Hello Guest it is May 06, 2021, 06:16:00 PM

Author Topic: Pulse per turn  (Read 3376 times)

0 Members and 1 Guest are viewing this topic.

slovenec

• 56
Pulse per turn
« on: June 06, 2018, 08:31:14 AM »
Greetings,

i have a 4th axis and i would like to determine the pulse per turn.

I have a 1/8 degree stepper motor and harmonic Drive gear head reducer 1:21.
So if i correcty understand  for one turn must the motor make 21 rotations.

What is then the pulse per turn?

Best regards

STEPER MOTOR SPECS : http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0ahUKEwimtoWshb_bAhXJD8AKHTiQDdAQFggnMAA&url=http%3A%2F%2Fwww.sorotec.de%2Fwebshop%2FDatenblaetter%2FSchrittmotoren%2FESM.86114.56MD%2Fesm8611456md.pdf&usg=AOvVaw2nym_QQ8xlCZQU58GZ_DBx

joeaverage

• 6,400
Re: Pulse per turn
« Reply #1 on: June 06, 2018, 09:42:12 AM »
Hi,
I assume you mean 1.8 degrees per step, that is the norm for a two phase stepper. If so that means 200 steps are required for ne revolution of the motor.

For one revolution of the output shaft of the gearhead:
200 x21=4200 per rev.

You don't mention microstepping, its a common technique to improve the smoothness of rotation. So if you employ '8 step microstepping' your calculation is:
4200 x8 =33600 per revolution.

Note that if using for an A axis the units of an A axis is degrees:

33600 / 360 =93.333 steps per degree. This is the figure you would enter in the 'steps per unit' value of the A axis motor tuning page if using an 8 microstep regime.

Craig
My wife left with my best friend...
and I miss him!

slovenec

• 56
Re: Pulse per turn
« Reply #2 on: June 13, 2018, 01:05:28 PM »
Thank you for the answer. But I still cannot figure this out.

If I set the micro-stepping to 1600 pulse/rev on my leadshine driver DM422

Then I calculate the 4200 per rev. Now I multiply 1600 with the 4200 and divide everything with 360 to get puls per turn?

Is this right?

joeaverage

• 6,400
Re: Pulse per turn
« Reply #3 on: June 13, 2018, 02:37:25 PM »
Hi,
not quite. You have obviously set the microstep per full step to 8, resulting in 1600 pulses required to turn the motor one turn.
With the gear reduction:
1600 X 21 = 33600 pulses per rev of the gear reducer output shaft or:
33600 / 360 = 93.333 pulses per degree.

I think your confusion is about microstepping. Without any microstepping your motor is 200 pulse/rev. If you apply an 8 microstep per full step regime
then you need 8 pulses to achieve what one pulse used to do, so no longer 200 pulse/rev but 1600 pulse/rev.

Craig
My wife left with my best friend...
and I miss him!

slovenec

• 56
Re: Pulse per turn
« Reply #4 on: June 28, 2018, 07:32:05 AM »
Thank you. I think I understand what are you saying.

Another question, regarding the pulse per turn for mach.

If I have a servo motor with encoder 2500PPR how do I calculate the pulse per for mach3. I now that with one turn of the spindle the machine moves 4,73 mm
Between servo motor and ball screw is gear transmission and I don't have a clue what is the gear ratio and I cant count the gears because I would have to disassemble half of the machine.

Any suggestion how could i determine that?

I have a mikron wf3 dcm and cslab csmioipa  controller, and i cant get right the pulse per...

Thank you!!!

JohnHaine

• 356
Re: Pulse per turn
« Reply #5 on: June 28, 2018, 01:32:42 PM »
I believe that there is a Mach 3 function to measure the pulses per feed unit, can't remember where to find it.

joeaverage

• 6,400
Re: Pulse per turn
« Reply #6 on: June 28, 2018, 02:53:19 PM »
Hi,
a 2500PPR encoder has 2500 lines, this is now the entry level standard for an incremental encoder. An incremental encoder has an A and B phase in phase quadrature, ie
90 degrees apart. That means you get four encoder counts per line, so 10,000 counts per revolution.

If you put 10,000 in your axis motor tuning 'steps per unit' ( this assumes no electronic gearing within the servo) and then issue:
G0 X1    Mach will try to move one unit by producing 10000 pulses.

If in fact your axis moved 2.345 mm rather than the 1mm you requested you can use that to recalculate the 'steps per unit' value:

steps per (true)= (unit distance)/ (actual distance moved)  X (estimated steps per)
= 1 / 2.345 X 10000
=4246.4

As a sanity check....if Mach issued 10000 pulses and the axis moved 2.345mm then if I wanted it to move 1mm only
then Mach will need to issue fewer pulses, in which case 4246.4 ( being less than 10000) sounds about right.

Craig
« Last Edit: June 28, 2018, 02:59:45 PM by joeaverage »
My wife left with my best friend...
and I miss him!