1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Triangular Wave Guidance

Discussion in 'Electronic Projects Design/Ideas/Reviews' started by blah2222, Aug 9, 2011.

  1. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    What current are you trying to measure? Is it R7 or R9 or both? If you're trying to measure IR7, then you don't need U5. Either way, I don't think your curcuit will work as shown. Help me to find what exactly you're trying to do, and I might be more help for you.
     
  2. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Yep, I'm trying to measure the current flowing through both R7 and R9. The parallel resistor/capacitor impedances are models for what is happening at the electrode/solution interface, it called the Randles circuit. The values for R7/R9, as well as the capacitors change with time and bias voltage applied. I picked these values as they are the average of the range.
     
  3. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    The problem is R8, R9 & C3 load the output of U2A, and created a voltage divider circuit in the feedback circuit, which in turn overloads your amplifiers. You need a follower between them to prevent this from happening. But you'll still measure a current through R7/R9. The follower won't mess that up.

    BTW, is R7 and R9 two different probes?
     
    Last edited: Aug 15, 2011
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. Warpspeed

    Warpspeed Member

    Joined:
    Sep 12, 2009
    Messages:
    182
    Likes:
    3
    Location:
    Melbourne, Australia

    You are on the right track with this, you just have a fault somewhere in your hardware that needs to be fixed.
     
  6. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Sorry for my lack of understanding, but where am I to connect the voltage follower? R7 and R9 are connected together by a 100 ohm resistor R8, if that's what you meant by probe?
     
  7. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    If you insert a follower ahead of R8, inline with the bottom-most conductor in your drawing, that should go a long way to solving your problem. Never mind my question about the probe. I went back to post #35 and got the answer.
     
  8. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Hey, so I tried what you said and I'm still getting a weird result for the voltage (green) at the non-inverting input of the transimpedance (U5A).

    [​IMG]

    Without Follower - 5000 minimum time points

    [​IMG]

    With Follower - 200 minimum time points

    [​IMG]

    For some reason I had to lower the minimum time points down to 200 for the follower circuit simulation or else all the voltages would oscillate at a very high frequency between +/- 2 V. The green voltage still doesn't want to stay down. Did I set up the circuit correctly?

    Red: triangular vbias
    Purple: negative vbias at reference electrode
    Green: non-invering input of U5A
    Blue: output of U5A

    Thanks,
    JP
     
    Last edited: Aug 16, 2011
  9. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    Ok, I didn't know what those traces were before. I though blue was the output of U3A. Looks like you have too much gain in U5A. Try making the 5K resistor = 1k for unity gain.

    I can't explain the high frequency oscillation.
     
    Last edited: Aug 16, 2011
  10. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    The oscillation might be created with the U3A/U4A loop along with the capacitor.
     
  11. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    It's weird, for some simulations if I want to use a minimum of 1000 points or more, I get a 4 Vp-p ~10 MHz oscillation... As soon as I lower it to around 500 - 1000 points it's fine, but I can't tell whether the signals are getting clipped due to the circuit, or that there aren't enough points to represent the peaks.

    For now I did some testing with keeping Rf = 10k on U5A and playing around with the working and counter electrode resistances. I made a mistake with my ranges beforehand, those resistances can be between 10k and 100k in reality. So what I noticed is that if Rwork is larger than Rcount, the reference voltage signal (-Vbias) comes out nicely (not clipped), but when Rcount > Rwork, the signal is majorly clipped and distorted.

    Again, because of this weird oscillation I can't fully tell whether the signals are behaving normally because I am limited to 500 - 1000 points.

    Any thoughts?
     
  12. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Another thing I noticed was that the inverting input of U3A would be held at zero, except when the generated triangular wave was in the range: 1.5 < |Vtri| < 2.

    I put a 1M resistor between the inverting and output of U3A to keep that grounded, and it seemed to work, but I can't fully tell because Multisim is oscillating like crazy now...
     
  13. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    I think the oscillation is caused by U1A. The capacitor phase-shifts the feedback signal, which at 180 degrees will apply positive feedback to the control amp. Before you added the 2nd follower, maybe you were attenuating the feedback signal so you weren't getting the oscillations.
     
  14. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    To show you what I mean about the voltages clipping:

    [​IMG]

    [​IMG]

    Vbias
    -Vbias (non-inverting input U4A)
    inverting input U5A
    inverting input U3A

    The green signal clips by the blue signal. I need to figure out how to keep the blue signal grounded without affecting the rest of the circuit...

    Getting closer, very slowly...
     
    Last edited: Aug 16, 2011
  15. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Hm, that might be it, but the main problem where I notice oscillations is if I choose to have more data points simulated on the plot. If I stay under 1000 points, then I don't see an oscillations, otherwise it is apparent. I think it might be the plotting software tripping up.
     
  16. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    The blue signal is not grounded because U3A is overloading as I described earlier. It should have a gain of 1, but it's over 1 because U5A is loading the output of U3A. What is Vbias? Can you put labels on your schematic that correspond to your signal names?
     
  17. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Okay, sorry about that, I edited the schematic above and put labels corresponding to the coloured traces.
     
  18. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Any possible way of getting around U5A overloading U3A to keep the blue signal grounded?
     
  19. BrownOut

    BrownOut Banned

    Joined:
    Jun 10, 2009
    Messages:
    4,630
    Likes:
    60
    Location:
    Huntsville, Alabama USA
    Yes. Use a follower.
     
  20. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    Hrm, I have tried that but I get the same result, if this is the configuration that you mean, U6A is the added follower:

    [​IMG]

    [​IMG]
     
  21. blah2222

    blah2222 New Member

    Joined:
    Jul 29, 2011
    Messages:
    41
    Likes:
    0
    Location:
    Canada, eh?
    I am not sure if that is where you mean the follower should go. Is that what you mean, based on my schematic above?
     

Share This Page