Page 1 of 3 123 LastLast
Results 1 to 10 of 25

Thread: disgusted, back from Selem

  1. #1
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    8,898

    Default disgusted, back from Selem

    While I know its difficult to make a linear DPSS, I feel you folks have been badly ripped off by so called analog lasers. I brought a ADAT (yeah, so called old school technology) and after so minor debate about how to wire a a ILDA cable (more on that later) hooked up to a couple of different projectors. Max adat color signal is 2.5 volts, so you'd think on a 0-5 input that would mean at least some illumination coming out of the lasers. Wrong!

    I was beginning to think I had a bunch of dead ADAT channels until we put the signal on a scope. (nice scope Spec!) So I went off to the shack to get some parts to build a gain/loss box. Magicly, suddenly the blue/green worked. It seems you have a huge KNEE voltage around 3.3 volts (I could live with .6 to 1.2 volts for various reasons ) before turnon and a very small linear range. Not to mention the timing issues between blue, green, and red.
    The folks at SELEM alomsot universally used Marconi's MAXY modules, so red is not a issue here, they turn on around ~ 1.2 volts. Now my ADATs are set to lag color about 3 points worth, due to the fast PCAOM, but still many of the results were darn ugly. Also many people were dependant on Pangolin for their image size and offset, this is a no-no in the field, its fine for home or planetarium work, but you need KNOBS. Digital systems chop resolution when they reduce image size ...........

    So I propose a few things.

    One. The SELEM board , aka a couple of differential to single ended and single ended to differtial converters, a lead/lag network for color correction and offset/size knobs with the capability for both gain and loss, and a small test square built in for sanity. The test square consists of a 7473 dual JK flipflop and a 555 wired as a walking ring counter, set for a 30 hz square. This board needs to be somewhat standardized. David Zurcher is building Bill Benner's color adjust circuit for evaluation. There is a commercial DSP solution available in Europe, but its way out of price range for a hobbyist.

    Two, the photon lexicon standardized risetime measurement, done with a standardized, published ,digital pulse circuit and a standardized common small area photodiode like a bpw34 and a dual trace scope. Thus we can compare different manufacturers and know what range of lead/lag we need. There will have to be some rules attached to this, like how to define the pulse envelope, ie most of you here have no experience with FWHM, ie full width half maximum measurements. FWHM is not always a good measurement, anyways, as we need to know fall times too.

    Three, we need to make sure we all wire the same ILDA connector on our projectors for SELEM/FLEM I traditionally use the pin 25 gound when running single ended, this was BAD when trying to interchange signals with almost everyone else, who had variations of single ended ground on their color and used signal minus tied to signal ground for single ended graphics. I'm glad I made up a hackable cable before I left, and it got hacked many, many, times. Most people didnt have knobs because of the differential signal issues.

    FLAMES and SUGGESTIONS WELCOME

    STEVE ROBERTS

  2. #2
    Join Date
    May 2006
    Location
    Native Floridian
    Posts
    3,080

    Default

    Quote Originally Posted by mixedgas View Post
    While I know its difficult to make a linear DPSS, I feel you folks have been badly ripped off by so called analog lasers.
    Well, I believe this is certainly a valid opinion, though, you have to admit, just 5 years ago for someone like me to make a projector equivalent to my current dpss rig would require at least a 220VAC single phase water cooled gas tube and a pcaom. Looking at costs, I've saved several thousand dollars as well as completely cut the need for major utilities while only sacrificing some color capability. "Badly ripped off"? I wouldn't say badly. Especially when 99.9% of your audience will never know the difference. Heck I was surprised last saturday night when my blue laser fell to 10mW output from 180mW. I was surprised that I seemed to have been the only one to notice that occured about 2 minutes into a 25 minute show. Perhaps the reason was because I doubt anyone there had seen that show before and were completely oblivious to the missing frame data. Yet another reason why solid state is the best way to go over gas, imagine if my projector were running a gas tube. I imagine everyone would have noticed then when my one and only laser died.

    Also many people were dependant on Pangolin for their image size and offset, this is a no-no in the field, its fine for home or planetarium work, but you need KNOBS. Digital systems chop resolution when they reduce image size ...........
    I admit I'm guilty of this. But at SELEM my output was set for between 80 and 100% and never anything less. Ultimately I do plan on a small console between the projector and pangolin but if I ever install this system somewhere it will not go with it. I designed it to be a bit idiot proof (ie. no knobs or buttons for an operator to screw up!)

