Design questions on LED circuit w/ quadrature oscillator

Discussion in 'The Projects Forum' started by MikeL, Aug 12, 2008.

  1. MikeL

    Thread Starter New Member

    Aug 12, 2008
    Ok, so a premise...
    I have an 8"x4" piece of glass that I etched sometime last weeek and was planning to light with LEDs. However, since the etching is of a shark, (I can provide a pic to those who are curious) I thought of trying to do a lighting scheme where the LEDs slowly brightened and dimmed out of phase with eachother, in the hopes of simulating light coming through waves.

    It took me a while, but evntually I found a design for a quadrature oscillator (link 1)and combined it with another design (link 2) I had found earlier in my search. I just have a few

    link 1 :
    Link 2:

    First big thing is I havent been able to find a way to properly calculate the values needed for the passive components for the oscilator section of the circuit. I am shooting for around .25Hz output as a starting point. The webpage metioned just having to adjust the capacitor value C.

    THe values for the Resistor R, and the voltage V i am pretty sure i know how to get, though I am not 100% sure if I am using the transistors properly. the number of LEDs I am going to use hasnt been set, but I hope to have at least 8.

    WHile i recently graduated with a bachelors in Electrical Engineering, lately I've had to wonder

    I guess I am not asking for all the answers (unless my "design" obviously wont work :p ), though I am hoping for guidance on how to from this point to a functional circuit. So, if you need more information from me, please ask away.

    Thanks in advance.

  2. thingmaker3

    Retired Moderator

    May 16, 2005
  3. MikeL

    Thread Starter New Member

    Aug 12, 2008
    Most Excellent....
    Thanks a bunch, I think i have some decent values I can use now... the only trick is finding an ( i know i may not need it, but its a good excuse to get one )
    thankfully there is an electronics store just down the road that may just have what i need...
    There are a few other concerns I have about this design, mostly how the transistors are going to operate with these inputs, and whether the outputs of the oscillator will be 0 to +V or -V to + V. But those answers will come in time i am sure.

    As for using PWM, I had thought of it before, and am still considering knowledge of PWM applications is not all that great at the moment, so it will likely be my next project :)

    Thanks Again,
  4. MikeL

    Thread Starter New Member

    Aug 12, 2008
    Ok... need some help....
    This is the circuit I currently have set up.
    I've tried to simulate it in TINA-TI, though cant seem to get that to work

    I've attached a schematic of what was currently set up, the only thing missing was the two diodes (which could be my problem, dont know.)

    LED set 4 would come on as soon as power is applied, Set 2 would flash breifly, then Set 1 would slowly brighten (not as slowly as i had hoped, but Set3 Never did anything. The hope was to get the Voltage across the LEDs to follow a curve like that in the wave.jpg, where each color represents a different LED set. (lol did that make sense? sometimes I dont describe things very well)

    Originally I had powered the op amps with +9V.
    I tried to set up a +18 V, which is what the sheet specified, but (stupidly) confused the terminals on the 9V batteries i was using and applied the +9 to the -V pin and the ground to the +V pin. Yeah... dumb mistake.... the popping, sizzling, and ensuing smoke was a good indicator i had botched

    I guess the big question I have: Is there a better OP Amp to use than the one I managed to let the "magic smoke" out of?

    and is there anything else inherently wrong with my design that I am just not seeing?

    Feel free to contact me through Yahoo messenger, or by email as well as this thread.

    I appreciate the help guys.


    PS: These are the links to the OPAmp i was using and for the Pot that I am using.
  5. beenthere

    Retired Moderator

    Apr 20, 2004
    Any dual op amp currently available is going to be better than that horrible (sorry to be judgmental) NTE. Try an OPA2130, although any dual ought to work. Do get it in an 8 pin DIP case - that would be an OPA2130PA. You won't find many in TO-5 cans anymore.

    Your circuit is still going to make the op amp hot with no current limiting resistors between the op amp outputs and the transistor bases. As a guess, use separate 1000 ohm resistors to each base.
  6. SgtWookie


    Jul 17, 2007
    The NTE941 is a drop-in replacement for a LM747, which is a dual 741 opamp, which is a 40+ year old design, and extremely "long in the tooth". When it first came out, it was the greatest thing since sliced bread. Nowadays, almost anything else is better. 741's are still around because they're used in acadamia, and they're cheap - unless you buy them from NTE. NTE's cross references only go one way (to their equivalent replacement part) and are usually many times the price of the original part.

    Have a look at LF353's; they are a decent and somewhat more modern dual JFET-input opamp that's available for under $0.50/each.

    An LM324/LM2902 is a single-supply quad opamp whos' common mode input goes down to ground. The LM2902 is an automotive (high-temp) version of the LM324. It's an old design, but still available and inexpensive.
    Last edited: Aug 18, 2008
  7. MikeL

    Thread Starter New Member

    Aug 12, 2008
    lol.... yeah i thought something was odd in that there were no cross references to that part anywhere else. and $5.16 US did seem a bit steep for that thing.... just goes to show that i wont be buying much from that store again.

    BUt yes, thank you for the suggestions, I'll try giving them a shot sometime this week. lol here's hopin.....

    thanks again guys.