• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 47 subscribers
  • Views 237 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

OPA691: Opa691 noise model seems not correct

Jean-Luc Bolli
Prodigy 190 points
Part Number: OPA691
Other Parts Discussed in Thread: TINA-TI

Tool/software:

Hello,

I am using le OPA691 Spice model with the LTSpice to simulate my design. All seems to work correctly, except the noise behaviour.

Simulating a non inverting configuration with a gain of 5.0 (see attached) files, I get an output noise density of 600 nV/sqrt(Hz). It is obviously about 100 time too much! Could you please check this model on your bench?

Thank you in advance for your insight, Jean-Luc Bolli

OP691NoiseIssue.zip

  • 0
    TI__Genius 17045 points

    Hi Jean-Luc,

    Unfortunately, we cannot offer support with LTSpice simulations due to licensing. However, I was able to simulate both current and voltage noise in TINA-TI and the values are in line with the actual device. Broadband voltage noise was slightly lower than the datasheet spec and the non-inverting current noise was around 9pA/Sqrt(Hz) higher. Inverting current noise matched the spec in the datasheet. Is there an image of the circuit you can share where you are seeing the higher noise value?

    Best Regards,

    Ignacio

  • 0 in reply to Ignacio Vazquez Lam1
    Prodigy 190 points

    Hi Ignacio,

    Sure, I did not expect from you LTSpice support! I was just suspecting the OPA691 Spice model. Now you have validated this model with Tina Spice, it seems more complicated...

    Within the files I have uploaded, the image OPA691_noise.emf shows v(onoise) at 600 nV/sqrt(Hz), which is obviously wrong. Did you see it? My hand calculation for the circuit given with image OPA691.emf, using formulae (5) from OPA961 data sheet, gives 12.4 nV/sqrt(Hz) for the output noise density.

    Best regards, Jean-Luc

  • 0 in reply to Jean-Luc Bolli
    TI__Genius 17045 points

    Hi Jean-Luc,

    I was a little confused with the calculation. Is the calculation you are referring to the .Meas that integrates the voltage noise? If so, I believe this calculation gives you the total noise in RMS. I was able to calculate the output referred voltage noise using the equations in the datasheet and got a very similar result in TINA and to the value you also calculated.

    Best Regards,

    Ignacio

  • 0 in reply to Ignacio Vazquez Lam1
    Prodigy 190 points

    Hi Ignacio,

    No, I was referring to a calculation with the data sheet formulae. You are right, the .MEAS expression gives the total RMS noise.

    In between, I have downloaded Pspice for TI and run a noise simulation of the same circuit with the OPA691. I get also results close to the hand calculations, like you. Thus for me this issue can be closed, the problem lies into a strange interaction between the OPA691 model and LTSpice. Many thanks for your implication in analysing this problem.