TMS320F28062F: Vibration with IPM motor

Can you provide a detailed description of your question? What do you mean  "particular frequency"? How much? And any load on the motor? What's the Ld and Lq?

Did you set the Ld and Lq in user.h seprately according to the motor parameters in its datasheet?

Hi Yanming,

Sure I will provide you with a detailed description. Since I am working on a proprietary project, it would be good if could share your mail ID.

Regards

Samir

Hi Yanming, waiting for your reply.

Regards
Samir

Did you try to set the value of Ld and Lq sepratly according to the datasheet of the motor, and see what happens?

You may request a friendship with me, and then you can just share the message with me.

Dear Yanming

I am aware of the fact that this one is high saliency motor and MTPA needs to be implemented in order to utilize both the torques magnetic as well as reluctance. So I have already implemented MTPA in my code with the help of SPRACF3 and Lab-13 code. I have attached the code snippet below, but the issue remains the same.

inline void CTRL_runOnLine_User ( CTRL_Handle handle , const HAL_AdcData_t *pAdcData , HAL_PwmData_t *pPwmData ) {
CTRL_Obj *obj = (CTRL_Obj *) handle;

_iq angle_pu;

MATH_vec2 phasor;

_iq gIs_mtpa_pu_u = _IQ(0.0);
_iq gId_mtpa_pu_u = _IQ(0.0);
_iq gIq_mtpa_pu_u = _IQ(0.0);

// run Clarke transform on current
CLARKE_run(obj->clarkeHandle_I, &pAdcData->I, CTRL_getIab_in_addr(handle));

// run Clarke transform on voltage
CLARKE_run(obj->clarkeHandle_V, &pAdcData->V, CTRL_getVab_in_addr(handle));

// run the estimator
EST_run(obj->estHandle, CTRL_getIab_in_addr(handle), CTRL_getVab_in_addr(handle), pAdcData->dcBus, TRAJ_getIntValue(obj->trajHandle_spd));

// generate the motor electrical angle
angle_pu = EST_getAngle_pu(obj->estHandle);

// compute the sin/cos phasor
CTRL_computePhasor(angle_pu, &phasor);

// set the phasor in the Park transform
PARK_setPhasor(obj->parkHandle, &phasor);

// run the Park transform
PARK_run(obj->parkHandle, CTRL_getIab_in_addr(handle), CTRL_getIdq_in_addr(handle));

// when appropriate, run the PID speed controller
if (CTRL_doSpeedCtrl(handle)) {
_iq refValue = TRAJ_getIntValue(obj->trajHandle_spd);
_iq fbackValue = EST_getFm_pu(obj->estHandle);
_iq outMax = TRAJ_getIntValue(obj->trajHandle_spdMax);
_iq outMin = -outMax;

// reset the speed count
CTRL_resetCounter_speed(handle);

PID_setMinMax(obj->pidHandle_spd, outMin, outMax);

PID_run_spd(obj->pidHandle_spd, refValue, fbackValue, CTRL_getSpd_out_addr(handle));

#ifdef MTPA
float gIs_ref_float;
gIs_ref_float = _IQtoF(_IQmpy( CTRL_getSpd_out_pu(handle), _IQ(USER_IQ_FULL_SCALE_CURRENT_A)));

float g_Flux = obj->motorParams.ratedFlux_VpHz;
float g_Diff_Lq_Ld = (obj->motorParams.Ls_q_H - obj->motorParams.Ls_d_H);
float g_Id_Ref = (g_Flux - (float) (sqrt((g_Flux * g_Flux) + (8 * (g_Diff_Lq_Ld * g_Diff_Lq_Ld) * (gIs_ref_float * gIs_ref_float))))) / (4 * g_Diff_Lq_Ld);
float g_Iq_Ref = (float) (sqrt((gIs_ref_float * gIs_ref_float) - (g_Id_Ref * g_Id_Ref)));
gId_mtpa_pu_u = _IQ(-g_Id_Ref/USER_IQ_FULL_SCALE_CURRENT_A);
gIq_mtpa_pu_u = _IQ(g_Iq_Ref/USER_IQ_FULL_SCALE_CURRENT_A);
#endif

}

// when appropriate, run the PID Id and Iq controllers
if (CTRL_doCurrentCtrl(handle)) {
_iq Kp_Id = CTRL_getKp(handle, CTRL_Type_PID_Id);
_iq Kp_Iq = CTRL_getKp(handle, CTRL_Type_PID_Iq);
_iq refValue;
_iq fbackValue;
_iq outMin, outMax;

