WO2016118419A1

WO2016118419A1 - Pump assembly and method for assessing valve conditions in pump

Info

Publication number: WO2016118419A1
Application number: PCT/US2016/013634
Authority: WO
Inventors: Gulshan Singh; Thomas Jaeger; Erik N. LEE
Original assignee: Baker Hughes Incorporated
Priority date: 2015-01-19
Filing date: 2016-01-15
Publication date: 2016-07-28
Also published as: US20160208794A1

Abstract

A method of monitoring and assessing valve conditions in a pump includes collecting data regarding timing of a pump piston and vibration of a pump fluid end, processing the data to form a filtered and transformed vibration signal, banding the vibration signal into high energy bands and low energy bands, and comparing a ratio of at least one high energy band to at least one low energy band of the vibration signal with at least one constant to determine valve condition.

Description

PUMP ASSEMBLY AND METHOD FOR ASSESSING VALVE CONDITIONS IN PUMP

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Application No. 14/599622, filed on January 19, 2015, which is incorporated herein by reference in its entirety.

BACKGROUND

[0002] In the drilling and completion industry, the formation of boreholes for the purpose of production or injection of fluid is common. The boreholes are used for exploration or extraction of natural resources such as hydrocarbons, oil, gas, water, and alternatively for C02 sequestration. Drilling fluids can be pumped downhole to assist in the formation of boreholes. Also, to increase the production from a borehole, the production zone can be fractured to allow the formation fluids to flow more freely from the formation to the borehole. The drilling operation and the fracturing operation sometimes include pumping fluids at high pressure downhole and towards the formation.

[0003] To pump fluids at the high pressures required for downhole operations, crankshaft driven positive displacement pumps are used, which include an engine, cooling system, transmission, power end and fluid end. The power end includes or is attached to a pump powering mechanism also known as a prime mover, commonly a diesel combustion engine or alternatively an electric motor, which connects to a pinion shaft to drive the power end. The transmission connects the engine and power end and provides speed control and dampening. The fluid end is connected to the power end. The fluid end includes a number of plungers driven by a crankshaft toward and away from a chamber in order to affect a high or low pressure on the chamber. The fluid end receives relatively low pressure fluid, and pressurizes the fluid to provide higher pressurized fracturing fluid at the required pressure for fracturing within the borehole. The valve and valve seats are some of the most commonly replaced items in the fluid end.

[0004] Typically in the well service industry, the valves and seats are replaced on a regular basis without knowledge of their condition at the time of removal. Generally, the valves are removed twice as frequently as compared to the seats. This is because the fluid end can be damaged if valves and seats are not removed before excessive wear occurs. Excessive wear leads to valve damage and valve damage leads to fluid end washout. Fluid end washout is one of the major causes of fluid end failure. Fluid ends are a lot more expensive to replace as compared to valve and seat assemblies. It is therefore desirable that valves and seats be removed before wear can result in damage sufficient to impact the fluid end. However, it is not always possible to ensure that replacement occurs in time yet such precautionary approach may still lead to premature removal of valves and seats.

[0005] The art would be receptive to improved apparatus and methods for valve and seat condition detection for well servicing pumps.

BRIEF DESCRIPTION

[0006] A method of monitoring and assessing valve conditions in a pump includes collecting data regarding timing of a pump piston and vibration of a pump fluid end, processing the data to form a filtered and transformed vibration signal, banding the vibration signal into high energy bands and low energy bands, and comparing a ratio of at least one high energy band to at least one low energy band of the vibration signal with at least one constant to determine valve condition.

[0007] A pump assembly includes a pump including a power end, a fluid end, a piston, and a plurality of valves, a timing sensor configured in the power end to sense movement of the piston, an accelerometer in the fluid end to sense vibrations within the fluid end, and a system configured to collect data from the timing sensor and accelerometer, process the data to provide a filtered and transformed vibration signal, and compare a ratio of high energy bands and low energy bands of the vibration signal to at least one constant to assess valve condition.

