Find pulse width for stepper motor

[joe_1]

Hello,

I am trying to find the proper pulse width for stepper motor from a given flow rate
Here is what I am doing:
1- User enters flow rate. As an example FL=10 ml/s
2- RPM(motor) = FlowRate/PUMP DISPLACEMENT
PUMP DISPLACEMENT for this motor = 0.0676
so, RPM = 10/0.067 = 147.93
3- Frequency = RPM/60.0 = 2.4655 Hz --- convert from minutes to seconds
4- Period = 1/frequency = 1/2.4655 = 0.4056 sec
5- Period = t_ON + t_OFF
assuming duty cycle = 50% then t_ON = t_OFF
t_ON = pulse width = Period/2.0 = 0.2028 sec

Please let me know if my calculations are correct.
Thank you in advance.

 

[Tony Stewart]

You did not state units of Pump displacement.

 

[Papabravo]

Is it the pulse width that you need or the repetition rate of the pulses which would determine the angular velocity of the stepper motor.

 

[rjenkinsgb]

You don't appear to have allowed for the motor pole count?

Frequency would be:

Required RPM x number of motor poles to get cycles per minute,
/ 60 to get frequency in Hz.

Also, common four-wire (or 6 wire) dual winding steppers need a four stage sequence to move one full step;

A+
B+
A-
B-
(A+,B+ etc.)

That's often done using a counter, so 4x the number of full step pulses in to give the required movement, and those pulses only clock the counter rather than setting the output duration, which is set by the puse frequency.

If you are using a dedicated stepper driver module, that likewise may only need clock pulses?

 

[joe_1]

Is it the pulse width that you need or the repetition rate of the pulses which would determine the angular velocity of the stepper motor.

I am trying to use a motor controller chip from NXP (PCA9629A), and they are using Pulse Width as a way to control the motor speed. By varying the pulse width, the motor speed changes. So, I guess longer pulse width means higher applied voltage which will translate to faster running motor.
My problem is: I am not sure how to figure out what pulse width I should select for a given motor RPM/frequency.

 

[rjenkinsgb]

I am trying to use a motor controller chip from NXP (PCA9629A), and they are using Pulse Width as a way to control the motor speed.

OK, with that IC, each pulse duration is 1/4 of a full motor step.


For 10mL per second with a pump displacement of 0.0676 ml per motor revolution ?????

That needs 147.93 (call it 148) revolutions PER SECOND; 8880 RPM ????

Is the pump actually rated for that speed? It's extremely high for a stepper and would need significant ramp up and down times. That seems impractical.


If that displacement is per step, then it needs 148 steps per second, which is reasonable (though still probably needing ramps).

The time would be 1/148 = 6757uS per full step, or 1689uS per quarter step, depending on how the motor volume steps are counted.

 

[joe_1]

OK, with that IC, each pulse duration is 1/4 of a full motor step.

For 10mL per second with a pump displacement of 0.0676 ml per motor revolution ?????

That needs 147.93 (call it 148) revolutions PER SECOND; 8880 RPM ????

Is the pump actually rated for that speed? It's extremely high for a stepper and would need significant ramp up and down times. That seems impractical.


If that displacement is per step, then it needs 148 steps per second, which is reasonable (though still probably needing ramps).

The time would be 1/148 = 6757uS per full step, or 1689uS per quarter step, depending on how the motor volume steps are counted.

I thought the 147.93 is RPM NOT RPS

 

[Pommie]

I thought the 147.93 is RPM NOT RPS

You stated 10ml/s where (assumed) s=second.

Mike.

 

[Tony Stewart]

I thought the 147.93 is RPM NOT RPS

Joe pls restate the problem without omissions or errors so the units match correctly.