More speed required

[Luke Lawcock]

Hi all,

I have an Arduino Due running the following code, I require more speed however. 

• If i remove the setMinPulseWidth I lose torque and gain no speed, if I increase it, i lose speed
• If i increase setMaxSpeed, it stays almost the same, i have tried increments upto 20000
• I cannot increase the acceleration due to the weight and inertia of the gearbox and adjoined parts.

Any advice?

Kind Regards

Luke

//Stepper Motor 23HS33-3008D

//Driver M542T

#include <AccelStepper.h>

#include <MultiStepper.h>

AccelStepper motor(1, 7, 8);

const int buttonPin = 3;     // the number of the pushbutton pin

// * pushbutton attached to pin 3 from +5V

// * 10K resistor attached to pin 3 from ground

// variables will change:

int buttonState = 0;         // variable for reading the pushbutton status

void setup() {

  // initialize the pushbutton pin as an input:

  pinMode(buttonPin, INPUT);

  motor.setMaxSpeed(12250);

  motor.setAcceleration(61250);

  motor.setMinPulseWidth(20);

  motor.setCurrentPosition(0);

}

void loop() {

  // read the state of the pushbutton value:

  buttonState = digitalRead(buttonPin);

  // check if the pushbutton is pressed.

  // if it is, the buttonState is high:

if (buttonState == HIGH) {

    motor.moveTo(36750);

    motor.run();

} else {

    motor.setCurrentPosition(0);

    motor.stop();

}

}

[gregor]

have you tried increasing the motor driver supply voltage, too? higher supply voltage means higher maximum step speed. you can supply up to 50V to the M542T (according to this page: http://www.omc-stepperonline.com/24-phase-nema-23-stepper-motor-driver-2450vdc-15a45a-256-microstep-m542t-p-293.html )

[Luke Lawcock]

Yes the power supply is 48V, 4.4A

[gregor]

I'll check your sketch with my oscilloscope as soon as I have time. what speed do you need to reach?

[Luke Lawcock]

I'd need to calculate it, I will get back to you this evening. 

On Wednesday, 9 November 2016 12:22:02 UTC, Luke Lawcock wrote:

[Hugh Gilhespie]

Hi Luke,

I am running AccelStepper with a Teensy 3.2 with a 96 MHz clock and it has no problem reaching about 18000 steps per second. This is with a Nema23 motor (23HS30-2804S_PG47) with a 47:1 reduction gearbox. The driver is a Gecko G251X.

I have not tried to run my setup any faster than 18000 steps/second but I would expect it to run to around 24000 based on the known  limit of ca. 4000 steps per second with a 16 MHz Arduino chip and by similar reasoning, you should expect about 21000 steps/second from your motor.

I know the M542 driver says up to 200 kHz input pulse rate but I have had at least one bad experience with the Chinese stepper drivers so if your code is OK - which it certainly seems to be, I think I would suspect the driver.

Regards, Hugh

--
You received this message because you are subscribed to the Google Groups "accelstepper" group.
To unsubscribe from this group and stop receiving emails from it, send an email to accelstepper+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[gregor]

with the sketch from your original post I was able to achieve a maximum speed of ~12,500 steps / second. you can still gain a bit of speed by setting the minimum pulse width lower - 20 is default, i was able to achieve ~16,000 steps / second with pulse width 0. (I set the maximum speed much higher for this test)

you should move the run() outside the if block, towards the end of your loop function, or the motor will not decelerate after you call stop().

[Luke Lawcock]

Hugh, gregor,

Thanks for the information, I looked at the teensy but went for the Due as it is 84MHz without overclocking, maybe I should have gone for the teensy. I have tried two M542T drivers and both give the same result, not to say that they both aren't a bit rubbish. 

I have today calculated the steps and speeds and I can say that when I have changed 'motor.moveTo(36750);' to 3000 which is the steps I require, I am achieving this. However, the speed I require it to do it at is 0.3 seconds, which i am not achieving. Is there a formula to work out the amount of steps travelled in the accelerating period? I am awfully close to where I need to be, I'd imagine in the region of less than 0.1, possibly 0.04 seconds.

KR Luke

On Wednesday, 9 November 2016 12:22:02 UTC, Luke Lawcock wrote:

[Luke Lawcock]

Hi gregor,

I will try again with no pulse width, however i think it lost all torque when i tried it before, i have posted more information a few minutes ago. Thankyou for your help.

