Optical SCR triggering.

Discussion in 'The Projects Forum' started by Artikbot, Nov 9, 2010.

  1. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Hi,

    I'm building an optical trigger for a SCR featuring both an IR emitter and an IR phototransistor rated for 850nm wavelength.

    I'll explain a bit:

    My circuit uses the broken beam method, triggering the SCR when the IR link between the LED and the phototransistor is broken.

    I use the phototransistor to drive the non-inverting input of a high speed comparator (0,2uS response time), and a voltage sample (out of my reliable 7805-driven low power LMPS) to drive the inverting one. When the beam is broken, the comparator output (an LM311) goes high, triggering the SCR's gate.

    And here kicks in my question. Since the comparator doesn't go LOW with a 0V value, will my SCR's gate be off? It's a super overrated (both voltage and amperage, but it's what I have. No, I don't want to change it) TYN640RG P-type gate 600V 25A SCR.

    If it doesn't, maybe I need a transistor right after the gate that has a threshold voltage above the low voltage state from the comparator, so it only goes ON when the voltage state is truly HIGH?


    My current schematic is attached.

    Thanks!
     
  2. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    Problem here is taking the gate voltage back to -0- isn't going to turn the SCR off, if I recall once they're on they're on until the load is broken.
     
    Last edited: Nov 9, 2010
  3. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Yep, I knew that, the idea is:

    If the comparator's LOW output isn't at 0V, will the SCR stay off until the comparator goes HIGH?
     
  4. KMoffett

    AAC Fanatic!

    Dec 19, 2007
    2,574
    230
    As I recall the LM311 is an open-collector output, and needs a pullup resistor between the output and Vcc.

    Ken
     
  5. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    There are sucker for a penny
     
  6. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    The LM311 is open collector AND open emittor.
    The LM393 and LM339 are open collector only.
    See the datasheets for more info.

    Bertus

    PS what will be tha application of the circuit?
     
  7. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    @KMoffett:

    I've been reading something about open collector comparators /pullup resistors on Wikipedia... From what I've read, the pullup resistor avoids the logical high/low levels to be distorted by the rest of the circuit's state, alright?

    @Bertus:

    Wow, open emitter is new for me. Could you explain a bit that part, please?

    Regarding the LM393 and the 339, they seem to fit better my circuit for the characteristics (only open collector) especially the LM393, since it's a dual comparator (what would I do with the 3 remaining of a quad? o_O).

    The problem with them is response time (1.3us, pretty high), although it might make the trick, I want to keep that as low as possible (LM211 had some great .2us) since I want minimal delays on the SCR triggering.

    The circuit serves the same purpose a high power SSR would, just it's WAY faster, WAY cheaper, and for me, it's more situable (I would have to trigger the SSR optically anyways). Ah, it's a multi-purpose circuit, by the way. I'm experimenting with low DC voltages to eventually develop a much higher one to turn on my house lighting when crossing doors, passing the hands over a specified zone of the wall, and so on.

    I'm kind of a geek, so I'm a lover of all this craps that just look cool :p
     
  8. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    I think your SCR takes 0.85V to trigger on (can't recall if that was a max or min quoted in the data sheet)
     
  9. retched

    AAC Fanatic!

    Dec 5, 2009
    5,201
    312
    How long is the total propagation time for the optical trigger?
     
  10. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Umm...

    If you mean the total circuitry delay since the beam is broken until the signal is delivered to the SCR, I believe it's about 210ns or so.

    Comparator has an intrinsec delay of 200ns, and the phototransistor has a fall time of 2'3ns.

    Add some more ns for any other possible delays, and there you go.

    So yep, I believe it's about 210ns.

    @Marshallf3:

    So I'll need a transistor with a collapse voltage higher than 0.85V to convert that LOW output on a real 0V signal... high will be high anyways, so I don't think there's gonna be a problem with it.
     
  11. bertus

    Administrator

    Apr 5, 2008
    15,648
    2,347
    Hello,

    Here is another datasheet of the LM311.
    It shows the use of pin 1 (the emittor output) driving a transistor.
    (see the pages 16 and 17)

    Bertus
     
  12. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Hmm...

    Emitter driving a base? I suppose it's not dangerous since the power of the transistors inside a comparator IC is very low...

    Looking at the page 16 schematic:

    So, essentially it grounds the pin 7 (output as I can see) and instead, it uses the emitter to drive another transistor. Does this make much clearer logical levels, such as 0V and Vin-Vdrop?

    In that case, it would be really good to have an open emitter comparator, wouldn't it?


    Thanks a lot Bertus!
     
  13. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    I've made a decision.

    For obvious convenience reasons, I'll drop the SCR design and use an IGBT one instead.

    This lil' beastie is gonna be my new victim this time :p

    Its a beefy 600V 140A IGBT manufactured by IXYS.
    http://www.datasheetcatalog.org/datasheets2/46/46419_1.pdf

    I've found that it would be good to control the time of opening/closing the system, since I doubt that my supply will stop providing current (a small wall transformer), therefore I doubt the system will ever stop on itself.

    On the good side, I have that the IGBT is a transistor, so it possibly needs a specific continuous voltage applied to turn on, unlike the SCR which needed a simple and plain pulse to work.

    I can't interpretate the datasheet well enough to locate the triggering voltage, can someone halp me in that aspect, please?


    Thanks a lot and more :)
     
  14. R!f@@

    AAC Fanatic!

    Apr 2, 2009
    8,754
    760
    IGBT's are faster indeed.

    I am learning from this, as I have something similar in mind.
     
  15. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Good to know that my project is being useful to someone else ^^

    Feel free to ask anything you might want to know in this thread, since it's about the same topic, might as well turn useful for my project :)
     
  16. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Sorry for doubleposting, but I've got an issue.

    I've received the parts I ordered from Farnell in order to build this trigger circuit.

    Well, instead of the SCR (no IGBTs, way more expensive. Ugh, 4 times the price for 1/2 rating :S) I've hooked up a LED to the output of the comparator, just to see if it works or it does not.

    Well, the circuit is exactly the same as I posted previously, but with a LED wired on the output.

    My issue is: The LED (it does work - tested :p) doesn't light up. Doesn't matter how high (or low) I set my pot between +7V rail and collector of my phototransistor, it won't light up. Already tried switching resistors - still doesn't work. Have rechecked and rebuilt the circuit several times (solderless breadboard), but it doesn't do anything either if I cover the phototransistor, if I light it with direct IR light... Nothing.

    By the way, it is a BPW85, peak wavelenght 850nm.


    Thanks.
     
  17. KMoffett

    AAC Fanatic!

    Dec 19, 2007
    2,574
    230
    Maybe you could post a new schematic? There was not 7v rail or pot in the original one.

    ken
     
  18. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    This, but powered with 7.4V and with a 470k pot on the place of the 470k fixed.
     
  19. KMoffett

    AAC Fanatic!

    Dec 19, 2007
    2,574
    230
    measure the voltage at the phototransistor's collector with the light path open, and with it blocked.

    Ken
     
  20. Artikbot

    Thread Starter Member

    Nov 7, 2010
    45
    1
    Oh darn... 5.04V in both states.

    I guess I have something improperly wired (protoboard issue?) I've moved the circuit to a piece of perfboard to ensure optimal connections. I'm aswell starting to build a 5V linear regulated module to power my inverting input and my LEDs, built in the same piece of perfboard.

    Hope is what I can't lose xD
     
Loading...