[johnt]

Please help me determine if my plasma table is jerky because of the mechanical design or motor tuning.
The problem is when making rapid small movements the torch holder shakes too much. Experimenting with motor tuning in Mach3 helps but hasn't eliminated the problem....unless I just havent found the right tuning yet.
I am using rack and pinion direct drive.
Rack has 5.1 teeth per inch and the pinion has 16teeth on 1.125" diameter.
Is this combo going to inherently give me problems or should I continue to play with motor tuning?
Thanks, John

[DSL PWR]

It could be any number of things.

Tuning:
First pull the motors away from the rack and run a problem program and see if the motors sound good and give smooth movments.

Rack and pinion:
The pinion needs to be square to the rack, and the rack needs to be level. If all of this is perfect, you may need a little more pressure pushing the pinion into the rack. On mine the spring was a little weak and the pinion would walk out of the rack and cause shaking. The problem here is your rack is coarse, and will require more pressure to keep the pinion seated, and more pressure will suck up more power. Are you sure the pressure angle is the same on you rack and pinion?

Gantry:
You don't make any mention of your motor config, so I will go with 1 on each axis. Is the shaking worse far away from the motor on the shaking axis? If it is you probably have a rigitity problem.

Motors:
Are you using servos or steppers? If you are using servos you will need to gear down the pinion. If you have steppers, with 200ppr you are moving your gantry about .018" per step, which is pretty coarse, and could be giving you some issues.

Good luck.

[johnt]

Thanks, dsl_
I didnt think to run a program with the motors out of the rack. I'll try that.
I have 2 stepper motors on my long axis, 1 stepper on short axis.
Gecko 201's, microstepping.
I may have a rigidity issue with the z.
That is where the shaking happens.
The torch holder hangs down below the gantry and tends to act like a pendulum.
I will try to beef up the torch mount.
I just dont want to rework a bunch of parts if I will never get smooth moves from this rack and pinion. I would however consider another form of transmission. Rack and pinion just seemed easiest to build.
Thanks again, John

[vcooney]

Johnt,
Post a picture and maybe someone could figure out your problem with a visual.

Vince

[ger21]

Are you running in CV mode, or Exact Stop mode?

[johnt]

I'm not sure if it's cv or exact stop (I'm not at that computer right now)
I'll check.
Which should it be?
Also, I'll work on getting a couple pics.
John

[Greolt]

Yeah definitely try CV mode. What version of Mach are you using? There was a bug with "CV feedrate" turning itself on that was fixed very recently. (needs to be off)

R&P drives have been used very successfully many times so there is not a problem per se.

However there can be large variation as to how well it is implemented. Pics would be good.

Many fall into the trap of making the clearance under the gantry too large. Especially on a plasma this can be quite small. In principal the more the torch hangs below the gantry the less rigid it will be.

Again pics will make comment more relevant. Overall view, R&P drives, etc.

Some with less rigid machines are successfully using Mach Quantum, though it is only in beta release with no support. Would not recommend that yet.

Greg

[Jcar]

The torch holder hangs down below the gantry and tends to act like a pendulum.

I had the same problem with my system, except my pendulum effect was above the gantry. I added some bracing and the problem went away.

Assuming you are running in CV mode, I would check the acceleration values in my motor tunning. If the values are too high, even a rigid system will experience some jerking/vibration.

[johnt]

Thanks guys,
I will try to mess with it today (being Sunday, my wife may make me watch football all day
you all certainly gave me some good ideas to try.
I think I will :
shorten the z axis/torch holder,
add bracing,
raise the material rack,
check for cv mode,
continue to play with motor tuning.

I will keep you posted.
Thanks

[Torchhead]

Thanks guys,
I will try to mess with it today (being Sunday, my wife may make me watch football all day
you all certainly gave me some good ideas to try.
I think I will :
shorten the z axis/torch holder,
add bracing,
raise the material rack,
check for cv mode,
continue to play with motor tuning.

I will keep you posted.
Thanks

Motor tuning will only do so much. The problem in my estimation is the direct pinion drive design. Look at it this way: The motor has 200 distinct "detents". So in one rotation moves through 200 positions. Now, direct coupling a 1.125 dia gear you move approx 3.5 linear inches with each rotation and makes the resolution (smallest distance you can move in one step) at .018". The microstepping in most modern drives will help smooth the moves but cannot be counted on to increase the raw resolution. Remember that microstepping becomes less effective the faster you spin the motor. The problem comes when your artwork generates moves (like an arc) that calls for lots of small moves smaller than your resolution is capable.

For this resaon and others we recommend a belt reduction on rack and pinion of at least the pinion step up ratio (in this case 3.5:1) to gain back torque and resolution.

In doing effective cutting, sharp turns and quick direction reversal with minimum backlash is important. To get that you need good acceleration. To get good acceleration you need good torque. With steppers it's hard to have both speed and torque. Things flopping around on your gantry will cause vibrations and resonance problems so the Z needs to not have a lot of flex.

If you move your motor tuning down to low levels and things smooth out you have your answer. Speed you can't use is a wasted resource.

Set things up with low velocity and low accel numbers and see if it smoothes out.

Get a nice smooth circle (low node count) and use that to test. CV has less effect on long arcs so it helps to rule out the artwork as the problem.

TOM CAUDLE
ww.CandCNC.com

[plain ol Bill]

Guys this is not an attempt to hijack this thread by johnt. I would really like to know what drive system you folks consider the best type and why. I am considering making my own machine and would really like to know which type and why if you don't mind.

[millman52]

Guys this is not an attempt to hijack this thread by johnt. I would really like to know what drive system you folks consider the best type and why. I am considering making my own machine and would really like to know which type and why if you don't mind.

Well Bill, that's a pretty loaded question. Like most things in life, There is pros & cons to each.


I am answering on the assumption you are asking about stepper or servo. & giving only a very general & brief answer.

Stepper systems are less expensive, will handle moderate loads at low to moderate speeds & usually don't require shielded wireing.

Servo systems are better for higher speeds, have more torque at higher speeds thsn low, cost more, you will have to take good care that the entire system is kept away from or shielded from outside electrical "noise"

Think these few thoughts over & try to ask a little more specific questions.

You may want to start your own thread or just do a little searching. This very topic has been covered numerous times here. You'll get plenty of help from many of us here.

Welcome to the forums.