P/T accuracy

I'm sure this has been discussed before, but how does one deal with a college and repeatability? I was to do a 2x6 college (which looked great) but then the client also wanted another look around the room for opening. I didn't think about it until rehearsal and clicked back to the college and all the alignment was off. Not by a lot, but enough that the college couldn't work.

Thought it was my programming, updated and tried again and same situation. Any good advise (besides not moving them)
  • Stephen,

    The best thing to do in this situation is to update each position palette in sequence of the show/cue list.

    For example...

    If cue 1 is your walk-in and cue 2 is your shift to the main look...

    Play cue 1.

    Play cue 2.

    Make very minor position modifications to the position palette in cue 2.

    Repeat.

    This helps to overcome possible hysteresis that is occurring in the position motors. They key here is that the position of a fixture can be affected not only by the absolute position palette but also by the path it takes to get to that position palette. (This is very noticeable in mirrored fixtures, as small discrepancies in pan/tilt motor positions result in much larger/more noticeable position inaccuracy) If you're getting a bad alignment, getting frustrated, releasing your cuelist, sending the fixtures to their 50/50 position, and then updating the position palette, you will probably continue to get a bad alignment coming out of your walk-in look.

    A related note (and something that you may have done already) is that it might be wise to make the position parameters on the DLs non-releasable so that they do no return to their 50/50 positions when you release your list.

    In the case of DLs, I believe that the hysteresis issues were more prominent in DL.1s and early DL.2s than in DL.3s, but I'm not sure on that statement.

    Hopefully this will help a little.

    Good luck and good show!

    PG
  • Its not an issue of the release, its an issue that the college is on the stage wall, the walk in look was 180 degrees in the other direction, much lower down. So its not a minor move, its a major one. Second, the simple, minor position mod, isn't that easy when I have 2000 people in the room after the lights go down and need to show a grid so the moving animated text doesn't mis align. We ended up hanging additional fixtures to combat this, but most of us (I included) though that was one of the point to hanging these instead of fixed projectors.
  • On each side of the Tilt mechanism there are three screws that can loosen enough to allow some play or backlash in the mechanism.

    With that said, making sure the fixture is tight ahead of time is a great start.
  • Hi,

    Now i do program mostly DML-1200 but a lot with Collage and my experience is that when i have this problem i replace all position information as hard values and NONE in pallets.
    Now you could say that this should not matter but i find that the workmethod supports the hysteresis of the fixture, forcing a return from the same direction every time for the new position.

    Can you still try this? Or is the event over?

    I do not think there is anything wrong with the fixtures.

    I hope it works out.
  • Stephen,

    I totally understand. Just trying to offer a one helpful idea on how to improve the recall.

    PG
  • I have found that using a combo of touch & replace with
    the correct mask layers in place when
    updating DL2/DL3 pallettes seems to give a better
    result than updating pallettes,along the same lines as
    what abbé is suggesting.
    Totally replacing pan/tilt values works better IMHO.

    Best c
    I
  • if I have to make large movements I use a fade time like 10secs from A to B, to prevent trussing or lamp-housing from moving/shaking.

    You might also adjust your most important position for the fixtures in the "edit-fixture"-sheet so they always come back in grid when released. If you then have a off-position you know for def it is in the fixture itself
  • It is also a good idea to do a Calibrate Motors when prepping your DL gear.....this will ensure that the encoders are tracking the motion properly.

    Hope this helps. :)
  • Stephen,

    A couple of things I always recommend to make sure you get the best accuracy/repeatability out of the fixtures:

    1) Clean the encoders. Because the DL.3s/DL.2s use an optical encoder for high accuracy, they are also more sensitive to dust and other debris. This is why we added dust covers to the encoder wheels, but they can still get gunk on them. A soft cloth with isopropyl alcohol is best, and make sure you clean both the top and the bottom.

    2) Make sure the pan belt is tensioned properly. Over time, the rubber teeth on the belts wear away creating a bit of slack in the belts (this also creates some of the dust that gets on the encoders!). All you have to do is loosen the 4 bolts that hold the pan motor plate to the chassis, allow the spring attached to the plate to tension the belt, and re-tighten the 4 bolts.

    3) After the encoders are clean and the pan belt is re-tensioned, run motor calibration as Marty stated. Be sure the fixture is very still for this as the calibration routine is very sensitive (no swaying truss, people bumping the fixtures, etc). Calibration should take less than 2 minutes.

    Also, as Chris stated above, it's not a bad idea to make sure that the set screws on all of the pinion gears are tight to the motor shaft.
  • Thanks for the advise. As they are all rented, I'll pass the info on.