U.S. patent number 3,859,602 [Application Number 05/338,050] was granted by the patent office on 1975-01-07 for device for simulating the original shape of a signal which is distorted by peaks.
This patent grant is currently assigned to U.S. Philips Corporation. Invention is credited to Frits Jacques Janssen, Arnold Lehmann.
United States Patent |
3,859,602 |
Janssen , et al. |
January 7, 1975 |
**Please see images for:
( Certificate of Correction ) ** |
DEVICE FOR SIMULATING THE ORIGINAL SHAPE OF A SIGNAL WHICH IS
DISTORTED BY PEAKS
Abstract
A device for simulating a distorted, exponentially decaying part
of an input signal, comprising an RC-network having a controllable
RC-time and a measuring member which converts the difference
between the input signal and the generated signal into a control
signal for the RC-network.
Inventors: |
Janssen; Frits Jacques (Waalre,
NL), Lehmann; Arnold (Emmasingel, Eindhoven,
NL) |
Assignee: |
U.S. Philips Corporation (New
York, NY)
|
Family
ID: |
19815609 |
Appl.
No.: |
05/338,050 |
Filed: |
March 5, 1973 |
Foreign Application Priority Data
Current U.S.
Class: |
327/100; 708/824;
327/165; 327/344; 703/3 |
Current CPC
Class: |
G06G
7/60 (20130101); A61B 5/026 (20130101) |
Current International
Class: |
A61B
5/026 (20060101); G06G 7/00 (20060101); G06G
7/60 (20060101); H03b 001/00 (); H03k 005/00 ();
H04b 001/10 () |
Field of
Search: |
;328/144,145,60,13,15,127,142,143,162,164,165 ;235/183,92FL
;128/2.5F,2.1R |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Miller, Jr.; Stanley D.
Attorney, Agent or Firm: Trifari; Frank R. Franzblau;
Bernard
Claims
What is claimed is:
1. A device for simulating the shape of an input signal which
includes a peak followed by a decaying part, at least the first
portion of said decaying part decreasing exponentially and a
subsequent portion of said decaying part being distorted, said
device comprising a first switching member, a signal generator
adapted to generate an exponentially decreasing signal that
simulates the decaying part of the input signal without the
distorted portion, means for selectively connecting an outgoing
conductor via said first switching member to an incoming conductor
which serves to supply the input signal and to the output of said
signal generator, said signal generator comprising a capacitor
which can be discharged via a resistance element, a second
periodically operating switching member for connecting said
capacitor to the incoming conductor, means for adjusting the time
constant of the network formed by the resistance element and the
capacitor so as to match this time constant to that of the decaying
part of the input signal, and means for connecting the incoming
conductor and the output of the signal generator to first and
second inputs, respectively, of a measuring member which derives a
difference signal determined by the difference between the
amplitudes of the signal generated by the signal generator and the
input signal.
2. A device as claimed in claim 1 wherein the measuring member
comprises a first integrator for integrating said difference
signal, and means connecting the output of said first integrator to
a control input of the signal generator, the time constant of the
resistor-capacitor network of the signal generator being dependent
on the voltage at the control input.
3. A device as claimed in claim 1 further comprising, an amplifier
connected between the output of the signal generator and the first
switching member, the amplification of said amplifier being equal
to two, means connecting the output of the amplifier to an input of
a voltage comparison unit, means connecting the other input of the
comparison unit to a storage element connected in circuit so as to
store the value of the input signal at the instant at which the
first switching member interrupts the connection between the
incoming and the outgoing conductor, a third switching member,
means connecting said voltage comparison unit to actuate the third
switching member upon equality of the voltages at its two inputs,
and means connecting the third switching member to the outgoing
conductor to render same voltageless upon actuation of the third
switching member.
