1hz-1khz sinusoidal signal -90 degree phase shift

Now, i want to achieve the 1hz-1khz sinusoidal signal -90 degree phase shift, and the signal is not interfered by noise after phase shift (the measurement is the thermal noise of the input resistance, and the voltage is very small and easy to be interfered), if i use the following circuit, is there any recommended operational amplifier(now i used LT1357) or should pay something attention to the circuit?


1. After phase shift, try not to add miscellaneous noise

2. - 90 ° phase shift

3. Are there any operational amplifiers or other circuits recommended?

Really appreciate for your help.

  • 0
    •  Analog Employees 
    on Oct 6, 2021 3:04 AM

    Hi Zijie,

    Apologies for the late response. Can you share your exact schematic? (with the RC values of the above circuit if possible) Analog Devices has a wide portfolio of precision and low noise amplifier. Details like how low is your signal going to be and specific noise level needed would be of help in selecting a suitable op-amp for your application. If its not too much to ask, can you also share what application this is for?

    Thanks and regards,


  • C = 1uF and R = 1kΩ,then i want to use this circuit to shift a voltage which frequency is about ~100hz for 90° (Vin,involved 1mA*10kΩ and  the thermal noise 4kbTRΔf(how to define the Δf is also a question)),but i dont want this amplifier's noise to damage the thermal noise,so i want to choose a low noise amplifier. 

  • Hi Zijie,

    The equivalent scheme you have given raises even more questions. Which electrical quantity is a useful signal here - current or voltage? Because it looks as if the internal resistance of the signal source is 10 kohm and the signal is voltage and you want to apply it to a phase-shifting device with an input resistance of less than 1 kohm, which is strange.



  • Hi Kirill,

    The useful signal here is a voltage signal(actually i want subtract this ac voltage and then extract the thermal noise only).I am curious about why you think it is strange to apply to a phase-shifting device with an iput resistance of less than 1 kohm.If the input resistance should be much larger than the signal resistance?



  • Hi Zijie,

    I guessed that the signal in your case is the voltage, since you specified the Nyquist formula that describes exactly the RMS noise voltage. But if the input impedance of the amplifier/other device is much lower than the internal impedance of the signal source, the signal will be reduced. The input impedance of the phase-shifting circuit with the values of resistors and capacitors specified by you depends on the frequency as follows (for OP27, for example):

    Meanwhile, you specify the resistance of the signal source as 10 kOhm, and it is unclear how you will extract the noise voltage here.