BACKGROUND OF THE INVENTION

1. Field of the Invention

The present invention relates generally to surge detection devices, and, more particularly, to methods and apparatuses for detecting surge in a compressor of a compressor-driven system, such as a refrigeration system.

2. Description of the Related Art

As is generally known, surge or surging is an unstable condition that may occur when compressors, such as centrifugal compressors, are operated at light loads and high pressure ratios. It is a transient phenomenon characterized by high frequency oscillations in pressures and flow, and, in some cases, the occurrences of a complete flow reversal through the compressor. Such surging, if uncontrolled, causes excessive vibrations in both the rotating and stationary components of the compressor, and may result in permanent compressor damage. In particular, at surge conditions there exists a momentary reduction in flow and pressure developed across the compressor. Further, there is a reduction in the net torque and mechanical power at the compressor driving shaft. In the case where the drive device is an electric motor, these oscillations in torque and power will cause oscillations in motor current and excessive electrical power consumption.

While a number of surge detection devices and systems have been utilized, there exists a need for an improved surge detection arrangement, particularly but not exclusively, for use in refrigeration systems having a centrifugal compressor driven by an electric motor.

SUMMARY OF THE INVENTION Accordingly, the present invention is directed to improved methods and apparatuses for detecting surge in a compressor of a compressor-driven system by sensing the suction and discharge pressures of the compressor, or representative pressure of these values, and more particular, by reading the original value of these

sensed parameters into a microprocessor and periodically sampling the magnitude of the pressure developed across the compressor. The present invention is further directed to an improved method and apparatus for detecting surge by sensing the magnitude of the current applied to a variable speed motor drive which drives the electric compressor motor, or a parameter representative of the input power to the motor and, more particularly, by applying that sensed current value, preferably a DC current, to a microprocessor that periodically samples that sensed parameter. In a preferred embodiment of the invention, the sensed parameters are periodically compared with earlier values and the microprocessor applies a preselected routine designed to determine when a surge condition exists. The preselected routine is designed for the particular compressor system to which the invention is applied and is divulged according to the principles described in this application.

The most preferred embodiment of the invention utilizes both the pressure sensing and the current sensing techniques, described herein, to detect a surge. When surge is detected in the refrigeration system, the system then changes the frequency of the drive supplied to the compressor motor, to eliminate or minimize any adverse consequences of the surge. It has been found that such a system substantially obviates one or more of the problems due to the limitations and disadvantages of the related art. Additional features and advantages of the invention will be set forth in part in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention. The objectives and advantages of the invention will be realized and attained by the methods and apparatuses particularly pointed out in the written description and claims hereof as well as the appended drawings.

To achieve these and other advantages and, in accordance with the purpose of the invention, as embodied and broadly described herein, the invention comprises a surge detection system for use in a fluid system having a load driven by a centrifugal compressor, the compressor preferably being driven by an electric motor powered by a voltage source variable speed drive. In one embodiment, the surge detection system includes means for sensing a differential pressure developed across the compressor.

In other embodiments, the system includes means for sensing a current representing an input power of the variable speed drive. In yet another embodiment, the system includes both such sensing means. The surge detection system further comprises a computing means responsive to the differential pressure change, and/or the sensed current, wherein the computing means periodically samples one or both of the sensed parameters and compares the sensed values with one or more predetermined thresholds in order to determine when a valid surge is occurring in the compressor.

In one aspect, the surge detection system includes means for sensing a condenser pressure and means for sensing an evaporator pressure. These sensed pressures are directly related to the discharge and suction pressures of the compressor. In that embodiment, a computing means is responsive to the condenser and evaporator pressures, periodically compares these pressures, performs a further periodically applied routine, and indicates when a valid surge is occurring in the compressor. In still a further aspect, the invention includes a method for detecting surge in a refrigeration system having a centrifugal compressor driven by an electric motor powered by a voltage source variable speed drive and controlling the system to eliminate or mimmize adverse conditions that can be caused by uncontrolled compressor surge. The method includes the steps of sensing a condenser pressure, sensing an evaporator pressure, sensing a current representing an output power of the variable speed drive, periodically sampling the sensed values to determine when a valid surge is occurring in the compressor, and applying a corrective control signal to the variable speed drive when a surge is detected. The method can be practiced so that it determines surge based on only the sensed values of pressures, or current, or it can be designed so that it will determine surge only when both the sensed pressures and sensed current values indicate that a surge is occurring.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed. The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification,

illustrate an embodiment of the invention and together with the description, serve to explain the principles of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of the a surge detection apparatus of the present invention for use in a refrigeration system having a centrifugal compressor driven by an electric motor powered by a voltage source variable speed drive;

