mud pump condition monitoring free sample

The 2,200-hp mud pump for offshore applications is a single-acting reciprocating triplex mud pump designed for high fluid flow rates, even at low operating speeds, and with a long stroke design. These features reduce the number of load reversals in critical components and increase the life of fluid end parts.

The pump’s critical components are strategically placed to make maintenance and inspection far easier and safer. The two-piece, quick-release piston rod lets you remove the piston without disturbing the liner, minimizing downtime when you’re replacing fluid parts.

A comprehensive range of mud pumping, mixing, and processing equipment is designed to streamline many essential but time-consuming operational and maintenance procedures, improve operator safety and productivity, and reduce costly system downtime.

If you run a mud rig, you have probably figured out that the mud pump is the heart of the rig. Without it, drilling stops. Keeping your pump in good shape is key to productivity. There are some tricks I have learned over the years to keeping a pump running well.

First, you need a baseline to know how well your pump is doing. When it’s freshly rebuilt, it will be at the top efficiency. An easy way to establish this efficiency is to pump through an orifice at a known rate with a known fluid. When I rig up, I hook my water truck to my pump and pump through my mixing hopper at idle. My hopper has a ½-inch nozzle in it, so at idle I see about 80 psi on the pump when it’s fresh. Since I’m pumping clear water at a known rate, I do this on every job.

As time goes on and I drill more hole, and the pump wears, I start seeing a decrease in my initial pressure — 75, then 70, then 65, etc. This tells me I better order parts. Funny thing is, I don’t usually notice it when drilling. After all, I am running it a lot faster, and it’s hard to tell the difference in a few gallons a minute until it really goes south. This method has saved me quite a bit on parts over the years. When the swabs wear they start to leak. This bypass pushes mud around the swab, against the liners, greatly accelerating wear. By changing the swab at the first sign of bypass, I am able to get at least three sets of swabs before I have to change liners. This saves money.

Before I figured this out, I would sometimes have to run swabs to complete failure. (I was just a hand then, so it wasn’t my rig.) When I tore the pump down to put in swabs, lo-and-behold, the liners were cut so badly that they had to be changed too. That is false economy. Clean mud helps too. A desander will pay for itself in pump parts quicker than you think, and make a better hole to boot. Pump rods and packing last longer if they are washed and lubricated. In the oilfield, we use a petroleum-based lube, but that it not a good idea in the water well business. I generally use water and dish soap. Sometimes it tends to foam too much, so I add a few tablets of an over the counter, anti-gas product, like Di-Gel or Gas-Ex, to cut the foaming.

Maintenance on the gear end of your pump is important, too. Maintenance is WAY cheaper than repair. The first, and most important, thing is clean oil. On a duplex pump, there is a packing gland called an oil-stop on the gear end of the rod. This is often overlooked because the pump pumps just as well with a bad oil-stop. But as soon as the fluid end packing starts leaking, it pumps mud and abrasive sand into the gear end. This is a recipe for disaster. Eventually, all gear ends start knocking. The driller should notice this, and start planning. A lot of times, a driller will change the oil and go to a higher viscosity oil, thinking this will help cushion the knock. Wrong. Most smaller duplex pumps are splash lubricated. Thicker oil does not splash as well, and actually starves the bearings of lubrication and accelerates wear. I use 85W90 in my pumps. A thicker 90W140 weight wears them out a lot quicker. You can improve the “climbing” ability of the oil with an additive, like Lucas, if you want. That seems to help.

Outside the pump, but still an important part of the system, is the pop-off, or pressure relief valve. When you plug the bit, or your brother-in-law closes the discharge valve on a running pump, something has to give. Without a good, tested pop-off, the part that fails will be hard to fix, expensive and probably hurt somebody. Pop-off valve are easily overlooked. If you pump cement through your rig pump, it should be a standard part of the cleanup procedure. Remove the shear pin and wash through the valve. In the old days, these valves were made to use a common nail as the shear pin, but now nails come in so many grades that they are no longer a reliable tool. Rated shear pins are available for this. In no case should you ever run an Allen wrench! They are hardened steel and will hurt somebody or destroy your pump.

One last thing that helps pump maintenance is a good pulsation dampener. It should be close to the pump discharge, properly sized and drained after every job. Bet you never thought of that one. If your pump discharge goes straight to the standpipe, when you finish the job your standpipe is still full of fluid. Eventually the pulsation dampener will water-log and become useless. This is hard on the gear end of the pump. Open a valve that drains it at the end of every job. It’ll make your pump run smoother and longer.

