JPS59176891A - Method and apparatus for correcting output signal from digital converter for measuring physical quantity - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for correcting the frequency of an output signal from a digital converter that oscillates pulses at a frequency approximately proportional to the physical quantity to be measured, and an apparatus for implementing the method.

When using a measuring device, it is desirable that there be an exact proportional relationship between the quantity being measured and the output signal.

Generally, the curve that shows the relationship between the measured quantity and the signal is called the characteristic curve of a measuring instrument, but it is desirable that this curve be a straight line passing through the origin of the Cartesian coordinate system necessary to express the characteristic. .

However, in practice, the output frequency is rarely perfectly proportional to the true quantity to be measured, due to friction, other physical conditions, or the inherent performance of the measuring instrument. In other words, although the characteristic curve of a measuring instrument is a straight line in its normal operating range, its extension usually does not pass through the origin of the coordinates.

This property is particularly disadvantageous if the converter output is a pulse train directly counted by a counter.

The object of the invention is to add or subtract correction pulses to the pulse train coming from such a transducer, over the entire operating range of the transducer, at a frequency exactly proportional to the true quantity to be measured. The purpose is to obtain a pulse train having .

A method for achieving the above objectives and an apparatus used for implementing the same will be described in detail with reference to the accompanying drawings.

Figure 1 is a typical example of a transducer characteristic curve in which the frequency of the transducer pulse corresponding to the measurand is shown as a straight line as the - order number of the true value of the measurand. Fig. 3 is a diagram corresponding to Fig. 3, which shows how the above-mentioned characteristic curve is corrected in the present invention; Fig. 3 is a pulse horizontal diagram, and correction pulses are mixed with the converter pulse train to correct FIG. 4 is a block diagram for explaining the principle of the method according to the invention; FIGS. 5 and 6 show an example of how the circuit according to FIG. FIG. 7 is a pulse horizontal diagram showing an example of a case where a pulse train and a correction pulse train are mixed, and FIG. 7 is a detailed block diagram showing an embodiment of a circuit implementing the method of the present invention.

The problems with the prior art described above are briefly illustrated in FIG.

The characteristic curve in Figure 1 shows the characteristics of a type of measuring device or transducer suitable for applying the present invention, and shows that the frequency of its output pulse varies linearly with the measured quantity. It shows.

Such a transducer has, for example, a flow rate q n? /s
There are rotor-type or circulating sphere-type flow transducers or transmitters for measuring . The rotational speed of these flow needles is converted into a pulse signal with a proportional frequency by any suitable method known in the art.

Its characteristics are generally shown by the solid line in FIG. Starting from q. In other words, q
win indicates the lower limit of the operating range of the measuring instrument.

Now, if we extend the straight line downward as shown by the dotted line, it will intersect the ordinate axis at a point p below its origin. This deviation from the origin is due to hydraulic and mechanical losses.

The thing to keep in mind here is the chain double-dashed line A in Figure 1.
The linear characteristic may also intersect the ordinate axis at a point p/ above the origin, as indicated by . As far as flow meters are concerned, there are several types of transducers that exhibit these characteristics, a typical example being a hydrodynamic vibrating flow needle, or a vortex flow needle having no moving parts. Of course,
In the present invention, as will be easily understood from the following description, it is not an important issue whether the transverse point on the ordinate axis is above or below the origin.

Integral or direct indicating devices, such as adders for counting the incoming pulses, are used to measure the frequency f of the transducer. Since line A is a straight line, the transmitted frequency f has a constant differential value with respect to the flow rate q, but since the straight line does not pass through the origin, f is not exactly proportional to q.

On the other hand, counters or adders directly add and record the received signals, their own characteristic curve starting from the origin. In order to compensate for this proportionality deviation, the adder is adjusted to have the given characteristic of the slope α, indicated by the dash-dotted line B in FIG. approximately the center of the predetermined operating range, i.e. q m
It is crossed by ed. Therefore, a measurement error naturally occurs, and the error increases as the distance from q +ned increases.

Normally, the tolerance range for this error is very small, so the normal operating range of the converter is f+'', denoted by α1 in Figure 1.
If A < -, it will be limited to a very small area in the vicinity of qlned. This is extremely inconvenient as a practical matter.