FIG. 2 is a more detailed block diagram illustrating the adaptive capacity control board and the pressure inputs of FIG. 1; FIG. 3 is a more detailed block diagram illustrating the adaptive capacity control board, the voltage source variable speed drive, and the variable speed drive current input of FIG. 1;

FIGS. 4 and 5 are graphical illustrations useful in understanding the principles of operation of the present invention in FIGS. 1-3; and FIGS. 6(a)-(d) are more detailed diagrams illustrating the operations performed by the microprocessor of the adaptive capacity control board for the pressure inputs of FIG. 2 and for the current input of FIG. 3, by way of a flow chart.

DESCRIPTION OF THE PREFERRED EMBODIMENT Reference will now be made in detail to the present preferred embodiments of the invention, examples of which are illustrated in the accompanying drawings. Wherever possible, the same reference numbers will be used throughout the drawings to refer to the same or like parts.

It is to be distinctly understood at the outset that the present invention was developed for a refrigeration system and the control of such a system. However, while the association of the present invention with a refrigeration system represents its present preferred application, that disclosure is not intended to serve as a limitation upon the broad scope or teachings of the invention. The present invention likely has applications in other fields and apparatus since the invention pertains to a method and apparatus for detecting surge in a compressor of a compressor-driven system.

The exemplary embodiment of the surge detection apparatus of the present invention, for use in a refrigeration or cooling system having a centrifugal compressor driven by an electric motor, is shown in Fig. 1 and is designated generally by reference numeral 10. The components of the refrigeration system and self-optimizing control system are the same as those disclosed in the teachings of U.S. Patent No. 4,608,833 [hereinafter "the '833 patent"], which is assigned to the assignee of the present invention, and which is hereby incorporated by reference. The '833 patent illustrates and explains details of one such refrigeration system which can be adapted to utilize the surge detection apparatus disclosed herein. In the '833 patent, the self-optimizing capacity control system is shown for inverter-driven centrifugal compressor based water chillers wherein adjustable inlet guide vanes and compressor speed are both automatically regulated in response to a continually updated "learned" chiller surge surface so as to realize minimum overall chiller energy consumption. In order to obtain the "learned" surge surface, the control system includes a microprocessor for initiating a "learning" mode in which the compressor motor speed is continually decreased incrementally and the pre-rotational guide vanes are moved to a more open position until an operating point is found where the compressor is surging.

For convenience of illustration of the present invention, the surge detection apparatus 10 is depicted in FIG. 1 in connection with a simplified version of a refrigeration system. The refrigeration system includes a centrifugal compressor 12, a condenser 14, an expansion device 16 and an evaporator 18, all being interconnected in series to form a conventional closed refrigeration circuit. Refrigerant gas is compressed in the compressor 12 and the compressed gas is delivered to the condenser 14 where a cooling medium, such as water from a cooling tower, causes the compressed gas to condense to a liquid refrigerant. The liquid refrigerant expands as it is passed through the expansion device 16 to the evaporator 18. As the liquid refrigerant flows through the evaporator 18, circulating water from a building is in a heat exchange relationship with the refrigerant so as to cause it to assume a vaporized state for delivery to a suction inlet of the compressor. In this manner, the water is chilled in the evaporator 18 for cooling the building. In order to

vary the amount of cooling imparted to the building in response to changes in the cooling requirement or load, the capacity of the compressor 12 is adjusted by regulating the flow rate of the refrigerant through the refrigeration system to vary its capacity. The refrigeration system of the present invention further includes a chiller control panel 34 having a main microprocessor board 36. An electric induction motor 20 powers the centrifugal compressor 12. The electric induction motor 20 is preferably driven by a voltage source variable speed drive 22. As shown in FIG. 3, the voltage source variable speed drive 22 preferably includes an AC to DC converter section 24, a low pass DC filter section 26 which smooths out a DC ripple, a DC to AC inverter section 28, and an inverter logic board 30 having a microprocessor 32 and an A/D converter 33.

