• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 82 subscribers
  • Views 810 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Lab05b question

Shay Stockstad
Prodigy 240 points

Other Parts Discussed in Thread: TIDA-00643, MOTORWARE, INSTASPINUNIVERSALGUI

I'm using a custom TIDA-00643 design and working through the steps as outlined in the tiduak1.  I'm using a hobby motor that is very similar to the one in the tiduak1 appnote except that it is a 960kV motor with Rs=.07 & L=4.7uH  The issue is that in lab05b is that since I don't know the inertia values of the motor and as a result can't calculate the Kp_spd/Ki_spd values, I attempt the trial and error method but I'm unable to find values of Kp_spd/Ki_spd which give the motor any sort of "spring" behavior.  It just rotates with little to no resistance with no sort of spring back behavior.  

Since this design uses a F2802xF device that means I can't use lab05c to identify the inertia of the motor correct?  If that's the case would running lab05a and carrying out the procedure as described here be a better option?

And in general is it preferred to run these labs using the universal gui webapp or through CCS debug mode?  I've tried both methods with lab05b.

Finally, I've been running these labs with the prop on as tiduak1 doesn't explicitly state whether the prop should be on/off for these labs.  Is it recommended to have the prop on?

52520.user.h

Best regards,

Shay 

  • 0
    TI__Guru** 119680 points
    you should use proj_lab02c to ID the Rs, Ls, and Flux of your machine. But it sounds like you've already done that.

    for proj_lab05b I wouldn't attempt to follow the lab guide for how to tune the speed controller.

    we will be introducing a lab in the next motorware that shows you how to do step response testing using the graphs inside CCS. In general I can tell you that for your type of motor, a good starting point is to take the default values for the speed Kp and Ki and divide by about 4 to 8. This should get you pretty close to a well tuned speed controller (which you can test at various speeds).

    The INSTASPINUNIVERSALGUI is just an instrumentation of the embedded code. I prefer to use it so I can see more variables at once in a way that makes sense, but I use it inside of CCS. You always need to build the project in CCS.

    I would run the motor first through all the labs you want to use w/o a prop, then go back to 5b or 10a and attach the prop. Note that with propellers you should try to attach the motor hanging OUT in free space. If you have it on a table top the airflow bouncing back off the table makes a very large difference in the loading on the propeller.
  • 0 in reply to ChrisClearman
    Prodigy 240 points
    Yes, I ran lab02c to get these values and they come out fairly close to what I've measured from an LCR meter as well.

    Any timetable on when the new lab will come out? I'm planning to test with other motors mostly in the lower kV ranges 250kV - 800kV most of which will have a lower RoverL value than what I'm currently testing.

    I agree the GUI is more readable/convenient, so far I've been running it standalone (copying/renaming .out file from CCS).
  • 0 in reply to Shay Stockstad
    TI__Guru** 119680 points
    trying to get it out by end of October.
    try zipping up the folder
    C:\ti\guicomposer\webapps\InstaSPIN_F2806xM_UNIVERSAL

    then in CCS add GUI Composer from the app store and import the zip
    You can then run it inside CCS and not have to deal with copy/paste/rename/launch