The invention is intended to achieve a complete modification of the transducer signal, and the aim of the invention is to make the measured value directly proportional to the measured quantity, in other words, the extension of the characteristic curve is The goal is to cross the origin.

This correction is simple in principle and can be obtained by mixing a fixed but adjustable correction pulse train with the converter pulse train, in which the frequency of the correction pulse is added to the frequency of the converter pulse. and the characteristic curve is displaced upwards,
Its extension will cross the origin.

This is illustrated in Figure 2, where modification of the transducer
The unmodified frequency is the line fg (corresponding to the line entry in the fi1 diagram), and the modified frequency is the line f g kor
Each is indicated by r. Denoted by a constant frequency factor fk added to the frequency of the transducer pulse, the corrected frequency is thus obtained by the following equation.

f g korr = f g + f k α in Figure 2
from the minimum (Itfmin to the desired upper limit f ma
In the operating range over x, the corrected transducer frequency f
B korr will be strictly proportional to q.

From the above explanation? It will be understood how the correction can be carried out if the point at which the uncorrected characteristic curve intersects the ordinate axis is above the origin (line A' in FIG. 1).

In this case, the required correction frequency has to be subtracted from the uncompensated frequency of the transducer pulse instead of being added to it. The term "mixing or addition" and its conjugations are used in this specification and claims in a common algebraic sense.

That is, by "mixing or adding" a pulse having a negative frequency to a signal pulse train, a part of the pulses of the signal pulse train is erased, and the frequency thereof is subtracted.

This frequency reduction is achieved by using a known arithmetic circuit, such as an up/down counter.

To achieve the desired translation of the transducer characteristic curve,
In the present invention, a pulse generator is connected to the output of the converter and is configured to send out a pulse train with a desired modified frequency fk.

Specific conditions apply to this connection. This condition is to add frequencies so that errors are correctly compensated.

There are mainly three conditions for this.

The first condition is that no correction pulses are sent when no transducer pulses occur. This condition is such that a correction pulse can only be sent when a transducer pulse is sent.

The second condition is that the transducer pulse and the correction pulse must not coincide. This is because the adder only records such matched pulses as one pulse, but to correctly count mixed pulses it is necessary to record them correctly as two pulses.

Therefore, the converter pulse and the correction pulse must not coincide or overlap in time, even partially, and both pulses must be separated so that the adder can distinguish between the two pulses.

The third condition is that correction pulses must not continue to be sent after the output pulses of the transducer or rotary indicator have stopped, and that if the transducer frequency is below f min, i.e. outside the lower limit of the operating range, the individual This means that it is necessary to limit the number of correction pulses that occur in response to a transducer pulse.

The above three conditions will be discussed in more detail with reference to FIG. In FIG. 3, α represents the transducer pulse, b represents the correction pulse, and C represents the mixed pulse, and the horizontal axis represents time t.

This figure is an example where the frequency of the transducer pulses decreases below its lowest value, where gl indicates the penultimate pulse, g2 the last pulse with frequency f win,
g3 indicates a pulse that may occur outside the operating range.

Correction pulses such as ni1, k2, etc. are sent at regular time intervals according to basic assumptions.

Regarding the last situation mentioned above, the correction pulse after the last transducer pulse g2 must not disappear at the same time as g2, and in fact there are n correction pulses corresponding to the interval between pulses g1 and g2. It is stopped after being sent.

Here, if the frequency of the converter pulse is fiin, the maximum interval tmaX between converter pulses within the normal operating range is tLIlax = 1/fn+in.If the frequency of the correction pulse is If f i <, then the number n of correction pulses in this time is tmax−
It is shown as fk=fk/fmin.

As shown by the dashed line S in FIG. 3, the sending of the correction pulse is stopped after the last transducer pulse g2, and at f#, at which point a further rl pulse is sent. In FIG. 3C, a scaled pulse train is shown up to M'A Nj, at which the transducer stops.

In general, it can be said that f min defines the lowest limit point of the linear portion of the converter characteristics. As already mentioned, the transducer pulses in the actual case may be sent at a frequency lower than fmin, ie in the 1 mm range outside the operating range. In particular, the above-mentioned flowmeter body has high viscosity, so 1. At a point just before the flow stops, the transducer or rotary indicator shows the value f mi as shown in FIG.
S is to send a pulse at a frequency below the lower limit of the linear characteristic indicated by n.

