TMS320F28069M: SPINTAC status problem in position control

Part Number: TMS320F28069M

Hello,

I am using 28069M with my customized PCB board.  Initially, I write my source code based on the TI example proj_lab13b.  This example project is used to realize position control, originally it works. But when I add more and more my source code on it, the position control does not work when servo on. In position control mode, the current noise on the motor can be heard when servo on, the motor rotor is free, it looks that there isn't any torque generated on the rotor, at this moment, when I examine the Spintac.posctlstatus, it is "st_ctl_idle", but not "st_ctl_busy".  Since when I initially start the source code programming on the board, I have examined that when servo on, the Spintac.posctlstatus is "st_ctl_busy", the motor work in position control mode. So, currently, it is "st_ctl_idle", I believe something wrong. But I have looked through the whole source code, the most of them are similar to the original source code of proj_lab13b. The figure below is the screen shot when servo on.

I have tried every methods to try to fix that, but I failed. I believe something wrong in the source code, because when I run the motor in velocity mode, it works. 

The CCS9.0 and incremental encoder is used in the project. 

Looking forward to your response and appreciated for any comment.

Thank you.

Hi,

I have merged these two threads as they are the same request.

Regards,

Vince

Thank you. I am stuck  here for more than one week. Expect TI expert to help me on that.

Make sure that the motor parameters are identified correctly with lab02, and the inertia and friction are identified with lab12a before running lab13b.

1. Tune the control bandwidth (USER_SYSTEM_BANDWIDTH) according to the system.

2. Tune the trajectory curve.

Thank you. Yanming,

Actually, I know the motor parameters and defined them directly in user_j1.h without necessary to identify them, shown as below:

#define USER_MOTOR_TYPE MOTOR_Type_Pm
#define USER_MOTOR_NUM_POLE_PAIRS (4)
#define USER_MOTOR_Rr (NULL) 
#define USER_MOTOR_Rs (0.165)
#define USER_MOTOR_Ls_d (0.00045)
#define USER_MOTOR_Ls_q (0.00045)
#define USER_MOTOR_RATED_FLUX (0.081) 
#define USER_MOTOR_MAGNETIZING_CURRENT (NULL) // not needed for PM motor, always NULL.

#define USER_MOTOR_RES_EST_CURRENT (1.5) 
#define USER_MOTOR_IND_EST_CURRENT (-1.5) 

#define USER_MOTOR_MAX_CURRENT (15.0)

#define USER_MOTOR_RATED_CURRENT (7.5) 

#define USER_MOTOR_FLUX_EST_FREQ_Hz (20.0) 
#define USER_MOTOR_MAX_SPEED_KRPM (5.0)
#define USER_MOTOR_RATED_SPEED_KRPM (3.0)

#define USER_MOTOR_FREQ_LOW (20.0)         // Hz - suggested to set to 10% of rated motor frequency
#define USER_MOTOR_FREQ_HIGH (200.0)       // Hz - suggested to set to 100% of rated motor frequency (3000rpm = 200 ElecHz)
#define USER_MOTOR_FREQ_MAX (240.0)        // Hz - suggested to set to 120% of rated motor frequency
#define USER_MOTOR_VOLT_MIN (10.0)            // Volt - suggested to set to 15% of rated motor voltage
#define USER_MOTOR_VOLT_MAX (36.0)           // Volt - suggested to set to 100% of rated motor voltage

#define USER_MOTOR_ENCODER_LINES (2500.0)
#define USER_SYSTEM_INERTIA (0.00488)        //default value 0.01; unit: A/(krpm/s)
#define USER_SYSTEM_FRICTION (0.01)           //default value 0.01;

As for USER_SYSTEM_BANDWIDTH tuning, yes, I have tried it but useless to the issue.

for " Tune the trajectory curve" you mentioned, how to tune? in original proj_lab13b exmaple, where can I find it so as to tune? 

In addition, I have thought about to comment out some my source codes so as to make the project similar to the original proj_lab13b to have a test whether it can works in position mode, but I find it is hard to me to do that, since I have added and modified a lot of codes and related definitions including .h files based on proj_lab13b example. Now the whole code size is so big that it have to be implemented in Flash.

This is my modified F28069M.cmd:

I am stuck here for a long time, hope TI expert help me to solve the issue.

Continue Asking TI expert to help me to solve the issue.

Currently, all jobs are stuck here. including firmware coding, next version PCB layout due to the position control does not work.

I have tried all kinds of methods to try to solve the issue. for example, I have tried different CCS version, like CCS8.0, CCS9.0, etc.  I have tried using different compiling tool versions in CCS, like version 6.2, version 15.2, version 20.xx, etc. But all fails.

Hope someone comment on that and may be your single word can notify me to do something about it.

Thank you.  

user5912823 said:

I know the motor parameters and defined them directly in user_j1.h without necessary to identify them

Recommend you'd better identify the motor parameters on your own board.

user5912823 said:

Now the whole code size is so big that it have to be implemented in Flash.

Do you copy the time-critical code and Flash setup code to RAM?

ok, thank you. To identify the motor parameters on my board, I need to run proj_lab2 that TI provided. I will run it.

For the time-critical code and Flash setup code to RAM, I think it is copied because I haven't changed any code related it.

The below is the code before main():

The below is code in the start of main()

The below is the code in F28069M.cmd

I have use proj_lab2a to identify motor parameters.

But the results looks big error than the true values from motor datasheet.

The identified Rs=0.38 ohm; the true Rs = 0.165 ohm;

The identified Ls = 4.1*10^-6 H; the true Ls=  0.00045 H;

The identified flux = 0.39; the true flux = 0.081 V/Hz;

And sometimes, the identified Ls = 1.2*10^-9 H; 

Sometimes, when flux identified is over, the "CTRL_STATE_ERROR" AND "EST_STATE_ERROR"  are reported, then the identification is halted.

I have tried increasing PWM freq. but the identification result looks unchanged greatly.

Is the identification process normal?

You might try to use the lab02c since the motor is a low inductance motor. And tune the following identification parameters according to the motor specification.

#define USER_MOTOR_RES_EST_CURRENT     (1.0)                               // A - 10-30% of rated current of the motor

#define USER_MOTOR_IND_EST_CURRENT     (-1.0)                             // A - 10-30% of rated current of the motor, just enough to enable rotation

#define USER_MOTOR_MAX_CURRENT           (5.0)                               // A - 30~150% of rated current of the motor

#define USER_MOTOR_FLUX_EST_FREQ_Hz     (20.0)                           // Hz - 10~30% rated frequency of the motor

OK, thank you . Actually I have tried yesterday using lab02c for low inductance motor identification. Today I will try again and tune the parameters. Any result will let you know.

I have a test using lab02c, the pwm freq.=45kHz;

The identified motor para. are as below:

Rs= 0.09 ohm (datasheet value is: 0.165 ohm)

Ls=0.0002 H (datasheet value is 0.00045 H)

Flux=0.04 vPhz (datasheet value is 0.081)

I use the identified motor parameters in my project, the position control loop does not work as before. The "PosCtlStatus" is still  "ST_CTL_IDLE" but not "ST_CTL_BUSY".

So the root cause of position control loop fails are still not known. Now it looks that is not caused by wrong motor parameter settings at least.