Getting Started Guide

 

This web page is going to detail the minimum steps need  to  install, setup, and instigate your first robotic Foucault test.  Advanced options, theory of operation, and other minutia will be documented on other WebPages

 

Here’s how to get started:

 

  1. Make sure your video capture system is working, try using the capture software that came with the hardware as a test bed
  2. Make sure Mel’s driver board is OFF until after the code is initialized (windows likes to turn on parallel pins on its own, these will be turned off once RTAFT initializes)
  3. Download and run setup.exe from this package http://lerch.no-ip.com/atm/rtaft_release/rtaft.zip
  4. Go to the windows Start / Programs / Rtaft / and Select Rtaft or Rtaft_2K_Xp depending on your OS
    1. If you get this:
    3. Your running Win NT / 2K / XP and selected the wrong start menu link!
  5. At this point RTAFT should start loading, and you will hear that sexy Star Trek computer voice J
  6. Next you should get this Warning:
    2. After Clicking “OK” you’ll be taken to the main configuration dialog box that looks like this:
  7. All right, now were getting somewhere J
  8. Next step will be to turn on Mel’s Driver board
    1. After powering up the board, reach over and try to manually turn one of the stepper motors!
    2. If the stepper motor is hard to turn (The coils are all turned on!)

                                                              i.      Check the “Invert Output”

                                                            ii.      Re-Check the motors, they should turn easily now!

    1. Were done with this tab on the dialog screen
  1. Next, select the “Motor Speed” dialog tab, it will look like this:
    2. ** Moving the slider bar LEFT makes the motor spin Faster!  Moving it RIGHT causes it to spin slower!
    3. Note the Purple Arrow; this is the MOST important slider.  This variable is the one used during Automated testing, the other two only work during manual platform slews. For the most part, you can ignore the top and bottom slider for the time being.
    4. To set the Top Speed slider, press and hold  a motion button on Mel’s handpad, or press one of the keyboard keys I/J/K/M
    5. BTW, don’t worry about what motors are going on/off with handpad or keyboard presses, we’ll fix that later
    6. For the moment, your only goal is to ABSOLUTLY, POSITIVLY set the top speed of the motor BELOW the point where they start skipping steps or stalling!

                                                              i.      If you going to make an error, MAKE sure the error is on the SLOW side!

                                                            ii.      A decent test is to lightly hold the motor shaft while slewing at top speed, and see if it stalls.

    1. For all intents and purposes you can ignore the top and bottom sliders, you can come back and play with these later if you want J
  1.   Next on the list of things to do is get the motor axis and directions setup correctly
    1. Selct the “Axis Setup” tab, the dialog will look like this:
    3. Now its time to sort out the motor controls

                                                              i.      First lets make sure pressing keyboard I/M (or Handpad Up/Down) actually moves the longitudinal motor (ya know the one that makes the camera and knife edge move TOWARDS or AWAY from the mirror!) 

                                                            ii.      Once you get the right motor assigned to the right keys, we need to get them rotating in the right direction!

                                                          iii.      Press the keyboard I key (or handpad UP), the camera and knife edge should move Towards the mirror! If it doesn’t select the “Reverse To/From Mirror Motor” option and try again

                                                           iv.      Next we need to get the lateral motor spinning the correct direction.  Press keyboard J key, the camera and knife should move to your LEFT as viewed while standing behind the Foucault platform, looking downrange towards the mirror.  If it doesn’t, selsct the “reverse L/R Motor” option, and try again J

                                                             v.      Now, if you’ve done everything correctly, Handpad UP moves platform towards the mirror, Down moves it away,  Left moves it left, and right moves it right!  BTW, if you start doing some troubleshooting later on, purposely changing this variables in an effort to “Fix” a problem, most likely will not solve the problem, but only make it worse!

  1. OK, were almost done with setup, only  a few more options to run thru.  Of which next on the list is the Foucault Setting dialog box.
    1. Here’s the dialog box, which has a LOT of options! (sorry, this stuff only appears complicated, fortunately you only need to do this once!)
    3. Lets start with the “Set Distance per Full Step” button, which will take you to the following dialog box:

                                                              i.     

                                                            ii.      You’ll need to calculate the distance traveled for ONE FULL STEP of your stepper motor on each axis.

                                                          iii.      Enter this data into the corresponding text box

                                                           iv.      *** Make sure you select if the measurements or in INCHES or Millimeters! ***

                                                             v.      OH, BTW J  Selecting which unit to use will force this measurement system for all subsequent data entries!!  IE You can’t select Metric on this screen, then enter mirror diameters in INCHES later on (Well, you could but the saying “Garbage in / Garbage out” will become very apparent to you J

                                                           vi.      Once completed, click “OK” and you will be returned to the previous dialog box.