4. A device as claimed in claim 2 further comprising an amplifier
having an amplification factor of two and connected between the
output of the signal generator and the first switching member, a
voltage comparator having first and second inputs and an output,
means connecting the output of the amplifier to the first input of
the comparator, a storage element selectively coupled to the
incoming conductor so as to store the input signal appearing
thereat at the instant at which the first switching member
interrupts the connection between the incoming and outgoing
conductors, means connecting the second input of the comparator to
the storage element, and a third switching member connected to the
output of the comparator and to the outgoing conductor so as to
remove the voltage from said outgoing conductor upon actuation of
said third switching member caused by equality of the voltages at
the first and second inputs of the comparator.
5. A signal simulation device for an input signal having a waveform
with a peak followed by a decreasing exponential portion the
terminal portion of which is distorted, said device comprising an
input line at which the input signal appears and an output line, a
signal generator adapted to generate a decreasing exponential
signal that simulates the decreasing exponential portion of the
input signal but without the distorted terminal portion, first
switching means for selectively coupling the output line to the
input line and to the output of the signal generator in mutually
exclusive time intervals, said signal generator including an RC
time constant network with a control input for adjusting the value
of said time constant to vary the exponential waveform of said
signal generator, a switching member for periodically connecting
said RC network to the input line, and means coupled to the input
line and the signal generator output and responsive to the signals
thereat for deriving a control signal determined by the difference
between the amplitudes of the signal generator output signal and
the input signal.
6. A device as claimed in claim 5 further comprising means
responsive to said control signal for signalling said switching
means to switch the connection of the output line from said input
line to the output of the signal generator when said control signal
achieves a predetermined value.
7. A device as claimed in claim 6 further comprising means for
modifying said control signal to derive a second control signal,
and means for applying said second control signal to the control
input of said RC network to vary the RC time constant thereof as a
function of said second control signal.
8. A device as claimed in claim 7 further comprising means
controlled by the input signal appearing on the input line at the
instant said switching means switches the output line to the output
of the signal generator and by the output signal of the signal
generator for coupling the output line to a point of fixed
potential upon equality of the input signal and the signal
generator output signal.
Description
The invention relates to a device for simulating the shape of an
unknown signal which consists mainly of a peak which is followed by
a decaying part, at least the first portion of said decaying part
decreasing exponentially and a subsequent portion of said decaying
part eventually being distorted, the said device comprising an
output conductor which can be connected, via a first switching
member, to either an input conductor which serves to supply the
unknown signal, or the output of a signal generator which is
adapted to generate an exponentially decreasing signal so as to
simulate the decaying part of the unknown signal without the
distortion, the said signal generator comprising a capacitor which
can be discharged via a resistance element and which can be
connected to the said input conductor.
A device of this kind is known from U.S. Pat. No. 3,375,701. The
known device is intended for the simulation and subsequent
integration, by means of an integrator, of signals whose shape and
instant of occurrence of the disturbing peaks are accurately known,
such as the signals which are produced by a gas chromatograph in
which a sample is analyzed whose constituents are known or
anticipated, but whose quantities must be measured.
However, the known device is not suitable for the simulation of a
signal about which it is known only that it consists of a peak,
followed by an exponentially decaying part, the said decaying part
having an unknown time constant which, moreover, can differ from
one case to another, while the decaying part can be distorted, for
example, by peaks, the instant of occurrence of which is not known
either. A signal of this kind is obtained, for example, with given
methods of determining the heart minute volume of humans or
animals, that is to say, the volume of blood which is displaced by
the heart per minute. According to these methods, a quantity of
some indicator substance (for example a dye) is injected into the
venous part of the blood circulation. This indicator gradually
mixes with the blood during its passage through the circulation
system so that it becomes increasingly less defined and further
diluted until the indicator is finally homogeneous with the blood.
If the variation of the indicator concentration is measured
somewhere else in the blood circulation by means of a suitable
measuring instrument, and the measuring result is plotted as a
function of time on a graph, generally a dilution curve is obtained
which comprises a first peak, followed by an exponentially decaying
part on which one or more subsequent peaks appear. These latter
peaks, referred to as recirculation peaks, are caused by the fact
that the circulation system is closed and the indicator apparently
performs a few rounds before it has been completely mixed with the
blood.
