square wave oscillator problems

Discussion in 'The Projects Forum' started by bob800, Dec 28, 2013.

  1. bob800

    Thread Starter Member

    Dec 10, 2011
    48
    3
    I'm trying to build a 4MHz square wave oscillator as part of a solid state tesla coil--the oscillator is shown at the bottom left of the following schematic (bottom transistor is tied to 12V): http://uzzors2k.4hv.org/projectfiles/4MHzclassE1/proper_4MHz_classE_SSTC.gif

    The first picture is a scope reading where the base of the top 2N3904 would be (I had not soldered in any transistors at this point). As you can see it is already quite ringy.

    The second picture is a reading at the same point after soldering in the transistors. I'm not sure I understand why the majority of the signal is now negative?...

    Finally the "4 MHz Drive" output is shown in the last picture, with no load attached (ie I haven't soldered the next transistor stage). What could cause such a distorted waveform?

    Other information:
    --I am using a 20MHz oscilloscope
    --The circuit is powered by the +12V rail of a computer PSU. I put a 470uF capacitor across the supply on my circuit but it only helped suppress the ringing slightly (as shown).
    --I used a tiny surface mount 180pF capacitor before the base of the 2N3906. Are these easy to damage with normal soldering irons? The first time I attempted soldering it I messed up and had to resolder it...
    --I attached a screenshot of the PCB layout I made.
     
  2. MrChips

    Moderator

    Oct 2, 2009
    12,453
    3,371
    What is the make and model of the oscilloscope?
    What are you using as a scope probe?
     
  3. bob800

    Thread Starter Member

    Dec 10, 2011
    48
    3
    Hi,

    The oscilloscope is a Goldstar OS-7020. I am using a cheap 20 MHz oscilloscope probe at the 10X setting.

    The input impedance of the scope is 1Mohm, with 25pF capacitance.
     
  4. MrChips

    Moderator

    Oct 2, 2009
    12,453
    3,371
    The oscilloscope probe should have a grounding clip.
    How are you connecting the grounding clip to the circuit?
     
  5. bob800

    Thread Starter Member

    Dec 10, 2011
    48
    3
    I clipped it to the circuit ground...
     
  6. MrChips

    Moderator

    Oct 2, 2009
    12,453
    3,371
    Do you observe a difference in the scope trace when the clip is disconnected vs connected to circuit ground?

    You should see a difference. If not, check that the ground clip wire is not broken. This is a common fault.
     
    bob800 likes this.
  7. bob800

    Thread Starter Member

    Dec 10, 2011
    48
    3
    Yes I observe a change when disconnecting it. Also when I connect the scope to a 9V battery the voltage jump indicates that the GND wire is certainly fine. I don't think anything is wrong with the scope...
     
  8. MrChips

    Moderator

    Oct 2, 2009
    12,453
    3,371
    I am not suggesting anything wrong with the scope. Usually when I see ringing like that it is because of improper probe ground or termination.
     
    bob800 likes this.
  9. k7elp60

    Senior Member

    Nov 4, 2008
    478
    69
    I don't see anything wrong with the waveforms. The first one appears normal with the oscillator circuit. The second one shows a reduction in amplitude as the 2n3904 transistors load the circuit. The last one is an inversion of the drive as in a common emitter the output is inverted.
    If I read the schematic correctly the bottom 2n3904 will either turn on the top one or turn if off. If the 1.2k resistor on it's base is connected to 12VDC then the drive is a continuous drive signal.
     
    bob800 likes this.
  10. bob800

    Thread Starter Member

    Dec 10, 2011
    48
    3
    OOOOHHH I think I see the problem now. I originally started this thread because the next stage of transistors (the push-pull stage) were getting hot and popping, which I somehow attributed to the slow rise/fall time of the first transistor waveforms.

    However now that I look at the schematic, at the push-pull transistor stage (top left), I think the 2n3906 and 2n3904 transistor labels are swapped. Does that sound right?

    EDIT: It works now!!! Thanks k7elp60 for preventing more wasted time on the oscillator, and MrChips for ruling out measurement errors.
     
    Last edited: Dec 29, 2013
  11. k7elp60

    Senior Member

    Nov 4, 2008
    478
    69
    Yes you are right the labels are reversed as the 2N3906 is PNP and the 2N3904 is a NPN.
    Glad I was able to help
     
Loading...