Demodulators - Ametek 7280 Instruction Manual

Chapter 3, TECHNICAL DESCRIPTION

3.3.12 Demodulators

3-10
It is important to note that this type of phase-locked loop, unlike a conventional edge-
triggered type using a clean reference, does not automatically re-acquire lock after it
has been lost. Lock can be lost as a result of a signal channel transient or a phase
reversal of the signal, in which case it may be necessary to repeat the lock acquisition
procedure. However, if the measurement system is set up with sufficient precautions,
particularly ensuring that the full-scale sensitivity is maintained at a suitable setting
in relation to the signal level, then the virtual reference mode is capable of making
signal recovery measurements which are not possible with other lock-in amplifiers.
When virtual reference mode is in use, the signal at the OSC OUT connector is a
sinusoid which is phase-locked to the signal. Naturally, this cannot be used as a
source for the measurement.
The essential operation of the demodulators is to multiply the digitized output of the
signal channel by data sequences, called the X and Y demodulation functions, and to
operate on the results with digital low-pass filters (the output filters). The
demodulation functions, which are derived by use of a look-up table from the phase
values supplied by the reference channel DSP, are sinusoids with a frequency equal
to an integer multiple, n × f, of the reference frequency f. The Y demodulation
function is the X demodulation function delayed by a quarter of a period. The integer
n is called the reference harmonic number and in normal lock-in amplifier operation
is set to unity. Throughout this chapter, the reference harmonic number is assumed to
be unity unless otherwise stated.
The outputs from the X channel and Y channel multipliers feed the first stage of the
X channel and Y channel output filters, implemented as Finite Impulse Response
(FIR) stages with selectable 6 or 12 dB/octave roll-off. The filtered X channel signal
drives a 16-bit DAC that, for short time-constant settings, generates the signal at the
instrument's CH 1 analog output connector. Both signals are combined by a fast
magnitude algorithm and a switch then allows either the filtered Y channel signal or
the magnitude signal, again when using short time-constant settings, to be passed to a
second 16-bit DAC to give the signal at the instrument's CH 2 analog output. The
significance of the magnitude output is discussed later in section 3.3.15.
In addition the X and Y channel signals are fed to further low-pass filters before
subsequent processing by the instrument's host microprocessor.
The demodulator output is digitally scaled to provide the demodulator gain control.
As discussed earlier in section 3.3.04 this gain is adjusted as the AC Gain is adjusted
to maintain a given full-scale sensitivity.
In dual reference and dual harmonic modes, the demodulators generate two sets of
outputs, one for each of the two references or harmonics, and includes two sets (four
channels) of initial output filtering. These outputs are passed to the host processor for
further processing and, when the time constant is less than 1 ms, the X
outputs are also converted by fast DACs to analog signals that appear at the CH 1
and CH 2 analog output connectors.
and X
1 2