Remotely Calibratable Instrument System

This invention relates to instrument systems. More particularly, this invention relates to remotely calibratable admittance-responsive instrument systems useful, for example, for detecting the level of materials in a vessel.

The art has recognized for many years that instruments can be made, to measure the level of materials in a vessel, which operate on admittance responsive principles, whereby the admittance between a probe in a vessel and the vessel is measured, typically by driving the probe with an rf signal and comparing the admittance of the probe to a reference admittance, e.g., in a bridge circuit. Many prior art patents show this subject matter generally, including commonly-assigned U.S. Pat. Nos. 4,146,834 and 4,363,030. Typically, calibration of these prior art systems has been performed manually; the level of materials in the vessel is controlled to bear some predetermined relationship to the probe, and the reference admittance carried by the transmitter at the vessel is manually adjusted until the transmitter's output signal is appropriate for the level of the materials. Frequently, plant conditions make it difficult or hazardous for a person to gain access to the transmitter to perform calibration manually, so that while this approach is workable, it would be desirable if calibration could be accomplished in a more efficient fashion; in particular, it would be desirable to provide a system in which calibration could be accomplished automatically and remotely, to eliminate the necessity of an operator's having to physically approach the transmitter.

It is accordingly an object of the invention to provide a remotely-calibratable admittance-responsive instrument system.

The art has generally standardized on so-called two-wire instrument systems, in which a receiver including a power supply is connected to a remotely-located transmitter by a pair of conductors which both supply power to and convey the output signal from the transmitter. The transmitter draws current in a predetermined range for its operations; the amount of current drawn by the transmitter provides the output signal.

It is an object of the invention to provide a remotely-calibratable two-wire instrument system, in which a transmitter located at a vessel draws a current responsive to the level of materials within the vessel.

In many plant situations, the transmitters used for level sensing are not readily accessible. Therefore, it would be desirable to provide a calibrator for calibrating such a system which could be attached to the transmission line connecting the transmitter and receiver, preferably at any point thereof, to enable the maximum flexibility in calibration operations, and to provide such is an object of the invention.

Some prior art admittance-responsive measurement systems have included remote or automatic calibration schemes. Typically these have involved comparison of a variable signal, such as the admittance measured by a probe, to a fixed reference admittance, and adjustment of some circuit component until the actual admittance signal sensed by the probe bears a predetermined relationship to the reference signal which calibrates for an air capacitance as well as preload. Thereafter it is assumed that no further variation in the circuit takes place. It will be appreciated that this is not necessarily the best approach inasmuch as the reference admittance may not be correct, may itself drift over time, and, furthermore, that this does not permit ready variation of the reference admittance with variation in the materials, the level of which is to be measured, for example.

Accordingly, it is an object of the invention to provide a remotely-calibratable level-measuring system in which the reference admittance is established during a remotely-initiated calibration operation, by variation of the reference admittance compared to an actual admittance signal determined in response to the existence of a predetermined condition of the materials (for example, the materials being at a level beneath any probe used), such that the correct reference admittance is determined regardless of the characteristics of the material being measured.

Some prior art admittance responsive measurement systems are known in which a reference admittance is varied and compared to an actual admittance signal in a calibration operation in accordance with this object of the invention. However, other aspects of these systems are not as useful as would be desired. For example, it is usual to provide "preload" in level-measurement systems. "Preload" is the difference between the operating point admittance (i.e., the admittance at which an associated control system responds) and the air capacitance (i.e., the admittance sensed with respect to an empty vessel). Thus, for example, the instrument may detect an admittance of 20 pf when the materials are above the level of a probe in their vessel and 10 pf when the materials are well below the level of the probe. It would be desirable to set the instrument to change output states (e.g. to cause the vessel to be filled) at 15 pf (i.e., preload=5 pf)--which is halfway between the full and empty capacitances. This choice of operating point would minimize the possibility of false indications by the transmitter due to minor variations in characteristics of the material and due to transmitter performance variations over time. To date, all remotely calibratable instrument systems of which the present inventors are aware require selection of preload at the transmitter itself, thus defeating the aim of remote calibration.

It is therefore an object of the invention to provide a remotely-calibratable admittance responsive instrument system in which preload may be established remotely as well.

It is a further object of the invention to provide a remotely-calibratable admittance sensitive instrument system in which a number which corresponds to the value of reference admittance, once having been determined, is retained in non-volatile storage means, so that the instrument remains calibrated during power outages and the like.

In addition to the non-volatility of the stored calibration information, it would similarly be desirable if the calibration information was secure from tampering or from spurious environmental influences, and such is additionally an object of the invention.

It is a further object of the invention to provide a remotely calibratable instrument system in which the calibration operation is performed in as simple and foolproof a manner as possible.

It is an object of the invention, therefore, to provide a remotely-calibratable instrument system in which the calibration operation is accomplished automatically by the operator's pushing a single button (preload selection and calibration mode switches having been set), whereupon calibration is accomplished automatically with respect to a known predetermined condition of materials, such that a correct reference standard is derived for use in subsequent comparison to actual level signals, as opposed to systems in which it is assumed the reference signal is never to be varied.

It would be desirable if the same piece of calibrating hardware, used to initiate calibration operations in which a reference admittance is varied and compared to an admittance sensed with respect to a known condition, could be made useful for calibration of a number of such transmitters, so that the hardware costs could be shared over a larger system.