It is generally known that a surge may cause permanent damage to the compressor if it is allowed to occur frequently and for long periods of time. Thus, in any application it is important to know when the compressor is surging. The surge detection apparatus 10 of the present invention will now be described in detail with reference to the refrigeration system of FIG. 1. Generally, the surge detection system 10 includes an adaptive capacity control board 44 which is provided within the chiller control panel 34 of the refrigeration system. In particular, the surge detection system 10 includes a first pressure transducer

38 placed in the condenser 14 to produce a signal which is a function of either the absolute or gauge pressure in the condenser 14. A second pressure transducer 40 is placed in the evaporator 18 for producing a signal which is a function of either the absolute or gauge pressure in the evaporator 18. The transducers 38 and 40 could, in the alternative, be located in the respective discharge line and suction line of the compressor 12. In either event, the transducers will sense pressures that are representative of the discharge and suction pressure of the compressor 12.

It has been discovered that when surge in the compressor occurs, the discharge pressure (or pressure at the condenser) decreases and tends toward the suction pressure (or pressure at the evaporator). The present invention applies the actual pressures sensed by the two transducers to a microprocessor, where the sensed

pressures are used to calculate a differential pressure which is periodically compared with delayed values to determine when a surge has occurred. In the preferred embodiment, these delayed values are periodically changed, depending upon a comparison of the difference between the sensed pressures and the then present delayed value. Through empirical and analytical efforts, a preferred routine has been developed to detect surge based on these broad principles. An example of a preferred routine developed for the system shown in FIG. 1 is described below, but the invention is not limited solely to that specific routine.

In the preferred embodiment, the adaptive capacity control board 44 is located within the chiller control panel 34, and is connected to the main microprocessor board 36 of the chiller control panel 34 and to the first and second pressure transducers 38, 40. As shown in FIG. 2, the adaptive capacity control board 44 includes a multiplexor 46 for receiving the condenser pressure and evaporator pressure signals in analog form and allowing each pressure signal to be output separately. Each pressure signal is then output to an A/D converter 48 for converting the selected pressure signals into digital signals which are finally output to a microprocessor 50. The microprocessor 50 receives the digital pressure signals, and periodically determines the difference between the condenser and evaporator pressure signals in order to produce a differential pressure value (dp). The microprocessor 50 periodically compares this differential pressure value with predicted control values and applies a routine designed to detect surge, based on the sensed pressures and threshold values generated through empirical and analytical efforts directed to the particular compressor and to the system to which it is attached. The operation of the microprocessor 50 of the preferred embodiment and routine is shown generally in FIG. 6 and is described below.

In order to provide an understanding of the operation of the preferred embodiment of the present invention, reference is now made to FIG. 4. In FIG. 4(a), there is shown a typical measured differential pressure signal (dp) which appears in the compressor 12 which is fluctuating and may be representative of a surge event. The difference between the discharge and suction pressure remains relatively constant when no surge is occurring. In Fig. 4(a) a dp_lag value, which represents a variable

created by the microprocessor 50 of the adaptive capacity control board 44, is used to detect surge and is automatically set equal to dp if dp exceeds dp lag. If dp drops below dp lag, then dp lag is preferably lowered toward dp at a preselected rate, one working rate being 0.1 psi per second. This rate may vary depending on the application. As shown in FIG. 4(a), it has been discovered that this variable lags relatively close if not equal to the actual dp values, when no surge is occurring.

FIG. 4(b) depicts a typical measured differential pressure signal during a valid surge event, as well as how the microprocessor 50 adjusts the dp lag variable. As shown in FIG. 4(b), the preferred embodiment of the routine requires four conditions to be met in order for the change in the differential pressure to qualify a valid surge. Two occur in what has been arbitrarily designated software state zero, at least one condition occurs in a software state one, and one occurs in a software state two. The first and second conditions are met (in software state zero) when the value (dp lag- dp) exceeds a preselected threshold [threshold 1 in FIG. 4(b)] which indicates that a negative transition has occurred on dp. In the preferred embodiment applied to the refrigeration system of FIG. 1, this value must then remain below the threshold continuously for at least a preselected time (preferably 100 milliseconds) to qualify as the second condition. When these two conditions are met, a five second timer (the timer may vary from two to ten seconds, depending on the application) is set and the microprocessor automatically enters into software state one.

