resolver and kflop - Page 2


Page 2 of 2 FirstFirst 12
Results 21 to 25 of 25

Thread: resolver and kflop

  1. #21
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4042
    Downloads
    0
    Uploads
    0

    Default Re: resolver and kflop

    Hi RUSJIaN,

    Very good plots. Do you see how to calculate the Sine/Cosine Ratios now?

    But the amplitudes are low. Y is very low:

    50 / 2048 x 10V = 0.25V

    Is the Y Output voltage really that low?

    What are the time and voltage scales of your oscilloscope?

    It could be that we are sampling the ADC near the zero crossing of the signal. Sampling the signal near its peak will result in a higher resolution measurement.

    Do you have a dual trace scope to look at when the stimulus (SetBit/ClearBit) switches and the output signal at the same time? Or use the stimulus as a trigger?

    You might shift the phase angle of the stimulus forward to see if that helps. Move the SetBit and ClearBit up before one of the WaitNextTimeSlice(). That should be about a 180us shift in time.

    HTH
    Regards


    Does your

    Regards
    TK http://dynomotion.com


  2. #22
    Registered
    Join Date
    Oct 2015
    Posts
    15
    Downloads
    0
    Uploads
    0

    Default Re: resolver and kflop

    Quote Originally Posted by TomKerekes View Post
    Hi RUSJIaN,

    Very good plots. Do you see how to calculate the Sine/Cosine Ratios now?
    #define RATIO0 (233.0f/312.0f) // size j/size k
    #define RATIO1 ----------------------------------------
    #define RATIO2 (222.0f/266.0f) // size p/size q

    Quote Originally Posted by TomKerekes View Post
    Is the Y Output voltage really that low?
    What are the time and voltage scales of your oscilloscope?
    time= 50 us/cell
    voltage=0.1 V/cell

    Resolver Y-axis actually defective. I put a spare motor on the axis Y, the problem disappeared. Now all axes give comparable data. I checked oscilloscope.

    Quote Originally Posted by TomKerekes View Post
    Do you have a dual trace scope to look at when the stimulus (SetBit/ClearBit) switches and the output signal at the same time? Or use the stimulus as a trigger?
    No, my oscilloscope has only one channel.

    What concrete steps are taken to increase the resolution of measurements?



  3. #23
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4042
    Downloads
    0
    Uploads
    0

    Default Re: resolver and kflop

    Hi RUSJIaN,

    Your Ratios look correct.

    With the oscilloscope showing 0.26V and the ADC reading 0.25V the ADC sampling must be close to the peak. So that is good.


    I'm not sure what you can do to increase the resolution if you are sampling near the peak. What are the amplitudes of XYZ now?


    222 counts x 10V / 2048counts = 1.1V


    What does the scope show? You might increase the amplitude of the stimulus if the resolver can handle the voltage/current. Otherwise you would need to add an amplifier.


    I suspect the amplitude and resolution are enough. The resolution should be:

    222 * 2 * PI = 1400 "counts" per revolution.

    With a lead screw of 5mm pitch that would be 3.5um

    Does it all work acceptably?

    Regards

    Regards
    TK http://dynomotion.com


  4. #24
    Registered
    Join Date
    Oct 2015
    Posts
    15
    Downloads
    0
    Uploads
    0

    Default Re: resolver and kflop

    Hi Tom.
    In the past, the code had two consecutive command WaitNextTimeSlice (). When JOG command to position all axes added 490 pulses, but at the end of the movement 490 pulses also removed. As a result, the axis jerk.
    Now I have added code three consecutive command WaitNextTimeSlice ().
    Now, when the command added JOG 3-10 pulses, that is acceptable.
    As this change has affected RATIO all axes, now their size reaches about 1000 which will increase the accuracy of movements, it is now the maximum error is 30 pulses. Lead screw of 10 mm, one turn of 5000 pulses (resolver is connected via a multiplier of 1: 5) => error is 0.06 mm, it is enough.



  5. #25
    Member TomKerekes's Avatar
    Join Date
    May 2006
    Location
    USA
    Posts
    4042
    Downloads
    0
    Uploads
    0

    Default Re: resolver and kflop

    Hi RUSJIaN,

    I believe you are saying it is working. Congratulations!

    Happy Holidays

    Regards
    TK http://dynomotion.com


Page 2 of 2 FirstFirst 12

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

resolver and kflop

resolver and kflop