When the present invention is actually adopted, fIIlin is determined as the lower limit value of the linear portion of the converter characteristic. It is further pointed out that, in principle, nothing would change if frequency multiplication were used and the output signal of the converter was multiplied, for example, by a factor of U. In fact, fk
is also multiplied by U, but the ratio indicated by fk/fmin=n is not changed.

FIG. 4 is a block diagram for explaining the present invention in detail.

In the figure, a transducer 1 sends out a pulse train of a frequency depending on the measured quantity, ie the flow rate to be measured.

The output of the variable 4IA device 1 is connected to the input terminal of the modification unit 2 and the control input terminal of the gate circuit 3, and the output of the gate circuit 3 is connected to the second input terminal of the modification unit 2. Pulse generator 4 for generating correction pulses
is connected to the second control input terminal of the gate circuit 3.

For ease of understanding, the four elements are considered here as separate units, although in practice these four elements would be provided as a single integrated circuit.

The circuit is such that the converter pulses from the converter 1 and the correction pulses from the pulse oscillator 4 are mixed in a modification unit 2 and transmitted to a counter or adder (not shown).

The two pulses generated simultaneously from the converter 1 and the pulse oscillator 4 must not be simply combined and supplied as an IM Sumibe signal; these two pulses must be supplied separately. The branching condition, that is, the second condition mentioned above, can be solved by providing a circuit that generates two timing pulse trains that are different in phase and do not overlap in time. Ru.

Thus, the correction pulse is synchronized with the first timing pulse, and the converter pulse is converted into a burst signal by intermittent the second timing pulse train in the correction Hounisoto 2, which causes the adder to output two pulses. be able to completely identify. Two subsequent pulses are then sufficiently replicated for the adder to respond.

The remaining conditions, ie, the first and second conditions above, are the gate I
It is satisfied by θ1 path 3. That is, on the one hand, the correction pulses are not deposited on their own, but are transmitted continuously following the transducer pulse after it has occurred;
On the other hand, there are regulations that limit the number of correction pulses that can be transmitted following each transducer pulse if the flow rate falls below the lower tolerance limit of the measurement range and the interval between transducer pulses becomes large. , is satisfied by the gate circuit 3.

That is, the condition regarding the independence of the converter pulse is easily met by providing a gate in the gate circuit 3 which blocks or allows the correction pulse to pass.

In order to satisfy the above conditions 11L and 2, this gate stops transmitting the correction pulse train if there is no converter pulse, i.e. if there is no converter pulse g2 followed by <[3 in FIG. and therefore for each such transducer pulse a predetermined number of correction pulses are passed. This gate +;+: can be configured in two fundamentally different ways, one based on time measurement and one based on counting pulses.

The gate is controlled by a retriggerable one-shot multivibrator, with each transducer pulse triggering the one-shot multivibrator for a certain maximum time t a
The gate is kept open for a time equal to +ax, the reciprocal of f min. Therefore, the converter frequency is f
When exceeding Wlin, the gates are opened continuously.

FIG. 5 illustrates the above situation.

The converter output signal is denoted α and the gate function is denoted d, where d is the time interval t during which the gate is open.
It shows wax. Furthermore, b indicates the correction pulse and C indicates the output signal mixed and sent out by the correction unit.

As shown in the figure, converter pulse g1 opens the gate, allowing subsequent correction pulses to pass through 1.

The next transducer pulse g2 is assumed to come with some delay because the transducer is operating below the lower limit of its normal operating range, and the gate has time to close, but g2
2 passes through the gate on subsequent correction pulses as it reopens. It is assumed that for a further 3 subsequent correction pulses the transducer is stopped and the nox g3 does not arrive, so that the gate 3 remains closed so that the nox g3 does not pass through the gate.

ff16121 exemplifies a state where the gate 3 is designed to have a counting function.

Here, a correction pulse determined by n=fk/fmin is shown. In this example, n=3.

As before, the transducer pulse is denoted by α, the correction pulse by b, and the mixed pulse by C, respectively.
There is.

