• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 38 subscribers
  • Views 6670 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TINA/Spice/TLE2037: First Order Active Low Pass Filter

Gareth Shelley
Prodigy 60 points
Part Number: TLE2037
Other Parts Discussed in Thread: TINA-TI

Tool/software: TINA-TI or Spice Models

Hello, new to TINA and first post on this forum!

I am designing a cascade filter system for uni, and have got the first part (high pass) done, however I have been having severe issues with the second (low pass). I seem to be getting a strange response at low frequencies, which I should not be. I have contacted my lecturer who tried the circuit at home, and works as it should. It therefore must be something to do with my settings, either in TINA's general setup, or individual components. I was stuck for two days with errors for example when trying to simulate this filter, and I was then told that changing the capacitor's parallel resistance to 1G would fix the feedback loop... So I am unaware if there are any other small tricks that I am missing here. I have uploaded a few images, one of how it is supposed to look (my lecturers), and my one. My one is obviously the one with the initial gain all the way down below -40dB, where it should be nowhere near. I know he has started his frequency graph at 1Hz, and me at 1mHz, but the first pole is still visible on my graph even at 1Hz.

Thanks in advance!

 

  • 0
    TI__Intellectual 1490 points

    Hi Gareth Shelly,

    Adding a value of 1G Ohm resistor in parallel to C2 (by setting Rpar parameter in the component) introduces a "zero" at origin and "pole" at 2*pi*1G*180p (~ 1Hz). So the work-around solution to help you with your errors (DC operating point / convergence issues) is also causing the difference you are seeing versus expected value.

    I suggest the following options:

    1. You can try increasing Rpar value to 1000G and use them for both C1 and C2 - this will give close to ideal results upto 1mHz - going below this will again cause the same issue of gain dropping.

    OR

    2. Set Rpar to Infinity (the default setting) for both C1 and C2. But this might cause errors / convergence issues - so to solve these issues try adding a large resistor (1000G) from -ve input node of amplifier to ground.

    If you need further assistance please share your TSC file.

    Regards,

    Srikanth Pam | Online Design Tools

  • 0 in reply to Srikanth Pam
    Prodigy 60 points

    Hi Srikanth,

    Thank you very much for your reply, option 1 did indeed fix the issue! I now have the correct response using the ideal op amp. I have now changed this for a real op amp however, and it looks completely different? Also, cascading the two ideal filters together, the gain is not what it is supposed to be? On the graph, Vout is supposed to be in line with VF1 in terms of gain, but seems to be offset. I have attached images and the TSC files.

    Thanks again,

    Gareth

    Low Pass Filer Real.TSCCascade Ideal.TSC

  • 0 in reply to Gareth Shelley
    TI__Intellectual 1490 points

    Hi Gareth,

    The above low-pass filter structure has a couple of issues, as you have already noticed.

    1. The low-frequency response is too sensitive to the Cap parallel resistor.

    2. As there is no proper DC feedback - amplifier is not biased correctly. Small input offset voltage of the amplifier will cause the amplifier to saturate close to positive rail / negative rail. As the amplifier is saturated the input to output gain will be close to 0 (-150dB in the case shown above).

    Due to these issues, though the circuit seems to work fine with an ideal OpAmp - it will have problems with real implementation.

    I suggest the following modification to your circuit to make it more robust - less sensitive to Cap parallel resistor and provide a proper DC bias. Added resistors R3=1Meg and R4=10Meg provide 20dB DC gain (R4/R3=10) and also help with the amplifier bias. Note that, to not affect the pole and zero locations R3, R4 should be significantly larger than R1, R2 while maintaining the DC gain ratio.

    Best Regards,

    Srikanth Pam | Online Design Tools

    7041.Low Pass Filer Real.TSC

  • 0 in reply to Srikanth Pam
    Prodigy 60 points

    Hi Srikanth,

    Thank you very much for your help, you are very knowledgable, and have resolved both of my issues. One last question however, and I will stop annoying you, I promise. I have calculated, and created two active filters, both with a gain of 20dB, however with different minimum limits. These look fine when separate, but when cascaded together the response follows the designed curve and frequency responses, but has been moved up the graph to a higher maximum gain of 40dB. The minimum limits have also therefore moved up to a higher number. I have attached images. How is this possible when my resistors and capacitors are responsible for the gains, and are not changed?

    Thanks a million,

    Gareth

  • 0 in reply to Gareth Shelley
    Prodigy 60 points

    Never mind, I figured it out from the graphs. The gain has already been set by the first filter, so the gain of the second should be set to 1. Thank you again for your help, you have been fantastic.

    Gareth