[0008] A method of monitoring and assessing valve conditions in the pump assembly described above includes collecting data regarding timing of the piston and vibrations in the fluid end, processing the data by the system to form a filtered and transformed vibration signal, banding the vibration signal by the system into the high energy bands and the low energy bands, and comparing a ratio of at least one high energy band to at least one low energy band of the vibration signal with the at least one constant to determine valve condition.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The following descriptions should not be considered limiting in any way. With reference to the accompanying drawings, like elements are numbered alike:

[0010] FIG. 1 shows a cross-section of a power and fluid end of one embodiment of a pump assembly; [0011] FIG. 2 shows a sectional view of a fluid end chamber of the embodiment of the pump assembly of FIG. 1;

[0012] FIG. 3 shows a flow chart of one embodiment of a method of data collection and condition monitoring of valves and seats of the pump assembly of FIG. 1;

[0013] FIG. 4 shows a Fast Fourier Transform chart of vibration signals from new, medium worn, heavily worn, and destroyed valves; and,

[0014] FIG. 5 A shows a chart of frequency filtered, noise filtered, and transformed signal from the valves, and FIG. 5B shows the chart of FIG. 5 A banded with overlapping high energy spectral bands and low energy spectral bands.

DETAILED DESCRIPTION

[0015] A detailed description of one or more embodiments of the disclosed apparatus and method are presented herein by way of exemplification and not limitation with reference to the Figures.

[0016] FIG. 1 illustrates one embodiment of a pump assembly 100. The pump assembly 100 may be applied to a variety of well servicing pumps, reciprocating positive displacement pumps, and may have one or more cylinders such as, but not limited to, quintuplex pumps with five cylinders or triplex pumps with three cylinders. The pump 10 of the pump assembly 100 includes a power end 14 and a fluid end 16. A prime mover 12 of the pump assembly 100 could be an electric motor or an internal combustion engine, such as a diesel engine, and other prime movers can be incorporated. The prime mover 12 drives the pump 10. The prime mover 12 can be located at the power end frame or housing 20 or at another convenient location, and is employed to rotate the crankshaft 18 within the housing 20 about its longitudinal axis 22. The prime mover 12 may also rotate a pinion shaft that engages with the crankshaft 18 to rotate the crankshaft 18. While not depicted, it should be understood that the crankshaft 18 may include a number of eccentrically arranged crankpins (or alternatively a plurality of first eccentric sheaves, not shown). A connecting rod 24 is operatively engaged with the crankshaft 18, and a connecting rod 24 may be connected to each crankpin or sheave of crankshaft 18. The connecting rod or rods 24 connect the crankshaft 18 to the pony rod(s) 26 and plunger(s) 28, which may be interconnected by pony rod clamp 30 and/or plunger clamp 32 (FIG. 2). The pony rod 26 and plunger 28 will together be referred to herein as the piston(s) 34. The connecting rod 24 may be connected to a crosshead 36 using a wrist pin 38 that allows the connecting rod 24 to pivot with respect to the crosshead 36, which in turn is connected to the piston(s) 34. The longitudinal axis 40 of each of the piston(s) 34 is perpendicular to the longitudinal axis 22 of the crankshaft 18. When the crankshaft 18 rotates, the crankpins (not shown) reciprocate the connecting rod(s) 24. Moved by the connecting rod(s) 24, the crosshead 36 reciprocates inside fixed cylinder(s) 42. In turn, the piston(s) 34 coupled to the crosshead 36 also reciprocate between suction and power strokes in the fluid end 16. Thus, the crosshead 36 converts rotational movement of the power end 14 into reciprocating movement to actuate the pistons 34 of the fluid end 16. While only one internal piston 34 to pump the fluid in the fluid end 16 is depicted in the cross-section shown in FIG. 1, a plurality of pistons 34 may be provided in the pump 10. Further depicted in FIG. 1 is the suction cover assembly 44, inlet and outlet valve assemblies 46, 48, and stay rods 50. FIG. 1 also shows the pressure chamber 52 that is used to create high pressure for well servicing. Sensors 54, 56, shown schematically in the power end 14 and fluid end 16, are further employed in the pump 10 for the method of assessing valve conditions as will be further described below. Sensor location shown in FIG. 1 is indicative in nature. Sensor is located on the fluid end 16.

