Stange op-amp simulation

Discussion in 'The Projects Forum' started by GRNDPNDR, Mar 26, 2012.

  1. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    Alright heres another op-amp related question that's boggling me, and I'm hoping it is nothing more than just an odd simulation thing.

    Using multisim, with the attached circuit, I'm getting these odd voltages. For the keen of eye you will see there is no power to this.

    I would expect that there would be 0V across ground and ANY input/output so why would there be voltage in this circuit?

    Secondly, in a real life scenario would I require diodes on the outputs to stop voltage from leaking back? As shown in the other attachment.
    EDIT - sorry the other image is the wrong one. There was no diagram with the blurb about the dioe being needed, but here is the quote.

    Anyone know why the schotkey diode is needed in the third diagram though?
     
  2. mlog

    Member

    Feb 11, 2012
    276
    36
    Usually the LM324 has its non-inverting (+) input biased at 1/2 of the op-amp supply.
     
  3. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    Meaning what exactly? The input should be at 4.5V?
     
  4. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    Some op amp macromodels have internal sources. Some of these models don't respond well to zero volt or unconnected power supplies. I have had similar problems with LM324 sims.
    It looks to me like your power connector (HDR1X3) is designed for dual (= and -) supplies.
     
  5. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    Ya thats where the main power will go, and DIP 6 will be the connection to the main power switch.

    I've noticed some problems and fixed them but still having the same overall problem.

    I'm getting mixed responses about using capacitors on the supply and I don't honestly know why they are there, and also been told the rails need to be pulled to ground? remove the caps and use 1M resistors, etc.

    This is the updated version but not much has changed except the addition of the op-amp power supply, and the AC supply to emulate the drumpad, and a couple of minor things found and fixed.
     
    • omg.JPG
      omg.JPG
      File size:
      237.3 KB
      Views:
      7
  6. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    This is with the power turned off.

    One other thing I noticed is that if I connect ch2 of the scope to the second op-amp, with NO signal input, it's sitting at 7V output.

    This cannot just be simulation can it? no input, so i should have no output.
     
    Last edited: Mar 26, 2012
  7. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    I assume you are simulating with power off because you are looking for a way to bypass the circuit. Is that correct?
     
  8. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    that is one reason, but before I move on to that I'm trying to figure out if my circuit is actually correct and working, or if these things i'm seeing with the power off, or certain things disconnected are going to happen in real life.

    With no power I should not see any voltage on the outputs of the op amps?
    no power on the rails (0V?)?

    but basically it's doing strange things when there is no power applied.

    As I mentioned just previously, if I provide NO input to the op-amp, I'm seeing 7VDC on the output. This can't be normal.

    It didn't do that when I tested a single op-amp in real life.

    Not sure if it will help or be of use to anyone but here is also the MS11 file for use in multisim.
     
    Last edited: Mar 26, 2012
  9. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    With no power, you won't get any DC voltage on the output, regardless of what the sim says.
    With a floating input, or a piezo pad with no resistor to ground, you will get about 7.5V on the op amp output. The op amp must have a DC path to ground for input currents. DC cannot flow through a piezo element. LM324 inputs float high. I would connect 100k from each input to ground.

    EDIT: An AC source will not simulate a piezo pad, because an AC source in spice has a DC path to ground. Add a capacitor in series, like 1uF, to simulate the piezo pad.
     
  10. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    well, that's interesting. Adding a 100K resistor as you state, drops that 7.6V right down to 1mV. using a 20K resistor gets the 7V down to 40uV

    I would consider this a positive change and it seems like it could be solving the problem?!

    This actually makes sense though now that you explain it like that. Op-amps flow no current....ideally, so it has to go somewhere.

    However when I turn it "off" IE cut the battery power, this is what happens.... I'm reading the voltage on the VMM across virtual ground/+ rail.

    Now obviously in real life this can't happen just like you say, so again it goes back to wondering if I do have things connected properly and this is just a simulation error, or if I still might have something incorrect.

    What I've done now though is to just disconnect the second half of the circuit (The right side).

    Anyway this could be correct now, it would appear to my meager mind that anything happening when it's turned off simply cannot happen in real life, and therefore this looks like it may be working.
     
  11. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    I ran a sim on the circuit snippet you posted on your other thread. With no power, I was getting megavolts on some nodes. Spice blows up sometimes if the models aren't designed for that condition. If an op amp is modeled with the actual internal circuit (it could have dozens of transistors), a sim will take a LONG time, and may not even converge, so the mfrs provide macromodels to make the sim times manageable, and to protect their IP. I was running LTspice, so I switched over to an LT op amp, and the sim with no power ran fine.
     
  12. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    So what your saying is, the circuit works :)

    I didn't know that about simulations though, thanks for clearing that up.

    Should I continue with the LM324 or should I look into a different op-amp?

    (I was reading some stuff, apparently these are "old" and there are better technologies now?! better amps?)
     
  13. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    LM324 isn't for hi fi audio, but it should be fine for a drum trigger, if that's what you're using it for.
     
  14. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    Ya nothing fancy, it's just to boost the drum signal, and alone it works just fine.

    Thanks for the help here ron, I really appreciate it, and you are one hell of a knowledgable guy.
     
  15. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    You might change your tune before this is over.:rolleyes:
     
  16. mlog

    Member

    Feb 11, 2012
    276
    36
    A more accurate statement is that op amp inputs usually aren't biased at either rail, but I get the impression that linearity isn't a concern with your project.
     
  17. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    And since his rails are ±9V, biasing the input at ground is good.
     
  18. GRNDPNDR

    Thread Starter Member

    Mar 1, 2012
    435
    7
    biasing means what exactly? I'm at a loss of understanding to what this means.

    I swear I'm not an idiot, I'm in school for electromechanical engineering but circuits have always been my bane. Electronics is quite the involved subject and with everything else were doing it's hard to fully absorb some of this stuff, which is exactly why I'm working on this stuff because not only do I have an interest, it's actually teaching me a few things. :)
     
  19. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    656
    See the wiki on biasing. Come back if you still have questions.
     
Loading...