Using differential Amplifier on LTC1968, and the input voltage changes; please help

I am currently using LTC1968 to output DC RMS value from a sine wave signal input, and the output of the LTC1968 is connected to the differential amplifier. The issue is after connecting the LTC1968 output to the differential amplifier, the LTC1968 signal gets affected (the voltage decreases). Can anyone please explain me why this is happening, and how can I fix this issue?

LTC1968: https://www.analog.com/media/en/technical-documentation/data-sheets/1968f.pdf

I hope to hear your response.

Thank you



fixed mistakes.
[edited by: shurp at 7:10 PM (GMT -4) on 21 Sep 2021]
  • +1
    •  Analog Employees 
    on Sep 21, 2021 8:58 PM

    Hi Shurp:  Yes, this happens simply because the LTC1966/7/8 output is not buffered.  The data sheet spec for output impedance of LTC1968 is 12.5k Ohms typically.   So you need to buffer the outputs before doing the difference amplifier.  Or, you could use a much higher impedance difference amplifier (such as LT1991), except that the + and - inputs of a difference amplifier present different loads upstream.  The +input presents 2R to ground, while the -input presents 1R to +input.

  • Just looked at the LTC1991, and it looks promising as it comes as a differential amplifier with a higher input impedance. Could you please explain me slightly more about your quote "+ and - input of a difference amplifier present different load upstream."  I have looked at the output impedance of LT1991; since, the differential amplifier will be operating with a DC signal coming from the LTC1968, the output impedance of LT1991 will be close to 0.1ohm (based on the Output Impedance vs Frequency). I should be able to connect it directly to a microcontroller. Would you agree with this?

  • I have looked into instrumentation amplifier as they have high input impedance. Can you please recommend me one that you believe works well with LTC1968. I have ordered LTC1991 differential amplifier but I do want to use an instrumental amplifier if possible. I hope to hear your response.

  •  Per your recommendation, I used LT1991, and I increased the gain until the input was not affected. Below is the diagram I chose to use; the diagram below is on Page 19 of the datasheet. The issue I currently have is that the output is not stable. Even after using a 1uF capacitor on the output, there is about 10mV fluctuation, measured by a microcontroller (Arduino nano). Just to reiterate, the input to the LT1991 are signals from LTC1968 which seems stable. Not sure why the output of LT1991 fluctuate. I hope to hear your thoughts on this.

    LT1991:https://www.analog.com/media/en/technical-documentation/data-sheets/1991fh.pdf