In order to determine the heart minute volume, it is assumed that
the circulation system is not closed, and the exponentially
decaying part of the dilution curve is asymptotically extrapolated
to zero. The curve thus obtained is the primary dilution curve. The
heart minute volume V can now be determined from the relation of
the injected quantity of indicator I and the surface area below the
primary dilution curve according to the formula:
V = K.sup.. [I/Sc(t)dt]
Therein, K is a constant and Sc(t)dt is the surface area below the
primary dilution curve c(t).
An object of the invention is to provide a device which is capable
of simulating the primary dilution curve on the basis of the
dilution curve comprising disturbing recirculation peaks. The
invention is based on the recognition of the fact that between the
first peak and the first recirculation peak the dilution curve
generally varies in accordance with the primary dilution curve for
some time, and that this part of the dilution curve can be utilized
to determine the time constant of the exponentially decaying part.
It is then possible to utilize the said time constant in the
subsequent generation of the remainder of the decaying part of the
primary dilution curve.
The device according to the invention is characterized in that the
time constant of the network formed by the resistance element and
the capacitor can be adjusted so that this time constant is matched
to that of the decaying part of the unknown signal, a second
periodically operating switching member being provided so as to
establish the connection between the input conductor and the
capacitor, the input conductor and the output of the signal
generator being each connected to one of the inputs of a measuring
member which is adapted to determine the difference between the
signal generated by the signal generator and the unknown
signal.
As soon as the measuring member indicates that the variation of the
generated signal is the same as that of the decaying part of the
unknown signal, the first switching member can be operated, after
which the generated signal instead of the unknown signal is applied
to the output conductor. It is obvious that the time constant of
the signal generator is no longer varied as of that instant.
In a further embodiment of the device according to the invention,
this switching operation is effected fully automatically. This
embodiment is characterized in that the measuring member comprises
a first integrator for integrating a signal which is proportional
to the difference between the unknown signal and the generated
signal, the output of said first integrator being connected to a
control input of the signal generator, the time constant of the
resistor-capacitor network of the signal generator being dependent
on the voltage on this control input.
As already stated, in many cases the surface area below the curve
representing the signal is required. This surface area can be
determined by connecting the output conductor of the device to an
integrator as is already done in said U.S. Pat. No. 3,375,701. A
drawback of a device thus obtained is that a long period of time is
required before the exponentially decreasing signal becomes so
small that it no longer makes a significant contribution to the
integration. This waiting time can be substantially reduced without
a loss of information by constructing the device such that an
amplifier is provided between the output of the signal generator
and the first switching member, the amplification of said amplifier
being equal to two. The output of the amplifier is connected to an
input of a voltage comparison unit, the other input of which is
connected to a storage element which is adapted to store the value
of the unknown signal at the instant at which the first switching
member interrupts the connection between the incoming and the
outgoing conductor. The said voltage comparison unit is adapated to
actuate a third switching member as soon as the voltages on its two
inputs are equal, the said third switching member being adapted to
render the output conductor voltageless upon actuation.
This construction is based on the following considerations. The
variation of the decaying part as a function of time can be
represented in general by the formula
S(t) = S.sub.o.sup.. e.sup.-.sup.t/.sup..tau.
The surface area below this decaying part is equal to ##SPC1##
Integration of the twice amplified signal during a finite time t
results in ##SPC2##
This must be equal to A, so
.tau.S.sub.o = 2.tau.S.sub.o - 2.tau.S.sub.o
e.sup.-.sup.t/.sup..tau.
or
e.sup.-.sup.t/.sup..tau. =1/2
from which it follows that
t = .tau.ln 2
The value of S(t) is then
S(t) = S.sub.o e.sup.-.sup.ln2 = 1/2S.sub.o
Consequently, the amplified signal must be integrated until it has
decreased to half its initial value, i.e., until it is equal to the
initial value of the original, non-amplified signal. This operation
can be automatically performed by means of the device having the
described construction.
