in the controllers parameters.
Sounds like a backwards motor wire on description.

 

[cngbrick]

The pinout for the 17 pin connector should be as follows (please confirm by pulling your connector and checking if this makes sense before applying voltage):

Pin--> Signal

A--> A
B--> B
C--> +5V
D--> /A
E--> /B
F--> Z
G--> /Z
H--> Shield (?)
J--> +5 V
K--> +5 V
L-->
M-->
N--> Ground
P--> Ground
R--> Over Temp
S--> Over Temp
T--> Ground

Supply power to the +5 V pins and grounds (one of each is probably fine as the pins on my connector were just connected together).

The quadrature signals consists of A, /A, B, /B, Z, /Z. "/A" being the complementary (opposite polarity) of "A" and so on. Put a scope on A and B to verify that they produce a clean square wave with 90 degree phase difference (B changes state in the middle of a peak on A). Check that the changes of state are "clean", i.e. no "bounce" from high to low before settling into a new state. Then compare A to /A, B to /B and Z to /Z and verify that the changes of state occur at the same time and are clean. Z will only have one pulse per rev and is not likely responsible for runaway behavior.

RT

 

[camaro_dan67]

Can I check signals with power on and the estop in if I unbolt the motor and lay it over on its side?

 

[atex57]

Can I check signals with power on and the estop in if I unbolt the motor and lay it over on its side?

That is the best as you can test everything, even with power applied. Strap it down so it can't chase you.

Ed.

 

[atex57]

Can I check signals with power on and the estop in if I unbolt the motor and lay it over on its side?

That is the best as you can test everything, even with power applied. Strap it down so it can't chase you.

Ed.

 

[camaro_dan67]

Ok here is the latest. Tonight I took the motor off and looked at it carefully. I also found an older picture I had of the machine. I noticed in the picture the nameplate on the stator was in the front of the machine and the cables went twards the back of the machine.
What I actually had was the nameplate to the side of the machine and the cables out the back. I straightened this out and aligned my scribe marks as close as possible. I put the motor back on and the servos came on with no problem. I held the qp and cancel buttons down and homed the machine. Everything seemed to function normal but when I move the z up or down at different spots in the travel the z gets a buzzing sound. If I move the z up or down an unspecified amount it would stop buzzing. I could handle jog or rapid to different z locations sometimes it would buzz sometimes not. What is causing this? Do I need to try tweaking the parts of the motor more? Is my motor shot? It did have oil in it the first time I took it off.

 

[cngbrick]

Ok here is the latest. Tonight I took the motor off and looked at it carefully. I also found an older picture I had of the machine. I noticed in the picture the nameplate on the stator was in the front of the machine and the cables went twards the back of the machine.
What I actually had was the nameplate to the side of the machine and the cables out the back. I straightened this out and aligned my scribe marks as close as possible. I put the motor back on and the servos came on with no problem. I held the qp and cancel buttons down and homed the machine. Everything seemed to function normal but when I move the z up or down at different spots in the travel the z gets a buzzing sound. If I move the z up or down an unspecified amount it would stop buzzing. I could handle jog or rapid to different z locations sometimes it would buzz sometimes not. What is causing this? Do I need to try tweaking the parts of the motor more? Is my motor shot? It did have oil in it the first time I took it off.

Is the buzzing actually motor oscillation? Is the encoder to motor shaft connection tight? Encoder mounting screws tight? Can you put a scope on the encoder output to see what's happening when stable vs buzzing?

You may also need to time the magnets to the commutation brushes after having reassembled the motor as you may end up with a difference in rpm depending on rotation direction.

RT

 

[camaro_dan67]

How do I time the magnets to the brushes.

 

[cngbrick]

How do I time the magnets to the brushes.

Disclaimer: I haven't done this on a Fanuc motor so I don't know if adjustment is possible on these motors.

You need a way to measure the motor RPM with the motor disconnected from any mechanical load. If you have a scope then you can just probe one of the encoder outputs. If you don't have a scope, some multimeters have a frequency counter mode that you could use to monitor the encoder instead.

Once you are set up to monitor RPM, apply a voltage to the motor and note the frequency/rpm of the motor. Apply the same voltage in reverse polarity and compare the frequency/rpm to the first reading. If its really close then you are ok. If not, loosen the motor screws and try to shift the brushes or magnet a bit (note the direction of the shift). Repeat the test to see if the situation improved. You can try different voltages as well to see if performance is consistent.

RT

 

[atex57]

Buzzing but responding without running away could be tuning at the amp.
Check your brush timing as above, hopefully that cures it.

Ed.

 

[camaro_dan67]

I want to say happy holidays to everyone and a huge thanks to all the responses. I have a used servo motor on the way that I am going to try before I try any tuning or motor timing. I will keep all of you posted as to how the replacement motor works out. One other thing I was wondering after reading posts about buzzing servos is people were saying that excessive backlash can cause hunting and buzzing. Is it possible that something in the ballscrew assembly loosened up after I started the servo those few times and it slammed the head into the prop board. I want to try the other motor but if the buzzing continues its back to troubleshooting.

 

[Vancbiker]

A buzzing servo can be a result of lack of damping by the preload of the ballscrew. If your screw/nut fit has been loosened by the crashes it may be the cause.

On Fanuc DC servos the encoder and motor are coupled with a version of an oldham coupler. Lash in the coupler fit is a common cause of buzzing.

Gain adjustment on the drive could be used to reduce buzzing, but not advised for many applications. Reducing gain increases lag. On a contouring application, lag must be pretty closely matched between all axes else inaccuracy of contour will result.

 

[camaro_dan67]

I received the used servo motor and installed it. The z axis movement seems very smooth and no buzzing at any positions. I didnt check any movements for positional accuracy yet but the newest motor seems to have fixed the problem. The only thing left is getting tool changer programs in the control. There is no automatic operation of the tool changer right now.