Stepper motor control IC not quick enough for low inductance motor ?

[snarkysparky]

We are using Sanken  SLA7073  step motor driver.   It has a current reference input and controls the motor winding current accordingly.

Except for about 1ms at startup.   We tried a new low inductance motor  1.5mh  and 1.1 ohm.

Power supply is 24V.

With this new motor the IC will trip on current fault before it can start controlling the current.  That is what it seems to me.

Any ideas or thoughts are appreciated.

attached scope images have the current as measured at the sense pin in yellow.
One image has a near trip and the other an actual trip.  Trip occurs when this voltage crosses 0.7V

Thanks

[moffy]

Really nice captures. What is your startup sequence? Are all voltages applied before you command the chip to step?

[Doctorandus_P]

Low inductance motors have fewer windings in each coil, and therefore need more current for the same torque output.

You can also do a test with a 12V (or lower) power supply to reduce the rate in which the "inrush current" changes.

[thm_w]

Why is the new low inductance motor necessary? It might not be suitable for your power supply voltage.

https://www.eevblog.com/forum/projects/stepper-motor-question-low-voltage-high-current-vs-low-current-high-voltage/

[snarkysparky]

All voltages are applied long before the motor is commanded to run.
The purple trace is the Vref  Value.   Above 2.0 volts puts the driver into standby.   So to run the motor Vref is dropped down to the current setpoint.

We needed to run the motor much faster for a test.  Double the speed.  The old motor with inductance of 5.5mh  didn't have the torque at speed.  It would stall.
Consulting the catalog it seems that only lower inductance motors had better torque at speed. We are clocking the IC at 3000 pulses per second in half step mode.

I can drop the supply down to 12V and run the motor.  But that isn't an option now without a lot of redesigning.   

I guess I will look again at the motor options.  Maybe I can find a motor with higher inductance that has the required torque.

[moffy]

I am assuming that the blue trace is the falling edge of the reset pulse. There is an interesting section in the data sheet under "REF/SLEEP1 PIN":

"Ref/Sleep1 Pin
The Ref/Sleep1 pin provides access to the following functions:
• Standard voltage setting for output current level setting
• Output Enable-Disable control input
These functions are further described in the Truth Table section,
and in the discussion of output disabling, above.
Range A . In this range, control current value also varies in
accordance with V REF . Therefore, losses in the IC and the sense
resistors must be given extra consideration."

Could the internal value of Vref take longer to settle? Could you add extra delay between Vref falling and the blue trace falling edge, by a couple of ms?

[snarkysparky]

Finally.

Trouble was the Ref pin  had a RC filter.  Made the transition too slow.

Changing the timing so that the ref signal could settle before the mode pins went active fixed it.

Thanks all

[mikerj]

Thanks for posting the solution