The invention will be described in detail with reference to the
accompanying drawing in which.
FIG. 1 shows a block diagram of a device according to the
invention,
FIG. 2 shows a circuit diagram of a part of the device shown in
FIG. 1.
FIG. 3 shows a diagram to illustrate the operation of the part of
the device shown in FIG. 2,
FIG. 4 shows a circuit diagram of another part of the device shown
in FIG. 1,
FIGS. 5 a-d show a number of diagrams to illustrate the voltage
variation at different locations in the device shown in FIG. 1,
and
FIGS. 6 a-6c show a number of block diagrams representing different
feasible embodiments of a part of the device shown in FIG. 1.
The embodiment of the device according to the invention, shown in
FIG. 1 in the form of a block diagram, comprises an input conductor
1 which can be connected to a detector (not shown), for example,
for measuring the concentration of a quantity of dye injected into
the blood of a human or animal to be examined. The output signal of
such a detector is a voltage which is the electrical analogue of
the variation of the quantity to be measured. This unknown signal
generally consists of a peak which is followed by a decaying part,
the first part of which decreases exponentially, and a next part of
which can be distorted by disturbing peaks.
The device further comprises an output conductor 3 which is
connected to a first switching member 5. In the position of the
first switching member 5 shown in FIG. 1, the output conductor 3 is
connected to the input conductor 1. In the other position of this
switching member, the output conductor 3 is connected to the output
7 of a signal generator 9 which is adapted to generate an
exponentially decreasing signal so as to simulate the decaying part
of the unknown signal without the disturbing peaks. The signal
generator 9, to be discussed in detail with reference to FIG. 2,
comprises a capacitor 11 which can be discharged via a resistance
element 13 (both denoted by broken lines in FIG. 1). The resistance
of the resistance element 13 is dependent on the voltage at a
control input 15 of the signal generator 9, so that a variation of
this voltage causes a variation of the time constant of the network
formed by the capacitor 11 and the resistance element 13.
The capacitor 11 is connected to the incoming conductor 1 via an
interruption circuit 17. The interruption circuit 17, to be
described in detail herinafter with reference to FIG. 2,
constitutes a second, periodically operating switching member which
each time establishes a connection between the input conductor 1
and the capacitor 11 for a given period of time, and which
subsequently interrupts this connection again.
Connected between the output 7 of the signal generator 9 and the
switching member 5 is an amplifier 19, the function of which will
be described hereinafter. The output of the amplifier 19 is also
connected to an input 21 of a measuring member 23 which is adapted
to determine the difference between the signal generated by the
signal generator and the unknown signal. The latter signal can be
applied to the measuring member 23 via a second input 25 which is
connected to the input conductor 1.
The measuring member 23 comprises a differential amplifier 27, the
two inputs of which are connected to the inputs 21 and 25 of the
measuring member. By a suitable choice of the resistors 29, 31, 33,
35 and 37, it can be achieved in known manner that the (unknown)
signal which enters via the second input 25 is amplified more than
the (generated) signal entering via the first input 21, the
relationship between the amplification of the two signals being
exactly equal to the amplification of the amplifier 19. As a result
the voltage appearing on the output 39 of the differential
amplifier 27 is proportional to the difference between the signal
generated by the signal generator and the unknown signal.
The output 39 of the differential amplifier 27 is connected, via a
coupling resistor 41, to an input 43 of a first integrator 45 which
consists of a known network, comprising an amplifier 47, a
capacitor 49 and a resistor 51. The output 53 of this integrator
also forms the output of the measuring member 23, and is connected
to the control input 15 of the signal generator 9.