[0017] Turning now to FIG. 2, the fluid end 16 is shown in more detail. The fluid end 16 includes an inlet valve 58 at an inlet of the fluid end 16 and an output valve 60 at an outlet of the fluid end 16. Withdrawal of the plunger 28 of the piston 34 during a suction stroke (by moving the piston 34 closer to the power end 14) pulls fluid into the fluid end 16 through the inlet valve 58 via the inlet. Subsequently pushed during a power stroke, the piston 34 then forces the fluid that has entered the pressure chamber 52 under pressure out through the outlet valve 60 and to the outlet. The inlet valve assembly 46 further includes a valve seat 62 that receives inlet valve 58, valve spring 64 that biases the inlet valve 58 to a closed condition within the valve seat 62, and valve spring retainer 66 that maintains positioning of the valve spring 64. Likewise, the outlet valve assembly 48 includes valve seat 68 that receives the outlet valve 60 and valve spring 70 that biases the valve 60 to a closed condition. Discharge cover 72 can serve to retain the valve spring 70 in place, and discharge cover retainer 74 secures the discharge cover 72 in place. FIG. 2 further shows a packing nut 76, stuffing box 78, packing assembly 80, and seal carrier 82 arranged for receiving the plunger 28 of the piston 34 and maintaining fluid within the pressure chamber 52. FIG. 2 further shows suction cover retainer nut 86 and suction cover 84 arranged to access pressure chamber 52. The pressure chamber 52 includes a cross bore (horizontal and vertical bore). The plunger 28 comes into the chamber 52 from the power end side of the horizontal bore. The other end of the horizontal bore, a suction cover side, is mainly used to repair and maintain chamber components. The bottom end of the vertical bore is called suction side and has the inlet (suction) valve 58. While only one sensor 56 is schematically depicted within the fluid end 16, and one sensor 54 within the power end 14, more than one sensor may be provided in each of the fluid end 16 and power end 14 of the pump assembly 100. The sensors 54, 56 are employed in the pump 10 for the method of assessing valve conditions, as will be further described below.

[0018] The suction and discharge valves 58, 60 are pressure-operated valves. Valve opening and closing depend upon the weight of the valves 58, 60, spring force of the springs 64, 70, valve geometry of the valves 58, 60 and pressure difference on the two opposing sides of each valve 58, 60. Valve weight, spring force and suction pressure inside the chamber 52 closes the discharge valve 60 and opens the suction valve 58. Valve weight, spring force and discharge pressure within the chamber 52 closes the suction valve 58 and opens the discharge valve 60. Each valve 58, 60 opens and closes once every revolution of the piston 34.

[0019] Under normal circumstances, valves 58, 60 and their seats 62, 68 are removed well ahead of utilization of their useful life. A premature removal of valves 58, 60 and seats 62, 68 leads to suboptimal expenses in repair and maintenance cost. However, it is difficult to devise a common valve and seat change schedule for a large fleet because their life consumption depends upon many factors. Some of these factors include suction pressure, discharge pressure, speed, type of fluid pumped, fluid end maintenance, quantity, quality and dimensional integrity of the valves 58, 60 and their respective seats 62, 68. Valve and seat material properties and surface hardness also play a role.

[0020] The system and method described herein allow a pump operator to assess the valve and seat conditions without opening the fluid end 16. The knowledge of valve damage level can mitigate fluid end washout while preventing premature removal of valves 58, 60 and seats 62, 68. A valve and seat operation in good condition is assessed and recorded for its particular timing and vibration signature. Time duration taken for opening and closing a valve 58, 60 and vibration signature of fluid end 16 are different in good condition as compared to worn or damaged conditions, thus providing a comparison.