It is therefore an object of the invention to provide a remotely-calibratable admittance sensitive level measurement system, in which a calibrating device is adaptable to calibrating a large number of differing transmitters.

As mentioned above, it is desirable that the transmitter and receiver be connected by a two-wire transmission line. For simplicity's sake, it would be desirable if the calibrator could be detachably connected for calibration to the same two-wire line, preferably at any point between the transmitter and receiver, and such is accordingly an object of the invention.

The invention has been described heretofore in connection with a two-wire system in which two wires connect a receiver, comprising a power supply and load, and a transmitter, the calibrator being adapted to be attached to the pair of wires. It will also be appreciated by those of skill in the art that, in some circumstances, "receivers" are not necessary. Instead, the transmitter comprises means for controlling a local condition, e.g., a transmitter monitoring the level of materials in a vessel may comprise means for controlling the flow of materials into the vessel. Similarly, if the transmitter is used only for providing a condition indication, no discrete receiver is required. Further, the transmitter may be connected to a local power supply, such that it is not connected to a receiver by a two-wire system. Nevertheless, all such transmitters require calibration, and it would be desirable if the same calibrator used to calibrate such receiver/transmitter systems could be used to calibrate such receiverless systems.

It is accordingly an object of the invention to provide a remotely calibratable transmitter in a receiverless system. The term "receiverless system" as used herein, of course, includes a power supply although it does not otherwise include a receiver.

It will be appreciated that a calibration-enabling signal, sent from the calibrator to the transmitter to cause initiation of a calibration operation, must be distinguishable by the transmitter from noise, accidental short or open circuits and the like occurring in the transmission channel.

It is therefore an object of the invention to provide a remotely-calibratable instrument system in which a calibrator transmits an encoded calibration enabling signal and in which a transmitter is adapted to decode the encoded signal and initiate calibration operations in accordance therewith.

Having determined that remote calibration would be desirable, it is furthermore desirable that such a remote calibrator be as versatile and as useful as possible. For example, in the preferred embodiment, in a calibration operation, the variable capacitance used by the transmitter as the reference admittance is caused by the calibrator to be incremented through a number of predetermined steps; in each step, the reference capacitance is compared to the admittance sensed by the probe in response to a given condition of materials for detection of a predetermined relation therebetween. It would be desirable if the number of the step at which the varied reference admittance bore the predetermined relationship to the actual admittance were displayed on the calibrator in readable form for the use of the operator for future reference and, in some cases, to determine the appropriate preload. To provide such a display is an object of the invention.

Similarly, it would be desirable if the remote calibrator were adapted to cause the transmitter to transmit its setting to the calibrator, to enable confirmation of its setting at the correct reference admittance value, such that an operator could remotely calibrate a transmitter, make a note of the reference admittance value selected, and be able to return some time later and use the calibrator to cause the transmitter to output its reference level in order to check that the proper calibration value was set in the transmitter. Accordingly, provision of a remote calibrator with this capability is an object of the invention.

It would further be desirable to provide a remotely-calibratable instrument system in which the calibrator could be operated to cause the transmitter to vary its reference signal a predetermined amount from a known starting value, regardless of the conditions of materials in the vessel, and to provide such a system is accordingly a further object of the invention.

All of the above objects are met by the present invention.

In one of its forms the instrument system of the present invention comprises a power supply at one location and a two-wire transmitter at another location. The power supply and two-wire transmitter are connected by a pair of transmission lines. The transmitter includes an admittance-sensing probe for generating an input signal representing the sensed condition and corresponding admittance of materials, an admittance-responsive network coupled to that probe for comparing the input signal to a stored reference signal, and an output means coupled to the admittance-responsive network for generating an output signal based upon that comparison. The transmitter further includes an admittance-calibrating means which sequentially generates calibrating signals on demand. The admittance calibrating means is coupled to the admittance-responsive network such that the output means generates an output signal whenever a predetermined relationship exists between the input signal and one of these sequentially generated calibrating signals. Calibration enabling means, typically at a location remote from the transmitter, are coupled to the transmission lines. The calibration enabling means transmits enabling signals over the transmission lines for enabling the admittance calibrating means. Thus, in one form, the present invention employs remote calibration in a two-wire transmitter. Power to the transmitter and output signals from the transmitter travel on the same pair of transmission lines. Moreover, calibration enabling signals which direct the transmitter to begin a calibration operation also travel on these same transmission lines as does an output signal indicating that calibration has been achieved. The foregoing arrangement is desirable because only a single pair of transmission lines is needed for both operating and calibrating a given instrument.

Moreover, in accordance with an important aspect of the present invention, the calibration enabling means may be coupled to the single pair of transmission lines at any point between the transmitter and the power supply. Because level sensing transmitters are often at inaccessible locations which may be great distances from plant control rooms, the foregoing arrangement is particularly desirable in not only reducing wiring costs and difficulties, but in also making it simple and reliable to calibrate the instrument.