    Bill Benner's color adjust circuit for evaluation. There is a commercial DSP solution available in Europe, but its way out of price range for a hobbyist.
    Yeah its on my list to get done, it certainly will clean up some of the artifacts of solid state analog modulation. Sadly, the DSP solution you speak of, if its the same one we're talking about, is no longer on the market and hasn't been for some time now. It was an absolutely ideal solution to this problem, but they were not able to make it cost effective and had little interest in it. So the analog circuit route from William may be the only way to go unless someone out there knows how to build around DSP circuits. Really though, a delay action circuit should be built into the dpss driver circuit from the dpss manufacturer.

    Two, the photon lexicon standardized risetime measurement, done with a standardized, published ,digital pulse circuit and a standardized common small area photodiode like a bpw34 and a dual trace scope. Thus we can compare different manufacturers and know what range of lead/lag we need. There will have to be some rules attached to this, like how to define the pulse envelope, ie most of you here have no experience with FWHM, ie full width half maximum measurements. FWHM is not always a good measurement, anyways, as we need to know fall times too.
    This would really be a good idea. If all projectors were made to the ILDAPL standard then anyone could copy a show to adat and it would play on any projector the same as it would also take into account a standard delay.

    Three, we need to make sure we all wire the same ILDA connector on our projectors for SELEM/FLEM I traditionally use the pin 25 gound when running single ended, this was BAD when trying to interchange signals with almost everyone else, who had variations of single ended ground on their color and used signal minus tied to signal ground for single ended graphics.
    I couldn't agree with this more....

  3. #3
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    8,898

    Default Proposed test #1

    Proposed solid state test number one.

    "Repeating TIMING Test"

    At 30K PPS, one point is .0000333 seconds long

    The idea is to generate a pulsed waveform that can be up to 8 multiples of this, so use a 74193 upcounter, which is a 4 bit counter, thus gives you 16 possible states, and a 7485 chip, which is a 4 bit magnitude comparator. The 7485 has three outputs, A less then B, A equals B, and A greater then B. B is programed by a 4 bit dip switch. We clock the 74193 at 480 khz, so when it divides the 480 khz by 16, we get one point time. This gives us 14 different timing periods, not counting full on and full off, the circuit effectively is a 4 bit PWM with the waveform being multiples of the point time. This is mointored with a fast photodiode and a scope. A ILDA frame can be constructed to do the same thing, but when you can do it with 3$ worth of off the shelf chips, it is better to do it in hardware. 480 Khz is easy to get. Digikey stocks a 4.8 mhz crystal, and a 7490 divide by 10 decade counter is a standard chip. 4.893 mhz is another standard clock module.
    if we wish to relax the standard by a few percent. So is 4.43 mhz.
    This test can evaluate "jellybeaning" We test at a few different PWM settings using the dipswitch.


    next test:

    COLD TIME TEST

    A microprocessor is used to generate a pulse with a period of .000033 seconds, every few seconds. It outputs a different pulse on another pin just before this to trigger the scope. This simulates a laser being off most of the time, the response will be very different from a laser that has been on or used frequently.


    Both tests are monitored by a small area photodiode set up to be linear as possible over the power range of the laser under test. The photodiode should have as high a bandwidth as possible and be a standard part. The photodiode transimpedance amplifier (if used) shoudl have a minimum bandwidth of 300 kilohertz. for analog laser testing, the 0-5V signal to the laser shall have a rise and fall time of 100 nanoseconds or less. The TTL laser test signal from the circuit shall be be buffered by a 74HCT04 inverter powered by 5 volts (about 5 nanoseconds rise/fall.) Laser is heatsinked by a 1/4" AL plate of large size.

    Please note these tasks are tough to meet, that is deliberate.
    The actual circuits are easy to build and easy to standardize. A 10 mhz cheap dual trace oscilloscope would easily handle this.


