ORIGINAL CONTRIBUTIONS
Nuclear resonance absorption circuit
By
F.
N. H.
ROBINSON,*
M.A.,
D.Phil., Clarendon Laboratory, Oxford
[Paper
first
received
13
April,
and in,finalform 14 May,
19591
The new circuit described in Part 1 has the sensitivity and flexibility of conventional circuits using
a separate oscillator, combined with the convenience and freedom from microphonics of marginal
oscillator circuits. In particular, it permits the use of very low levels of oscillation.
Because the factors governing the sensitivity of conventional circuits are not widely known,
this first paper is largely devoted to their delineation.
Only the principles and theory of the
new circuit are given in Part 1. Practical design details are relegated to the second paper.
Part
1. Sensitivity considerations
Nuclear resonance absorption in a specimen surrounded by
a coil forming part of a tuned circuit changes the quality
factor Q by
where
77
is a filling factor and
x”
is the imaginary part of the
nuclear susceptibility.
The excellence of any circuit for
detecting nuclear resonance (n.r.) absorption can therefore
be specified by giving that change 6(l/Q) which yields a
signal equal to noise.
The simplest n.r. circuit (Rollin(’)) is shown in Fig. 1, in
which a constant current generator
I
drives the resonant
circuit
(L,
C with loss G) at its natural frequency.
N.R.
absorption changes the shunt impedance of the circuit and
thus the voltage across it.
This change is detected by a
receiver of bandwidth
B
and noise figure
F.
6(l/Q)
=
4~77~”
(1)
ffl
Fig. 1. Simple nuclear resonance circuit using a separate
oscillator
A
change 6(l/Q) results in a fractional change Q6(l/Q) in
the shunt impedance and so, if the r.m.s, voltage in the
absence of signal is VI, it is changed by:
SVi
=

V:Q6(l/Q)
(2)
The effective rms. noise voltage across the coil is
V,,
=
(4kTFB/G)’12
=
(4kTFBQjwc)lI‘
(3)
Thus unity signaltonoise ratio is achieved when
(4)
Identical results are obtained for bridge circuits. the only
purpose of which is to give the experimenter freedom to
minimize the receiver noise figure by working with the
optimum input.
The voltage Vi cannot be indefinitely increased because of
nuclear saturation. If the maximum permissible r.f. magnetic
*
English Eiectric Research Fellow.
field is Hl then energy considerations yield the following
relation
where
U
is the specimen volume.
Y:Cg H?U/4q
(5)
At the optimum level a value of
satisfying
gives a signal equal to noise. Both H1 and
x’’
are beyond
the experimenter’s control. To optimize the signaltonoise
ratio one can only increase Q,
U,
7
and decrease
F
and
B.
Of these only Q,
F,
B
pertain to the circuit. Factor Q is
limited partly by the experimental arrangement and partly
by the physical properties of materials.
Bandwidth
B
is
limited by the stability of the equipment and by the tinir
available for measurement.
Thus, all that the circuit designer
can do is to attempt to ensure that
F
does not appreciably
exceed unity.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '07
 SCHIANO
 Shot noise, JOURNAL OF SCIENTIFIC INSTRUMENTS

Click to edit the document details