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.

OPA1637: noise simulation using PSpice for TI

Part Number: OPA1637
Other Parts Discussed in Thread: TINA-TI

Tool/software:

Hi,

I want to calc OPA1637 noise using PSpice for TI. Looking at datasheet Fig. 6-1, Input-Referred Voltage Noise seems under 4nV/sqrt(Hz) @ 1kHz.

PSpice output over 10nV/sqrt(Hz).

Could you let me know the tips, hints for this? Thanks.

  • Hello, 

    General Resources:

    I used Tina Ti to compare. Tina Ti is a free download from Texas Instruments. I have attached my simulation below as well. You can see my curve below is below 10nV/rtHz. It still isn't perfect but much closer. 

    I made the resistors noiseless resistors. In TINA-TI, a noiseless resistor can be simulated by disabling the noise contribution of a standard resistor. This is achieved by setting the temperature of the resistor to absolute zero (-273.15°C), which effectively eliminates thermal noise. TINA-TI allows for noise analysis, and by manipulating component attributes, individual noise contributions can be isolated and evaluated

    1.  Standard Resistor Noise Model:

      In TINA-TI, resistors are modeled as ideal noiseless resistors in parallel with a thermal noise current source. This noise source is a white noise, meaning it has a uniform distribution across all frequencies. 
    2. Disabling Resistor Noise:

      To simulate a noiseless resistor, you need to effectively turn off the thermal noise generated by the resistor. This is done by setting the temperature of the resistor to absolute zero (-273.15°C). This can be achieved by modifying the component's properties within the simulation environment. Double click the resistors and make the settings as shown below:
    3. Noise analysis in TINA-TI:

      TINA-TI offers a dedicated noise analysis feature (Analysis > Noise Analysis). This analysis allows you to examine the noise contributions of individual components, including resistors, within a circuit. 
    4. Isolating Noise Contribution:
      To isolate the noise contribution of a specific resistor, you can disable the noise of all other resistors in the circuit by setting their temperatures to absolute zero. This allows you to focus solely on the noise generated by the resistor of interest. 

    Here is a noiseless resistor macro:

    Using the Analog Engineers Calculator we can calculate the noise contribution of the 2k ohm resistors. I believe if you set your resistors to noisless resistors to isolate the noise to the amplifier you will get much closer results like I did. 

    OPA1637.TSC

    Best Regards, 
    Chris Featherstone

  • Hello Chris,

    Thank you for helpful reply. I can understand that need sqrt(s(v(onoise)*v(onoise))) to convert V/sqrt(Hz) to Vrms. And will check TINA-TI later.

    Unfortunatelly I can't find use 'noiseless' resister for PSpice. Is there more information? 

    Best,

    Yasunori Okutsu

  • Hello, 

    I found a video from Cadence on this topic that I believe will be helpful. Let me know if this worked for you or if I can provide additional guidance. 

    Pspice Noisless Properties

    Best Regards, 

    Chris Featherstone

  • Hello Chris,

    Added option:

    @PSPICE:
    .OPTIONS PRINT_NOISE_CONTRIBUTORS=10

    seems work well. Thank you.

    But adding "NOISELESS=YES" to the resistors properties, seems to be ignored. The result is:

    **** Top Contributors To Output Noise [DeviceName = TotalDeviceNoise (Contribution percentage)] ***
    X_U1.X_U1.D1 = 6.390E-15 (49.26 %)
      R_R1 = 3.315E-17 ( 0.26 %) 
      R_R3 = 3.315E-17 ( 0.26 %)
      R_R2 = 3.315E-17 ( 0.26 %)
      R_R4 = 3.315E-17 ( 0.26 %)
    X_U1.X_U1.R4 = 2.497E-17 ( 0.19 %)
    X_U1.X_U39.D1 = 2.389E-18 ( 0.02 %)
    X_U1.X_U40.D1 = 2.389E-18 ( 0.02 %)
    X_U1.X_U39.R4 = 2.404E-19 ( 0.00 %)
    X_U1.X_U40.R4 = 2.404E-19 ( 0.00 %)

    In stead, I added "R_NOISELESS " before the value:

    The result is same as your's TINA-TI.

    Best regards,

    Yasunori Okutsu