My Z axis only moves at high speed...

[Goemon]

I have a strange issue with my Z axis. When I set it to move at 2200mm / minute (or faster), it moves very well. When I try and make it move at lower speeds, it stutters or won't move at all, as if the load is too great for the motor.

Most of what I read indicates that this is the reverse of what should happen. I.e. The suggested fix for a struggling motor is to gradually reduce speed and acceleration until it stops struggling.

My Z axis is moving a 35lb spindle and I installed a 35lb gas spring from little machine shop to counter weight it. I have 20mm pitch ball screws on all axis. I thought that maybe the spring is not enough but it struggles to move in both directions (at low speed).

Does anyone know what sort of problem makes a stepper motor struggle at 800mm / minute but move easily at 2200mm / minute? I keep reading that steppers produce the most torque at lower speeds...

The motor is a Nema 34 3.5a 465 oz / in.

Similar Threads:

• Mechanism techniques used for a high speed z axis
• X-axis chatter at high speed travel
• 2-axis cut control - Speed/feed calculator RPM too high
• High Speed CMS FOAM 5-axis milling machine
• High-Speed 3 (4?) axis cnc router

[davida1234]

I don't know why your axis works on higher speeds and not on lower but I tried using gas springs on my machines for their constant spring rates and most of them failed rather quickly.

This is what I have found out about gas springs on the internet:

They are high friction devices and not meant for continuous operation. They are best for intermittent operations like hatches and doors where they can cool down between operations.

That still doesn't explain your problem. You have a rather large lead on your screw (20mm) and moderately sized steppers. Maybe the inertia of higher speeds helps to overcome the resistance of the ball nuts?

I would try a slower screw/ball nut or gear down if possible.

[PCW_MESA]

A couple of guesses:

1. You have a resonance issue, increasing the u-step ratio may help this
2. Marginal step timing or drive levels sometime cause this kind of behavior (try doubling the step pulse on time if that's possible)

[ger21]

Did you adjust the trimpots on the G540?

[cncmakers001]

I have a strange issue with my Z axis. When I set it to move at 2200mm / minute (or faster), it moves very well. When I try and make it move at lower speeds, it stutters or won't move at all, as if the load is too great for the motor.

Most of what I read indicates that this is the reverse of what should happen. I.e. The suggested fix for a struggling motor is to gradually reduce speed and acceleration until it stops struggling.

My Z axis is moving a 35lb spindle and I installed a 35lb gas spring from little machine shop to counter weight it. I have 20mm pitch ball screws on all axis. I thought that maybe the spring is not enough but it struggles to move in both directions (at low speed).

Does anyone know what sort of problem makes a stepper motor struggle at 800mm / minute but move easily at 2200mm / minute? I keep reading that steppers produce the most torque at lower speeds...

The motor is a Nema 34 3.5a 465 oz / in.

Not sure for the stepper motor, but if for AC servo motor, the matching AC servo driver might need to adjusted to tune with the motor.

[Goemon]

Did you adjust the trimpots on the G540?

No. I didn't touch anything inside the g540.

I have the same size motor on my X axis and it works fine so I assumed my issue was mechanical. Would adjusting the trimpots make a difference to how a motor performs at low speed?

If so, should I try adjusting them?

[Goemon]

A couple of guesses:

1. You have a resonance issue, increasing the u-step ratio may help this
2. Marginal step timing or drive levels sometime cause this kind of behavior (try doubling the step pulse on time if that's possible)

I don't recall seeing any options to adjust the U-step ratio or marginal step timing in Linuxcnc or on the Gecko g540. Where should I look?

[Goemon]

I don't know why your axis works on higher speeds and not on lower but I tried using gas springs on my machines for their constant spring rates and most of them failed rather quickly.

This is what I have found out about gas springs on the internet:

They are high friction devices and not meant for continuous operation. They are best for intermittent operations like hatches and doors where they can cool down between operations.

That still doesn't explain your problem. You have a rather large lead on your screw (20mm) and moderately sized steppers. Maybe the inertia of higher speeds helps to overcome the resistance of the ball nuts?

I would try a slower screw/ball nut or gear down if possible.

The gas spring was only meant to be a temporary solution as I happened to have one spare. Replacing the ball screw is a last resort as I'm using NSK linear actuators where everything is nice, neat and precise. I need to first figure out what the issue is before ordering new parts...

