Page 2 of 10 FirstFirst 123456 ... LastLast
Results 11 to 20 of 99

Thread: Open Source Scanfail Unit (phase 1)

  1. #11
    Join Date
    Jul 2005
    Location
    Lancashire UK
    Posts
    1,379

    Default

    Great start guys ...

    it all looks like we are on the same track .....

    bill, i see your point regarding the safe zone window and hobbyists, but im guessing that as this project matures, there could be the possibility of designing this as an add on or an enhanced version

    i think that a simple analogue board is the ideal starting point

    one other thing that i would like to possibly add, ( its not really high on the list for safety though ) is .....

    LED indication of scanfailure active ..... nothing worse than not being able to get any output from the diodes and not knowing if it is software / diode PSU etc...

    Blanking isolation ...
    Idealy to maintain compatability with users exsisting equipment, and also that most people would want blanking of 3 ( RGB ) diodes, and because of the variations in Diode manufacturers prefered blanking ( ttl / analogue / inverse blanking etc.. )
    these signals would be best isolated, this would mean the use of small relays, ( more likely solid state relays )
    as Bill suggested ... the ideal response time to blank the laser is between 1 and 10 ms .... we would have to check response time and any losses through the relays etc ....

    all the best ... Karl

  2. #12
    Join Date
    Nov 2005
    Location
    Melbourne, Australia
    Posts
    3,702

    Default

    Quote Originally Posted by Banthai View Post
    this would mean the use of small relays, ( more likely solid state relays )
    Optocouplers might be a better choice..

    Led indicating status of the board is worth doing, perhaps it would be a good idea to have it on a header, so the led can be mounted somewhere off the board, on the case of the laser..
    KVANT Australian projector sales
    https://www.facebook.com/kvantaus/

    Lasershowparts- Laser Parts at great prices
    https://www.facebook.com/lasershowparts/

  3. #13
    Join Date
    Jul 2005
    Location
    Lancashire UK
    Posts
    1,379

    Default

    dave said
    Optocouplers might be a better choice..
    doh.... of course dave ... my minds not thinking small scale

    its really weird at the moment ... im currently in preparing for a job in my other industry ( high speed packaging ) so i can only think along the lines of large electrical systems .... suprising that i have not suggested that we thown in a couple of circuit breakers, a PLC and a PID temperture contoller

    which brings me to .....

    im flying out to Mexico this afternoon, to supervise the installation of a new production line for nestle in Toluca ...... which means over the next 5 weeks my visits here will be very sparse, so sorry guys for the quick departure but keep this thread going and hopefully by the time i get back it will be in manufacture


    all the best ..... Karl

  4. #14
    Join Date
    Nov 2005
    Location
    Melbourne, Australia
    Posts
    3,702

    Default

    So we have worked out a little of Bill's riddle, what else are we looking at?

    I assume the signals from the differentiators will be summed, then what? We will need to square them, or at least precision rectify them with a full wave rectifier? It seems a full wave rectifier would account for the other half of the quad opamp....
    KVANT Australian projector sales
    https://www.facebook.com/kvantaus/

    Lasershowparts- Laser Parts at great prices
    https://www.facebook.com/lasershowparts/

  5. #15
    Join Date
    Apr 2006
    Location
    Orlando, FL - USA
    Posts
    1,770

    Default

    Quote Originally Posted by greenalien View Post
    In the interests of simplicity and not having to re-adjust the projector for every venue, how about an 'above / below horizon' solution, where the scanfail device only operates below the 'horizon'. Then, when setting the projector up in a venue, you align the horizon with audience head clearance height, thus allowing fail-safe audience scanning 'below the horizon' and aerial effects like stationary beams above the horizon. You could then design all your show frames with this in mind, with a 50:50 split.
    Yes, this is much better. In fact, in the ILDA show standard (there isn't a formal document, but many past discussions, and you will find various show principals documented on the ILDA web site), they suggest that the horizon be naturally above the heads of the audience.

