Relation between frequency and noise

[priya]

The active low pass filter cut off frequency is at 2.34HZ (0-10Hz) range. Is there any relationship between noise and frequency? As the frequency range is very low here, the signal can be effected by various noises .These will be amplified when given to an operational amplifier. Could you please tell me how can i remove noise to some extent without changing my cut off frequency? I can't change the cut off frequency because my project is related to measurement of heart sounds which are at very low frequency and amplitude .

[suicidaleggroll]

"Noise" at <2 Hz is typically caused by things like thermal drift, board flexing, breezes/drafts in the room, etc.  Make sure you use parts with very low thermal drift, including op-amps (look at offset voltage and bias current temperature drift in the datasheet).

That's a very low cutoff frequency for an active filter (any filter really), what kind of resistances/capacitances are involved?  Things that would normally not be an issue start to become problematic when you're talking about megaohms and millifarads.

[Zero999]

It's quite likely to be pink noise at that frequency band. Flicker noise is also a source but modern op-amps are very good.
https://en.wikipedia.org/wiki/Pink_noise

[DaJMasta]

Low frequency noise is often dominated by thermal and current noise, and this will come from your input amps (and the first stage with the lowest input signal will be the most sensitive to noise and the thermal noise in nV*sqrt(Hz) will rise at lower frequencies), but it will also come from the resistors and other parts used.

Thermal noise from resistors goes up with temperature and goes up with resistance value, current noise comes from the fabrication process for the resistors - thick film is pretty bad, thin film is better, wirewound is better again, and metal foil is the top tier - similar to wirewound, but without all that pesky inductance and with lower drift.

So aside from the noise inherent to your amp, very low noise front ends are going to be very sensitive to component choice.  Even leakage through caps or across the surface of the board can be an issue if you're looking at very low levels (and something like a polyester cap and guard traces can help overcome those).

There's a surprising amount of detailed information about low noise design available in app notes, so I'd take a look at notes for very low noise op amps and instrumentation amps, as well as app notes on thermal and current noise.

[Audioguru]

Maybe you are using noisy ordinary opamps instead of low noise audio opamps? My Electronic Stethoscope-2 project uses TL072 low noise audio opamps (there are newer opamps now with much less noise) with a simple first order highpass filter that has a cutoff frequency of 2.8Hz and noise (rumbling?) is not heard. Maybe your input cable is not shielded and it is picking up mains hum?