The output 39 of the differential amplifier 27 is not only
connected to the input 43 of the integrator 45, but also to an
indicator 55 which can be, for example, a voltmeter, a lamp or a
circuit which is adapted to actuate one or more switching members.
The indicator 55 gives an indication when the difference between
the unknown signal and the generated signal becomes zero, or
reaches another predetermined value.
The amplification of the amplifier 19 is adjusted to the value two
by means of resistors 57 and 59. The output of this amplifier is
connected to an input 61 of a voltage comparison unit 63, the other
input 65 of which is connected to a capacitor 67. The output of the
voltage comparison unit 63 is connected to a third switching member
69 in the form of a relay. When the relay 69 is energized, the
output conductor 3 is grounded so that this conductor no longer
carries a voltage. The operation of the voltage comparison unit
will be described with reference to FIG. 4.
The device shown in FIG. 1 also comprises a fourth switching member
71 which can establish a connection between the incoming conductor
1 and the electrode of the capacitor 67 which is connected to the
second input 65 of the voltage comparison unit 63. The device
further comprises a fifth switching member 73 which connects, in
its closed position, the interruption circuit 17 to the capacitor
11 of the signal generator 9, and a sixth switching member 74 by
means of which the input 43 of the integrator 45 can be
short-circuited. As is shown in FIG. 1, the output conductor 3 is
preferably connected to a second integrator 75 which is constructed
in the usual manner including an amplifier 77, a resistor 79 and a
capacitor 81. The output 83 of the second integrator 75 can be
connected, for example, to a write unit, an indicating measuring
instrument, or an apparatus for further arithmetical processing
(not shown).
FIG. 2 shows a more detailed circuit diagram of the interruption
circuit 17 and the signal generator 9. The interruption circuit 17
consists of a switching transistor 85 which is controlled by an
astable multivibrator which is formed by the transistors 87 and 89,
the capacitors 91 and 93 and the resistors 95, 97, 99 and 101. The
operation of such a circuit is generally known and will not be
described in detail in this context. The time during which the
transistor 87 is conducting is determined by the values of the
capacitor 93 and the resistor 101, and the time during which the
transistor 89 is conducting is determined by the values of the
capacitor 91 and the resistor 99. The choice of the values of these
components is, of course, completely dependent on the nature of the
unknown signal to be analyzed. For heart minute volume
determinations by means of a dye dilution curve, a multivibrator
frequency of approximately 10 Hz was found to offer satisfactory
results, the transistor 89 then being conducting for about a
twentieth part of each period. When transistor 89 is conducting,
the base of the switching transistor 85 is connected to the
positive supply voltage via a resistor 103, with the result that
during this time the switching transistor 85 is also conducting
which means that the input conductor is connected, via the fifth
switching member 73, to the capacitor 11 of the signal generator 9.
The signal generator 9 comprises the capacitor 11 and the
resistance element 13 which is connected parallel thereto and which
consists of a transistor 105 with a resistor 107 in its collector
lead. The base of the transistor 105 is connected, via a resistor
109, to the control input 15 of the signal generator 9 and, via a
resistor 111, to a sawtooth generator which also forms part of the
signal generator and which is formed by the transistors 113 and
115, the capacitors 117 and 119, and the resistors 121, 123, 125
and 127. The operation of the sawtooth generator will not be
described in this context as this circuit is generally known. The
voltage generated by the sawtooth generator is supplied, via the
resistor 111, to the base of the transistor 105. This voltage
V.sub.g is shown in FIG. 3 as a function of time. Added to the
voltage V.sub.g is a direct voltage V.sub.o which is applied via
the control input 15 and the resistor 109. The transistor 105
starts to conduct as soon as the base voltage exceeds a given value
which is denoted in FIG. 3 by V.sub.s. At a given amplitude and
frequency of the sawtooth voltage V.sub.g, the time during which
the transistor 105 is conducting depends on the control voltage
V.sub.o as is obvious from FIG. 3. This time is denoted in FIG. 3
by t.sub.3, and the time during which the transistor 105 is not
conducting is denoted by t.sub.2. The sum of t.sub.2 and t.sub.3 is
the duration of one period of the sawtooth generator. The effective
resistance of the resistance element is then equal to
R.sub.eff = [(t.sub.2 + t.sub.3)/t.sub.3 ] .sup.. R = (1 + [t.sub.2
/t.sub.3 ]).sup.. R,
where R is the value of the resistor 111. The time constant of the
network formed by the capacitor 11 and the resistance element 13
is
.tau. = CR.sub.eff = (1 + [t.sub.2 /t.sub.3 ]) RC,
where C is the capacitance of the capacitor 11.
