AnsweredAssumed Answered

Johnson noise amplification and filter circuit design review

Question asked by dhhsieh on Mar 26, 2017

Dear EZ community,


I have been working on a Spice deck of a Johnson noise amplifier and filter circuit. Its purpose is to simulate the results of a research project to demonstrate measurement of thermal conductivity, for a sample which has a microheater line resistor laid on top. Current is passed through the resistor, heating the sample, causing the temperature and resistance of the resistor changes, producing a voltage which contains thermal information about the sample. It is known as the 3-omega method in published literature, but this project is to use the Johnson noise of the resistor instead of the temperature coefficient of resistance of the resistor. This deck simulates the amplified and filtered voltage noise of the resistor. Since I am a mechanical engineer working outside my domain, I am requesting that someone or multiple people help me check and improve the circuit.
The inputs to the Johnson noise circuit (first attachment) are two signals which include noise, one of which is slightly different in its noise. The noises are subtracted from each other and the result is then high-pass filtered, gained, bandpass-filtered and analog multiplied by itself, producing a squared voltage signal in units of volts which can be sent to a lock-in amplifier. Since I don't know how I haven't fed realistic noise inputs, but I found that Spice could take AC inputs which result in measurement of the circuit's own noise. In the second attachment, the inputs come directly from the 3-omega circuit as is proper.
Please let me know if you need more information or if there is some other issue. Thank you in advance for any help or advice.
attachment1: Johnson noise circuit alone
attachment 2: 3-omega and Johnson noise circuit combined