When the gate is opened by transducer pulses g1, g2,
In each case, three correction pulses are allowed to pass, after which no transducer pulses occur, so that the input stops as shown to the right of line S.

If transducer pulse g3 occurs, it passes three correction pulses by itself. This situation is shown in dotted lines to the right of line S.

Note that in both FIG. 3 and FIG. 6, for convenience of explanation, the correction frequency is several times larger than the converter frequency.
In actual cases, around the lower limit of the normal operating range, the image frequency is approximately the same size, that is, n-1, and sometimes fk is f
In some cases, it is much smaller than g.

The method described in FIG. 6 represents an approximation.

This is because the number of pulses passed through the gate corresponds to the number defined earlier as n=fk/fmin, and this quotient must be rounded to the nearest positive integer. It is possible, without difficulty to point out, that the average number of correction pulses passing through the gate for each pulse when the converter is operating below the lower limit of its normal operating range is the desired value n. The gate can be changed to For this purpose, the gate is a number N, sometimes larger than n,
Sometimes it is supplemented with a memory to pass a number of correction pulses smaller than n, so that the ratio of the number of output pulses to the input pulses converges to n.

This principle can be easily demonstrated mathematically. If the quotient fk/f
Assume min, the desired number n of correction pulses per transducer pulse, to be 2.6 (7 times 1).

At this time, the gate is configured so that N=3 and K=2, that is, when the gate is activated, three correction pulses are passed out of the five times, and only two correction pulses are passed the remaining two times. The long-term average value of the ratio of converter pulse to correction pulse then clearly amounts to 2.6.

A circuit illustrating an embodiment of the present invention that can achieve the above three conditions regarding pulse train mixing will be described with reference to FIG.

In the exemplary embodiment, the converter pulse is converted into a secondary converter pulse which does not overlap in time with the correction pulse and is emitted.

This secondary transducer pulse and the correction pulse are mixed and further processed to obtain the final corrected output signal as described in detail below.

In FIG. 7, 60 is a converter, 64 is a burst generator, 66
is a pulse train transmitter, 68 is a basic pulse oscillator, 74 is O
R gate, 78 is an AND gate, 80 is a multi-high break, 82 is a frequency divider, 86 is a first ratio multiplier, 92 is a second ratio multiplier, 96 is a pulse averaging circuit, 100 is a counter drive circuit , 102 is an A/D converter or a recording needle;
104 is a counter.

The pulse train transmitter 66 is supplied with a pulse train having a high fundamental frequency of, for example, IMHz from a fundamental pulse oscillator 68 .

The pulse train transmitter 66 is composed of, for example, a known decimal ring counter, and output terminals provided for each bit of the pulse train transmitter 66 receive one pulse frequency input from the oscillator 68, respectively.
10 timing pulse trains having a frequency of /10 and each having a phase shift corresponding to one cycle of the output pulse of the oscillator 68 are output.

From one of said output terminals a first timing pulse train 76 is obtained, in this case having a frequency of 100Kilz;
This pulse train 76 is connected to an AND gate 78 and is used to generate correction pulses.

On the other hand, from an output terminal of the pulse train transmitter 66 that is different from the terminal connected to the AND gate 78, a second timing pulse train 70 (this train has the same frequency of 100 KIIz as the first timing pulse train '76) is output. Yes?)
- the pulses are staggered in time and do not overlap.

) is directed to burst generator G4 for chopping.

The analog signal is suitably amplified in a transducer 60 (eg, the vortex flow needle described above), converted to a square wave, and provided as a pulse signal 62 to a burst generator 64.

The burst generator 64 receives the second timing pulse train from the pulse train transmitter 66 and has the characteristic of outputting a series of, for example ten, pulses each time it receives a pulse from the transducer 60.

Thus, as burst generator 64 receives each transducer pulse, it delivers a burst signal 72 consisting of ten second timing pulses in this example. The frequency of pulses within each burst is 100Kllz.

Burst signal 72 is directed to OR gate 74.

The two functions of the gate for generating correction pulses have already been explained with reference to FIGS. 5 and 6.