If V.sub.o .gtoreq. V.sub.s, the transistor 105 is continuously
conducting which means that t.sub.2 = 0. In that case, .tau. = RC.
This is the minimum value of .tau. .sup.. The values of R and C
must be chosen such that this minimum value is always smaller than
the smallest anticipated time constant of the decaying part of the
unknown signal. The frequency of the sawtooth generator must, of
course, be so high that t.sub.2 + t.sub.3 is always very small with
respect to RC. A suitable combination for the recovery of primary
dilution curves is found to be formed by an RC-time of 1.5 seconds
and a sawtooth frequency of 1,000 Hz.
The voltage comparison unit 63 shown in FIG. 4 comprises a known
comparator circuit which is composed of the transistors 129 and 131
and the resistors 133 and 135. The collector connections of the
transistors are interconnected by a sixth switching member 137
which can be opened so as to start the comparator. The operation is
as follows. As long as the voltage on the first input 61 is higher
than that on the second input 65, the transistor 129 is conducting
and transistor 131 is cut off. This means that the voltage on the
base of an output transistor 139 is comparatively high, so that
this transistor is also cut off. If the voltage on the first input
61 becomes lower than that of the second input 65, the transistor
129 is cut off and transistor 131 starts to conduct. When the
values of the resistors 133 and 135 are suitably chosen, for
example, 100 K.OMEGA. and 12 K.OMEGA., respectively, the base
voltage of the output transistor 139 decreases very strongly so
that this transistor becomes conducting. The collector lead of the
output transistor 139 incorporates a coil 141 which forms part of
the said relay 69. A limiting resistor 143 is connected in series
with the coil 141. As soon as the transistor 139 starts to conduct,
the collector current of this transistor flows through the coil 141
with the result that the relay is energized and the output
conductor 3 is grounded.
The operation of the device shown in FIG. 1 will now be described
with reference to FIG. 5. FIG. 5a shows, by way of example, a
dilution curve as usually measured in the determination of the
heart minute volume. A signal voltage V.sub.1 (t), forming the
electrical analogue of the dilution curve, is applied to the
incoming conductor 1. This signal has a first peak 145, followed by
an exponentially decreasing decaying part having a number -- in
this case two -- of disturbing recirculation peaks 147 and 149.
Via the interruption circuit 17 and the switching member 73, this
voltage is periodically applied to the capacitor 11 with the result
that the latter is each time charged until its voltage is equal to
the instantaneous value of V.sub.1 (t). The voltage on the
capacitor 11 is denoted by V.sub.11 (t) in FIG. 5a.
The output voltage of the signal generator 9 (the capacitor voltage
V.sub.11 (t)) is amplified by a factor of two by the amplifier 19
and is subsequently applied to an input of the differential
amplifier 27, the second input of which is connected to the
incoming conductor 1. Consequently, the original signal V.sub.1 (t)
is applied to the second input. As the differential amplifier 27 is
adjusted such that the voltage which is applied to its second input
is amplified twice as much as the voltage applied to its first
input, a voltage V.sub.39 (t) appears on the output 39 of the
differential amplifier 27 which is proportional to V.sub.1 (t) -
V.sub.11 (t). This voltage is shown in FIG. 5b.