A reciprocating pump has positive displacement, which means, it takes a fixed volume of liquid under suction conditions, compresses it (pressure increase) and it ejects it through the discharge nozzle. In this equipment, pumping the fluid is achieved by the alternating movement of a plunger, piston or diaphragm.

The reciprocating pump is not kinetic as it is the centrifugal one, and it does not require speed to generate pressure as high pressures can be obtained at low speeds. This is one of the advantages of the reciprocating pump, particularly to handle wash and abrasive pulps, and very viscous liquids.

The reason to select a reciprocating pump instead of a centrifugal one or a rotating one must be the cost; not only the initial total cost, but also energy and maintenance costs. Figure 1 shows a typical assembly of this type of pumps.

Another application where the reciprocating pump is practically mandatory is to handle wash and abrasive pulps, or very viscous materials at more than 500 psig. Examples include from carbon wash pulp to peanut butter.

The reciprocating plunger pumps have a pumping element that alternates in and out of the pumping chambers when operating. Each chamber includes, at least, a suction and a discharge valve. These are retention valves that open due to the differential pressure of the fluid and most of them operate under a spring load.

The Fluid End is the part of the pump where the pumping takes place. The components which are common in all of them are the cylinder for the liquid, the pumping element and the valves.

The cylinder for the liquid (case, cylinder liner) is the part that retains the pressure in the fluid end, and it is the most important part of the pumping chamber. It usually includes or supports all the other components of the fluid end.

A plunger is a flat and cylindrical disc installed on a rod, and it usually has some type of sealing rings. A plunger is a plain stick, and it can only be single action in its normal configuration. A plunger must seal against the cylinder or the case inside the pump.

The plunger pump usually has a replaceable case that absorbs the wear of the plunger rings. The sealing between the pumping chamber and the atmosphere is achieved by means of a cable gland, which includes packaging rings that adapt and seal against the inner diameter of the cable gland and the bar or stem.

In this type of pumps, the usual transmission is by belts/pulleys in more than 90% of the applications, while the type of supports is usually roller bearings in the entry shaft where the sprocket is installed (fast shaft), and in the exit shaft where the gearwheel is installed (crankshaft or slow shaft); however, in a few applications flat bearings can also be used.

As we have addressed in a general way in our article Where to Place the Vibration Sensor, in a reciprocating plunger pump, the vibration inspection points must correspond with the entry shaft or fast shaft, where the sprocket is installed, both on the pulley side and the opposite, as well as on the crankshaft, where the gearwheel is installed to measure on both ends. In addition, when possible, measurements must be taken in the horizontal (H), vertical (V) and axial (A) directions of each bearing. In total, it would be 4 measuring points or bearing supports at least (2 on each shaft), where 8 radial measurements would be taken (4V + 4H) and 2 axial measurements (1 on each shaft) at least. See Figure 4.

Safety is the priority when selecting the vibration monitoring points. A lot of care must be taken when placing sensors on both shafts, mainly in the area closest to the pulley with the bigger diameter (which is installed on the entry shaft) of the reciprocating pump, since there is a high entrapment risk. Based on this point, it is recommended to take axial measurements on the area opposite to the pulley.

Verifying the assembly of the pump at the workshop, ensuring the correct interaction between the teeth of the sprocket and the gearwheel (contact width and height between teeth, right backlash).

Checking the lubricant levels, the condition of the lubricant (oil analysis) to determine possible replacement, as well as knowing which are the components rub and wear generation (wear particle analysis).

Internal inspection of the pump during workshop interventions, either for partial or total maintenance, measuring looseness between components and replacing wearing parts.

Deficient attachment between the structural base and the foundation, or between the pump base and the supporting structural base. Loose bolts, deteriorated silentblocks.

Monitoring the condition by means of vibration analysis, ultrasound, among other techniques, in order to extend the life span and determine the right moment to replace the bearings.

Obstruction in the suction pipe due to sediments or damage in the valves; low suction flow rate. Internal overheating in the pump due to a lack of working fluid.

Continuous monitoring of the volumetric flow of the equipment, comparing it against the ideal operating curve; in case there is not a flow meter in the line to measure constantly and automatically, the flow rate periodic measure must be planned and performed by means of portable ultrasound equipment. This action will allow to calculate the efficiency of the pump and plan the replacement of plungers and/or casings of the equipment, repairs of the valves, among other aspects.

