Modify 120v AC circuit to 12v DC by changing resistor kitsrus.com DIY kit Music 2 light Modulator

Discussion in 'The Projects Forum' started by Rolland Elliott, Sep 19, 2015.

  1. Rolland Elliott

    Thread Starter New Member

    Sep 19, 2015
    3
    0
    I own a small company that manufacturers portable photo kiosks used in photo booths that are ipad based.
    I want the ipad to trigger lights when the photos are taken so the pictures look nice and found kitsrus.com Music to Light modulator.
    The ipad will play a sound through the headphone jack and this will trigger the lights to go on and off.

    Here is where I please ask for your help. Due to the way the kiosk is designed it would be much easier to wire if the circuit took a 12v 18amp DC input power source instead of 120V.
    The instructional pdf located here
    http://www.kitsrus.com/pdf/k12.pdf
    that I've attached shows that by using a different resistor value one can change the power source from 220v to 120v AC
    I am wondering what value resistor LABELED R3 IN THE CIRCUIT, would I need to run the circuit on 12V DC power?
    Parts List:

    Resistors
    1K PCB mount pot..............R1
    330R 1/2W carbon ..............R2
    2K7 1W carbon ...................R3
    5K6 1W carbon ...................R3
    Semiconductors
    MOC3021 Opto-triac ..........IC1
    2N6075 or BT136 Triac......Q1
    Miscellaneous
    2.5mm audio jack ............... X1
    2 way screw terminal ......... X2, X3
    IC socket, 6 pin
    Heatsink
    3mm screw and nut
    K12 Printed Circuit Board
     
  2. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,773
    1,103
    Welcome to AAC!
    That kit won't work as you want it if you use DC power. Once triggered by the music the lights will stay on instead of pulsing in time with the music. Even if you could use it I don't think you'd be happy with the result, since it's such a primitive circuit.
     
  3. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    With DC, you might just as easily use an SCR instead of the triac, but you need some way of turning it off again - Commutating in the technical jargon.

    One of the simpler commutating circuits involves adding a second SCR and a commutating capacitor. Both SCRs have grounded cathodes, the switching SCR has the load in series with the anode. The commutating SCR has a resistor from anode to +V.

    When the load SCR is triggered, its anode pulls one end of the commutating capacitor down to 0V, the other end of the capacitor goes to the anode of the commutating SCR and is held at +V by the anode resistor. When you trigger the commutating SCR it pulls the "hot" end of the capacitor down to 0V - that tries to pull the load SCR anode below 0V, that takes it below its holding current and switches it off.

    How you arrange the commutating SCR trigger pulse depends on how you want the circuit to work, probably the simplest option is to have a pulse delayed by a monostable to trigger the commutating SCR.
     
  4. Rolland Elliott

    Thread Starter New Member

    Sep 19, 2015
    3
    0
    Thanks for much for the replies folks. I guess I will just ho. ok up a 12V transformer to the AC connections. The only issue with a transformer is there is a bit of a delay added, but that's ok . I wish I understood all that technical stuff. Again many thanks.
     
  5. #12

    Expert

    Nov 30, 2010
    16,252
    6,750
    No more delay than the other voltages. You need to change R3 to about 300 ohms. The triac will trigger in not more than about 10 milliseconds, depending on what country you are in. (Triacs are very patriotic.) I expect your load must operate on 12 volts or you wouldn't be considering how to alter this circuit. Right?
     
  6. Rolland Elliott

    Thread Starter New Member

    Sep 19, 2015
    3
    0
    Yes my load I am using are 12V led lights. I live in USA. Thanks for your reply.
     
  7. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    Whatever you feed in, is most likely to be DC by the time it gets to the actual LEDs.

    If you just need to control 12V DC, a MOSFET would be much easier.
     
  8. #12

    Expert

    Nov 30, 2010
    16,252
    6,750
    That would be a different circuit. To get the hard switching that the triac circuit provides, an n-channel mosfet would be configured with a grounded source. It would still fire in the 3 volt range, but it wouldn't operate on both polarities of signal. So, basically, I disagree that changing to a mosfet would be easier and work much the same.
     
