Other Parts Discussed in Thread: TMDXIDDK379D
Hello, I have a few questions about the fast current loop library parameters. I have customized the tmdxiddk379d example solution code for my own hardware and motor. I'm having an issue where I can only get the motor speed up to 3100 RPM when I give a 3600 RPM speed command. The motor is rated for 3600 RPM. At 3100 RPM, There is a large phase current that is close to the rated value, even at light loading levels. My suspicion is that this is a back EMF issue and that at these high speeds the phase voltages are no longer able to overcome the back emf produced by the pmsm, which prevents the motor from going any faster. I have a few questions about a couple of fast current loop parameters that will hopefully help me resolve this issue.
1) First, in the fcl_f2837x_tmdxiddk_settings.h file, where the motor parameters are defined, what unit for FLUX does the control algorithm assume? I've pasted below the #define motor parameter code for the TEKNIC_2310PLN04K.
// // Define the electrical motor parameters (Teknic Servomotor) // #elif(USER_MOTOR == TEKNIC_2310PLN04K) #define RS 0.381334811 // Stator resistance (ohm) #define RR NULL // Rotor resistance (ohm) #define LS 0.000169791776 // Stator inductance (H) #define LR NULL // Rotor inductance (H) #define LM NULL // Magnetizing inductance (H) #define FLUX 0.0398557819 // BEMF constant (V/Hz) #define POLES 8 // Number of poles #define ENC_SLOTS 1000 // Numer of slots in the encoder #define M_ID_START 0.2 // alignment reference d-axis current #define M_IQ_LI5 0.10 // reference q-axis current for level5 #define M_IQ_LN5 0.10 // ref q-axis current for no level5
Is the value in V line to neutral pk, V line to neutral rms, V line to line rms, V line to line pk, etc.? It seems that if I get this unit wrong it would effect my performance at higher speeds. I plug this value into the FCL_params.BemfK value in my main script, as you can see in the code below. My FLUX has currently been set at 0.5178 V line to neutral pk / Hz.
2) My second question relates to the calculation of the maxModIndex. From the initFCLVars() function below, my maxModIndex value is 0.879. This is calculated from a switching frequency of 60 kHz using single sampling mode and the default FCL_COMPUTATION_TIME of 1.00 (us). This seems like it would also be an issue at higher speeds with larger back emfs generated by the motor, because the modulation index directly impacts the inverter output phase voltages. It seems to me that maxModIndex should be as close to 1 as possible, but the only way to do that is to further decrease the FCL computation time to be as close to zero as possible, which obviously can't reach zero.
//Function that initializes the variables for Fast current Loop library
void initFCLVars()
{
#if(SAMPLING_METHOD == SINGLE_SAMPLING)
maxModIndex = (TPWM_CARRIER - (2 * FCL_COMPUTATION_TIME)) / TPWM_CARRIER;
FCL_params.carrierMid = maxModIndex * INV_PWM_HALF_TBPRD * 0x10000L;
#elif(SAMPLING_METHOD == DOUBLE_SAMPLING)
maxModIndex = (TPWM_CARRIER - (4 * FCL_COMPUTATION_TIME)) / TPWM_CARRIER;
FCL_params.carrierMid = INV_PWM_HALF_TBPRD * 0x10000L;
#endif
FCL_params.adcScale = ADC_PU_PPB_SCALE_FACTOR * LEM_TO_SHUNT;
FCL_params.sdfmScale = SD_PU_SCALE_FACTOR * SDFM_TO_SHUNT;
FCL_params.cmidsqrt3 = FCL_params.carrierMid * sqrtf(3.0f);
FCL_params.tSamp = (1.0F / SAMPLING_FREQ);
FCL_params.Rd = RS;
FCL_params.Rq = RS;
FCL_params.Ld = LS;
FCL_params.Lq = LS;
FCL_params.BemfK = FLUX; // 0.8 default
FCL_params.Ibase = BASE_SHUNT_CURRENT; // LEM sensing is scaled to match
// with shunt sensing
FCL_params.Wbase = 2.0 * PI * BASE_FREQ;
FCL_params.wccD = CUR_LOOP_BANDWIDTH;
FCL_params.wccQ = CUR_LOOP_BANDWIDTH;
return;
}
So a few questions about this.
What is the range on which the modulation index is defined? Does it go from 0 to 1?
Where does the equation come from the calculates the maxModIndex? I would like to learn more about this.
Can I overwrite the equation and specify my own maxModIndex (e.g., set it to 1)? What will happen if I do so?
Finally, what is the carrierMid value and what does it do? My value calculates to ~24 million.
Any assistance or direction towards further TI reading resources on this would be much appreciated. Also, I just noticed that 3100 RPM is ~86% of my full speed (3600 RPM), and given that my maxModIndex is 0.88, it seems that the modulation index is what is limiting my speed?
