My Pspice circuit won't oscillate?

Discussion in 'The Projects Forum' started by mjturnqu, Mar 3, 2007.

  1. mjturnqu

    Thread Starter New Member

    Mar 3, 2007
    3
    0
    Hello all -

    I am working on a BJT multivibrator (emitter coupled) circuit. I am having trouble getting the circuit to oscillate. :mad:

    Things I have tried:
    1. added a pulse to VDD
    2. added a transient current source in parallel to the output load.

    Any tips on getting this oscillator to actually oscillate in Pspice? or Aplac?

    Thanks,
    matthew
     
  2. rwmoekoe

    Active Member

    Mar 1, 2007
    172
    0
    hi matthew,
    can u post the schematic plz?
    i can learn from you :)

    thx,
    robert
     
  3. mjturnqu

    Thread Starter New Member

    Mar 3, 2007
    3
    0
    thanks robert ... see attachment
     
  4. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    The capacitors appear to be 1 picoFarad. That seems awfully small. Have you tried using larger capacitor values? At what frequency are you looking for the circuit to oscillate?
     
  5. rwmoekoe

    Active Member

    Mar 1, 2007
    172
    0
    i've looked over the circuit, matthew.

    i understand if u can't agree with one or more of my opinion. still, you're willing to consider, right? thanks. :)

    the evaluation goes like this:

    1. the whole circuit form a perfect positive feedback loop. so it won't oscillate. it will be stable in one or the other position. a monostable it is allright.
    you could make it an astable multivibrator in maybe many ways, one of them maybe to invert the function of q3 and q4 into common emitters of pnp's.
    it seems to me this configuration is what's intended, because the diodes d7 and d6 is acting as current mirror with the complementary q3 and q4.
    you see the q7, q8 and q9? they are current mirrors.

    2. the values for the resistors could be right in their proprositions, only you better scale them up, say 20 to 100 times higher. in the real life, this circuit would generate heat. in spice heat doesn't matter, but i'm afraid it would matter if the saturation current limit of the transistors are exceeded.

    3. you may wanna let go off the loads (1p and 600 ohms), but some resistive load i think may be suitable, at least in one side of the vibrator, also to make sure the initial condition won't be in equilibrium, so it can oscillate.

    hope this helps. thx alot, matthew, i'm learning from you.

    friendly,
    robert
     
  6. mjturnqu

    Thread Starter New Member

    Mar 3, 2007
    3
    0
    Thanks for looking.

    The circuit is similar to a model on p. 672 of Analog Integrated Circuit Design, by David A. Johns so I think it should oscillate with the correct values and excitation in pspice.

    Sorry for not giving the specs... I am trying to get an oscillation of 915kHz.
     
  7. rwmoekoe

    Active Member

    Mar 1, 2007
    172
    0
    mjturnqu, it turn out i was wrong.

    by changing q3 and 4 to pnp's, the circuit will also be stable too. so my suggestion won't do.
    so i ran over it - the original configuration - a few times more thoroughly again this time. over and over.
    and i still found the circuit won't vibrate. 'cause it's stable too.

    but wait, i'll find out using micro cap this time. i'll get to you again after that, my friend.

    thx,
    robert
     
  8. n9352527

    AAC Fanatic!

    Oct 14, 2005
    1,198
    4
    You need to put an initial condition (.IC in SPICE) on one of the node or probably two nodes on the feedback loop. SPICE always has trouble in simulating circuit with positive feedback because it can not determine the initial DC condition at time zero to start the analysis.

    Refer to your SPICE manual on how to put initial condition. Also, search this forum, this topic had came up several times previously.
     
  9. rwmoekoe

    Active Member

    Mar 1, 2007
    172
    0
    i haven't checked it out in microcap yet, mj.
    maybe tomorrow i'll post the result. it's my bedtime now :) (i'm in the other hemisphere to you guys).

    now i only wanna share this thing i've just observed: you mentioned the frequency. i didn't noticed. it's almost 1MHz! maybe it is the problem?
    parasitic capacitance of the transistors must be much higher than 1pf i guess.
    the values of the resistors are too low too. it has to be that way to reach the frequency though. but the current it needs may have nearly saturated the transistors.

    have you tried an LC colpits or hartley oscillators? at that high a frequency, usually an LC-controled oscillator.

    thx,
    robert
     
  10. rwmoekoe

    Active Member

    Mar 1, 2007
    172
    0
    i've tried the schematic you're testing in microcap and as you can see in the first picture, it won't oscillate either. it stabilizes at about 3.46v.
    i've tried my suggestion of reversing the npn to pnp and a few other ways i can think of, but it still won't oscillate.

    i also ran the colpitts. the second picture is the screen capture of the colpitts oscillator. the f is about 2.5MHz.
    i hope this will be of use for you. :)

    friendly,
    robert
     
  11. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    I don't have access to the book you site.

    Is there anyway you can scan or sketch a schematic of the circuit you site so that we may take a look and compare it to the one you are using?

    hgmjr
     
Loading...