L6470 High Speed Issue

Martiska.Greg · ‎2014-09-18

Posted on September 18, 2014 at 21:39

Dear Motor IC Forum:

I am using an L6470 to drive a stepper motor attached to a gearbox and am not able to achieve step rates greater than roughly 700 steps/sec with any microstepping setting. The motor gets loud and stalls as speeds increase. At speeds below 500 steps/sec the motor works perfectly. The Rm of the motor is 4.5 ohms. The Lm is 3.6mh (spec sheet). The Ke is 0.143 v/hz (approximately). I have played with all setting except pmt and have not been able to get more speed. I need roughly 2000 steps/sec.

I also tried the L6472 on the same setup and had no issues reaching the required speeds. The problem is I also need to run at very low step rates of 5-10 steps/second at 1/16th steps. With the L6472 the motor is not smooth enough and vibrates quite a bit.

What I need is the high speed range of the L6472 and the low speed smoothness of the L6470. So my questions are:

1) Am I doing something radically wrong with the L6470 that is preventing me from reaching higher speeds or is this normal for this chip?

2) Can I connect the motor to both chips and activate one or the other depending on the speed requirements? I do not want to put in a switching system to switch between chips but simply connect them both to the motor and put one chip in the high impedance state while I am using the other i.e 6470-1A to 6472 1A to Motor A etc.

Thanks for your help,

Greg

#l6470-l6472

larken · ‎2014-09-20

Posted on September 20, 2014 at 22:40

What is your supply voltage ? Speed is proportional to voltage.

The motor you have (4.5ohm /phase) is a fairly high inductance, low speed motor. Get a motor with a much lower inductance and a phase resistance of less than 1.5 ohms. You will be able to run it much faster, also you should have a PS voltage of at least 30 volts.

Larry

Martiska.Greg · ‎2014-09-20

Posted on September 20, 2014 at 23:13

Larry:

Thank you for your response. I am using 24V. It would be difficult to get more voltage. I chose the motor based on torque requirements. I will look for options with lower phase resistance and inductance. I can also try a parallel connection as the motor has 8 wires. Not sure how much that will reduce the low seed torque though.

Martiska.Greg · ‎2014-09-21

Posted on September 21, 2014 at 16:40

Larry:

I was thinking about your response to my speed question and I am not sure increasing PS voltage will help the problem. The L6470 controls the drive voltage to achieve the target current unlike typical chopper drivers which control the current at the drive voltage. The way I understand it is the drive voltage is independent of the PS voltage for the L6470. In my application the drive voltage is around 9V due to the resistance of the motor and the 2A current. This will be the same with a 24V PS or a 36V PS. Therefore there seems to be a 9V speed limit with the L6470 i.e whatever speed can be achieved with 9 volts drive current. I am not sure how motor inductance and back emf factor in with the L6470. The L6472 uses 24V to drive the motor so there is no problem with the speed.

Am I missing something here with the L6470 driver? I think you are correct that the drive voltage is too low for higher speeds with the L6470. I just do not see how to correct the problem with the steppers available in my torque range.

Enrico Poli · ‎2014-09-23

Posted on September 23, 2014 at 09:16

Hi Greg,

the considerations of Larry were right. In general the maximum speed you can achieve with the motor is proportional to the supply voltage, whatever the driving mode you are using (voltage mode, current control, etc.).

In your case the issue with the L6470 could be related to the mid-point resonance of the motor. At about 700\800 step/s most of the stepper motors generate a mechanical resonance causing vibrations and distorting the BEMF. The voltage mode driving of the L6470 is more sensitive to this effects compared to the current control of the L6472.

When a mechanical load is connected to the stepper shaft, the resonance spot is damped and its effect is no more critical for the L6470.

My suggestion is to test you stepper adding a small load on the shaft.

This is a good advice for all the motors and in particular for the steppers (whatever device you are using to drive them).

Enrico

Martiska.Greg · ‎2014-09-23

Posted on September 23, 2014 at 17:34

Dear Enrico:

Thank you for your reply. The speed question does appear to be due to resonance at 700-800 rpm. However I am testing the motors are under load when seeing the resonance. It does not appear to change much with load although I will do more testing on this. I have also tried to accelerate quickly through the resonance speeds but this has not helped.

I have a question on the voltage. I have a couple of different motors that I am using in the application. One motor uses KVAL's of 14%. This indicates to me a motor drive voltage of around 3.4 volts with a 24V power supply. Is this correct? If so how does this voltage limit the maximum speed for the motor?

I really want to use the L6470 chip as it is so much smoother at the low speeds I need. However I must be able to also reach 2000 full steps/second . What motor parameters and drive voltage are best for this. I am using 24V for power which would be difficult to change but I can change motors or winding styles.

Thanks for your help.

Greg

Martiska.Greg · ‎2014-09-25

Posted on September 25, 2014 at 15:53

Enrico:

I re-read the application note on the L6470 and I think I have answered my own question. The KVAL is only the starting voltage and the applied voltage is the starting voltage plus the BEMF voltage at the target speed. With my motors I am limited by the high BEMF. One motor has a BEMF of 0.15V/hz so I only achieve my 2 amp current at up to 400 steps/second with a 24V supply. At 2000 steps per second the current is negative. Is this correct?

Greg

Enrico Poli · ‎2014-10-03

Posted on October 03, 2014 at 15:55

Correct :)