SCR firing circuit? newbie needs help~~~

Discussion in 'The Projects Forum' started by adamsadam, May 2, 2007.

  1. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    Hi all!

    I'm currently designing a project which is to build a SCR firing circuit with 24VAC, 50HZ supply, and the output requires to have a delay of triggering (alpha) that can be controlled from 0~180 degrees.

    I'm thinking of using a 555 timer to generate pulses at 50HZ, but I don't get it very much. I saw some suggestions such as TCA785, too, but I'm not sure how to use these chips.

    Could someone give me a hint please?:)
     
  2. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    The lower limit of pulses from a 555 monostable is about 10uSec, so you could get 0.18 to 180 degrees, but not 0 to 180.
     
  3. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    Thanks for reply.

    0.18 is close enough^^
    If I use monostable, how could I generate frequency pulses (syncronous with AC source)? I thought 555 monostable takes DC input and generates only one pulse?
     
  4. mozikluv

    AAC Fanatic!

    Jan 22, 2004
    1,437
    1
    i believe the illustrated circuits from the siemen datasheet explains it quite simply.

    however if u still want to make a go for the 555, since 0.18 is acceptable to you, why not draw a schem of your design, post it so we can point out what is wrong with the circuit or help you in improving the circuit.;)

    moz
     
  5. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
  6. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    Thanks for reply.

    I've found the Siemens TCA785 datasheet, the explaination excited me, but... I'm not quite sure how the circuit is going to be built, the sample circuits on the datasheet are too complex for my project.:)

    My partners and I have tried a 555 monostable design which looks like:
    [​IMG]
    Instead of a push button at "trigger", we tried to connect the required AC sinusoid input into it so that it triggers 555 constantly, sadly it didn't work, as well as square wave input, it just simply gives some DC voltage at the output, I need pulses~~:)

    (The choice of chips does not matter):)
     
  7. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    My supervisor told me that chips like TCA785 are not allowed to use for my project, because they solve all the problems directly (too simple). I think I'll have to go for 555.
     
  8. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    [​IMG]
    I tried the 555 monostable circuit drawn on this webpage (as above) today, by taking square wave as trigger input at pin2, the output looks pretty much like the input square wave, but not pulses as I desired.

    Does 555 timer works fine under AC signal? Or any other suggestions? Please it's really urgent...
     
  9. Spoggles

    Well-Known Member

    Dec 2, 2005
    67
    0
    Are you committed to using a 555? I can think of much easier ways to do this.

    A quad comparator (Lm339) configured to integrate the 50Hz 'clock' and a pot to adjust the turn on point should work. You have 4 devices to work with in one package.

    Square the 50Hz, Integrate it to a ramp, compare ramp to pot voltage, Output a normally low signal thru some appropriate circuitry to the SCR / TRAC and you should be there.

    LM 'slew rate' should be better than 10uSec

    Spoggles
     
  10. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    Thank you for reply.

    My supervisor rejects the ideas of using powerful chips:( , but I'll give a try on it, however I have no idea how to build a circuit like what you have mentioned above (I'm a newbie, remember?:) ), could you please give me a circuit diagram which helps me to understand easier?
     
  11. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    Try using a potentiameter in place of R. Watch what happens to the o/p pulse width as you vary the potentiameter.;)
     
  12. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    I think I got some negative going pulses (2 or 3) in each duty cycle this time, the pulse width is correct, but too many, lol, and at each time a pulse occurs, there is a pulse at the trigger input, too, like some distortion.

    Instead, I tried another 555 circuit (I don't know which mode it is), by adding resistance between AC input and pin 2, also another tiny resistance between pin 5 and ground, connecting pin 4,6,8 to Vdd, pin 1 to ground, and pin 3 with some resistance to output. So the voltage at pin 5 is a control voltage which compares with the voltage at pin 2, once V(pin2) is less than V(pin5), the output will be high, and since V(pin5) is very small, so the output looks like a pulse each time.

    From this implementation, I got clean pulses, however it occurs twice a cycle (doubling the frequency), is there a way to make it once a cycle?
     
  13. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    What are you using for triggering? The trigger pulse must be more narrow than the o/p or the 555 will retrigger.
     
  14. adamsadam

    Thread Starter New Member

    May 2, 2007
    8
    0
    Thank you for reply.

    My triggering signal is just the AC input (reduced voltage), so it triggers once a cycle.

    For the monostable case, the length of trigger signal is 0.01s (a half cycle of 50HZ), so I had to choose values of R and C from T=1.1RC to make T to be longer than 0.01s and shorter than 0.02s (a full cycle), where T is the output pulse length (long, say 0.018s), then use an inverter to switch it to low, and the rest 0.002s will be high (which is the real pulse we want). However I can't get the inverter work.

    For the new design I described, it works like a comparator or something, its output depends on the comparison of voltage drop at pin2 (trigger) and pin5, the problem is that I had got better pulses but of doubled frequency (twice a cycle), I think it should be a better way (don't need inverter) if you could help me out on this one.:)

    Sorry about the length of my post, I hope it is clear...
     
Loading...