Page 1 of 2 12 LastLast
Results 1 to 10 of 17

Thread: How to test/measure new projector's specs?

  1. #1
    Join Date
    Nov 2019
    Posts
    26

    Default How to test/measure new projector's specs?

    I just bought a budget projector from China as sort of a starter level unit that should be able to serve my needs for a little while. I want to now test it to see if it meets the specs the manufacturer claims, and learn something in the process.

    Looking for best practices on how to measure.

    I plan to set low power (so as to not be overwhelmed by brightness) to measure the beam diameter with a caliper or ruler. Maybe there's a better way?

    Divergence I can measure by measuring the beam diameter at a long distance. Makes sense to me.

    But galvo speed has me stuck. I believe it's supposed to be 36kpps at a certain angle. Would I create a square or line pattern and see if it flickers or rounds the corners? Pretty lost on this one.

    For reference I have lasershowgen software and a Helios DAC.

  2. #2
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    You need to be able to display the ILDA test pattern to check the true speed of your scanners. You also must ensure that your software does not try to optimize the ILDA test pattern file before displaying it. I'm not familiar with the LaserShowGen software, so hopefully someone else can comment on whether this applies to that software or not.

    First, set your scan speed to 36,000 points per second, which is the advertised speed of your projector. Then turn down your size control to around 10% or so for starters. Position the projector exactly 2 meters from a flat wall or a display screen. Now display the ILDA test pattern.

    You must adjust the size control in your software until the height and width of the test pattern are both exactly 28 cm. With the projector aperture exactly 2 meters from the display surface, a pattern width of 28 cm equates to a scan angle of 8 degrees, which is the spec for the ILDA test pattern.

    Now look at the displayed image. If it looks like this:

    https://i.imgur.com/wbbb8Ml.jpg

    Then congratulations! Your scanners are properly tuned and the projector is displaying the ILDA test pattern correctly at 8 degrees scan angle at the advertised speed of 36Kpps.

    At this point you may want to increase the size control in software to see how far beyond 8 degrees you can push them before the center circle starts to pull away from the sides of the center square. But don't get carried away. You might get as far as 11 or 12 degrees if you are lucky. (9 degrees is 31.4 cm wide, 10 degrees is 35 cm, 11 degrees is 38.6 cm, 12 degrees is 42 cm)

    As soon as you see the circle starting to pull away from the sides of the square, stop. Pay attention to the size setting in your software. This is your "safe" size setting for virtually any image, as the ILDA test pattern is designed to be excessively hard on the scanners. Note that you can still scan most images at a much wider setting than this "safe" setting, because most images are nowhere near as difficult to display as the ILDA test pattern is.

    On the other hand, if your initial display of the ILDA test pattern at 8 degrees looks more like this:

    https://i.imgur.com/30dWjMY.jpg

    Then your scanners are *not* meeting the ILDA spec for 36Kpps. In this case you must reduce the scan speed in your software until the test pattern looks correct at 8 degrees.

    Once you've found the speed where the test pattern looks OK at 8 degrees scan angle, that is the true speed of your scanners. And it would be a good idea to always use that speed setting even if the image flickers more due to the slower speed. Pushing your scanners too hard will cause them to overheat and wear out much faster, and it can lead to catastrophic failure if the temperature gets too high or if they fall into resonance while being overdriven.

    On a more general note, once you have identified the actual speed of your scanners and marked the size setting in software that gives you 8 degrees of scan angle, you can switch over to a regular laser show file and start experimenting with increasing the size control. Now that you've seen how the ILDA test pattern starts to deform when you push your scanners too hard, you'll be able to see similar signs of distortion in the images of a regular show when you are over-driving your scanners. Go slow and get a feel for what works for most of the shows you have, and then note that size setting as your maximum size setting for graphics.

    Finally, I should point out that simple beam shows can often be scanned with the size control set to maximum, because the patterns in a beam show are normally quite basic and thus are very easy to scan. But be cautious. Some beam shows are more complex than others. If you start to see severe distortion in a beam show, immediately reduce the size.

    For more information on how the ILDA test pattern works, I suggest the following resources:

    http://www.laserfx.com/Backstage.Las...Scanning3.html
    https://pangolinlegacy.com/userhelp/scanangles.htm
    http://lasershowparts.com/files/Arti...20Tutorial.pdf

    Full disclosure, that last link was authored by me.

    Adam

    PS: Your plan to measure the beam diameter and divergence should work fine, although you should be aware that the beam from some lasers will tend to "bloom" as the output power increases, so you may wish to repeat the measurement at full power - assuming you have appropriate eye protection!

  3. #3
    Join Date
    Nov 2019
    Posts
    26

    Default

    Thanks Adam! Really appreciate the detailed response.

    I gave it a go, though I'm not sure exactly how to do it in LaserShowGen. I was able to import the ILDA test file into the software and display it, but the only settings I can tweak when I turn optimization off are the size and the FPS. Nowhere to set the scan rate to 36k.