    Static linearity test. Standard fixture: A 7805 regulator drives a 10K pot. The 10 K pot wiper drives a gain of one TL082 opamp for a standardized low output impedance. A SPST switch is in the wiper circuit. The pot is stepped from 0 to 5V in .1 V steps and the laser is allowed to settle for 10 seconds, then the output power is measured. The voltage is then switched off for 30 seconds while being adjusted to next setting. Laser is bolted to a 1/4" thick AL plate of significant size as to be nearly infinate heatsink.

    Dynamic linearity test, a 120 hz linear ramp from 0-5V is applied to the laser, result is a scope picture compared to static linearity points above after laser has ran for a few minutes with ramp. Similar heatsink to above.


    ILDA frame test, the ilda frame has a built in blanking test, but it would require a standardized pair of scanners , and that is nearly impossible unless the test is conducted at only one place.

    comments welcome and flames welcome, I learn new words that way.

    Steve Roberts
    Last edited by mixedgas; 08-23-2007 at 07:57.

  4. #4
    Join Date
    Mar 2006
    Location
    East Coast , Canada
    Posts
    285

    Thumbs up

    Hey Steve, I like it, Good ideas!! I think there should be just such a board that will allow all the adjustment. More than just gain and offset, but including gain and offset, and if each of us used it. It would allow much better compatibility...!! I think a lot of people are missing the idea here. Im on board... Lets make it happen!
    CREATIVITY AT THE SPEED OF LIGHT!
    www.laser-ad.com

  5. #5
    mixedgas's Avatar
    mixedgas is offline Creaky Old Award Winning Bastard Technologist
    Infinitus Excellentia Ion Laser Dominatus
    Join Date
    May 2007
    Location
    A lab with some dripping water on the floor.
    Posts
    8,898

    Default

    in progress
    using ADs SSm series 2:1 and 1:2 converters!
    4$-6$ a channel, but 3Db bandwidth is excess of 100Khz.

    Steve

  6. #6
    Join Date
    Mar 2006
    Location
    East Coast , Canada
    Posts
    285

    Default

    well, if you need beta tester or money, let me know, im willing to add
    CREATIVITY AT THE SPEED OF LIGHT!
    www.laser-ad.com

  7. #7
    Join Date
    Apr 2007
    Location
    The Netherlands
    Posts
    178

    Default

    I'm also willing to help on this great thing as far as I can !!

  8. #8
    Join Date
    Apr 2007
    Location
    Leicestershire, UK
    Posts
    137

    Default

    I've been trying to sort a modulation problem out on my laser. I've attached a jpg of a cicuit diagram. Does anyone know if this circuit would work?

    I need to convert the 0-5v analogue blanking signal into about 1-3v for the laser. I'm using NWC's Livewire to experiment with but I'm not sure I trust the simulation because when I put Bill Benner's "Ready for anything" colour filter circuit in I can't get it to work.

    In my circuit the gain pots are 10k and 1k and the offset pots are 1M. The offset pot doesen't shift the waveform up and down, it just sets the minimum voltage. The zener stops the output going above 5.1v which I asume is near enough not to blow anything up.

    It would be nice to add the delay function like on Bill's design but I can't seem to make it work.

    Any suggestions before I try this on Veroboard? Or shall I just make the Bill Benner design and trust he knows what he's doing!!

    Thanks Chaps,

    Matt.
    Attached Thumbnails Attached Thumbnails Laser Modulation PCB.jpg  


  9. #9
    Join Date
    May 2006
    Location
    Native Floridian
    Posts
    3,080

    Default

    I haven't had a chance to finish the delay circuit as I'm building my new office/workshop. I starting building the "ready for anything" circuit on a proto board but just got busy with other things. I'm a bit surprised that it didnt work in the software program that you have. The only way to know for sure, if there are no obvious flaws, which I didn't see, is to put it on a board and see what happens. I hope to have a circuit done up over the winter and maybe have something available early next year, that is of course, if Bill has no objections to me using he circuit design.

  10. #10
    Join Date
    Apr 2007
    Location
    Leicestershire, UK
    Posts
    137

    Default

    Yeah, that's the trouble with the computer simulation, it's sometimes easier just to make the thing on a proto board in the first place.

    If bill doesn't mind I think I'll try his design and maybe combine it with mine if his doesn't have a way of setting the miniumum output voltage.

    Cheers,

    Matt

Posting Permissions

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