[Luke Lawcock]

Also i want the motor to stop without deceleration, currently it stops dead which is what i want to achieve 

[Luke Lawcock]

//Stepper Motor 23HS33-3008D

//Driver M542T

#include <AccelStepper.h>

#include <MultiStepper.h>

AccelStepper motor(1, 7, 8);

const int buttonPin = 3;     // the number of the pushbutton pin

// * pushbutton attached to pin 3 from +5V

// * 10K resistor attached to pin 3 from ground

// variables will change:

int buttonState = 0;         // variable for reading the pushbutton status

void setup() {

  // initialize the pushbutton pin as an input:

  pinMode(buttonPin, INPUT);

  motor.setMaxSpeed(15000);

  motor.setAcceleration(75000);

//motor.setMinPulseWidth(20);

  motor.setCurrentPosition(0);

}

void loop() {

  // read the state of the pushbutton value:

  buttonState = digitalRead(buttonPin);

  // check if the pushbutton is pressed.

  // if it is, the buttonState is high:

if (buttonState == HIGH) {

    motor.moveTo(3000);

    motor.run();

} else {

    motor.setCurrentPosition(0);

    motor.stop();

}

}

From some napkin math, this code should achieve 3000 steps in 0.3 seconds, can anyone confirm?

KR Luke

 

[Hugh Gilhespie]

Hi Luke,

My napkin suggests a rather different answer.

Accel =	75000	pps^2		maxSpeed	15000	pps
	Accel	iniSpeed	endSpeed	avgSpeed	Total
	Time, s	pps	pps	pps	Pulses
	0.02	0	1500	750	750
	0.04	1500	3000	2250	3000
	0.06	3000	4500	3750	6750
	0.08	4500	6000	5250	12000
	0.1	6000	7500	6750	18750
	0.12	7500	9000	8250	27000
	0.14	9000	10500	9750	36750
	0.16	10500	12000	11250	48000
	0.18	12000	13500	12750	60750
	0.2	13500	15000	14250	75000
	0.22	15000	15000	15000	90000
	0.24	15000	15000	15000	105000
	0.26	15000	15000	15000	120000
	0.28	15000	15000	15000	135000
	0.3	15000	15000	15000	150000

I am not convinced this calculation is correct but if it is, you will have made a lot more that 3000 steps in 0.3 seconds.

My calculation - Excel spreadsheet - is based on the following assumptions:

Acceleration is linear wrt time

The 'mean' pulse rate over a time interval is simply half of the final rate - initial rate.

I would normally say Hope This Helps - but I'm not at all sure that it does.

Regards, Hugh

 

--

[Hugh Gilhespie]

Right, having used a nice, clean napkin, I find we are in total agreement!!

Accel	iniSpeed	endSpeed	avgSpeed	Total
Time, s	pps	pps	pps	Pulses

0.02	0	1500	750	15
0.04	1500	3000	2250	60
0.06	3000	4500	3750	135
0.08	4500	6000	5250	240
0.1	6000	7500	6750	375
0.12	7500	9000	8250	540
0.14	9000	10500	9750	735
0.16	10500	12000	11250	960
0.18	12000	13500	12750	1215
0.2	13500	15000	14250	1500
0.22	15000	15000	15000	1800
0.24	15000	15000	15000	2100
0.26	15000	15000	15000	2400
0.28	15000	15000	15000	2700
0.3	15000	15000	15000	3000

Apologies for my initial stupidity - I used an old spreadsheet I had set up where my time intervals were 1 second

so I missed out on the need to count pulses in fractional seconds.

Regards again, Hugh

[Luke Lawcock]

Hugh,

Thankyou so much for the clarification from your spreadsheet, I hadn't thought to do that, I was literally drawing graphs on paper lol! I'm rough with excel but I may have a go myself as it looks very handy.

I also think I have a small delay from my PLC as its a relay output which is 2Hz or 0.5Hz depending on resistive or inductive, which I need to look into. 

Do you have a link to your Teensy setup as I may need to increase the processor speed as gregor was only able to get ~12500 steps out of the arduino setup I have with the minPulseWidth set a 20 microseconds (if I go lower, I have zero torque). 

Thanks again,

Kind Regards 

Luke  

[Hugh Gilhespie]

Hi Luke,

The answer is yes but I am just about to go out so I will reply in more detail this evening.

Regards, Hugh

--

[Luke Lawcock]

Hugh,

Thankyou kindly sir, I look forward to it.

KR Luke