In accordance with another aspect of the present invention, the calibration enabling means includes means for transmitting calibration enabling signals to the transmitter, the transmitter being responsive to those calibration enabling signals by varying a stored reference signal and by comparison of that varied reference signal to the monitored condition signal. When a predetermined relationship exists between the monitored condition signal and the varied reference signal, the transmitter output so indicates to the calibration enabling means. The foregoing arrangement is particularly desirable because the instrument system of the present invention is calibrated against a varying reference signal which produces an error signal of zero amplitude when the predetermined relationship exists as opposed to a comparison of a fixed reference to an error signal of non-zero amplitude which may drift over time. In addition, in the foregoing arrangement, the output signal of the transmitter itself provides an indication to remote calibration means that the instrument is calibrated. Such an arrangement is desirable because the transmitter output not only indicates when calibration has been achieved, but further indicates that all portions of the transmitter are working properly.

In accordance with still another important aspect of the present invention, the calibration enabling means described above transmits an encoded calibration enabling signal to the transmitter to initiate a calibration operation. At the transmitter, the signal is decoded and compared against a stored reference before a calibration operation will begin. Such an arrangement is particularly desirable because it precludes the possibility of spurious environmental factors causing miscalibration of particular instruments.

Still another important aspect of the present invention resides in the fact that the calibration enabling means may be further adapted to detect variation in the transmitter output signal which indicates a predetermined relationship between a reference signal and the input signal indicating that calibration has been achieved, the calibration enabling means being further operative to continue to transmit calibration enabling signals until requisite preload has been established.

Still another important aspect of the present invention resides in the provision of a display means which provides an indication of the set point of a given instrument thereby providing the instrument operator with information as to the particular point at which a given instrument has been calibrated.

These and other objects of the present invention will be more fully understood by reference to the drawings and to the detailed description of those drawings which follow.

The invention will be better understood if reference is made to the accompanying drawings, in which:

FIG. 1 shows an overview of the system of the invention in a broadly-described embodiment;

FIG. 2 shows the system of the invention in a two-wire embodiment;

FIG. 3 shows a block diagram of a transmitter according to the invention;

FIG. 4 shows a block diagram of a calibrator according to the invention;

FIG. 5 comprising FIGS. 5a and 5b, shows the calibrator circuitry in detail;

FIG. 6 shows a first portion of the transmitter circuitry;

FIG. 7 shows a second portion of the transmitter circuitry;

FIG. 8 shows certain diagrams for clarification of the way in which proper preload is determined; and

FIG. 9 shows diagrams illustrating the calibration, calibration check, and preload operations.

FIG. 1 shows the system of the invention in broad conceptual outline. A probe 10 is disposed in a vessel 12 containing materials 14, the level of which is to be measured. The nature of the materials may vary widely, including insulating or conductive liquids, or granular materials; similarly the shape of the vessel can vary widely. The probe as shown is generally disposed horizontally in order to provide a large change in admittance for a given level change when the material level is near the height at which the probe is mounted in the vessel. However, those skilled in the art will recognize that numerous of the teachings herein will have applicability to systems including vertically disposed probes, such as are commonly used for continuous level monitoring.

The signal sensed by the probe is supplied to a transmitter 16, at which it is compared by a comparator 18 to reference data stored at 20 which, according to the invention, is derived in a calibration operation. The output of the comparator, termed "condition signals", is transmitted to a receiver 22 which provides an output indication. As noted, the receiver 22, in the preferred embodiment, provides power to the transmitter; power is supplied over the same communication channel 24 over which the condition signals are transmitted. The receiver 22 also may output control signals indicated at 23. Typically these would operate a valve controlling the flow of additional materials into a vessel or the like. However, it will be appreciated that, in some systems, the transmitter itself may be enabled to output such control signals, as indicated in phantom at 29. Similarly, in such a circumstance, the transmitter would not be powered from a receiver but would have a separate power input 21. Operation of the calibrator 26, which will be described below, is similar in either case.

In the preferred embodiment, the receiver 22 detects output or "condition signals" from the transmitter 16 which are in the form of the bistable current signals. The receiver 22 outputs "normal" or "alarm" indications, e.g., by lighting panel lights or the like and outputs suitable control signals. Typically, where the transmitter 16 is monitoring the level of materials in a vessel, the "normal" indication will be given whenever the level of materials is below the probe, i.e., when the admittance sensed by the probe bears a predetermined relation to the stored admittance. The "alarm" indication will be given whenever the level of materials reaches the probe; typically, a control signal would be output, closing a valve controlling flow of materials into the vessel or the like. This is the most common type of on/off level system operation. In the present specification, this arrangement will be referred to as High-Level Fail-Safe (HLFS) operation and the operation of the preferred embodiment of the system or the invention is described accordingly. However, Low-Level Fail-Safe (LLFS) operation is also possible, in which the "alarm" signal will be given when the level of materials falls below the probe. A fail-safe select switch 77 is described below, enabling the system of the invention to be used in either way.

In the preferred embodiment a calibrator 26, which detects the condition signals output by the transmitter 16 and outputs calibration signals as required, is also connected to the communication channel 24. The calibration signals may be referred to as "calibration enabling signals" in discussion of the preferred embodiment since in this case the calibrator's output signals cause the transmitter to vary the stored reference admittance for comparison to an actual admittance. The transmitter 16 comprises means 28 for detecting the calibration enabling signals and means 30 for variation of the reference data in response thereto. In a calibration operation, then, the calibrator 26 outputs calibration enabling signals which cause the reference data stored at 20 to be varied by 30 and compared at 18 to the signal received from the probe 10, responsive to a predetermined condition of materials. Typically, for example, calibration will be initiated only when the materials 14 are below the level of the probe 10. Then the problem is finding the reference data which corresponds to the "air" capacitance and storing this, together with additional "preload" data, at 20. Thereafter, this stored reference data can be compared by comparator 18 to the input data received from the probe 10 and used in generation of the condition signals.