  9. Alec_t

    AAC Fanatic!

    Sep 17, 2013
    5,773
    1,103
    Two FETs in series (drain-to-drain or source-to-source) could overcome that.
     
  10. #12

    Expert

    Nov 30, 2010
    16,252
    6,750
    Yes, I know, and adding that much more circuitry is the basis for my opinion that claiming the conversion of the circuit to mosfets is, "much easier" is false.
     
  11. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,515
    1,246
    What sound? Music could be made to work, but a continuous tone at a constant amplitude would be better.

    I think the original circuit concept can be reworked for this application, but we need a lot more information about the load. LED lights make/part number? LED power supply make/part number? LEDs like constant current, not constant voltage, so high power LED systems can get a bit tricky to manipulate. Not difficult, but some details must be covered. At its simplest it looks something like this:

    Requirement: iPad makes a loud 1 kHz tone continuously to turn on the lights, and total silence at all other times.
    1. Change the optocoupler to a standard 4N3x.
    2. Change the triac to an n-channel power MOSFET. Drain to LED low side, source to +12 V return. LED high side to +12V.
    3. The opto output is connected from +12V to gate.
    4. 10 K resistor from gate to source and +12V return.
    5. Add 0.33 uF cap across resistor from gate to source.

    The opto LED acts as a rectifier and level detector. The output is a saturated switch pulling the gate up to +12 V above the source. With a 0.33 uF cap, the gate never dips below 9 V during the input half-cycle off times. (0.22 uF by calc plus 50% margin.) Decay period to five time constants (99% decayed, Vgs=120 mV) is 17 ms. This is the linear region time for the MOSFET, short enough to prevent overheating.

    ak
     
  12. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    Well of course it would be a different circuit - that was the whole point.

    Changing the supply to DC with a latching semiconductor switch incurs the requirement for commutation (turning it off again).

    IMO: it would be easier to condition the signal to drive a MOSFET gate than add a commutating SCR/capacitor to the one that switches the load - then there's whatever circuitry as needed to trigger the commutating SCR. At simplest, that might be a monostable to delay the pulse that triggered the load switching SCR.
     
  13. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,515
    1,246
    I agree.

    ak
    PhotoLights-1-c.gif
     
  14. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    Presumably the iPad that triggers it, outputs an audio signal.

    Your circuit could be improved by using an AC input opto coupler.

    They used to be used as ring indicator detectors on PC modem cards, but most makes took to using a bridge rectifier to feed a standard opto.
     
  15. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,515
    1,246
    As I mentioned in post #11, a 1 kHz loud tone makes an excellent trigger, and the half-cycle period is so short that you don't need any other input conditioning or a relatively uncommon optocoupler.

    A little digging turned up this: http://www.kenrockwell.com/apple/ipad-1-audio.htm

    Unloaded max output: 0.9509 V RMS, or 1.34 Vp
    75 ohm load output: 0.8931 V RMS, or 1.26 Vp


    Here is an updated circuit. Besides driving a voltage doubler, the input capacitor presents a series impedance of 50 ohms at 1 kHz to current-limit the load on the iPad.

    EDIT: Wrong capacitor size, slipped a decimal point. 1.0 uF is a good compromise value between gate holdup time and input impedance.

    ak
    PhotoLights-2-c.gif
     
    Last edited: Sep 21, 2015
  16. dannyf

    Well-Known Member

    Sep 13, 2015
    1,769
    358
    The simplest would be to take out the triac and replace it with a relay, or a mosfet, for dc application.

    The primary side has nothing to do with the load side.
     
  17. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,515
    1,246
    Interesting conclusion. I'm not sure a modified DIAC makes a good DC relay coil driver, especially with a 3V max output Vf, 25% of the power rail. Also, adapting a circuit designed to make lights flicker intentionally to an application that requires steady brightness probably needs some attention to the input signal. A TRIAC load doesn't care because it latches on with the first input peak. A relay coil won't like kilohertz pulses, and MOSFET will track them.

    ak
     
Loading...