[0021] FIG. 3 shows a flow chart of one embodiment of a method 1 10 of data collection and condition monitoring of valves 58, 60 and seats 62, 68 of the pump assembly 100 of FIG. 1. The process involves data collection, as indicated by boxes 112, 114, 116, data analyses (signal processing) as indicated by boxes 118-124, and condition prediction as indicated by boxes 126-134. Signal processing involves multiple steps including noise filtering, signal conditioning, frequency filtering and signal transformations. The transformed signal is compared against good, worn out and destroyed condition signals to predict current state of a valve 58, 60 and valve seat 62, 68.

[0022] Data collection involves using two or more types of sensors 54, 56 to collect data and monitor valve condition. As shown in FIGS. 1 and 2, one or more timing sensors 54 and one or more time synchronous vibration sensors (accelerometers) 56 may be employed to determine the timing signal, pump speed, vibration, velocity and displacement per cycle 118 shown in FIG. 3. The timing sensors 54 (sensors for acquiring the frequency of the cycling of the piston(s) 34 within the pump 10) schematically depicted in FIG. 1, can be configured at power end 14 to sense piston movement back and forth to determine the timing and frequency of piston reciprocation and pump speed. The vibration sensor(s) 56 are optimally mounted at the fluid end 16. In an exemplary embodiment, an accelerometer 56 is employed as the vibration sensor to detect the vibration within the fluid end 16 by detecting movement in x, y, and z axes. While depicted at a particular location, the vibration sensors 56 may be positioned at various locations on or within the fluid end 16. Combining information from the accelerometer 56 and the pump speed provides the desired information regarding vibration in all three axes, velocity and displacement per cycle. While not required, additional sensors (not shown) for suction pressure, discharge pressure, power end vibration and acoustics may also be employed in the pump assembly 100.

[0023] A system 150 is provided within the pump assembly 100 to collect data at a high sampling rate, such as, but not limited to, more than 10,000 samples per second and up to 60,000 samples per second, including the data collection shown in boxes 112, 114, 116. The system 150 includes a controller 152 containing software for performing the method 110, a memory, and data acquisition, cables or wireless communication apparatus (not shown), and the sensors 54, 56. The system 150 may also include a display 154, such as a monitor, to provide indications of the conditions of the valves 58, 60 and valve seats 62, 68 to an operator. Optionally, if the system 150 determines that the valves 58, 60 and/or valve seats 62, 68 are in a dangerously worn out condition, the system 150 may further provide a signal to the prime mover 12 to halt operation of the pump 10 to prevent damage to the fluid end 16. Otherwise, utilization of the system 150 does not interfere with operation of the pump 10.

[0024] With the data acquired from the sensors 54, 56, or derived therefrom, the controller 152 of the system 150 conducts a noise filtering technique, which is a digital signal processing method such as spectral averaging as indicated by box 120 in FIG 3 to beat down any non-pump related energy. Spectral averaging can be used for noise filtering to increase the ability to detect pump synchronous related energy. This method entails computing many spectra over a time frame when the rotating equipment is at a steady state condition. By then averaging the spectra together the persistent energy (produced by the pump or other sources) remains evident while any stochastic process will collapse to its mean value. Spectral averaging can also be employed in addition to time synchronous averaging to further beat down any noise for the noise filtering operation employed in block 120 to further enhance the signal (frequency versus magnitude) that is truly attributable to valve 58, 60 and valve seat 62, 68 wear detection. For example, time synchronous averaging may be a pre-processing method performed prior to spectral averaging.