// read max voltage vector to set proper limits to current controllers
_iq maxVsMag = CTRL_getMaxVsMag_pu(handle);

// reset the current count
CTRL_resetCounter_current(handle);

// ***********************************
// configure and run the Id controller

// compute the Kp gain
// Scale Kp instead of output to prevent saturation issues
if (CTRL_getFlag_enableDcBusComp(handle)) {
Kp_Id = _IQmpy(Kp_Id, EST_getOneOverDcBus_pu(obj->estHandle));
}

PID_setKp(obj->pidHandle_Id, Kp_Id);

// compute the reference value
#ifdef MTPA
refValue = gId_mtpa_pu_u; //TRAJ_getIntValue(obj->trajHandle_Id) + CTRL_getId_ref_pu(handle);
#else
refValue = TRAJ_getIntValue(obj->trajHandle_Id) + CTRL_getId_ref_pu(handle);
#endif

// update the Id reference value
EST_updateId_ref_pu(obj->estHandle, &refValue);

// get the feedback value
fbackValue = CTRL_getId_in_pu(handle);

// set minimum and maximum for Id controller output
outMax = maxVsMag;
outMin = -outMax;

// set the minimum and maximum values
PID_setMinMax(obj->pidHandle_Id, outMin, outMax);

// run the Id PID controller
PID_run(obj->pidHandle_Id, refValue, fbackValue, CTRL_getVd_out_addr(handle));

// ***********************************
// configure and run the Iq controller

// compute the Kp gain
// Scale Kp instead of output to prevent saturation issues
if (CTRL_getFlag_enableDcBusComp(handle)) {
Kp_Iq = _IQmpy(Kp_Iq, EST_getOneOverDcBus_pu(obj->estHandle));
}

PID_setKp(obj->pidHandle_Iq, Kp_Iq);

// get the reference value
if (CTRL_getFlag_enableSpeedCtrl(handle)) {
#ifdef MTPA
refValue = gIq_mtpa_pu_u;
#else
refValue = CTRL_getSpd_out_pu(handle);
#endif

}
else {
// get the Iq reference value
refValue = CTRL_getIq_ref_pu(handle);
}

// get the feedback value
fbackValue = CTRL_getIq_in_pu(handle);

// set minimum and maximum for Id controller output
outMax = _IQsqrt(_IQmpy(maxVsMag,maxVsMag) - _IQmpy(CTRL_getVd_out_pu(handle),CTRL_getVd_out_pu(handle)));
outMin = -outMax;

// set the minimum and maximum values
PID_setMinMax(obj->pidHandle_Iq, outMin, outMax);

// run the Iq PID controller
PID_run(obj->pidHandle_Iq, refValue, fbackValue, CTRL_getVq_out_addr(handle));
}

{
_iq angleComp_pu;

// compensate angle delay
angleComp_pu = CTRL_angleDelayComp(handle, angle_pu);

// compute the sin/cos phasor
CTRL_computePhasor(angleComp_pu, &phasor);
}

// set the phasor in the inverse Park transform
IPARK_setPhasor(obj->iparkHandle, &phasor);

// run the inverse Park module
IPARK_run(obj->iparkHandle, CTRL_getVdq_out_addr(handle), CTRL_getVab_out_addr(handle));

// run the space Vector Generator (SVGEN) module
SVGEN_run(obj->svgenHandle, CTRL_getVab_out_addr(handle), &(pPwmData->Tabc));

return;
} // end of CTRL_runOnLine_User() function

Thanks and Regards
Samir

Could you please post some phase current waveforms of the motor to show the problem? What speed is the motor running at? And any load on the motor?

Did you try to use the graph tool with datalog to check the rotor angle and feedback speed? If not, you may refer to the lab01b and lab01c to use this graph tool.

Did you try to use the TI EVM board to identify and run the motor if possible? And compare the testing result with your own board? Please make sure that the current and voltage sensing signals are good enough on your own board.

Hi Yanming,

Thanks for all the suggestions.

I have tried plotting feedback-speed (single time-0), feedback-angle (single time-1) and motor-current (single time-2) in the graph tool using datalog method as suggested within Lab-1b. Below is the code implementation.