    This kind of "horizon" detector is better than having a window (from a reliability standpoint) and a non-adjustable horizon detector would not be difficult to implement. (If you need to adjust the horizon, you physically reposition the projector). Still, this needs to be done very carefully. One thing everyone needs to ask is -- what if something goes wrong.

    I see it all the time. People are always talking about their circuits and discussing the great merits when everything is working correctly. But, how does the circuit react when something goes wrong? If you have a horizon detector in there, and it fails such that it believes the whole audience is "above the horizon", then you don't have a scan-fail interlock any more. So, the system would be safer without such a thing because with it, you have one more point of failure.

    For hobbyist purposes, I would rather people think on a more minimalist basis. Is a horizon detector really beneficial for the power of lasers typically used by hobbyists? As I have said several times, if you are professional, then you probably have no business reading this post. Just go out and buy one of the half-dozen or so scan fail systems made by professional companies.

    Quote Originally Posted by dave View Post
    Optocouplers might be a better choice..

    Led indicating status of the board is worth doing, perhaps it would be a good idea to have it on a header, so the led can be mounted somewhere off the board, on the case of the laser..
    Regarding relays and opto-couplers -- it's just more to fail. Think about it. Relays have a definite number of open/close cycles before they degrade. (I am not sure you can use a solid-state relay in this kind of circuit.) And I have seen several so-called professional scan fail units use opto-couplers that "pass" the color signal through them. But -- what if they fail shorted? Then the signal is always passed.

    Keep thinking guys .

    Regarding the LED, yes, good idea. Worst case -- the LED fails and you just can't tell when the circuit is operational or not. So this kind of failure can not impact safety. No objections there .

    Quote Originally Posted by dave View Post
    So we have worked out a little of Bill's riddle, what else are we looking at?

    I assume the signals from the differentiators will be summed, then what? We will need to square them, or at least precision rectify them with a full wave rectifier? It seems a full wave rectifier would account for the other half of the quad opamp....
    You could take the velocity of each axis, and then square them, and then add the squares together. From there, you could either use a square-root circuit, or a simple comparison. But how many non-linear circuits have any of you seen at Radio Shack? And... it's just more BS to fail. If the whole point is to detect if velocity is above or below a threshold, then a simpler full-wave rectifier (aka absolute value) will certainly do the job. This can be done using some number of op-amps and four diodes.

    Bill

  6. #16
    Join Date
    Nov 2005
    Location
    Melbourne, Australia
    Posts
    3,702

    Default

    Quote Originally Posted by Pangolin View Post
    Regarding relays and opto-couplers -- it's just more to fail. Think about it. Relays have a definite number of open/close cycles before they degrade. (I am not sure you can use a solid-state relay in this kind of circuit.) And I have seen several so-called professional scan fail units use opto-couplers that "pass" the color signal through them. But -- what if they fail shorted? Then the signal is always passed.

    Ok, so how about using the opto to pull the color signal down in the event of a failure? Then if the opto does short, the color signal will be pulled down, meaning no output.. I think I'm thinking on the right track

    Is a opto-coupler even needed in the first place? If this was integrated with balanced receivers and perhaps color filters as well, would there even be a need for electrical isolation? In this case could a single transistor or fet be used to ground the color signals?
    KVANT Australian projector sales
    https://www.facebook.com/kvantaus/

    Lasershowparts- Laser Parts at great prices
    https://www.facebook.com/lasershowparts/

  7. #17
    Join Date
    Apr 2006
    Location
    Orlando, FL - USA
    Posts
    1,770

    Default

    Yes, all true. But think further. Eventually you have to compare something. Last time I checked, comparators only come in packages of four. Also, comparators have an open-collector output.

    You have three signals -- Red, Green and Blue. Then you have Karl's LED idea. So that's four things that magically need to be shorted sooner or later.

    So, in the words of Morpheus, when I see four comparators, and four signals that need to be shorted, I do not see coincidence, I see providence .