[0025] Another method for noise filtering that can be used to increase the ability to detect pump synchronous related energy is time synchronous averaging. Time synchronous averaging can replace or assist spectral averaging as indicated by box 120 in FIG. 3. For example, noise that is not persistent for a certain time period, such as but not limited to 30 seconds or one minute, may be considered non-pump related energy. Thus, the goal of the noise filtering technique is to remove any non-pump related data that would not be indicative of valve and valve seat wear. Rotational machinery operating at a steady state condition typically rotates at a fixed rotational frequency that is controlled by its prime mover. Even when a precise time invariant rotational frequency is desired, the actual rotational speed will vary slightly in time producing a time- varying rotational speed referred to as "dithering" in the field of rotational machinery. An example of this is would be when a controller dictating the drive frequency implements feedback control and must respond to varying torque and load conditions on the pump/drive. When conducting a fixed block size Fourier

decomposition, the frequency bins are equally spaced and the cutoff frequency for the Fourier filter bank is fixed. In this case, the synchronous energy produced by the pump rotational process may dither between several frequency bins over the time duration covered by one block size. This will result in smearing out the synchronous energy related to the rotational process. Changing the block size is not a feasible option as the problem with time resolution linked to frequency resolution associated with a Fourier analysis excludes a single "optimal" block size to properly represent the dithering energy spectra. One method to resolve this issue is to use time-synchronous averaging. With this method it is assumed that a key phasor signal is provided which specifies the duration of each revolution. Assuming a pump is dithering, the time required to complete one revolution will be a time-varying function. With knowledge of the duration for each revolution, the vibration signal will be segmented such that only one revolution is contained within one time sequence. Each of these time sequences will then be stretched or compressed in time such that all of the time sequences cover the same time duration. This in effect removes the known dithering process. These time sequences can then be averaged together which will remove any non- synchronous energy and a spectrum computed. Alternatively, these nonlinearly altered time sequences can be recombined to form a new time history with all dithering removed. There are also cases where a key phasor is not available and one must use a generalized hyper-coherence method or related methods to estimate a key phasor before proceeding with the time synchronous averaging process.

[0026] A low pass filter may be employed in the method 1 10 to narrow the range of the specific frequency of the signal that the system 150 analyzes, as demonstrated by box 122. As should be understood, a low pass filter passes signals with a frequency lower than a certain cut-off frequency, and attenuates signals with a frequency higher than the cut-off frequency, such that the low pass filter will set in a particular frequency range in an area of actual interest. The use of a low pass filter 122 also decreases the amount of processing required in the remainder of the method. In some embodiments, the low pass filter 122 may be eliminated since the transformation (box 124) is breaking energy down into individual bands. However, employing a low pass filter enables computation of the total energy over a specially selected band, e.g. 0-10 Hz. Thus, a low pass filter 122 may be used in addition to transformation 124 or as an alternative to the transformation 124. With regards to the transformation 124, Fast Fourier Transformation ("FFT") is employed to segment the measured vibration signals into components of different frequencies.

[0027] FFT of vibration signals in frequency domain from valves in as-new condition 160, medium worn out condition 162, heavily worn out 164, and destroyed conditions 166 are shown in FIG. 4. The signals 160, 162, 164, 166 demonstrate how the spectral signature from new, medium worn, heavily worn, and destroyed valves and seats are significantly different from each other. FIG. 5A depicts an example of a filtered and transformed signal 170 as detected by the system 150. Power spectrum density ratio (or difference) inside and outside of specific frequency bands, as will be further described below, are used to estimate valve health.

[0028] The signal 170 (as shown by example in FIG. 5 A) in frequency domain is used after low pass filtering 122, transformation 124 (e.g., fast Fourier transformation) and noise filtering 120 (e.g., time synchronous averaging or spectral averaging). Thus, FIG. 5A shows a frequency filtered, noise filtered, and transformed signal 170. As noted by box 126 in FIG. 3, multiple bands 172 are created within a frequency range based on the pump speed and the phenomenon of interest (i.e., valve wear). These bands 172 are called harmonic frequency response spectral bands 174 ("HFRSB") and non-harmonic frequency domain spectral bands 176 ("NHFRSB"). FIG. 5B shows two examples of each type of bands 174, 176, and show a schematic of a spectral banding process. In a normal pump operation, HFRSBs 174 will have higher energy (auto spectrum amplitude) as compared to HFRSBs 176. The total number of HFRSBs 174 and NHFRSBs 176 depends upon pump operating parameters and phenomenon of interest. High energy spectral bands 174 are { HFRSB i, HFRSB₂, ... , HFRSBi, ...