FIG. 2 shows the block diagram of a two-wire on/off remotely-calibratable admittance measuring system according to the invention. The receiver 22 at one location comprises a power supply supplying power to a transmitter 16 which outputs bistable current signals, both over a two-wire transmission line 24. The receiver comprises means responsive to these current draw signals. The transmitter comprises a controlled current sink 30 and draws current in one of two ranges, depending upon whether the probe admittance detected by a probe 10 is above or below a reference admittance stored at 20. The comparison is made by an admittance comparator 18 to be detailed below. A load is interposed between the transmitter and power supply, to limit the current drawn during short circuits and to provide a voltage change across the transmission lines responsive to changes in the current drawn. The setpoint stored as the reference admittance at 20 is set by a calibrator 26 which is similarly attached to the transmission lines 24 for calibration at any point between the load and the transmitter. The response of the probe 10 to materials is determined by, among other things, the admittance of the probe when uncovered, as discussed below in connection with FIG. 8. In such a system, the instrument should be calibrated when the level of the material 14 is below the probe 10 in the typical horizontal configuration shown. Preferably, calibration is initiated by a pushbutton switch 66 in the calibrator. This causes the calibrator to provide a calibration enabling signal to the transmission line 24 and thus to the transmitter 16. The calibration enabling signal is a digitally encoded voltage change on the signal wires; in the preferred embodiment the calibration enabling signal is essentially an encoded sequence of short circuits across the signal wires, such as, for example, the sequence 1011001, where the "ones" represent short circuits and the "zeroes" represent the transmission line's usual characteristics; each bit of this code word may last 0.125 millisecond. The transmitter can be provided with sufficient energy storage to keep running during the brief intervals of the short circuits. The transmitter 16 detects the voltage changes on the signal wires and, if it recognizes the code, increments its setpoint by one step in a sequence of varying reference admittances. During this calibration operation, the transmitter continues to compare the varying reference admittance to the actual admittance signal, as in normal operation. Therefore, when the varied setpoint reaches a predetermined relationship to the air capacitance, the transmitter's current signal will then change state. This causes a voltage change across the load and hence a change of the voltage across the transmission lines. This voltage change is detected by the calibrator. After supplying the first coded calibration signal, the calibrator 26 waits a sufficient time for the transmitter 16 to respond; if no output change is detected, the calibrator applies another calibration enabling signal. This is repeated until the transmitter 16 responds by changing the state of its bistable output current. Having thus tuned the transmitter to the probe's air capacitance, the calibrator 26 applies a number of additional calibration enabling signals (the number of which is selected by the operator using preload selection switches 84) to the transmission lines to cause the transmitter to further increment its setpoint, thus providing preload. The amount of preload provided is chosen by the operator and is determined in accordance with the probe vessel geometry and the material characteristics. The calibrator may be operated in other modes discussed below, selected by mode switch 67.

FIG. 3 shows a detailed block diagram of the transmitter 16. The transmitter is generally a bridge circuit design, as discussed in commonly assigned U.S. Pat. Nos. 4,146,834 and 4,363,030 incorporated by reference herein. However, the transmitter of the invention incorporates a digitally controlled capacitor 32 in one leg of the admittance measurement bridge, indicated generally at 33, to set the value of probe-to-ground admittance at which the bridge will be balanced and therefore at which the bistable output of the transmitter will be switched. The effective value of the capacitor 32 is controlled by the value supplied by a digital counter 34 which thus stores the calibration information. The counter 34 is supplied with a back-up battery 31 in order to retain the count in the counter 34 in the event of loss of power. In order to calibrate the transmitter, the counter contents are incremented when a pulse recognition circuit 37, comprising a pulse train recognition circuit 36, for examining a received sequence of pulses and a signal wire transition detector 38, recognizes an encoded signal received from a calibrator 26 (FIG. 2) which, as discussed above, transmits such an encoded signal over the two wire transmission line which is connected to terminals 40. When the counter 34 is thus incremented, the digitally controlled capacitor 32 increments the reference capacitance compared to the capacitance sensed by the probe, shown in FIG. 3 at 35 as the admittance being measured.

The other elements shown on the drawing of FIG. 3 are generally conventional, and most are shown in applicant's previous patents referred to above, these including a power supply 42, a radio frequency oscillator 44, a shield-referenced power supply 46, a transformer 48 coupling the oscillator 44 to the bridge circuit 33, an error amplifier 50, a phase sensitive detector circuit 52 responsive to the bridge circuit 33, a threshold detector 54 and an output current control 56.

