Design of a Logarithmic Amplifier for Partial Discharge Detection

Abstract

This paper focuses on the different features of logarithm Amplifier. This thesis explores the concept of applications and theory in relation to logarithm Amplifier. A logarithmic amplifier is an amplifier which provides the logarithmic function. Linear amplifiers produce an output which is equal to the equation y=kx, where y is the output, x is the input, and k is a constant. If the amp is unity-gain, y=x because k=1. If the amplifier linearly increases the input signal, k will be greater than 1. If the amplifier reduces the input signal, but retains the linear relationship, k will be less than 1. About 7 years ago, when the author's spectrum analyser project was entering its final construction phase, it was worth finding the simplest possible logarithmic intermediate-frequency amplifier. The limiting conditions were: 70dB dynamic range, band width at least 200 kHz. The existing solutions at that time were the expensive logarithmic intermediate-frequency amplifier from DL8XZ and a log amplifier using a Plessey integrated circuit, SL521 or SL523. The latter variant can be found in relatively simple radar equipment for marine navigation, and requires four SL523's. The intermediate-frequency amplifier from DL8XZ was eliminated, in spite of the high precision of measurement and the dynamics, due to the space required and the cost of smoothing. The author next worked out a variant of the DL8XZ amplifier with differential stages and high-frequency side summation, which was presented in issue 2/1987. It was certainly almost free of smoothing, but the cost of 18 transistors for 70dB dynamics was considerable.

In the second version, therefore, the author used an FM intermediate-frequency circuit with a logarithmic field strength measurement output. The choice fell on the TDA1576 from Philips. Very little expense is involved and the logarithmic precision is fully sufficient for radio equipment (owners of Japanese radio equipment can only dream of such S-meter precision), but the short waves still stand out in an unpleasant way on a gauge.

Acknowledgements

I would take this opportunity to thank my research supervisor, family and friends for their support and guidance without which this research would not have been possible.

Table of Contents

Abstractii

Acknowledgementsiii

Nomenclaturev

Chapter 1 Introduction1

Chapter 2 Literature Review14

Chapter 2 Literature Review14

Chapters 4: Single pulse calibration37

Chapter 5: Logarithmic Amplifier Illustration39

Chapter 6: Discussion41

Chapter 7 Conclusions & Recommendations/Future Work44

End Notess46

Nomenclature

LOG AMP: Logarithmic Amplifier

ADC: Analog-to-digital converter

IEEE: Institute of Electrical and Electronics Engineers

FPGA: Field-Programmable Gate ArrayChapter 1 Introduction

This project will involve the design of a logarithmic amplifier for a range of partial discharge amplitude signals so that an 8-bit or 10-bit ADC card will be able to cover a wide dynamic range of amplitude signals. The specifications for the amplifier will be as follows:

Frequency Bandwidth: 100kHz - 50MHz

Dynamic Range: 1mV to 5V (74dBm)

Input Impedance: 50 ohms;

Output Impedance: Selectable 50 or 1M

Background Study & Problem Statement

The detector described here has been developed in order to work in pulsed or very intense neutron fields.

The detector specifications have been defined in reference to beam loss monitoring in hadrontherapy accelerators ...