    The FPS defaults to 60, but I have to turn it down to about 15 to get the ILDA image to look reasonably close. At this point the image is flickering quite a bit.

    If I turn optimization on, I can set a Sampling Rate setting to 36000. But the display looks different obviously. The LSG manual says this:

    The sampling rate is the intended rate of points, measured in points per second. The total number of points in each frame is scan rate divided by FPS.
    Maybe I can calculate the right FPS if I can find the number of points in the ILDA file?

    Or maybe doing this with LaserShowGen might not be possible? I'm thinking of buying LaserShow Xpress but need to a bit more research before buying.

  4. #4
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    From what you are describing, I'm worried that you will not be able to properly test your projector with the software you are using. I don't understand why you can't adjust the scan speed after you turn off frame optimization.

    That being said, your idea to count the points and multiply by the frame rate to get the scan speed should work in theory. Remember that you need to count *all* the points in the pattern though - both the lighted ones and the blanked ones.

    LSX is very capable software, and there are several members here on PL who use it regularly and would be able to answer questions.

    Where are you located? It's possible you might be able to meet up with another member who could assist you in testing your projector.

    Adam

  5. #5
    Join Date
    Nov 2019
    Posts
    26

    Default

    Indeed it might not be possible. I posted on the LSG reddit forum to see if anyone can help. Any idea how I can count the number of points in the test pattern?

    I'm located in Bali, Indonesia.

  6. #6
    Join Date
    Nov 2019
    Posts
    26

    Default

    Ok, I got this from the LaserShowGen author. Makes sense as I expected. Will be trying it today.

    You can set the number of frames per second that results in the desired kpps by calculating it from the number of points in the test pattern frame. For the standard ILDA test pattern I think this is 1191. This means for 12 kpps you want 12000/1191 = 10.0755 FPS.


    I realize this is a pretty unintuitive way of doing this, but LSG was mostly designed to work with its own optimization that uses a fixed framerate, rather than variable framerates like many special ILDA files like the test pattern expect. So the feature that plays these files was hastily patched on later. In the future I plan to rework this to make it easier to play such files with intuitive settings and good results.

  7. #7
    Join Date
    Nov 2019
    Posts
    26

    Default

    Ok, got the test done. As I noticed before, the inner circle starts to pull away from the square noticeably above 16 FPS. So 16 FPS x 1191 points is 19056 pps, roughly half the advertised speed. I'm not too surprised, it was cheap.

    This speed should still be good for most beamshows, right? I don't plan to do much if any image scanning.

  8. #8
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    Yeah, 20K scanners (more or less) are fine for beams. You'll notice more flickering on graphics shows, but for beams you shouldn't be able to tell.

    Note that it's very likely you can replace your scanners yourself with a faster set. You'll need to change the scanners, the amps, and maybe the power supply as well, but depending on how much space you have inside the projector it might not be as big of a job as you think. Goldenstar sells the DT-40s for $250...

    Adam

  9. #9
    Join Date
    Nov 2019
    Posts
    26

    Default

    Actually, I may have been interpreting the test results incorrectly. Here is what the pattern looks like at 36kpps (30fps):

    Click image for larger version. 

Name:	IMG_3722.jpg 
Views:	1 
Size:	175.9 KB 
ID:	56575

    Excuse the quality... screenshot of a video, but it's pretty accurate.

    The pattern looks like a mess, but to be fair, the circle is pretty much where it should be inside the square.

    Oddly, if I set the speed very low, like 10fps (12kpps), the square is perfectly inside the circle.

    I'm not sure how to interpret this.

  10. #10
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Default

    The "tails" that are visible in the picture you posted can be corrected by adjusting the blanking offset in software. They are caused by a signal timing mismatch between the X/Y data and the color data. They are not related to your scanner's performance.

    If the picture you posted was being displayed at 8 degrees optical scan angle and 36Kpps, and if you are certain that the software optimization was turned off, then the fact that the circle is just touching the sides of the center square indicates that the scanners are performing adequately at their advertised speed of 36K. (Which is a good thing!)

    Slowing the scan speed down will cause the center circle to get larger, just as increasing the scan speed will cause the center circle to get smaller. The reason can be found by looking carefully at the points in the ILDA test pattern that make up the circle. The circle is made of 12 equidistant points that are all positioned well outside the center square. (They form a dodecagon.) Actually there are 36 total points, because this shape is scanned 3 times in a row.

    The reason the points are so far outside the square is because the intent of this part of the pattern is to test how the scanners behave when they are under constant, maximum acceleration. When drawing the circle they are never supposed to get anywhere close to the point they are headed for before being sent off towards the next one. When subjected to this continuous, maximum acceleration the scanners are said to be "ballistic". The test is meant to demonstrate how evenly the performance falls off when you drive the scanners to their limits.

    But if you slow the scan speed down, then the scanners can get closer to each of those outside points before heading off to the next one. So the circle gets bigger. If you continue to reduce the scan speed, eventually the scanners will be able to reach those points and the circle will stop getting bigger and will start to grow corners as it transforms into a dodecagon.

    Adam

Posting Permissions

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