How to change retract waveform?

Hardware
2005-10-27
2012-09-14
  • MarioGiacomino
    MarioGiacomino
    2005-10-27

    Hi, I would like to change the retraction waveform. I have looked at the DSP code and it seems as though the approach waveform is interconnected with the retract waveform. Do you have any suggestions on how to separate the retract waveform from the approach waveform?
    Thanks in advance,
    Mario

     
    • Percy Zahl
      Percy Zahl
      2005-11-16

      GXSM and the SR are now offering several mode of approach and coarse motion.

      The Cycloid is now also availabe and in addition any arbitrary waveform can now be loaded on the fly by the DSP. For that the sranger-hwi generates wavforms and download them to the DSP as needed. The user has currently the choice of the old style parabolum, a simple sawtooth, a sine wave (for whatever and testing purposes) and the cycloid!

      Also the output channel (AIC) can now simply set to XY-Scan, XY-Offset or X-Motor.

      It's all in the newest version of GXSM in CVS.

      And if needed, it's now very simple to add any custom waveform.

       
    • Adam Stieg
      Adam Stieg
      2005-11-21

      That is fantastic! The parabolic waveform has been working but the up motion was still a bit scattered. I will have to try the cycloid ASAP. Thanks so omuch for implementing this!

      -AZS

       
    • Percy Zahl
      Percy Zahl
      2005-10-27

      This will need extra code or the arbitrary wave forme download variant -- both new DSP code for that section.

      However, please let us figure out first some details to make sure it will work as expected later:

      1) I expect you know the parameter of you working wave form. So, 1st I do not care about the shape, but:

      a) The peak-to-peak amplitude you need form the DSP (before you amplifier). [V]

      b) The repetition frequency [Hz]

      c) minimal required slope at the "jump" section [V/s]

      2) The wave forme itself, do you have a math. expression for it? Or do you have a "sample" of one section of your shark-fin or cycloid?

      Some term like this can be implemented on the DSP:
      (((((C6t + C5)t + C4)t + C3)t + C2)t + C1)t + C0

      So, actually I just remembered some one did a cycloid implementaion with the DSP already. I'll try to get in touch with him. And do a CC to you. Does that sounds OK?

       
      • MarioGiacomino
        MarioGiacomino
        2005-10-28

        Hi Percy, thanks for the quick response. To answer your questions:
        1a) I would like to use the peak-to-peak amplitude (Set by the GUI of GXSM) that you use in your code. Thus allowing for adjustment during operation.
        1b) I would like to use the repetition frequency that you use in your code for the same reasons as 1a.
        1c) The waveform has been worked out by someone else. I am not sure what the "jump" section is but I am sure I can talk to them to find that out.
        2) I do have a mathematical representation of the waveform so I can put that on the DSP. The only problem is how to insert it into your DSP code such that it uses as many of your variables (set by the GUI of GXSM) as possible.
        Thanks again for any help, it is much appreciated.
        Mario Giacomino

         
        • Percy Zahl
          Percy Zahl
          2005-10-28

          Sorry, but this kind of answers do not help at all.

          I'll wait for the code from the MPG GXSM user (got a already response) and will see how he did it -- A Cycloide should be the best we can do for any slip-stick driven motion and thus I'm indending to replace the current parbolic approximation by it and all of us should be happy.

          -PZ

           
    • Adam Stieg
      Adam Stieg
      2005-10-28

      Good morning gentlemen! I have been following the continued communicade and just wanted to mentio that while a cycloid type waveform is effective for moving a slip stick drive it does also give larger step sizes than your parabolic wave due to the slope of the wave at the "jump" region. This might be fine in many cases, but if you are trying to find a generic solution it would be nice to be able to have the approach and retract be separate and possibly different. I am sure that this makes the process a bit more difficult but when the step sizes are meant to be small on approach, the cycloid is tricky, especially when moving vertically. In any case, I am at your mercy with this but wanted to give my tow cents. Have a great day.

       
      • Percy Zahl
        Percy Zahl
        2005-10-28

        Not sure if you know, but if you are configuring GXSM to the "Mover" Variant (Pref->User/SliderControlType = mover), then you even today get several folders with different Speed/Amp/Steps settings! And you can choose a different one for the (always) manual Tip-retraction!

        However, this will apply to all direction, but change is simple.

        -PZ

         
    • Adam Stieg
      Adam Stieg
      2005-10-29

      Yes I did notice that the other folders and their affect of the waveform. On a side note, what do the Z++ and Z-- timings do? I have been playing with my approach but am not sure what these are doing. Thanks!

       
      • Percy Zahl
        Percy Zahl
        2005-10-29

        Z++/-- are not used by the SRanger at all.
        Good hint, I'll remove them since in the newest version of Mover-Controls is SRanger specific only.

        -PZ

         
    • Adam Stieg
      Adam Stieg
      2005-10-29

      That makes sense. It didn't seem to be doing anything when changed. On another note, I posted a while back on the help forum regarding the mover pugin not being loaded when I use the newest version of the SR software i.e. ../testing/Debug/FB_spmcontrol.out when used in combination with the SPM Signal Ranger Testing HWI. If I use just the SPM HWI the module is loaded but I get a version mismatch error when I load Gxsm2. It seems that this error really doesn't affect anything in the functionality but I wanted to make sure. Thanks.
      -AZS

       
    • Adam Stieg
      Adam Stieg
      2005-11-01

      Well after crashing may tips I have found the correct parameters to get my slider to move up vertically with the current parabolic waveform.The step sizes aren't as clean or small as I would like but beggars can't be choosers. I am suprised it works, but it does! Thanks for all the feedback and help with this. Now onto imaging...

       
  • fuku70
    fuku70
    2011-04-11

    Hi, I am using a uni-polar stacking piezo for coarse approach of inertial
    slider. So I need the cycloid wave of only positive voltage (0 to 10V). When I
    press Z+ button, the wave appears positive voltage and O.K. But when I press
    Z- button, the wave appears negative voltage (-10 to 0V). So, I need to shift
    the negative voltage to positive by adding +10V. Do you have any suggestions
    how to do this? Thanks in advance.

     
  • Percy Zahl
    Percy Zahl
    2011-04-14

    So far you would have to adjust the one of the "user" wave forms generated
    (calculated) by the HwI plugin to meet your needs.

    see:

    sranger_mk2_hwi_mover.C