    1. Next on the list, click the “Set Test Repetition Options” button. 

                                                              i.      Select the option to do 4 sets of reading per test.

                                                            ii.      You can change this later by going to “Test Options / Repetition Options”

                                                          iii.      The reason for setting it to 4 for the moment is, when you do your first few tests, we want to make sure you can get repeatable numbers.  If there are any problems with backlash or Air Currents, these will show as changing readings for each zone.  BTW, the code should auto compensate for backlash, but ‘Should’ and ‘WILL’ are two separate things J

                                                           iv.      Click OK to exit and return to the previous dialog box.

    1. Next up is telling the code what side you knife edge is on.  Press the “Set knife edge location” button and you will get the following screen:

                                                              i.     

                                                            ii.      This is another one of those REALLY important setup questions (Not that all the other’s were not as equally important!)

                                                          iii.      If you knife edge (as viewed standing behind and above the Foucault platform) cover the LEFT side of the light source and camera lens, then you knife is on the left.  If the knife cover’s the RIGHT side of the light source and knife edge, your knife is on the right! (seems simple enough, but easy to get confused over!)

                                                           iv.      Select the image representing your system, then click OK (were almost done!)

    1. OH BOY, here’s a fun one ;)  This next dialog screen is how we tell RTAFT exactly where to save the data, HOW to save the data,  and what to do AFTER the data is saved!  Below is this ever so much fun to program dialog box!:

                                                              i.     

                                                            ii.      This screen is broken down into three categories

1.      Where to find figure45.exe, should the code save figure45 files, and after saving the file should it launch figure45?  BTW, figure45.exe doesn’t appear to run under Windows XP (which really sucks, but oh well, fortunately Figure45 and Sixtests aren’t required for RTAFT to record zone measurements!)

2.      Next section covers Sixtests.exe options, plus one other possibly entertaining option of Launching my 3D mirror surface code after exiting Sixtests

3.      Finally, the more important section, where to SAVE the test result data after a test is complete. 

4.      On any of these three options, you may either use the Browse button, or you can manually type the path in yourself. 

5.      One last note before were done with this section.  For each mirror you test, a new folder is going to be created in the “Save Data location”.  Inside each folder will be sub-folders containing the Figure45, Sixtests, Plain Text, and Image data.  If you do a LOT of mirrors, this folder can get rather large (FYI, in the last 12 months of testing each Saturday at our ATM lab, I’ve generated nearly 760MB of data, of which most of it is image data!)

6.      OK, that it for this page, click OK to return to the main setup dialog box

    1. Here’s a simple one for you, Tell the code if you have a fixed or moving light source! J (see that was easy)
    2. Next and Final Tab, “Zone Intensity”

                                                              i.      First off, leave this one alone for the moment!  It will probably be ok as-is.

                                                            ii.      However, you may need to revisit this page once you try testing your first mirror, so I’m going to try and give you the low down on what these variables do.

                                                          iii.      Its all about “Shades of Gray!”  Well, actually its about shades of gray and pixel intensities to be precise. 

                                                           iv.      What is a “Shade of Gray?”  Well, its that area between really BRIGHT and really DARK!  Unfortunately this is a pretty broad area! 

                                                             v.      For RTAFT, really bright means a pixel intensity of 255, while really dark means a pixel intensity of 0.  To do a Foucault test we need to work with gray values.  Of course this begs the question “Do we use dark gray, medium gray, or light gray?”

                                                           vi.      The answer to that is still rather vague, but I’m finding that dark gray works best.  More precisely a pixel intensity centered at 45 above the noise floor really seems to work well. 

                                                         vii.      However, there is a caveat.  We can’t tell RTAFT to use ONLY a pixel intensity of 45, as odds are it won’t ever find that precise level.  Instead we tell RTAT to find a level somewhere NEAR 45.  I’ve found that +/- 15 seems to work out rather well for my platform.  Consequently the default values for this page are 60 for max intensity, and 30 for min intensity.  This will mean RTAFT will adjust the knife to get a shade of gray that has a pixel intensity between 30 <-> 60 for determining if a zone is nulled or not. (See that was simple J YEA RIGHT!)

  1. CONGRATULATIONS!  Your done with setup!  If you ever need to get back to this screen, just select “Platform Settings / Platform and Foucault test setup”  Now all you have left to do is setup for your first test run.  Since this page is getting rather long winded I’ll refer you to the “Testing Procedure” page, found on the left.

 

James