In the embodiment shown in FIG. 7, pulse 6 from transducer 6o
2 is supplied not only to the burst generator 64, but also to the retriggerable monostable multivibrator 8o, and the AND gate 7 is supplied by the monostable multivibrator 8o.
8 is opened in response to transducer pulses 62, thereby continuously opening AND gate 78 to pass first timing pulse train 76 while transducer 60 is operating in its normal operating range. .

Furthermore, this monostable multi-pie break 8o satisfies the first and third conditions described above. That is, no correction pulses are allowed to be generated before the transducer pulse appears, and unlimited generation of correction pulses is prevented after the transducer pulse has ended.

The pulse train 76 that has passed through the AND gate 78 is passed through the frequency divider 82
Then, in this embodiment, the frequency is divided by 1/100. Therefore, the pulse train appearing on line 84 is IKIIz,
This pulse sequence number J first ratio multiplication! It is supplied to the E unit (RMP) 8b. RM P 86 is a decimal multiplier or pulse divider that calculates the desired average frequency (in this case 10
fk) pulse train 8 days is output. This output pulse 88
is synchronized with any of the first timing pulse trains 76 and is added to the secondary converter pulses 72 from the burst generator 64 at an OR gate 74 .

As mentioned above, the corrected converter frequency fQ korr
is equal to fg→−fk and is represented in the graph, and the straight line is transformed so that it passes through the origin.

In this case, the fkO value suitable for the instrument under consideration is set in the first ratio multiplier 86. However, in this case, it is necessary to take into account that the signal has a frequency of 720 or 10fg. (Therefore, in this case, 10j and k are set.

), and the output signal 90 of the gate 74 has a frequency of 4; t
lO([g + fk) J: fx le.

The second condition necessary for the correction of the converter signal, namely that the converter pulse and the correction pulse must be separated in such a way that they do not coincide in time, is:
This is achieved by configuring the output pulse of the first ratio multiplier 86 to always occur in synchronization with the first timing pulse.

In a iron ball configuration, the pulses from the same source, the fundamental pulse oscillator 68, are divided equally and staggered in time, two of which serve as the correction pulse and the second timing pulse, respectively. Used to generate secondary transducer pulses, said transducer pulses (or rather burst signal pulses dependent on transducer pulses) and correction pulses do not overlap and are completely separated. .

The corrected signal 90 is processed in a second ratio multiplier 92 which multiplies the signal by a meter factor K to adjust its frequency. The signal 94 from the second multiplier 92 is K.
10(fg+fk), and each pulse of the signal has a frequency of, e.g.
1j per pulse! A value that directly corresponds to the desired measured quantity is given.

The minimum spacing of their pulses is very close to C(
They will never match, but considering the basic pulse frequency of 1 MHz, in some cases there will be a minimum interval of 1 μs. ), many A/'D converters or recorders 10
2, a large increase/l pull occurs, and since the counter, especially the miss machine type counter 104, does not follow up, it is desirable to process the signal using the 94-pulse averaging circuit 96.

This means that in the -=-・ direction, the interval between pulses is equal to 1,
On the other hand, it is a circuit that changes the width or duration of the pulse itself without changing the total number of human pulses, and can be easily described. Such an objective may be achieved, for example, by the well-known up/town counting wrestling system.

The pulses sent from the flattening circuit 96 to the counter drive circuit 00 and 98 have an extremely large spatial interval and a uniform period, so the well-known A
/D converter or recorder 102 and mechanical or electronic counter 104 can process their input signals without difficulty.

The circuit of FIG. 7 includes elements that correct the frequency of the generated signal in a special way, such as a pulse train transmitter 66 and a frequency divider 82.
A pulse source with a common frequency of IMHz is also illustrated. It should be noted, however, that the above-mentioned values have been chosen merely as an example, and the circuit may operate with other values without changing the fundamentals of the circuit configuration, if necessary in implementation. It can also be configured.

A major advantage of the circuit shown in FIG. 7 is that the correction pulse frequency can be easily varied to compensate for varying external conditions that affect the measurements.

For example, in a flow needle, if varying temperatures change the viscosity of the flowing fluid, this will cause error variations that can be easily compensated for by varying the frequency of the correction pulses with temperature. This compensation is done automatically by introducing temperature as a control factor.

Furthermore, manufacturing errors between the same parts can be compensated for by changing the setting values of the first and second ratio multipliers, thereby improving the measurement accuracy of each measuring device.