Have defined the buffer in main.c itself

#define DATA_LOG_SIZE 1000

uint16_t datalogbuff0[DATA_LOG_SIZE];
float datalogbuff1[DATA_LOG_SIZE];
float datalogbuff2[DATA_LOG_SIZE];

#pragma DATA_SECTION(datalogbuff0,"graph_data");
#pragma DATA_SECTION(datalogbuff1,"graph_data");
#pragma DATA_SECTION(datalogbuff2,"graph_data");

In the mainISR

datalogbuff0[cntr] = (uint16_t)(_IQtoF(gMotorVars.Speed_krpm)*1000);
datalogbuff1[cntr] = _IQtoF(gMotorVars.est_angle_pu);
datalogbuff2[cntr] = _IQtoF(gAdcData.I.value[0]);

Also, I have attached the video of the graph tool while running in debug mode. In the video you will see, custom GUI is being used to command reference speed, which is also being used for monitoring few parameters. In the beginning, the motor is started at 80Hz and later on frequency is increased in steps of 2 to 5Hz. At around 88Hz motor vibrates and stop (at time 01:48 in the video). From the graph data you can observe that at this frequency, motor current amplitude starts oscillating, hence the ripples in feedback-speed increases.

Here are direct responses to your queries

Could you please post some phase current waveforms of the motor to show the problem? What speed is the motor running at? And any load on the motor?
You will find the current waveform in the video, Single Time-2 tab is used for output current. Till 88Hz motor runs fine. The motor is connected to the submersible pump.

Did you try to use the graph tool with datalog to check the rotor angle and feedback speed? If not, you may refer to the lab01b and lab01c to use this graph tool.
Yes, video is attached.

Did you try to use the TI EVM board to identify and run the motor if possible? And compare the testing result with your own board? Please make sure that the current and voltage sensing signals are good enough on your own board.
No, I haven't used any EVM board so far to run or identify the motor. Current and voltage sensing is good. Because with the same drive I am able to run the SPM motor with a similar power rating, up to 120Hz, 3.8Kw, 8.5A.

Let me know if any other data is needed.

Thanks and regards

Samir

Hi Samir,

Could you please post some captured current waveform directly since the Video can't be supported? Did you have the inductance vs. current curve of the motor? If not, you might ask for this data from the motor manufacturer to see if the inductance changes a lot with the increasing current. If yes, you have to change the motor parameters on the fly according to the spec. of the motor as the lab13 in MCSDK. The inductance table in lab13 is just for reference, you should set the inductance table based on the spec. of the motor.

Did you meet the vibration during the motor is tested with a full load? Or without loading at the motor is just connected to the pump?

Hi Yanming

Could you please post some captured current waveform directly since the Video can't be supported? 
 Am attaching the video in another format. Also attaching the screenshots of the video, in case if this time also the video doesn't work.

1: 

2:

3:

Video:

Did you have the inductance vs. current curve of the motor?
No, I don't have the Inductances Vs Current data right now. I have asked the manufacturer to provide the same. WIll share with you once I have it.

If not, you might ask for this data from the motor manufacturer to see if the inductance changes a lot with the increasing current. If yes, you have to change the motor parameters on the fly according to the spec. of the motor as the lab13 in MCSDK. The inductance table in lab13 is just for reference, you should set the inductance table based on the spec. of the motor.
What do you mean by lot here? How much variation is allowed?

Did you meet the vibration during the motor is tested with a full load? Or without loading at the motor is just connected to the pump?
Vibrations are there in all the three situations i.e only motor without pump, with pump but without load and with the pump on full load. But in each case, the frequency threshold at which vibration starts is different, till this threshold frequency there is negligible vibration in all the cases.

Regards

Samir

I don't think the rotor angle in the figure above is correct. You might try to use a lower Kp&Ki of the speed PI controller for the high speed, and tune the current controller as well.

Did you have the detailed specification of the motor? What is the rated frequency, current, and voltage of the motor?

Is it possible to increase the dc bus voltage to run the motor at the same frequency and load to see what happens? And please check if the "gMotorVars.Vs" equal or close to the USER_MAX_VS_MAG_PU.

// read Vd and Vq vectors per units
gMotorVars.Vd = CTRL_getVd_out_pu(ctrlHandle);
gMotorVars.Vq = CTRL_getVq_out_pu(ctrlHandle);

// calculate vector Vs in per units
gMotorVars.Vs = _IQsqrt(_IQmpy(gMotorVars.Vd, gMotorVars.Vd) + _IQmpy(gMotorVars.Vq, gMotorVars.Vq));