The voltage V.sub.39 (t) is integrated and its sign is reversed by
the integrator 45, the output voltage V.sub.53 (t) of which is
shown in FIG. 5c. This voltage is applied to the control input 15
of the signal generator 9, with the result that the time constant
.tau. at which the capacitor 11 is discharged changes as described
with reference to the FIGS. 2 and 3.
Before V.sub.1 (t) reaches the first peak 145, V.sub.1 (t) -
V.sub.11 (t) is continuously positive, so that V.sub.39 (t) is also
continuously positive and V.sub.53 (t) becomes increasingly
negative. As a result, V.sub.g (see FIG. 3) continuously remains
smaller than V.sub.s, so that t.sub.3 = 0, so
.tau. = (1 + [t.sub.2 /t.sub.3 ]) RC = .infin..
This means that the capacitor 11 retains its voltage during the
time that it is not connected to the incoming conductor 1, so
V.sub.11 (t) remains constant during this time and is only
periodically matched to V.sub.1 (t). As soon as V.sub.1 (t) has
passed the first peak 145, V.sub.1 (t) decreases so that now
V.sub.1 (t) - V.sub.11 (t) and hence V.sub.39 (t) is negative each
time. V.sub.53 (t) then starts to increase and at a given instant
it reaches a value at which t.sub.3 (see FIG. 3) becomes larger
than zero. This means that .tau. obtains a finite value and the
capacitor 11 starts to discharge as soon as it is no longer
connected to the incoming conductor 1. The variation of V.sub.11
(t) is then as shown at 151 in FIG. 5a.
At a given instant t.sub.o, .tau. is exactly equal to the time
constant of the exponentially decreasing decaying part of V.sub.1
(t) .sup.. At this time V.sub.1 (t.sub.o) - V.sub.11 (t.sub.o) = 0,
also after the connection between the capacitor 11 and the input
conductor 1 has been interrupted, so V.sub.39 (t.sub.o) = 0. As a
result, V.sub.53 (t) remains constant, so that .tau. no longer
varies either.
The disappearance of V.sub.39 (t) is indicated by the indicator 55,
in reaction to which the user of the apparatus can perform an
operation by which the switching members 5, 71, 73, 74 and 137 can
be simultaneously switched from the position shown in FIG. 1 and
FIG. 4, respectively, to the other position. The instant at which
this takes place is denoted by t.sub.s in FIG. 5. If desired, the
indicator 55 can be constructed to perform the said operation
automatically, without human intervention.
Because the switching member 73 is opened, the capacitor 11 can no
longer be connected to the incoming conductor 1, so the further
variation of V.sub.1 (t) no longer influences V.sub.11 (t). The
closing of the switching member 74 causes the input 43 of the
integrator 45 to become voltageless, so that V.sub.53 (t) remains
constant in any case.
Due to the opening of the switching member 137 (FIG. 4), the
voltage comparison unit 63 is actuated. This unit compares the
voltage on the capacitor 67 (equal to V.sub.1 (t.sub.s)) with the
output voltage 2V.sub.11 (t) of the amplifier 19.
By means of the switching member 5, the output conductor 3 and
hence the input of the second integrator 75 is connected to the
input conductor 1 until the instant t.sub.s. Consequently, on the
output 83 the voltage V.sub.83 (t) (FIG. 5d) varies as the integral
of V.sub.1 (t). At the instant t.sub.s, the switching member 5 is
switched over and the output conductor 3 is connected to the output
of the amplifier 19. Consequently, as of that instant, V.sub.83 (t)
varies as the integral of 2V.sub.11 (t). This integration is
continued until V.sub.83 (t) reaches a value V.sub.e. This is the
value which V.sub.83 (t) would reach after an infinite time if the
non-amplified voltage V.sub.11 (t) were applied to the integrator
75. V.sub.83 (t) would in that case vary in accordance with the
curve 153 which is denoted by a broken line in FIG. 5d.