The third condition continuously samples subsequent dp signals, after the first and second conditions are met, in software state one. In the present preferred embodiment, the dp signals of the third condition are met in software state one when the value (dp ag-dp) exceeds a second preselected threshold [threshold 2 in FIG. 4(b)] for at least 340 milliseconds. The time period that threshold 2 is to be exceeded may vary considerably from application to application. For systems like that shown in FIG. 1, the time period may vary from 200 to 750 milliseconds depending upon the type of system. If the dp signals of the third condition are met, the microprocessor automatically enters software state two, and dp lag is set equal to dp. By means of example, threshold 1 is preferably 0.5 psi on low pressure compressors and 3.4 psi on high pressure compressors. The threshold may vary depending on the application.

In the preferred embodiment, threshold 2 preferably is ten to twenty percent greater than threshold 1.

The fourth condition requires further sampling of the dp signal, after the first through third conditions are met. The dp signal of the fourth condition meets the requirements of software state two when at least two successive positive transitions [Δdp_lag > 0, see FIG. 4(b)] are detected. If the third and fourth conditions are not met within a preselected time period (in the preferred embodiment, five seconds), software state zero is automatically re-entered and the process starts over again. Also, by setting the dp lag to dp when the third condition is met, the dp_lag is automatically re-armed for the next surge event. If the four conditions are satisfied within the five second time interval, a valid surge is acknowledged.

The details of a computing device, such as the microprocessor 50 of the adaptive capacity control board 44, required to perform the functions just described with respect to the operation of FIG. 4 is shown in FIG. 6 in a flow chart fashion. The microprocessor 50 receives as input signals the condenser and evaporator pressure signals.

The surge identification process is initiated by a start block 70 which causes the reading of the input signals from the pressure transducers 38, 40. A logic or decision block 72 determines whether a 20 millisecond time period has elapsed since the last values were read. If 20 milliseconds have not expired, then the routine is exited. Otherwise, block 74 is entered to read the first input from the pressure transducer 38 attached to the condenser 14, and block 76 is entered to convert the first input value into "value 1." The second input from the pressure transducer 40 is then read and converted into "value 2" in blocks 78 and 80, respectively. The input values "values 1 and 2" are digital signals. Thus, in the conversion process, the microprocessor 50, searches a lookup table to correlate the digital values "values 1 and 2" with a corresponding pressure value times ten, i.e., the pressure values stored in the lookup table are lOx the actual values of the pressure.

The second value "value 2" is then subtracted from the first value "value 1 " to obtain "x" ("x" represents the differential pressure dp) in decision or logic block 82. If the value of "x" subtracted from variable "x lag" is positive, decision block 84

proceeds to counter block 86, in which a "lag count" is set equal to fifty counts, and then to block 88 in which the variable "x lag" is set equal to "x." One count is equal to twenty milliseconds, thus, fifty counts is equal to one-thousand milliseconds or one second. If the value of "x" subtracted from variable "x lag" is negative, decision or logic block 84 proceeds to decision or logic block 90 in which it is determined whether the "lag count" is equal to zero and whether the "surge state" is equal to zero. If both conditions of decision block 90 are true, then "x lag" is decremented by one unit in block 92, the "lag count" is reset to fifty counts (1000 milliseconds or 1 second) in block 94, and decision or logic block 96 is entered. The units in this application are equal to 0.1 psi per second, as stated above. In other words, the "x lag" is actually decremented by 0.1 psi per second.

If both conditions are not satisfied in decision block 90, then decision block 96 is entered where it is determined whether the "lag count" is greater than zero. If the "lag count" is greater than zero, then the "lag count" is decremented by one count (20 milliseconds) in block 98, otherwise decision or logic block 100 is entered. If the "surge_state" is determined to be equal to zero in decision block 100, then the routine branches to subroutine A, otherwise, the routine continues to decision or logic block 102. If the "surge state" is determined to be equal to one in decision block 102, then the routine branches to subroutine B, otherwise, the routine continues to decision or logic block 104. If the "surge state" is determined to be equal to two in decision block 104, then the routine branches to subroutine C, otherwise, the routine continues to set the "surge state" equal to zero in block 106 and then exits.