HFRSB_n}. Noise or low energy spectral bands 176 are {NHFRSB i, NHFRSB₂,

NHFRSB_j, ... NHFRSB_m}. FIG. 5B shows the same frequency filtered, noise filtered and transformed signal 170 as in FIG. 5 A, but with overlapping HFRSB and NHFRSB bands 174, 176. A signal inside a HFRSB band 174 (boxes drawn with dash lines) is called region of interest. A signal inside a NHFRSB band 176 (boxes drawn with dotted lines) is outside the region of interest.

[0029] For a normal operation (valves in good condition), on average the signal magnitude inside HFRSB band 174 is expected to be higher as compared to signal magnitude inside a NHFRSB band 176. The ratio of average magnitude inside HFRSB and NHFRSB bands 174, 176 is used to predict valve condition, as noted by box 128 in FIG. 3. An example range is given below in Table 1. The table shows that the valves 58, 60 are in good condition if the ratio is greater than Ki, and usage may be continued as indicated in box 130 of FIG. 3. The valves are worn if the ratio is between K₂ and Ki, as indicated by box 132 of FIG. 3. The valves are destroyed if the ratio is less than K₂, as indicated by box 134 of FIG. 3. The values of constants Ki and K₂, where K₂ is smaller than Ki, are decided based on the definition of wear out and destroyed valves 58, 60 and valve seats 62, 68.

[0030] Table 1 : Valve condition determination example:

[0031] (I a| HFRSBi) / (I a| NHFRSBi) > Ki Good condition

[0032] K₂ < (I a| HFRSBi) / (I a| NHFRSBi) < Ki Worn out condition (ideal time to change)

[0033] (I a| HFRSBi) / (I a| NHFRSBi) < K₂ Destroyed (Stop operation)

[0034] (I a] HFRSBi) indicates average magnitude of the spectrum in the ith HFRSB band 174. Similarly (I a] NHFRSB_j) indicates average magnitude in )i NHFRSB band 176.

Multiple bands 174, 176 are used to ensure the reliability of the prediction. The number of bands 174, 176 and band location is decided based on the pump speed and predictive conditions. [0035] Thus, a valve condition monitoring and assessment algorithm 110 has been described, as well as a pump assembly 100 incorporating a system 150 for employing the valve condition monitoring and assessment algorithm 110. Although more sensors can be employed, the system can advantageously utilize as few as two sensors 54, 56 to monitor and predict valve conditions in fluid ends 16. Pump speed, vibration (Gs) and the frequency domain signature is calculated from the collected data to predict valve condition. The data is filtered (frequency and noise) and transformed. Multiple filtering techniques (in time and frequency domain) may be used to filter the signals. Transformations are used to prepare the signal for harmonic frequency response spectral bands 174, 176. Because only specific signal bands are looked at, excessive computing processing time is not required. These bands are used to predict valve condition.

[0036] While the invention has been described with reference to an exemplary embodiment or embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed as the best mode contemplated for carrying out this invention, but that the invention will include all embodiments falling within the scope of the claims. Also, in the drawings and the description, there have been disclosed exemplary embodiments of the invention and, although specific terms may have been employed, they are unless otherwise stated used in a generic and descriptive sense only and not for purposes of limitation, the scope of the invention therefore not being so limited. Moreover, the use of the terms first, second, etc. do not denote any order or importance, but rather the terms first, second, etc. are used to distinguish one element from another. Furthermore, the use of the terms a, an, etc. do not denote a limitation of quantity, but rather denote the presence of at least one of the referenced item.