[Brief explanation of the drawing]

Figure 1 is a typical example of a measuring instrument characteristic curve in which the frequency characteristic of the converter pulse is shown as a straight line as a linear function of the true value of the measuring instrument, and Figure 2 is a diagram corresponding to Figure 1. A diagram showing how the measurement characteristic curve is corrected in the invention, FIG. 3 is a pulse model diagram, and shows an example in which correction is performed by mixing a correction pulse with a converter pulse. Figure, 4th
5 and 6 are block diagrams for explaining the principle of the method of the invention; FIGS. 5 and 6 are the circuits shown in FIG. 4 in which converter pulses and correction pulses are mixed according to two alternative methods of operation; FIG. 7 is a detailed block diagram showing an embodiment of a circuit for carrying out the method of the present invention. 1---・-------------m-converter 2--
−・・−−−−−1−−−・Correction unit 3−−−−
・−−−−−−−−・Gate circuit 4−・−−−−−
−11−−−−−・−Pulse oscillator 60−・−・・−−
−−m−−・−−−1 converter 64・−−1−・−・−
−−−−1・−Burst generator 66・−・−−−−−
−1−−−−・・−Pulse train transmitter 68・−・・−・−
・−・・−Basic pulse oscillator 74・・・−・−・・−
・−1−−・OR gate 78− ・・−〜−1・・−
------・-AND gate 80・----1...--
・・・・・・−・Multi Pie Break 82・−・−・−
−−−−・−−−One frequency divider 86−・−・−・−・・−−
m--・--first ratio multiplier 92--------
---・-----1----Second ratio multiplier 96...
−・−・−−−−1・−−−・−Pulse averaging circuit 1
00−−−・−m−−−−・・−・・−Force counter drive circuit 102−−−−−−−−−・−−−−−−・A/D
Converter or recording needle 104 -------------- Counter Patent Applicant: Gripers Thor.
Lennart Berndt agent (7524) Shotabu Mogami

Claims (11)