In subroutine A, decision or logic block 108 determines whether "x lag" less "x" is greater than a "surge threshold," i.e., threshold 1 in FIG. 4 ). If the "surge_threshold" is exceeded, the subroutine proceeds to block 110 where "neg_count" is incremented by one count (20 milliseconds), otherwise the "neg count" is set equal to zero in block 112. After block 112, the subroutine proceeds to decision or logic block 114 in which it is determined whether or not the "neg_count" is equal to five counts (100 milliseconds). If "neg_count" is not equal to five counts the subroutine is exited, otherwise, "surge count" is set equal to zero in block 116, "surge state" is set equal to one in block 118, "neg_count is set equal to

zero in block 120, and "test count" is set equal to "250," i.e. , five seconds (250 x 20 milliseconds = 5 seconds) in block 122. After block 122, the subroutine is exited.

In subroutine B, a decision block 124 is entered to see if "test count" is greater than zero. If "test count" is not greater than zero, "surge state" is set to zero in block 126 and the subroutine is exited, otherwise, "test_count" is decremented by one count (20 milliseconds) in block 128 and decision or logic block 130 is entered. In decision block 130, if "x_lag" less "x" is less than or equal to "surge_threshold + n", (i.e., threshold 2 in FIG. 4(b), with "surge threshold" representing threshold 1 in FIG. 4(b) and "n" representing ten to twenty percent of threshold 1) then the subroutine is exited, otherwise, "surge count" is incremented by one count (20 milliseconds) and decision or logic block 134 is entered. In decision block 134, if "surge count" is less than or equal to fifteen counts (300 milliseconds), then the subroutine is exited, otherwise, "x_prev" and "x lag" are set equal to "x" in blocks 136 and 138, respectively, "surge state" is set equal to two in block 140, "pos count" is set equal to zero in block 142, and the subroutine is exited.

In subroutine C, decision or logic block 144 determines if "test count" is greater than zero. If "test count" is not greater than zero, then "surge state" is set equal to zero in block 146 and the subroutine is exited, otherwise, "test count" is decremented by one count (20 milliseconds) in block 148 and decision or logic block 150 is entered. In decision block 150, if "x" less "x_prev" is less than or equal to zero, then decision block 152 is entered, otherwise, "pos count" is incremented by one count (20 milliseconds) in block 154 and "x lag" is set equal to "x" in block 156. If "pos_count" is determined to be not equal to two counts (40 milliseconds) in decision block 152, then the subroutine is exited, otherwise, a surge is identified in block 158, a surge LED is flashed in block 160, "surge_pos_transition" is set equal to "true" in block 162, "surge state" is reset to zero in block 164, and the subroutine is exited.

When a surge is detected, the microprocessor 50 of the adaptive capacity control board 44 generates a control signal via "link 1" (see FIG. 1) to increase the frequency supplied by the voltage source variable speed drive 22. The increase in frequency in the voltage source variable speed drive 22 increases the electric motor

speed, which in turn increases the rotational speed of the compressor 12, thus preventing another surge from occurring in the compressor 12. Such control techniques are outlined in the '833 patent incorporated by reference above.

In an alternative embodiment, as shown in FIG. 3, the surge detection system 10 may include a DC current transformer 42, or plurality thereof, located in between the DC filter section 26 and the DC to AC inverter section 28 of the voltage source variable speed drive 22 for directly sensing an output DC link current ^ of the DC filter section 26. The average value of the DC link current i_ _c is preferably reconstructed using three phases of motor current i _A , i _B and i^. supplied by the DC to AC inverter section 28 of the voltage source variable speed drive 22 and three output switch command signals A, B and C supplied by the inverter logic board 30 to switches S1-S6 of the DC to AC inverter section 28. The average value of the DC link current !_ • is indicative of the power delivered by the DC filter section 26, since the DC voltage v _DC generated by DC filter section 26 is constant, and the power delivered by the DC filter section 26 is calculated by multiplying the DC voltage v_ _c by the DC link current i _DC

As seen in FIG. 3, the adaptive capacity control board 44 is connected, via serial "link 2," to the main microprocessor board 36 of the chiller control panel 34 and to the inverter logic board 30, via serial "link 1," of the voltage source variable speed drive 22. The main microprocessor board 36 provides control for other stages of the chiller as well as digital display read-out of the chiller operating conditions. The variable speed drive's inverter logic board 30 includes an A/D converter 33 for converting the average value of the DC link current signal ]__ into a digital signal, and a microprocessor 32 for receiving the digital representation of the average value of the DC link current signal ir _j c. The variable speed drive inverter logic board 30 may process the DC link current ^ _c directly and send a surge flag to the adaptive capacity control board 44 during a surge event via "link 1." Alternatively, the inverter logic board 30 may send the digital value of the DC link current i_ _c , via "link 1," and permit the microprocessor 50 of the adaptive capacity control board 44 to process the DC link current i^ in order to determine surge. Thus, in an

alternative embodiment, the present invention, may preferably detect a surge event by utilizing the detected DC link current ipc.