FIG. 4 is a block diagram of calibrator 26. To eliminate the need for separate power input to the calibrator 26 and to simplify the calibrator connections, the calibrator 26 is adapted to be connected to the transmission lines 24 (FIG. 2) at terminals 60. A local power supply 62 provides a local regulated supply and a power-up RESET signal to initialize the circuitry when the calibrator is first attached to the transmission lines. The calibration sequence is controlled by a reset controller 64 and is initiated by a pushbutton switch 66. The reset controller prevents calibration enabling signals from being sent until a "start" button 66 is pushed. Pushing the button 66 causes the following actions. It resets the output latches 68A and 69A which disable a preload controller 70 and blank a display 65, the functions of which are discussed below. Two loop-current transition detector circuits 68 and 69 monitor the transmission lines 24 (FIG. 2) to detect changes in the state of the transmitter's bistable current output signal. At the same time, a user-preselected number ("PRELOAD INPUT") is loaded into preload controller 70 which will eventually determine the amount of preload. Pushing "start" button 66 changes the state of the reset controller 64, which then outputs the RESET signal to a pulse code generator circuit 72 which outputs a calibration enabling signal; as discussed above, in the preferred embodiment this is a sequence of short circuits across the transmission lines 24. Pulse code generator 72 provides the short circuits by closing switch 90 across the two-wire terminals 60. The sequence is controlled by counting and gating the output of an oscillator 74 to effectively generate the calibration enabling signal pulse train. These calibration enabling signal pulse trains are periodically generated at a duty cycle or repetition rate controlled by the output of the transition detector 68. While waiting for the transition of the transmitter's output signal to be detected, the repetition rate is relatively slow (e.g., 4 times per second). As noted above, the calibration enabling signals transmitted by the calibrator cause the digitally controlled capacitor 32 of the transmitter (FIG. 3) to increment its capacitance which is compared by bridge 33 to the admittance being measured. Eventually, when the varied reference signal bears a predetermined relationship to the actual admittance signal, the output of the threshold detector 54 changes state and the current controller 56 causes the transmitter's output signal to change state, which is in turn detected by the calibrator 26.

For reasons relating to versatility of use of the calibrator, which will be discussed more fully below in connection with FIG. 9, in the preferred embodiment, two loop current transition detectors 68 and 69 are provided. Detector 68 detects the low-to-high loop current transitions while detector 69 detects high-to-low transitions. Their outputs are latched by flip-flops 68A and 69A, respectively.

It should be understood that the following discussion assumes that the system is being operated in the high-level fail-safe mode as discussed above and assuming that the high current state corresponds to the normal condition state in the two-way embodiment. This will be the case throughout this specification unless distinctly indicated otherwise.

When the low-to-high loop-current transition detector 68 detects a transition indicating that the transmitter's adjusted setpoint has reached a predetermined relationship to the air capacitance value, the preload phase of calibration begins. The change of state of the output of the loop-current transition detector 68, latched by flip-flop 68A, DUTY CYCLE SELECT, speeds up the pulse code generator 72 to a higher rate (e.g., 16 times per second) and also enables the counter in the preload controller 70. At the start of the preload phase, this counter contains a user selected number indicative of the preload desired, i.e., a number ("PRELOAD INPUT") indicating the number of steps through which the digitally controlled capacitor is to be incremented for preload purposes. This counter is decremented by an "END OF PULSE TRAIN" signal provided by the pulse code generator 72 every time it transmits a calibration enabling signal. When the counter reaches zero--that is, when the desired number of calibration enabling signals have been transmitted--the preload controller 70 outputs an "END OF PRELOAD" signal which causes the reset controller 64 to generate a "RESET" signal to disable the pulse code generator 72. This completes the calibration sequence. FIG. 4 also shows the display 65, which shows the "step" at which the transmitter is currently set and its connections. This is discussed further below.

FIG. 5, comprising FIGS. 5A and 5B, shows the block diagram of FIG. 4 in greater detail, essentially depicting the circuitry of the calibrator 26. The two wire connection is made at terminals 60 and a full wave bridge eliminates polarity critical connection. A voltage regulating transistor 62A provides the correct voltage at terminal V+to power the remainder of the circuit elements. In order to prevent premature sending of the calibration enabling signals, an R-C network, comprising a resistor 62B and a capacitor 62C, prevents transistors 62D and 62E from being turned on until the capacitor 62C is charged. Therefore, on power-up, the power supply 62 sends a pulse to the reset controller 64, which sends a latched reset signal to the pulse code generator, so as to prevent calibration enabling signals from being sent until the pushbutton 66 is depressed by the operator. Subsequently, pushing button 66 changes the state of the "latched reset out" of controller 64 and allows the pulse code generator 72 to begin sending calibration enabling signals. The pulse code generator 72 gates pulses received from the oscillator 74 in accordance with the data input of an 8-channel data selector 78, gated by a twelve-stage binary counter 80. Thus, the 8-channel data selector 78 provides the desired sequence of bits, as indicated by its input connections D0-D7, which are used to control the sequence of 0.125 millisecond short circuits across the transmission lines 24. The frequency of transmission of these sequences is controlled by the DUTY CYCLE SELECT signal which is output by the latch 68a of the low-to-high loop current transition detector 68. Again, this assumes high-level fail-safe operation. The end of transmission of each of the sequences of short circuits which comprise the calibration enabling signals is indicated by an "END OF PULSE TRAIN" signal output by a multi-input AND gate 79 connected to the outputs of the counter 80. This signal clocks the preload controller 70. The preload controller 70 comprises several switches 84 which may be binary-coded decimal switches as shown or any other desired type of selector switch. These are set by the operator to provide PRELOAD INPUT signals to two cascaded, presettable binary down counters 85a and 85b which count down the number of times selected by the two switches 84 to provide a predetermined amount of preload. Thus, this additional number of calibration enabling signals are sent out after detection by the low-to-high loop-current transition detector 68 of the change of state of the transmitter's output current. Preload ends when the binary down counters 85a and 85b reach zero, whereupon an "END OF PRELOAD" signal is sent to reset controller 64 which resets 12-stage binary counter 80, which then terminates transmission of the calibration enabling signals. The display 65A and associated driver, counter and latch circuitry 65B are clocked by the "END OF PULSE TRAIN" signal as well. The display, which is mounted on the housing of the calibrator so as to be visible to an operator, indicates the "step" in the calibration sequence (discussed in connection with FIG. 9) which the digitally-controlled capacitor has reached.

