Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 49 subscribers
  • Views 1555 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.

OPA4348: OPA4348 application problem

昆 袁
Prodigy 50 points
Part Number: OPA4348
Other Parts Discussed in Thread: TINA-TI

Hi! When I was using  OPA4348AIDR, I had some problems.The output of op amp is oscillating, and the following are the circuit diagram and waveform.

Take circuit 2 as an example.Only power supply to the circuit, There was no input signal at the input end. C106 position weld 1nF capacitor.The MINSIGN-B produced an oscillating wave and it got back to normal when I removed C52.

The circuit 1 produced another oscillating wave.

How do I know the steady state of the circuit and the τ？Or how do I know the stable parameter. I had simulated the circuit in the TINA.

Can I get your swift support?

 

 

  • 0
    Prodigy 50 points

    Circuit 2- C52_wave

    Circuit 2:

    Circuit 1:

    Circuit1-C25_wave:

  • 0 in reply to 昆 袁
    Guru 140200 points
    Hi,

    oh yes, there's a heavy mistake in your circuit: C105, C106, C35 and C55 must be removed!! They introduce a dangerous phase lag, which erodes the phase margin of OPAmps. The result is instability and oscillation!

    Kai
  • 0 in reply to 昆 袁
    TI__Guru* 93105 points

    Hi,

    Kai's observation about connecting capacitance across an op amp's inputs is correct. Stability problems come about from the phase shift added to the op amp feedback loop by the feedback resistance and this added capacitance across the input. The feedback goes from being the intended negative feedback, to positive feedback at some frequency, resulting in oscillation providing the gain is 1 V/V, or greater, at that frequency. Adding capacitance across the feedback resistor can help compensate for the phase shift, but that capacitance has to be appropriately selected for correct compensation. 

    If you must have input filtering before the op amps and want to isolate the filter capacitances from the inputs, try splitting the filter input resistors into two separate resistors. Then bridge the filter capacitor across the circuit points where the input resistors are split into two separate resistors. For example, in your schematic R61 and R67 would be split into R61A, R61B and R67A, R67B, respectively. C55 would be connected across the R61A, R61B and R67A, R67B connection pints. The sum of R61A, plus R1B, would equal the original value of R61, etc. 

    The R and C values would have to be recalculated to satisfy the filter cutoff requirements and the gain requirements of the stage.

    Regards, Thomas

    Precision Amplifiers Applicaitons Engineering

  • 0
    TI__Expert 6700 points

    Hi,

    It seems your circuits are unstable, so output are oscillate by itself. According to my analysis in TINA(I have attached them), the root cause is the output caps on the each amplifier't output.

    Circuit1 is the first stage stability analysis circuit in TINA-TI. Figure 2 is the test result of this circuit. As you could see, due to the pole in Aol curve formed by output resistance of amplifier, your isolated resistor, output cap, this is a low frequency pole. And 1/belta cross Aol curve in -40dB region. So this lead to a poor phase margin 27 degree, much less than 45 degree.

    Use same method could get the result of second stage, as Figure 3 shows. The phase margin is 39 degree, also smaller than 45 degree. 

    To solve this problem, you have two choice, reduce the caps value or increase the resistor value. I have reduced the out caps in first stage to 470p, and get 48 degree phase margin. And the second stage, I reduced the output caps to 2.2nF, get a 63 degree phase margin. You could use this as a reference.

    For a detail stability analysis, you could visit the following link and watch our precision Lab - Amplifier stability video.

    Regards,

    Will

     

    Figure 1

    Figure 2 

    Figure 3

    First stage.TSC

    Second stage.TSC

  • 0 in reply to kai klaas69
    Prodigy 50 points

    Dear Kai,

         The aim  of connecting caps across an op amp's inputs is to reduce the interference of the interphone. I found that the circuit didn't work well when the interphone was using. 

  • 0 in reply to Will Wang47
    Prodigy 50 points
    The aim of connecting caps across an op amp's inputs is to reduce the interference of the interphone. I found that the circuit didn't work well when the interphone was using.
    So I want to know if caps can be added between op amp's inputs.
    Another question is that there are no input signals in the circuit board's input ports, but I see your simulation circuit's input end is connected to the GND . Does it have any influence on simulation?
  • 0 in reply to Will Wang47
    Prodigy 50 points
    Hi, why add L1& C6 in the circuit
  • 0 in reply to 昆 袁
    TI__Expert 6700 points
    Hi,

    Just as Kai and Thomas said, caps on amp's negative input pins should always be avoided, this cause stability problem. However, in your circuit, the big cap(100n) parallel with feedback resistor help compensate this. So it's ok.
    If you do stability, you should modify your circuit. The big L1 is use to provide DC route for circuit, let the circuit have on dc stable point. And the big C6 is used to provide AC route for circuit analysis based on the dc stable point.
    Again, if you want know more about circuit stability analysis, please refer to our TI Precision -Amps Stability analysis video.
    training.ti.com/ti-precision-labs-op-amps-stability-1

    Regards,
    Will
  • 0 in reply to 昆 袁
    Guru 140200 points
    Hi,

    as Will mentioned L1 and C6 are only needed for the special stability simulation.

    Connecting caps directly between the inputs of OPAmp is not wise. This can work in very limited situations. But it's no standard methode to make an OPAmp application noise free!!

    What is connected to the left of your circuit? What is the purpose of your filtering?

    I think your circuit can be improved to make it much more immune against the interphone radiation.

    Kai