In order to provide an understanding of the operation of the alternative embodiment of the present invention, reference is now made to FIG. 5. In FIG. 5(a) of the drawings, there is illustrated graphically a typical measured average value of the DC link current (designated as "ilnk") which appears in the voltage source variable speed drive 22 and has fluctuations which may be representative of a surge event. Also shown in FIG. 5(a) is a ilnk lag value which represents a variable created by the microprocessor 50 or 32 of the adaptive capacity control board 44 or the variable speed drive inverter logic board 30, respectively. The variable ilnk lag is automatically set equal to ilnk if ilnk exceeds ilnk lag. If ilnk drops below ilnk lag, then ilnk lag is slowly lowered toward ilnk at a rate of ten least significant bits (LSBs) per second. The A/D converter 33 of the inverter logic board 30 has twelve bits or a total of 2 ¹² (=4096) LSBs. The rate ilnk lag is lowered may vary depending on the application. FIG. 5(b) depicts a typical measured DC link current signal (ilnk) during a valid surge event, as well as how the microprocessor 50 or 32 adjust the ilnk lag variable.

In accordance with the DC link current i^ surge detection method, the conditions listed above for the differential pressure surge detection method, must also be satisfied for the DC link current surge detection method. Thus, the DC link current surge detection method is analogous to the differential pressure surge detection method. However, the sensed value of the DC link current (not the dp value) is used. Furthermore, the DC link current surge detection method's third condition requires the value (ilnk lag-ilnk) to exceed a second threshold [threshold 2 in FIG. 5(b)] for at least 350 milliseconds, rather than 340 milliseconds. The time period that threshold 2 is to be exceeded may vary from 200 to 750 milliseconds, depending upon the application. Furthermore, the DC link current surge detection method's fourth condition requires at least three successive positive transitions [Δilnk lag > 0, see FIG. 5(b)] to be detected, rather than two. Further in accordance with the DC link current surge detection method, the microprocessor 50 or 32 of the adaptive capacity control board 44 or inverter logic

board 30, respectively, follows the same surge identification process of FIG. 6 as was described above for the differential pressure surge detection method. However, in the DC link current surge detection method the "x" should be replaced with "ilnk, " and 10 milliseconds should be used in place of 20 milliseconds as shown decision block 72 of FIG. 6(a). This means that in the DC link current surge detection method, one count is equal to ten milliseconds, rather than twenty milliseconds. Furthermore, in the DC link current surge detection method, the microprocessor 50 or 32 does not read and convert a second input [blocks 78 and 80 of FIG. 6(a)], since the microprocessor 50 or 32 only reads in one input, i.e. , the DC link current i_ _c . Thus, the microprocessor 50 or 32 does not subtract "value 2" from "value 1 " in block 82 of FIG. 6(a), but only sets "x" equal to "value 1. " Also, the DC link current surge detection method lowers "ilnk lag' toward "ilnk" at a rate of ten LSBs per second.

In light of the above embodiments, various combinations of the surge detection system can be devised. For example, surge may be detected by sensing pressure change across the compressor 12 only, by sensing DC link current fc _c in the voltage source variable speed drive 22 only, or by sensing pressure change across the compressor 12 and/or DC link current i_ _c in the voltage source variable speed drive 22. In the present preferred embodiment, both the pressure and the current models are used to detect surge and surge may be indicated if one or both routines determine usage.

It will be apparent to those skilled in the art that various modifications and variations can be made in the surge detection method and apparatus of the present invention and in construction of this method and apparatus without departing from the scope or spirit of the invention. As an example, the DC link current method can be performed by the microprocessor 32 contained in the inverter logic board 30 of the voltage source variable speed drive 22, thus eliminating the need for the adaptive capacity control board 10. Furthermore, the compressor 12 need not necessarily be of the centrifugal-type, but may also be any other type of rotodynamic compressor.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.