As shown in FIG. 5, the low-to-high and high-to-low loop current transition detectors 68 and 69 comprise voltage comparators 61C and 63C and paired R-C networks. By proper selection of the values of the resistors and capacitors in the R-C networks 61a and 61b, 63a and 63b, respectively, these filters can be arranged to ignore the voltage changes on the two-wire transmission lines due to the short circuits of the calibration signals while passing changes in the voltage across the load (FIG. 2) due to variation in the current drawn by the transmitter. In this way, transition detection may be effected by the voltage comparators 61C and 63C.

Item 77 is a fail-safe select switch. This switch enables selection of low-level fail-safe or high-level fail-safe operation. It amounts to a DPDT switch, which determines whether low-to-high or high-to-low transition detection enables the preload controller, enables the display and varies the duty cycle. The outputs of transition detectors 68 and 69 are supplied to a mode select switch 67. The mode select switch 67 allows operation of the calibrator in calibration, recalibration and check modes. These modes are discussed below in connection with FIG. 9.

The actual shorting of the two-wire transmission line which generates the calibration enabling signals is performed by switch 90 which comprises a Darlington pair 92 which essentially connects the two terminals 60 together in a sequence controlled by data selector 78 outputting the "PULSE CODE" signal. As noted, the calibration enabling signal typically comprises an encoded series, i.e., a predetermined sequence of 0.125 millisecond short circuits which can be examined by the transmitter so that the transmitter is prevented from varying the reference admittance in response to spurious signals on the transmission line.

FIGS. 6 and 7 show more detailed circuit diagrams of the novel portions of the transmitter circuit; the remainder of the block diagram circuit of FIG. 3, being in accordance with the commonly-assigned patents incorporated by reference above, is not detailed in FIGS. 6 and 7. In FIG. 7, a digitally controlled capacitor 32 is operated under the control of a binary counter/divider 96 which is the primary component of storage of calibration information block 34. The digitally controlled capacitor 32 comprises a number of capacitors 98 which can be connected in parallel by actuation of a like number of switches 99, typically FETs or the equivalent, which are controlled in turn by the output of eight-stage binary counter/divider circuit 96. This, in turn, is incremented by an "ALTER CALIBRATION INFORMATION" signal received from the pulse train recognition circuit 36 (FIG. 6) by way of an optical isolator pair 100 (duplicated on FIG. 6). Accordingly, upon receiving a signal from the pulse recognition circuit of FIG. 6, the optical isolator pair 100 sends an "ALTER CALIBRATION INFORMATION" signal to the eight-stage binary counter/divider 96 which increments its counter and controls the digitally controlled capacitor 32 to connect a different parallel combination of capacitors 98 across the connection to the rest of the bridge 33 (FIG. 3), which connections appear at the left of the digitally controlled capacitor 32. Voltage regulator 46 and battery 31, which ensure that a voltage V+is always present at the binary counter/divider 96 so that it retains the calibration information (i.e., the positions of switches 99) during power outages and the like, are also shown in FIG. 7.

It will be recognized by those skilled in the art that the total range of capacitance to be detected by admittance-responsive measuring systems is on the order of only 40 pf. The smallest value of capacitor 98 used in the circuit of FIG. 7 is 10 pf because, for values below 10 pf, accuracy is not as good and the effect of stray capacitance is more significant. Accordingly, in order to subdivide the 40 pf total range into very small steps, e.g., 0.1 pf, it is necessary to attenuate the capacitance provided by the commercially available capacitors. Two attenuators, comprising a pair of capacitors 97 at the left side of the digitally controlled capacitor 32 and three more capacitors 97 at the right side, provide this function when connected as shown. These provide typically 100:1 attenuation of the leftmost three capacitors 98, thus resulting in the desired small step size of 0.1 pf.

A capacitor 101 is also shown in FIG. 7 in series with the probe. This provides some capacitance across the terminals of the transmitter so that it always "sees" some capacitance even when the terminals are shorted by conductive materials in the vessel.