[Claims] (1) A digital converter that oscillates a pulse with a frequency fg that is approximately proportional to the physical quantity to be measured, and within the range of its normal operation, the frequency rg of the output pulse is a linear function of the physical quantity, and both On the graph representing the relationship between
A method for correcting an output signal from a device having a characteristic such that an extension of a straight line showing the relationship between the two within the above normal operating range crosses the axis representing the frequency at a point other than the origin, the output signal having a constant frequency fk. The correction pulse train is mixed with the converter output pulse train of the frequency fg to form a corrected pulse train, the frequency f g korr is also a linear function of the physical quantity, and the frequency fk of the correction pulse is
is the frequency of the transducer pulse corrected above f g kor
A method for adjusting the straight line representing r or its extension so that it passes through the origin of the graph, characterized by comprising the steps (a) to (511) below. (al) Predetermined frequency (bl) Obtaining from the basic pulse train a first and second timing pulse train which are shifted in time so as not to overlap with each other; (C) using the converter to produce a correction pulse; (di) intermittent controlling said first timing pulse train by means of pulses. (e) Mixing the correction pulse and the secondary converter pulse, and adjusting the frequency of the mixed pulse as a function of the physical quantity by a straight line that substantially passes through the origin of the graph. Obtaining the final corrected pulse train as shown. (2) intermittent control of said first timing pulse train by said converter pulse to produce a correction pulse, the correction pulse being produced when the frequency of said converter pulse falls below its lowest allowable limit; 2. The method of claim 1, wherein the number of sources is limited. (3) the opening of the number of correction pulses, after each individual transducer pulse is emitted, emitting said correction pulses only during a period not exceeding the time between adjacent transducer pulses at said lowest allowed limit frequency; A method according to claim 2, comprising: (4) Claims in which the limitation on the number of correction pulses consists in limiting the number of emitted pulses to a number n or less that is not greater than the quotient of the frequency of the correction pulse and the above-mentioned minimum allowable limit frequency. The method described in Section 3. (5) in the case where the number of said correction pulses is limited, the correction pulses are emitted alternately with a suitable frequency equal to an integer N and an integer K, where N is an integer greater than said number n; 5. The method of claim 4, wherein K is an integer smaller than the number n, such that the number of correction pulses per individual transducer pulse approaches the number n on average. (6) A digital converter that oscillates a pulse with a frequency fg approximately proportional to the physical quantity to be measured, and within its normal operation range, the frequency fg of the output pulse is a linear function of the physical quantity, and both On the graph representing the relationship between
This is a method for correcting an output signal from a signal having a characteristic such that the extension of a straight line representing the relationship within the normal operating range crosses the axis representing the frequency at a point other than the origin, A corrected pulse train is formed by mixing a correction pulse train having the frequency fg with the converter output pulse train having the frequency fg, and the frequency f g korr is also a linear function of the physical quantity, and the frequency fk of the correction pulse is
is the frequency f g k of the above corrected converter pulse train
In the method of adjusting the straight line representing orr or its extension to pass through the origin of the above graph, the following (al to +f1
A method characterized in that it consists of the steps described in Section 1. fa) providing a fundamental pulse train of a predetermined frequency; (bl) Obtaining from said basic pulse train a first and second timing pulse train which are offset in time so as not to overlap with each other. (C) said first timing pulse train by said transducer pulse to produce a correction pulse. (d) intermittent controlling the second timing pulse train with the transducer pulse to generate a predetermined number of secondary transducer pulses for each individual transducer pulse; (+111) The above-mentioned correction pulse and the above-mentioned secondary converter pulse are mixed, and the frequency of the mixed pulse is finally corrected as a function of the above-mentioned physical quantity by a straight line passing through the origin of the above-mentioned graph. (f) without changing the number of pulses of the pulse train corrected by mixing the correction pulse with the secondary converter pulse (by averaging the inter-pulse intervals and the width of each pulse); step. (7) A device for correcting the output characteristics of a physical quantity/frequency converter whose output characteristics within its normal operating range are represented by a straight line that does not pass through the origin, which comprises the elements described in items (21) to (e) below. A device characterized by: (a) A fundamental pulse oscillator for generating pulses at the fundamental frequency. (bl) A pulse train transmitter connected to said basic pulse oscillator and producing first and second timing pulse trains which are offset in time with respect to each other and which do not overlap. By intermittently controlling the timing pulse train of , the correction frequency U fk (
A first gate that produces a correction pulse train (u is a positive number). (dl A second gate that responds to the converter pulse and controls the second timing pulse train intermittently to produce a secondary converter pulse having a frequency u-fg. (el The value of the frequency fk is Corrected transducer pulse frequency u−fgkorr=u°fk+u−fg, or f
1. A circuit that mixes the quadratic converter pulse and the correction pulse 7 to produce a corrected pulse train in order to obtain gkorr=f k + f g. (8) The first gate passes the correction pulse train only during a period not exceeding one period of the transducer pulse at the lowest operating pulse frequency of the transducer after receiving the transducer pulse. The device described in the ode. (9) When the first gate receives a converter pulse at a frequency lower than the converter's lowest operating pulse frequency, each pulse is approximately equal to the quotient of the corrected pulse frequency and the converter's lowest operating pulse frequency. 9. Device according to claim 8, in which an equal number of correction pulses are passed. (10) the second gate includes a burst generator responsive to receiving a transducer pulse, through which a fixed number of successive pulses from the second timing pulse train are transmitted each time a transducer pulse is received; Apparatus according to claim 7 for providing. (11) the first gate comprises a retriggerable monostable multivibrator responsive to the individual transducer pulse; and a gate opened by the output of the monostable multivibrator to allow the first timing pulse to pass through. , a first ratio multiplier connected to the output terminal of said gate and outputting a correction pulse train having said correction frequency; said second gate being responsive to said converter pulse and outputting said second timing pulse train. A circuit for mixing the secondary converter pulse and the correction pulse, comprising an intermittent controlled burst generator, has an input terminal connected to the burst generator and the first ratio multiplier; an OR gate that adds a frequency to provide a pulse signal having the corrected converter pulse frequency; a second ratio multiplier connected to the output side of the OR gate; and varying the total number of output pulses of the ratio multiplier. 8. The apparatus of claim 7, further comprising a pulse averaging circuit for averaging the inter-pulse spacing and the width of each pulse.