Claims

CLAIMS What is claimed is:

1. A method of monitoring and assessing valve conditions in a pump (10), the method comprising:

collecting data regarding timing of a pump piston (34) and vibration of a pump fluid end (16);

processing the data to form a filtered and transformed vibration signal (170);

banding the vibration signal (170) into high energy bands (174) and low energy bands (176); and

comparing a ratio of at least one high energy band (174) to at least one low energy band (176) of the vibration signal (170) with at least one constant to determine valve condition.

2. The method of claim 1, wherein, if the ratio of the at least one high energy band (174) to the at least one low energy band (176) of the vibration signal (170) is greater than a first constant amongst the at least one constant, determining the valve (58, 60) to be in good condition.

3. The method of claim 2, wherein, if the ratio of the at least one high energy band (174) to the at least one low energy band (176) of the vibration signal (170) is less than the first constant, but greater than a second constant amongst the at least one constant, determining the valve (58, 60) to be worn out.

4. The method of claim 3, wherein, if the ratio of the at least one high energy band (174) to the at least one low energy band (176) of the vibration signal (170) is less than the second constant, determining the valve (58, 60) to be destroyed.

5. The method of claim 1, wherein banding the vibration signal (170) into high energy bands (174) and low energy bands (176) includes banding the vibration signal (170) into harmonic frequency response spectral bands and non-harmonic frequency response spectral bands.

6. The method of claim 1, wherein processing the data includes noise filtering.

7. The method of claim 6, wherein noise filtering includes time synchronous averaging.

8. The method of claim 6, wherein noise filtering includes spectral averaging.

9. The method of claim 6, wherein processing the data further includes passing the signal (170) through a low pass filter.

10. The method of claim 6, wherein processing the data further includes fast Fourier transformation.

11. The method of claim 10, wherein the transformation prepares the signal (170) for banding.

12. The method of claim 1, further comprising configuring a sensor (54) to provide a timing signal of the pump piston (34).

13. The method of claim 1, further comprising arranging a sensor (56) to detect vibration within the fluid end (16) of the pump (10).

14. The method of claim 13, wherein arranging a sensor (56) includes utilizing an accelerometer (56).

15. The method of claim 1, wherein data is collected at more than 10,000 samples per second.

16. A pump assembly (100) comprising:

a pump (10) including a power end (14), a fluid end (16), a piston (34), and a plurality of valves (58, 60);

a timing sensor (54) configured in the power end (14) to sense movement of the piston (34);

an accelerometer (56) in the fluid end (16) to sense vibrations within the fluid end (16); and,

a system (150) configured to collect data from the timing sensor (54) and

accelerometer (56), process the data to provide a filtered and transformed vibration signal (170), and compare a ratio of high energy bands (174) and low energy bands (176) of the vibration signal (170) to at least one constant to assess valve condition.

17. The pump assembly (100) of claim 16, wherein the system (150) includes a user operable display (154) arranged to indicate the valve condition.

18. The pump assembly (100) of claim 16, further comprising a prime mover (12) operatively connected to the power end (14) to move the piston (34), wherein the system (150) sends a signal to the prime mover (12) to halt operation if at least one of the valves (58, 60) requires immediate replacement as determined by the assessed valve condition.

19. The pump assembly (100) of claim 16, wherein the pump (10) is a positive displacement reciprocating pump (10).

20. The pump assembly (100) of claim 16, wherein the valves (58, 60) of the pump (10) include at least one inlet valve (58) and at least one outlet valve (60).

21. A method of monitoring and assessing valve conditions in the pump assembly (100) of claim 16, the method comprising:

collecting data regarding timing of the piston (34) and vibrations in the fluid end (16); processing the data by the system (150) to form a filtered and transformed vibration signal (170);

banding the vibration signal (170) by the system (150) into the high energy bands (174) and the low energy bands (176); and

comparing a ratio of at least one high energy band (174) to at least one low energy band (176) of the vibration signal (170) with the at least one constant to determine valve condition.