Signals of this phenomenon are cracks, fractures of the structures, welds and bolted joints; the pulse generated by the alternative pumping of the fluids can also generate these faults.

Checking the possible existence of resonance in the pump and associated structures; by means of vibration techniques, verifying the natural frequencies (bode diagram, spectral cascades); applying vibration isolation or damping techniques if required.

Gear teeth fractures as a result of excessive working loads in the equipment and the mechanical transmission. This can be the consequence of a high discharge pressure of the working fluid, which exceeds the design conditions and the safety factors, for example.

Inspecting the inner gears of the pump, remove inspection caps from the casing and performing detailed inspections of the teeth, looking for cracks, fractures, abrasive wear, among other failure mechanisms.

Temperature patterns in the suction pipe that may be indicative of the condition regarding a possible obstruction and the level of accumulated sediments to plan the opening and cleaning of the pipe.

ISO 14224 and API 689 to determine the border conditions and analysis limits in pumps, the failure modes, the failure mechanisms and the cause of failure.

The complexity of today’s drilling projects, especially the need for sufficient pressure and flow rate for wellbore cleaning, challenge mud pumps manufacturers. Their efforts focus on the improvement of pump running time and efficient maintenance management to reduce or eliminate nonproductive time and HSE risks. Drilling rigs rely on mud pumps to efficiently circulate the mud, and synchronized pumps are employed to minimize mud pulsation effects. The mud circulation system is of major interest…Expand

The ballasted track currently remains one of the few leading types of railway track structures due to the advantages in construction and maintenance [1,2]. However, the particulate nature of ballast particles often leads to performance degradation of ballasted trackbed. For example, the abrasion and breakage of ballast particles intensify with increasing axle load and train speed, thus causing the unfavorable densification, fouling, and clogging (i.e., reduced drainability) problems in ballasted tracked [3,4,5]; consequently, mud pumps, among other commonly observed track problems, can be prompted within such fouled ballasted trackbed [6,7]. Mud pumps could seriously degrade track stiffness and thus endanger operational safety of railway trains [8,9,10]. Traditional manual inspection and detection of mud pumps and other track problems are often labor-intensive, time-consuming, and subjective in nature; therefore, it becomes indispensable to develop automated, intelligent, and accurate means for the early-age diagnosis and detection of mud-pumping risks in ballasted trackbed so that remedial maintenance measures can be timely taken according to real-time health condition rather than the fixed schedules.

The root cause of mud-pumping fault has remained a widely-studied but challenging topic. Tadatoshi [11] proposed a suction-driven model and showed that the main cause of mud pumps is the intrusion of fine particles from the subgrade generated by the suction of ballast bed during the loading and unloading cycles. Raymond [12] found that the freeze-thaw cycles can cause fine-grained materials to pump out of the ballasted trackbed in winters according to a field performance investigation of the North American railway geotextiles. Duong et al. [13,14] believed that the interlayer materials between the subgrade and the ballasted trackbed were mainly generated by broken ballast particles, which then penetrated into the subgrade surface. The formerly Transportation Technology Center, Inc. (TTCI) established a field-testing zone to further study mud pumps [10,15,16,17]. Despite a considerable number of research studies have been carried out to explore the mechanisms of mud pumping fault, there still lacks radical countermeasures to prevent and control it in railway engineering practices.