FIG. 6 shows the pulse train recognition circuitry 36 in detail. The signal wire transition detector 38 connected to terminals 40 comprises a transistor 38A, which ordinarily does not conduct with line power supplied to terminals 40. When the terminals 40 are short circuited, however, the transistor 38A then conducts current from V+toward ground. This pulse is passed to the pulse train recognition circuit as the "DATA INPUT" signal. Accordingly, the signal wire transition detector 38 outputs a "DATA INPUT" signal to the pulse train recognition circuitry 36 upon detection of a short circuit on the two-wire transmission line connecting the transmitter, calibrator and receiver. As discussed above, such short circuits are placed across the two wire line by the calibrator as calibration enabling signals. The oscillator 104 is enabled by an "OSCILLATOR ENABLE" signal passed from a pair of NOR gates connected for flip-flop operation at 106 and set by the first "DATA INPUT" signal; the oscillator 104 provides clock signals for synchronization of the incoming sequence of short circuits. Shift register 105 converts serial code received over the signal wires to parallel data. This output is then checked to see if it is good code, i.e., the appropriate sequence of shorts, by logic 102 and 103. The "CODE RECOGNIZED" signal output by the AND gate 103 is buffered by buffer transistors 108 from the isolated coupling 100, to which is connected the binary counter/divider 96 (FIG. 7). After the entire sequence of "DATA INPUT" signals has been supplied to shift register 105, the first "DATA INPUT" signal is shifted over to output Q8 of shift register 105 and resets the flip-flop 106; this disables oscillator 104 until the next incoming short circuit starts this sequence again.

FIGS. 8A, 8B and 8C explain selection of preload. The graphs of FIG. 8B and 8C show typical admittance output signals versus material level for two different probes, shown in FIG. 8A at the same height in the vessel, for various classes of materials. The empty admittance of the vessel is independent of the material but does vary with the probe size and shape, as well as with the configuration of the vessel and the proximity of any grounded members in the vessel such as agitators.

As shown in FIGS. 8B and 8C, for a given probe, the admittance change going from empty to full (for insulating materials) is related to the electrical properties of the material. In addition, the change for a given material is proportional to the empty admittance of the probe; that is, to the admittance sensed by the probe when the vessel is empty. The desired setpoints are indicated with an "X" on the various graphs. As shown, in general, the desired setpoint is about midway between the empty and full admittance values; this gives the best reliability, despite changes in the material's electrical properties or in the instrument.

In order to achieve these desired setpoints, one must include a component due to the empty admittance and a component due to the material's electrical properties. As mentioned above, in the presently preferred embodiment the amount of preload is determined by the operator manually setting switches 84 in the calibrator. It is possible however that, at some point, the desired preload may be determined by the calibrator itself. This could be done by having the calibrator determine the empty admittance component of the desired preload, by determining the number of calibration signals transmitted before the transmitter's bistable output changed state. To then determine the proper preload would require the operator only to input the material characteristics (e.g., to set a switch to "conductive", "insulating granular" or "insulating liquid" positions). Furthermore, the instrument could further be adapted to determine the material-related component of the desired preload by noting the admittance present when the material covers the probe for the first time, causing the transmitter's output to change state. The calibrator could then adjust the first approximation calibration which was determined as above and then automatically achieve the desired preload calibration. The operator in this case would be required to know nothing except to confirm that the vessel was empty when the calibration operation began. Reliable and obviously highly efficient calibration would thus be obtained. Such additional modifications are deemed to be within the scope of this invention.

It will be appreciated by those skilled in the art that what has been described may be referred to as a "blind" calibration operation in that one simply adds a predetermined preload value to the air capacitance rather than, for example, measuring the air capacitance, filling the vessel, measuring the filled capacitance and choosing some intermediate value as the setpoint. It is found by the applicants that such a blind calibration is quite effective where the setpoint is chosen properly, as explained in connection with FIGS. 8B and 8C. Moreover, of course, blind calibration is very advantageous in that one does not have to fill an empty vessel simply to calibrate the level measurement system; in many circumstances, this would be quite cumbersome, if not actually impossible. For this reason, the term "preload" has been used to refer to the difference between the setpoint and the air capacitance; the preload value is chosen by the operator when calibrating the instrument. In some circumstances, and according to others in the art, the term "setback" may also be used, although it is felt by the applicants that this may possibly imply a calibration of the type in which both air and full capacitances are measured and a value is established for the setpoint somewhere in between. However, description of the blind calibration operation herein, and use of the term "preload" in preference to "setback" herein, should not be interpreted to limit the invention except as required by the context; in fact, many of the teachings of the invention are quite applicable to systems in which both the empty and full capacitances are measured in calibration operations.

FIG. 9, comprising FIGS. 9(a)-(c), shows graphs of the capacitance of the digitally switched capacitor 32 in a transmitter versus time in calibration, calibration check and recalibration operations. In each case, the graph shows variation of the capacitance between a "zero" point and a 255 point. This refers to the division of a total variation in capacitance of some 40 pf provided by the digitally switched capacitor 32 into 256 steps, all controlled by the 8-stage binary counter/divider 96 of FIG. 7, operating the switches 99 which sequentially switch ones of 256 total combinations of capacitors 98 into the circuit. For example, in each of the graphs, the air capacitance C.sub.AIR is shown to be step 42, that is, the combination of capacitors of the digital capacitor 32 made at step 42 bears a predetermined relationship to the air capacitance, such that the bridge circuit is balanced. The transmitter's output changes state upon reaching this point. It is then presumed that an additional 148 steps are added as preload to reach a total setpoint at step 190. It should be appreciated that the steps are not necessarily linear so that, for example, the capacitance at step 100 will not necessarily be one-half that at step 200.