    So you can either use one comparator, then some extra BS after that (opto, transistor, FET, etc.) or.... just the comparator.

    Bill

  8. #18
    Join Date
    Nov 2005
    Location
    Melbourne, Australia
    Posts
    3,702

    Default

    So what do we have so far?

    Differentiators - summing - precision rectifier - comparators ?

    Bill, you made a reference to 10Hz, so I'm guessing this would be the cutoff value of the high pass filter?
    KVANT Australian projector sales
    https://www.facebook.com/kvantaus/

    Lasershowparts- Laser Parts at great prices
    https://www.facebook.com/lasershowparts/

  9. #19
    Join Date
    Jul 2005
    Location
    Lancashire UK
    Posts
    1,379

    Default

    Hi guys

    sorry for the lack of posting from me, but its been very difficult to type with a Corona extra in one hand and a Sol in the other

    Dave ..... looks like your last post had the basics for what we need

    sorry if im being a bit forward with this but would it be possible to knock up a circuit diagram for the basic circuit, i feel bad in asking as its only yourself and bill that is discussing the development at the moment

    but if we had a basic circuit perhaps others will help us out with this thread ( come on Guys n Gals ! ) more input the better

    all the best .... Karl
    Last edited by Banthai; 04-26-2008 at 04:12.

  10. #20
    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
    9,890

    Exclamation

    As I type this I'm waiting to hear from my Dr about checking into teh hospital for urgent surgery. So I wanted to get this to you guys. Its one of two designs I was working on, the other involves a pair of home made transistor based multipliers, teh multiplier works but I'm not done with the rest yet. I may be down for two weeks, so heres the rough drawings.
    As many of you know I have had chronic hypertension for two years and the meds didnt work. The docs finally found a growth on a gland and want to remove it, IF it hasnt matasticized I will be back. If not, please name this the "last circuit" and prayers would be appreciated either way. I've tried to reserve a space in the non smoking section, and if I have offended anyone here, I'm sorry and I ask for your forgiveness.

    And now for something completely different:

    you need one of these per channel. It fits BB's four opamp (ie a quad) concept.
    A differentiator drives a proven fast response absolute rectifier I've used before to 100+ kilohertz..
    The differntiator is referenced in David Terrells execellent book on OP-AMPs, its good to 10 Khz or more. use a opamp with at least 10 V per uS slew rate, ie a 741 need not apply.
    The absolute value circuit drives a compaitor with hysteresis, if I got the math right the switching points would be ~2 and 4.7V respectively, give or take a fudge factor. use 1% resistors for the 22ks in the rectifier/absolute value circuit. A 555 missing pulse detector smooths things out, set up its time constant for .05 to .1 seconds, to allow some beam effects that are somewhat stationary. use a jfet between two resistors as the means to short the modulation signal to ground. You may need a inverter between the 555 and the MPD unless you use a p channel fet. Nchannel is prefered, and many fets are bidirectional in their conductivity. Driven full on the fet should be about .01 ohms. thus pulling things down hard. The details for the MPD design (sorry I'm out of time) are in the 555 timer applications handbook (tab publishing or signetics) or the following web site. please note the inverted RC coupling circuit I have of the resistor to VCC after the coupling cap, this is needed or you'll have a devil of a time getting the 555 to retrigger. The MPD is a contraversal option, however I can see accidently terminating some effects. Hopefully there is enough here for you to flesh out the rest.

    simple hobby ist examples of MPDs

    SEE: http://home.cogeco.ca/~rpaisley4/LM555.html (way down the page)
    http://www.uoguelph.ca/~antoon/gadgets/555/555.html (figure 10A)



    See attached two PDFS

    see you soon

    Steve Roberts
    Attached Thumbnails Attached Thumbnails DC_2501864.pdf  

    DC_2501865.pdf  

    Last edited by mixedgas; 04-22-2008 at 16:41.

Posting Permissions

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