Lake Shore 475, Руководство пользователя

Страница: 17 / 148

Lake Shore Model 475 Gaussmeter User’s Manual
Background 2-1
CHAPTER 2
BACKGROUND
2.0 GENERAL
This chapter provides background information related to the Model 475 Gaussmeter. It is intended to give the user
insight into the benefits and limitations of the instrument and help apply the features of the Model 475 to a variety of
experimental challenges. It covers basic DSP concepts and how they are applied to the operation of the Model 475, flux
density and Hall measurement, probe operation, and an introduction to field control. For information on how to install
the Model 475 please refer to Chapter 3. Instrument operation information is contained in Chapter 4 and Chapter 5.
2.1 MODEL 475 THEORY OF OPERATION
2.1.1 Sampled Data Systems
Humans rely on analog signals to interact with their environment: individual wavelengths of light are converted to
colors, pressure waves are interpreted as sound, and the vibration of vocal cords creates speech. In the fields of science
and engineering, a variety of sensors are used to convert analog signals of interest into some electrical property, usually
voltage, so that they can be measured or used as an input to a system. Analog-to-digital converters (ADC) and digital-to-
analog converters (DAC) allow computers in the digital domain to interact with these analog signals.
Digital signals are different from continuous analog signals in the fact that they are sampled in time and quantized in
amplitude. Both of these properties limit the ability of the digital representation to match the original analog signal. An
ADC will sample a signal at fixed intervals of time. Quantization results from the fact that an ADC has a limited amount
of resolution. When both the sampling frequency and resolution are properly chosen however, the digital signal is an
accurate representation of the original analog signal. The sampling frequency of the Model 475 allows an accurate RMS
measurement to be made on signals of up to 20 kHz. The sampling and filtering in the Model 475 can allow realizable
resolutions of 20 bits, which is in the noise floor of the instrument.
2.1.2 Digital Signal Processing
Digital Signal Processing (DSP) is the science of manipulating digital data through the use of mathematics. The variety
of processing that can be done is almost endless, from simulating an analog filter, to enhancing a visual image, to
encrypting sensitive information. Digital Signal Processing is being used in more and more products due to its accuracy,
flexibility, and reliability.
The Model 475 Gaussmeter is an ideal product that can benefit from DSP technology. The Model 475 offers the user a
choice of 15 band-pass and 3 low-pass AC filters. Digital filters can easily be modified in software and in addition offer
a sharper roll-off, less ripple in the pass-band, and better stop-band attenuation. It would be difficult to implement all of
these filters in an analog form since changing the filter parameters would require extensive hardware manipulation. The
components that are used in analog signal processing can have different values from component to component and are
temperature dependent. Using Digital Signal Processing gives better measurement repeatability and increases the
temperature stability of the instrument.
2.1.3 Limitations of Sampled Data Systems
Sampled data systems do have their limitations, but if they are understood, they can be dealt with easily. The limitations
of sampled data systems come from the fact that a continuous analog signal is being sampled and digitized. This
inherently limits the frequency of the signal that can be read as well as the resolution at which it can be read. Typically,
the resolution is high enough and enough averaging is done that it does not present a problem. The frequency limitation
can cause unique problems. There are notches in the frequency response as the input signal approaches one-half the
sampling rate and its harmonics. As the measured signal approaches these harmonic frequencies, the reading will fall off
due to the null in the filter.