Showing results for 
Search instead for 
Did you mean: 

Difficulty with Pol. Pulse

Associate III


I am having difficulty getting my motor to start reliably and smoothly.  Sometimes it starts as expected, but sometimes, it just wiggles around it's start position.  If I spin it by hand in the direction it's supposed to go, it starts spinning on its own as expected.  Also, if it's already spinning in the correct direction when I enter speed control mode, it also spins on its own as expected.

I have attempted following the steps in HSO_Startup_Guide, but can't seem to get the results listed.

My setup:
Motor Control Board: B-G473E-ZEST1S
Power Board: LVLP01
Motor: MOONS R57BLB50L2 (recommended for zest eval kit)

Results of motor profiler:
RsDC: 0.359 Ohm
RsAC: 0.371 Ohm
Ls: 0.000690 Henry
Flux: 0.0130 Wb
Isc_A: 18.8A
kt_Nm_A: 0.0389 Nm/A

Pol Pulse Parameters
Pulse duration: 8 periods
pulse current goal: 4.4A
decay time 11 periods
Result PulseDuty: 0.97

To get oscilliscope traces:

Current: I soldered leads onto TP9, TP11, TP13 (power board op-amp outputs).  I read one of these at a time.

I tried using TP10 (op amp reference) as reference for oscilloscope measurement, but connecting my ground lead from my probe causes the processor board to crash, showing 2 red lights on the RGB LED's by the usb cord

Voltage: pulled directly from the Vout screw terminals.

My oscilloscope is set to record a single frame, triggered by a rising edge of the voltage going to the motor.  I used software to monitor the button User2, entering speed control mode on press, and going to observer mode on release.

What I expect to see: gradual current ramp up (+- Pulse current goal), down, up reverse direction, and back to resting point, as shown in HSO startup guide.

What I see: Chopped ramp current up to less than goal, and abrubt drop to zero current.  I only see current in one direction.

Peak current calculation:
Peak op-amp voltage Vp: 0.584V
Amplifier gain (from schematic) g = 15k Ohm / 1k Ohm = 15
Shunt Resistiance: 0.015 Ohm
Peak current: (Vp / g) / R = 2.59A


Gael A
ST Employee

Hello ccut93,

Concerning your oscilloscope issue, I fear you may have created a short-circuit while soldering the leads on test points
For the pole pulse setup, with the MOONS R57BLB50L2, the following should do the trick :
- Pulse duration : 4
- Decay time : 1
- Pulse current goal : 6 Amps

This setup will give you a 0.83 PulseDuty, which can be improved but is fine anyway in my opinion.

If you agree with my answer, please consider accepting it by clicking on 'Accept as solution'.

Hope this will help,
Gaël A.
Associate III

Hello Gael,

Thanks for your reply.

I checked my solder job, and don't see any shorts. (pics below)  I did check the probe on my oscilloscope, and it does connect directly to the power ground, so if signal ground isn't isolated from power ground on the power board, that would explain the crash on using ORef as my oscilloscope reference.

WIN_20240327_09_40_49_Pro.jpg   WIN_20240327_09_41_01_Pro.jpg   WIN_20240327_09_41_14_Pro.jpg

I thought shorting ORef may have damaged my power board op-amps These op-amps are not used by the code, as I use the internal op-amps on the processor board.  I used the circuit to trace internal op-amp outputs (prefiltered) to SB35, SB39, and SB22.  Putting my probe on these points, I was able to get a waveform similar to the one I had from the other points, minus the chopping on the ramp up. (Pic below)




I tried a pulse duration of 4 with a current goal of 6A, and decay time of 1.  It doesn't replicate the .83 duty cycle you report.  It shows -0.98 PulseDuty in red, so I am guessing I am saturating the pulse output.  I am running 24V supply.  What voltage did you run yours at?

The motor still shows the same problematic behavior.  I show a single up ramp on the op-amp output to 124mv, and an immediate drop to 0v.  This suggests an output current of about 0.5A, much less than the 6A current goal given on PolPulse Params






While the actual current on a phase is continuous, the shunt signal is 'chopped' as you say. The signal is only present when the relevant low side FET is on. I assume the HSO startup guide image was captured using an external current probe.

Hello ccut93,

Opref must not be connected to GND because it is the biasing to Vref/2 of the OPAMPS for measuring the current in the shunts in both direction. Setting this point to GND will probably not break the OPAMPS but risks to break the power stage depending on the reaction of the algorithm.





Concerning the Pol. Pulse, I believe I made the test around 22V. In any way you need to tinker the Pol Pulse setup values to get a good start-up. If you want to make running tests first, I recommend to switch to RsDC start-up before tuning Pol Pulse.

If you agree with my answer, please consider accepting it by clicking on 'Accept as solution'.

Hope this will help,
Gaël A.