OPA1656: Noise on DSP-Board Output

Hey Emanuel, 

Are you measuring high noise or is this an audible noise you are hearing? The reason I ask is that I see X7R capacitors in the signal path. X7R capacitors have very high Barium Titanate (BaTiO3) content that has the property of being highly piezoelectric. What this means to the signal path is there are little microphones in the signal path that will introduce distortion from movement or vibration. We have a document online discussing this effect and it's comparison to other capacitor types. Here are some resources on this topic:

Stress-induced outbursts: Microphonics in ceramic capacitors (Part 1):

• https://e2e.ti.com/blogs_/archives/b/precisionhub/posts/stress-induced-outbursts-microphonics-in-ceramic-capacitors-part-1

Stress-induced outbursts: Microphonics in ceramic capacitors (Part 2):

• https://e2e.ti.com/blogs_/archives/b/precisionhub/posts/stress-induced-outbursts-microphonics-in-ceramic-capacitors-part-2

More about understanding the distortion mechanism of high-K MLCCs:

• https://www.edn.com/more-about-understanding-the-distortion-mechanism-of-high-k-mlccs/

For more information on op-amp noise we have a training series dedicated to noise:

Ti Precision Labs:

• https://www.ti.com/video/series/precision-labs/ti-precision-labs-op-amps.html

I have simulated the circuit and attached the Tina file at the bottom of the post. The PCM5242 is a differential output voltage output and I have converted a single input to a differential voltage to represent the DAC. Since you are operating on a single supply with a 5 V common mode DC voltage (10V/2) you may need to keep the DC blocking caps and therefore may consider a different capacitor technology alternative to the X7R. The input referred noise of this circuit shows a broadband noise of 20.72 nV/rtHz. 

The resistor noise will be a significant factor. As a demonstration I have reduced the input and feedback capacitors by 10x. The broadband input referred noise is reduced 2.9x as seen below. 

OPA1656B_B.TSC

Best Regards, 

Chris Featherstone

Hey Chris,

Thank you!

We measure the noise with an audio analyzer in dBV. The noise is also audible via a connected amplifier.

So C0G types can be used to improve the noise influence of the capacitors?

According to the data sheet, the input voltage noise density at 100Hz is 11.8 nV/rtHz. If the resistors are reduced by a factor of 10, the input voltage noise density at 100Hz is 15 nV/rtHz. So the optimum has almost been reached here?

At the output of the DAC there are 390R resistors in range to the coupling capacitor. Do these have a negative influence on the noise? And is an impedance matching required on the part of the OPA?

Kind regards
Emanuel Lemnitz

Thank you for the support!

Anytime! Best of luck with your project!