Thanks to the amplifier 19, V.sub.83 (t) reaches the value V.sub.e
after a comparatively short time, i.e., at the instant t.sub.r,
when the voltage 2V.sub.11 (t) has decreased to half its value at
the instant t.sub.s. Because the value V.sub.11 (t.sub.s) is equal
to V.sub.1 (t.sub.s) at the instant t.sub.s, the instant t.sub.r
can be determined since at that instant 2V.sub.11 (t.sub.r) =
V.sub.1 (t.sub.s). This instant is determined by the voltage
comparison unit 63, one input 61 of which receives the voltage
2V.sub.11 (t), while its other input 65 is connected to the
capacitor 67 which was connected until the instant t.sub.s, via the
switching member 71, to the incoming conductor 1 and which,
consequently, has retained the voltage V.sub.1 (t.sub.s) since the
instant t.sub.s. At the instant t.sub.r, the relay 69 is energized
with the result that the outgoing conductor 3 is connected to
ground and the input voltage of the integrator, consequently,
becomes zero. V.sub.83 (t) remains constant and equal to V.sub.e as
of that instant.
The described operations take place only if the signal V.sub.1 (t)
indeed comprises an exponentially decreasing part of sufficient
length after the peak 145. If this is not the case, V.sub.39 (t)
will never become equal to zero, which is of course noticed by the
user. If desired, the indicator 55 can incorporate an alarm device
which gives an alarm when V.sub.39 (t) is still not equal to zero
after the expiration of a predetermined period of time. This alarm
unit can, for example, simultaneously short circuit the output 83
of the integrator 75 so that the device does not provide a
measuring result. This is an important advantage of the device
according to the invention over other devices which are based on
the assumption that there is always an exponentially decreasing
part so that they supply an incorrect measuring result if this is
not so, for example, because the first recirculation peak 147
appears very early.
If will be obvious that several of the described parts of the
device can be readily replaced by other functionally equivalent
parts. For example, for the variable resistance element 13 a
photosensitive resistor in combination with a controllable light
source can be used instead of the circuit described with reference
to FIG. 2. The interruption circuit 17 can be readily replaced by
an electromechanical switch, for example, a reed relay.
FIG. 6 shows some possible variations of the construction of the
measuring member 23. FIG. 6a corresponds to the diagram shown in
FIG. 1. The input receives a voltage V.sub.d (t) which is
proportional to V.sub.1 (t) - V.sub.11 (t). One output produces a
voltage V.sub.u (t) which is proportional to V.sub.d (t) and which
can be applied to the indicator 55, while the other output produces
a control voltage V.sub.r (t) which serves to control the signal
generator G and which is equal to
(R.sub.2 /R.sub.1) .sup.. (1/R.sub.o C.sub.o) .intg.V.sub.d
(t)dt.
In the example of FIG. 6b, the voltage V.sub.d (t) is directly
integrated without insertion of the amplifier 27. The advantage
thereof is that any drift in the amplifier 27 does not affect
V.sub.r (t). In this case
V.sub.u (t) = (R.sub.2 /R.sub.1) V.sub.d (t) and V.sub.r (t) =
(1/R.sub.o C.sub.o) .intg. V.sub.d (t)dt.
If desired, instead of a voltage which is proportional to the
integral of V.sub.d (t) a combination of such a voltage and a
voltage that is proportional to V.sub.d (t) itself can be used as
the control voltage V.sub.r (t). An example of a circuit for
obtaining such a voltage is given in FIG. 6c. Therein, V.sub.r (t)
= (R.sub.1 /R.sub.o) V.sub.d (t) + (1/R.sub.o C.sub.o) .intg.
V.sub.d (t)dt.
As a result, it can in some cases be achieved that the output
voltage V.sub.11 (t) of the signal generator 9 adapts itself
quicker to the variation of the unknown voltage V.sub.1 (t).
* * * * *