Page 9 of 9 FirstFirst ... 56789
Results 81 to 87 of 87

Thread: Info on scan fail systems

  1. #81
    Join Date
    Oct 2009
    Location
    sweden
    Posts
    116

    Default

    can i place a lens on the output of my spectra physics 168
    to get a higher divergence?

    for us who donīt like silicone

  2. #82
    Join Date
    Oct 2009
    Location
    sweden
    Posts
    116

    Default

    i mean can i put the lens before the scanners instead of after the scanners
    is there any product (lenses) to recomend?

  3. #83
    Join Date
    Oct 2009
    Location
    sweden
    Posts
    116

    Default

    what about this?
    http://www.dragonlasers.com/product....cat=285&page=1

    feels strange to combine a spectra physics with some laser pointer thingy

  4. #84
    Join Date
    Nov 2010
    Location
    Ilchester Somerset UK
    Posts
    60

    Default

    Hi everyone,
    Is this thread still live?
    I have a few ideas for practical circuits that I can share.
    Thanks, Si Bond.

  5. #85
    Join Date
    Mar 2007
    Location
    Rotorua New Zealand
    Posts
    528

    Default

    Hi
    I'm always interested in good ideas
    Thanks
    Ray

  6. #86
    Join Date
    Jan 2006
    Location
    Charleston, SC
    Posts
    2,147,489,446

    Cool Hooray for grave digging!

    Wow - this *is* an old thread!

    Sometime after this thread died down I started a project with Mo (Daedal) to build a digital scan-fail device based around the Arduino. The idea was to store the X/Y coordinates (sampled from the position feedback signal on the amp) in a large array and then compare the current value to the last 500 values in the array. If you got a hit, it tripped the scan-fail for 1 second - which would crowbar all the modulation lines to ground. (Why 500 values? Because that worked out to no more than 6 pulses per second in the same location, which made the MPE math easier.)

    We weren't sure if it would be a hobbyist product or a commercial product at the time, so we just banged away at it to see what we could accomplish. Very quickly we discovered that the Arduino wasn't fast enough, so we switched to a different processor (forget exactly which one, but I'm thinking it was the Beagleboard). Note that this was well before the Raspberry Pi was released. Mo finally got the code working at 10-bit resolution, which seemed OK at the time, but we soon ran into lots of other problems.

    Truthfully, the concept was fraught with conceptual errors, compromises, and performance issues from the very beginning, and in the end it would not have been robust enough to guarantee safety, so we abandoned it. Spoke with Casey Stack about the effort sometime later (after I had completed my LSO course) and he said that lots of people who design scan-fail systems try the digital approach at some point, but they usually drop it in favor of an all-analog one because they're more reliable. Tellingly, the gold standard for audience-scanning scan-fail systems (PASS) is all-analog.

    The idea of scan-fail is pretty simple. But implementing it in a truly fail-safe manner is more complicated than you might imagine. You've got to account for some really edge-case scenarios (including highly unlikely things like a single-rail failure on a dual-rail power supply!), and in the end I would always be worried about the liability issue if I were designing a commercial product.

    All that being said, if the PL community wanted to collectively work on something that was open-source, I think that would be cool. And so long as there is a disclaimer that it comes with no guarantees as to performance, I think it would be fine for hobbyist use. If a company decided to use it commercially - well, that puts the legal onus on them.

    Adam

  7. #87
    Join Date
    Nov 2015
    Location
    Northamptonshire (UK)
    Posts
    9

    Default

    This has been a really fascinating read - I ended up here after searching for things like "galvo safety" and "laser safety" etc. I have some Chinese lasers (had for some time) and although appalled at their safety approach, HAVE used one recently for a non-profit event where the laser was projecting a single image on to a high building (we used it to project a soldier on the church steeple for Remembrance Sunday).

    To make things as safe as possible, the beam (about 1 watt), left a locked building one story up, projected a couple of hundred feet up the church AND had a crude but effective metal "mask" secured to the front of the laser so that whatever happened, the laser could not possibly go outside of its "target zone". Short of any pigeons that might have got in the way, it took a huge stretch of imagination to see how there was any (credible) risk to anyone. Happy that is was "reasonably safe", we went ahead.

    Although the safety aspect of Chinese lasers is about as bad as it can get (VERY slow moving beams on some "built-in" programs and ones that collapse almost to a single point), they are a good starting point for a decent DIY laser if you regard them as a source of parts, rather than a finished product.

    The 30K galvos are often of a usable quality, the LD units not bad and some of them now have reasonable optical alignment mounts. Ran from the ILDA port and NOT audience scanning, I am guessing they could be used with some degree of safety without too much additional modification. A galvo monitoring board would seem a great addition.

    Before I found this thread, I have considered all sorts of weird ideas (pass beam after galvos through a slightly off-axis piece of glass or exit window and then have the 1% or so beam hit a lightly frosted piece of glass. Cheap camera then looking at screen and software (the stumbling point for me) identifying any "bright / hot spots". The idea was to use something like "frame subtraction" and trigger an output if a certain part of the frame became brighter than a given pre-set. Decided fairly quickly that it was too complicated - and too many ways it could fail.

    I though of a VERY simple (perhaps too simple) safety measure as a real fallback (so assuming the system is normally running safely anyway), where each galvo drive is spliced and (I did say this was crude), fed via a capacitor so if the drive failed HI or LO (anything other than moving = AC), then the laser would be switched off (TTL/analogue blanking) while a magnetic shutter release was used. This would offer (some) protection from a failed galvo driver but no use if the driver was ok but the galvo had failed.

    A better idea came to mind though which would allow for SOME feedback from the galvos themselves - rather than just making sure the drivers were working. If the galvo cable was cut (rather than spliced), current monitoring could be achieved (as well as confirmation of an AC signal) AND voltage. If current was too low for a given voltage (mirror fallen off, galvo coil failed, connector disconnected) then a failure would be triggered and if a galvo had started to lock up / foul / jam, the current would be too high for a given voltage - again triggering a failure.

    Although I have (just) started to play around with galvo monitoring, I would rather not try to re-invent a badly designed wheel if a much better design is already out there.

    As such I thought I would see if there are any open source designs which are fairly "ready to use". If not, I would be very interested in trying to help come up with an open source design if there is anything I could offer in terms of ideas. As a laser user goes, I am a complete amateur but have a reasonably healthy respect for eye safety and a modest ability with electronic prototyping.

    I will do some more searching in a bit on here - but if there is still an interest in making a basic (but better than nothing) automatic galvo monitor, then I am very keen to participate if I can.

    In the meantime, I am sure this method has been done before (but never seen it), I have found a way to give the ILLUSION of a high power audience scanning system without the obvious risks of actually doing it. I am about 99% sure this is something that everyone in the industry knows about, but I have never seen it mentioned OR seen it in practice despite "looking out for it".

    Take your Big Scary Laser - use whatever safety checks are needed and keep projection 3 Metres from the ground. On TOP of the BSL, and being driven by exactly the same signals, have a 200mW laser, via one or three beamsplitters pointing down THROUGH the high power overhead beams. You end up with 100mW or 50mW beams copying EXACTLY what the BSL does - so when you look up, you see all the BRIGHT powerful beams - and you kind of forget (even when you know), that the dinky little beams scanning you are a tiny tiny tiny fraction of the power from the main laser - you are just aware of the powerful beams above you - and that you are being scanned.

    As I say, this might be a very old trick in the industry, but I have never seen it used before.


    Off now to see if I can find any follow-up to the safety board.

    Guy

Posting Permissions

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