The accurate early-age diagnosis and detection of mud pumps are the key step on which timely and effective prevention and control measures depend. The late-stage mud-pumping fault manifested on the surface of ballasted tracked is relatively easy to detect through routine labor-intensive methods; however, it is quite challenging to directly identify the early-age mud-pumping problem initiated inside the thick ballasted trackbed. The ground penetrating radar (GPR) technology has been widely applied in the non-destructive detection of structural faults in railway ballasted trackbed and subgrade [10,18,19,20]. Hugenschmidt [21] successfully applied GPR in the detection of railway subgrade problems for the first time in 1998. Since then, many countries including China have conducted related field and laboratory studies in this field [22,23]. Trong Vinh Duong et al. [13] carried out physical model tests and analyzed the influencing factors of the mud-pumping problem occurring in the interlayer between the ballast bed and underlying subgrade, including particle size distribution, moisture content, pore water pressure, hydraulic conductivity, etc. Kuo et al. [24] developed a characterized grid method and a scoring method to assess the mud-pumping distribution with an accuracy rate of 80%. Although the GPR technology has been reported to successfully detect visible or hidden mud-pumping problems in ballasted railway tracks [21], the accuracy and reliability of different GPR equipment and supporting post-processing software programs still vary considerably, not to mention the fact that they are highly costly and unaffordable for routine applications. In addition to GPR, other techniques have also been widely used for non-destructive detection of railway track foundation problems in recent years, such as the digital image correlation (DIC) [25,26,27], Interferometric Synthetic Aperture Radar (InSAR) [25], impact-echo method (IEM) [28], and synthetic aperture focusing technology (SAFT) [29,30]. However, these methods all require costly equipment and/or highly specialized skills that railway engineering practitioners usually lack. Therefore, to diagnose the in-service health condition and detect invisible problems of the ballasted trackbed accurately and reliably, it becomes imperative to explore automated, intelligent, and universally applicable methods in lieu of traditional ones.

The occurrence of mud pumps could cause uneven (or differential) rail track settlement and increasing track irregularities [31,32,33]. The existence of track irregularities could not only compromise the operational safety of heavy-haul trains but also degrade track substructures [34,35,36]. Li et al. [37,38] proposed a data-driven method for infrastructure deformation identification based on the characteristics of track geometry data, as well as a spatio-temporal identification model for identifying high-speed railway infrastructure deformation by using four years of track geometry data. Li et al. [39] analyzed the time and frequency characteristics of track geometry irregularity signals at the locations of mud pumps and used a multi-scale signal decomposition method to extract defect-sensitive features and then realize automatic detection of mud pumping problems. The nearly continuous and real-time track health monitoring of the entire rail networks could be possibly accomplished in a timely and cost-efficient manner by mounting robust sensors on in-service trains. For example, the problematic sections of railway track sub-structures were reportedly detected by using the vertical acceleration responses of a moving train [40]. Zeng et al. [41] proposed a data-driven approach for identifying mud pumps in the railway track substructure based on vibrational acceleration responses and Long Short-Term Memory (LSTM) artificial neural networks. The vibrational responses of ballastless slab tracks were also compared to detect the locations of mud pumps and study the feasibility of technical countermeasures to rectify and control the mud-pumping damage [42]. Therefore, analyzing the vibrational responses of the ballasted trackbed appears to be potentially helpful and promising for intelligent detection of mud-pumping problems in railway tracks.

Particle movement is a meso-scale manifestation of inter-particle contact behavior of ballast assemblies within the ballast bed; therefore, investigating the meso-scale movement characteristics of ballast particles may emerge as a promising, effective alternative to diagnose and identify the mud-pumping problem of ballasted tracked. The use of motion sensors (termed as “SmartRocks”) has been reported in the literature to directly capture real-time movement of ballast particles and then evaluate the field performance of ballasted trackbed under different in-service conditions [43,44,45,46,47]. The applications of such so-called SmartRock sensors in effective and accurate measurements of the vibrational responses of unbound aggregate particles including railroad ballasts were demonstrated in laboratory scaled model tests and triaxial tests [43,48,49,50]. Liu et al. [51] compared the ballast particle motion data measured by SmartRock sensors against those simulated by the discrete element method (DEM) model. Preliminary studies [52,53] suggested that SmartRock sensors could be used as a potential tool to quantify ballast behavior without using invasive measurement devices or disrupting railroad operations and to reflect the variations of dynamic behavior of ballasted trackbed under different substructure conditions. However, the widespread, reliable field applications of this new smart sensing technology for detecting invisible track defects such as mud pumps within ballasted trackbed remains to be extensively explored.

The purpose of this paper was to further study and substantiate the feasibility of SmartRock sensors in real-world field applications to diagnose and identify mud-pumping risks in ballasted trackbed. Therefore, a typical section of heavy-haul railway ballast bed with severe mud pumping problems was chosen for investigation, where the SmartRock sensors were employed and instrumented accordingly to monitor particle-scale acceleration responses prior to, during, and after major maintenance operations including ballast-cleaning and tamping. The three-dimensional (3D) acceleration responses and associated marginal spectra of ballast particles recorded by SmartRock sensors in different positions were comparatively analyzed for the initial degraded and subsequent rectified scenarios of the ballast bed. The findings are expected to contribute to the optimization of maintenance operation parameters and smart track health monitoring.