The "receiver status" indication below each of the diagrams indicates which of the "normal" and "alarm" signal lights shown in the receiver of FIG. 1 is lit. As a rule, in the preferred embodiment, in high-level fail-safe operation, whenever the capacitance sensed by the probe is less than the reference value provided by the switched capacitor, i.e., the level of materials in the vessel is below the probe, the "normal" display is given.

In the example given of calibration operations, FIG. 9(a), the capacitance initially provided by the digitally switched capacitor is shown to be somewhat above the air capacitance sensed by the probe. Upon initiation of calibration at the point marked "Calibration Begun" (upon pressing of the calibration start button 66), the reference capacitance provided by the capacitor is gradually incremented. Eventually it reaches its maximum value at step 255 and drops to zero on the next calibration enabling signal. The alarm light is then lit because the capacitance sensed, C.sub.AIR, is greater than that then required to balance the digitally controlled capacitor. When the bridge subsequently balances at step 42, the alarm light goes off and the normal light is relit. A predetermined value for the preload may then be added. The amount added is selected by setting the switches 84 (FIG. 5). Here the amount of preload selected is 148 steps, resulting in a final setpoint at step 190.

Prior to the change in state of the transmitter's output upon the achieving of the predetermined relationship between the capacitance sensed and the varied reference admittance (step 42), at each step, the transmitter must compare the sensed admittance of the iteratively varied reference admittance. Hence, the iteration is at a relatively slow rate to allow the transmitter time to respond, e.g., 4/sec, controlled by the duty cycle of the sending of the "PULSE CODE" signals. Thereafter, during the addition of preload, the rate can be much faster, e.g., 16/sec, as no comparison is required. Hence the variation in the slope of the curve shown. The variation in duty cycle is controlled by the calibrator's detection of the change in the current drawn by the transmitter, performed by the high-to-low and low-to-high transition detectors 68 and 69, respectively, as discussed in connection with FIG. 5.

FIG. 9(b) shows the check operation, in which the operator confirms that the setpoint is indeed at the correct value. This might typically be done several weeks after calibration; for example, to verify that the original setpoint is still being retained by the transmitter. The operator will have recorded the preload value used in a notebook or the like, for use in such check operations. He selects the check mode and pushes the calibration button. In the check mode, the calibrator causes the transmitter to step through all 256 steps at once. Initially, the capacitance provided by the digitally controlled capacitor steps upwardly at the lower rate. It eventually reaches step 255 and drops to zero, causing the alarm light to be lit (in high-level fail-safe), and then is stepped up until the reference capacitance bears the predetermined relation to the sensed capacitance. If the vessel is, in fact, empty this will be when C.sub.AIR is reached. The rate will then increase until the setpoint at step 190 is reached as shown and the normal light is then relit. The display on the calibrator, which is controlled as noted by the "END OF PULSE TRAIN" signal output by the pulse code generator, will indicate this final setpoint value of 190, which will be the same as the recorded value if, in fact, the transmitter is properly calibrated.

Note in this connection that the display 65 on the calibrator, when it is initially attached to the signal lines, will be blank. However, when the high-to-low loop current signal transition detector 69 indicates that the transmitter output has changed state--that is, that step 255 has been reached and the transmitter is, therefore, now going to step 0--the display is set to correspond to the correct setting of the value, namely "zero". Thereafter, the display steps in accordance with transmission of the calibration enabling signals and therefore in accordance with the iteration of the digitally controlled capacitor, such that its final value at the conclusion of the calibration check is indeed the correct value for the step at which the digitally switched capacitor stops.

In the recalibration (re-cal) mode, FIG. 9(c), one sets the desired "re-cal" mode, sets the desired final setpoint on the switches 84 of the calibrator and presses the "start" button 66. This simply causes the digitally controlled capacitor to be set to this value by stepping upwardly the desired number of steps after "zero" has been reached. This is done, for example, where it is not convenient to empty the vessel for true calibration but where it is desired that a known setpoint be established. Accordingly, in this operation, one simply sets the mode switch to "re-cal" and selects the desired setpoint. In this example, one would set the switches 84 to "190". Upon the pressing the calibration button 66, the capacitance steps upwardly to step 255, drops to "zero" (lighting the "alarm" light) and then steps rapidly directly up to the setpoint of 190. In the "re-cal" mode, after the 255-to-0 transition is reached, one is not comparing the actual capacitance with the capacitance provided by the digitally-controlled capacitor so there is no need to increment the capacitance at the slower duty cycle rate.

It will be appreciated that there has been described a remotely-calibratable, admittance-responsive instrument system which satisfies the needs of the art and objects of the invention as above. By provision of a calibrator which outputs calibration enabling signals which cause the transmitter to repetitively increment its reference capacitance and to compare it to an actual admittance signal, all elements of the system are automatically calibrated at once, as opposed to a system in which the calibrating admittance were added to or subtracted from an actual sensed admittance being compared to a fixed reference admittance. Automatic inclusion of preload in the calibration operation has also been described. Further, it will be appreciated that the transmitter, calibrator and receiver may all be connected by a single two-wire transmission line and that both calibrator and transmitter are adapted to draw power from the receiver, thus further simplifying matters. Those skilled in the art will recognize that, while there has been described a preferred embodiment of an instrument system, further improvements and modifications thereto would be possible. Accordingly, the invention is not to be limited by the present exemplary disclosure by only by the following claims.