Investigation of Environmental Noise in Small Electrical Conductors

Abstract

Large excess resistance noise has been observed in small conductors surrounded by air. The conductor resistance was found to have a well-defined average power spectral density over the observed frequency range from 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-4</sup> to 200 Hz and the dependence of the noise was measured as a function of bias current, temperature, temperature coefficient of resistance (TCR), spatial correlation, and state of the surrounding air. In this paper, three different mechanisms were identified that produce the noise. The room-temperature fluctuations were measured and found to have a spectral density nearly proportional to f <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-2</sup> over the observed six-order-of-magnitude frequency range. The lowest frequency noise around 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-4</sup> Hz could be predicted from the measured temperature fluctuations using the TCR. Above 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-3</sup> Hz and below 1 Hz, enclosing the wire in a box greatly reduces the noise, and placing the wire in a vacuum eliminates the predominant noise. This noise was directly related to the temperature of the conductor, somewhat proportional to the TCR, independent of bias current, and has a correlation length smaller than the specimen size. The highest frequency noise does not depend on the conductor temperature, TCR, or the presence of air. It had a very strong dependence on bias current and had a long spatial correlation. The mechanism that generates this noise is not understood