I can't see why it would move fine at higher speeds if the issue was related to under-sized motors for the z axis weight and ball screw lead. And, if that was the issue, I can't think why it would not be alleviated by a counter-weight.

[Ntl]

What drivers are you using?

Sorry I didn't see the g540. Have you tried swapping motors or drives for the z? That would be a good starting point to possibly narrow it down a bit.

I personally would disconnect the helper strut as well, just do one thing at a time though so you can isolate what the problem was when it goes away.

Dan

[PCW_MESA]

I don't recall seeing any options to adjust the U-step ratio or marginal step timing in Linuxcnc or on the Gecko g540. Where should I look?

If your drive is a G540, you cannot change the u-step ratio, its fixed at 10

I would set the step time to 5 usec, If this is LinuxCNC the step time would be set in the ini file
I have see very similar behavior with G540s and 3.3V signals when using the default 1 usec
minimum step times

for example (in your .ini file)
[AXIS_2] # normally Z
...
...
...
change
STEPLEN = 1000
to
STEPLEN = 5000

[Goemon]

I found the issue. I am officially an idiot....

One of the 4 wires inside the db9 connector had been pulled out of the screw terminal. I had assumed (incorrectly) that the 4 wires were to cover forward and reverse power. I did a little reading on stepper motors and it turns out the wire configuration just sets the mode of operation. The things apparently work with just one of the poles connected but just not at all speeds...

Anyway, I reconnected the red wire and now it works perfectly at high and low speed. So far I tested it at various points between 300mm / minute up to 8000mm / minute and so far, no stalling.

[Ntl]

Glad it was something simple and it's up and running again.

[Goemon]

Glad it was something simple and it's up and running again.

Me too. Just another reminder of how much I have to learn....

I had a go of cutting some aluminum with it and seeing the chips fly made it all worth it.

[mactec54]

Me too. Just another reminder of how much I have to learn....

I had a go of cutting some aluminum with it and seeing the chips fly made it all worth it.

Something else that is a little unusual with what you have done is the 20mm pitch Ballscrew for the Z axes is a much higher pitch than the norm most don't go over 10mm pitch but I guess with the gas spring there is not much work for the motor to do so will be able to move supper fast in the Z axes

[Goemon]

Something else that is a little unusual with what you have done is the 20mm pitch Ballscrew for the Z axes is a much higher pitch than the norm most don't go over 10mm pitch but I guess with the gas spring there is not much work for the motor to do so will be able to move supper fast in the Z axes

Yeah speed is not going to be an issue with this machine. The Z axis seems to be able to move as fast as the others. I am impressed by how much torque and acceleration you get from those Nema 34 465 oz in motors. I have no doubt that they could handle a lot more.

I got mixed advice on choosing the ball screw pitch but I was under the Impression that using a slower pitch on one axis would limit the overall speed potential of the machine. I guess it doesn't matter now as all three of my 20mm pitch ball screws work well.

Now that I have done some test cuts, I can see that the limitations of the end mills and work holding make any other concerns about not having enough speed or torque irrelavent,

[skrubol]

Having the same speed on the axes applies more to the X and Y. As long as Z is at least half as fast as X and Y it won't impact things much.
Not a bad thing to have the max Z speed equal to X and Y though.

[Goemon]

Having the same speed on the axes applies more to the X and Y. As long as Z is at least half as fast as X and Y it won't impact things much.
Not a bad thing to have the max Z speed equal to X and Y though.

That's what I figured.

When I was doing my initial research, it sounded like "the more speed, the better". Now that it's built and I am making chips, I doubt I'll ever have any need to to use more than 1/4 of the available speed. The tool would break long before I ever ran up against the limits of the my motors or spindle. These little steppers can snap a 3/8" end mill like a twig if I don't listen to G wizard. G Wizard never suggests 9000mm / minute feeds...

I am glad I went with a fixed gantry design as it allows incredible stiffness while all the moving parts can be kept relatively light. It seems like a good way to get the most out of it when you only have enough in the budget for small steppers.

I had thought that I might regret buying a Gecko G540 due to it's 3.5amp / per motor limit but 3.5a seems to be plenty for aluminum milling and woodwork.