Different steppers on each axis?

[pulsar200]

This might be a FAQ, but i couldn't find it anywhere..

I have 3 almost identical motors.
They are all of the 6-wire unipolar NEMA 23 and 200 steps/rev.

However:
The first one is 5 Volts 1,8Amps
The second one is 4Volts 2Amps
The third one is 5,7Volts 1,6Amps

Okay. The question is if Im going to get in trouble for using 3 not completly identcal motors. Some people say that im going to have problems with syncronisation. Im planning on using the hobby-cnc driver board.

Regards
Martin Holm Pedersen who is on his first cnc-project and almost done with the building:-)

[randyf1965]

3 different motors on 3 different axis.. no problem. In Mach 2/3 you set each motor/axis up so as long as you configure each properly, no problems

[rippersoft]

I use 2 of the same motor on X and Y and a different size on the Z axis with a HobbY-CNC. No problems. Just adjust the amps per phase to match the motors and with whatever software you use, ie Mach, TurboCNC, etc., config each axis for the size of the stepper to maximize performance. For best results, make sure you drive the motors at 24VDC to 36VDC.

RipperSoftware

[pulsar200]

Okay. That was what i was hoping for..

Thanks
- Martin

[NC Cams]

FYI: different motors with different inertia's and current constants will accelerate differently when the same current is applied to each.

The "adjustment" features you indicated as being able to deal with the motors in a "no problem fashion" may not be so 'no problem'ish' under cases where fast moves are asked for in a simultaneous fashion.

I have a full feedback servo system with FULLY matched commercially blueprinted motors and we went thru hell tuning them to get them to respond in a simultaneous fashion - and they ALL HAD THE SAME INERTIA & ELECTRICAL COEFFICIENTS.

If find it hard, perhaps impossible to believe that 3 entirely different motors (probably salvaged or random buys off E-bay) can/will respond equally as well to each other.

Rippersoft's method at least matches the X and Y which should alleveiate some of the problems and this would be FAR superiour to the use of 3 different motors.

How to test/prove viability: mill a perfectly round circle as fast as you can. Check it for ovality and flats at the direction changes and also to see if it returns to 0,0 when done. Then do it slow, then do it at mid speed.

It is hard enough to make 3 IDENTICAL MOTORS work in concert with each other. But to intentionally ASK for the grief of getting 3 different ones to physically perform on que and exactly the same as each other when they electrically can't, doesn't seem like a wise thing to do IMO....

Do 3 different motors if you must - but forewarned is forearmed - try to at least match the X & Y, motor wise.

[Madclicker]

I've had nothing but happy results with my Ebay bought NOT "FULLY matched commercially blueprinted motors"! It was super easy to get them to work together at different speeds. Slightly different, but different.

Remember you are in a DIY forum. Many people want to use what they have, can scrounge, beg, or (God forbid) get off Ebay.

I'll pass on the "spend big bucks to go through hell" trip. Cheap and easy works for me.

[Madclicker]

NC,

Even though you use the same motors on each axis it doesn't mean they get tuned the same. The dynamics on each axis are determined as much (or more) by what is attached to the motor as the motor itself. So, buying perfectly matched motors to expect perfectly matched results on each axis is unrealistic.

[spalm]

I happen to use the same motors on X and Y. Two on the X and one on the Y with 8 tpi, 2 start screws. I use a different manufacture motor on the Z with a 12 tpi, 1 start screw. They all work just fine. Mach takes care of all of the math.

NC, I hear you. This works for me on my hobby machine. One might very well have to use different ramp up and ramp down speeds on each axis even if the motors are the same. The tuning for moving the entire gantry in the X might be different than just moving the Y left to right. I have test cut many pieces and believe that the results are acceptable for the wood that I cut.

Steve

[Madclicker]

You don't "tune" steppers the same way as servos. With steppers you just find the speed you can run them without losing steps and the acceleration they can handle. With servos you have to find the critically damped point for optimal tuning. This is dependent on the entire axis dynamics, not just the motor. It's easy (and pretty fun) to do with a spectrum analyzer. It gives you the system (ie axis) model and you just design the PID controller variables. With a scope you can do it by looking at the step response and adjusting the PID variables.

[dertsap]

i'd suggest using the strongest motor for the x axis and the weaker slower for the z axis , z doesn t travel as far and not as demanding as x y

[pulsar200]

Okay. So the next big question is. How to find out what acceleration they can handle. I don't think my stepper motors have the same torqe, but Im pretty confident that they have pretty much the same inertia. I am of course guessing based on the look of the three. They have the same housing and weigh pretty much the same.

But I think I can conclude that they probably won't accelerate equally fast. I also think that this is the case for any cnc-router setup, as the dynamics(read: weight,Inertia,friction) of each axis is almost never the same.

So what to do to get the best possible performance?
Well as Madclicker pointed out I can initially try to find my max speed without loosing steps.

Does anyone have an idea of how to do this?

Secondly I have to find the maximum acceleration that a given stepper can handle. This must also change depending on the material. In denser materials more torque is required and I must mill slower to get this and thereby the acceleration.

I don't have any sensors, so I can't measure the acceleration or speed. So I guess the conclusion will be that I simply turn down the speed/acceleration until I get acceptable results. Is it possible to set the default maximum acceleration in my milling program? or is it included in the g-code?

Best regards
- Martin H. Pedersen

[dertsap]

Originally Posted by pulsar200 until I get acceptable results. Is it possible to set the default maximum acceleration in my milling program? or is it included in the g-code? Best regards - Martin H. Pedersen

you set the max in the control software be it mach or what have you , for example , if the control is set to travel the motors at 100ipm ,and your program has a feed of 150ipm the max the motors will run at is 100 ipm

set the accelerate low and try the motor then bump up the accel until it looses steps , ,then back off the accel some ,that will be the max