A novel smart caliper foam pig for low-cost pipelineinspection—Part A: Design and laboratory characterization

G. Canavese a,b,n, L. Scaltrito b,e, S. Ferrero b,e, C.F. Pirri a,b, M. Cocuzza b,d, M. Pirola c,S. Corbellini c, G. Ghione c, C. Ramella b, F. Verga f, A. Tasso g, A. Di Lullo g

a Center for Space Human Robotics@PoliTo, Istituto Italiano di Tecnologia, Corso Trento 21, 10129 Torino, Italyb Dipartimento di Scienza Applicata e Tecnologia (DISAT), Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italyc Dipartimento di Elettronica e Telecomunicazioni (DET), Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italyd CNR-IMEM, Parco Area delle Scienze 37, 43124 Parma, Italye Microla Optoelectronics, Campus Tecnologico Località Baraggino, 10034 Chivasso, TO, Italyf Dipartimento di Ingegneria dell’Ambiente, del Territorio e delle Infrastrutture (DIATI), Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italyg Eni E&P, via Emilia 1, San Donato Milanese, MI, Italy

a r t i c l e i n f o

Article history:Received 14 April 2014Accepted 9 January 2015

Keywords:Oil pipelinePiggingUtility pigsPipeline integrity

a b s t r a c t

Pipeline pigging for in-line inspection is a fundamental practice in the oil and gas industry. Yet, the so-called “smart pigs” used for this purpose are expensive and delicate and the risk related to their possibleblocking inside the pipeline is non-negligible, hence their deployment is rather infrequent (generally,just once in several years). In this paper, we present a new, low-cost and low-risk foam pig withinspection capabilities similar to those of a multi-channel caliper pig (i.e. able to detect, locate and sizeinner diameter changes and deformations) together with additional features that allow to detect internalroughness changes (e.g. due to corrosion) and perform some pH/salinity determinations, also useful forcorrosion assessment purposes. One implementation of the new tool makes use of a foam pig “carrier”,providing the required push with a good capability to surpass restrictions, equipped with specializedsensors and modules for data acquisition and storage. Another implementation, called “skeleton caliperpig” and suitable to prevent the massive displacement of condensates from gas lines, deploys the lightplastic system without any foam pig carrier, pushed by the gas velocity alone. In the paper we willdiscuss the design, construction and field testing of this new low-risk inspection pig.

& 2015 Elsevier B.V. All rights reserved.

1. Introduction

The oil and gas industry has developed an efficient globaldistribution network for both crude and refined products invol-ving pipeline, tanker, barge, truck and rail transportation. Pipelinesare the most economical transportation method and are mostsuited to movement across long distances, for example, overcontinents (Trench, 2001).

Pipeline infrastructures are vulnerable to several degradationfactors, including construction defects, aging-related issues (corro-sion, creep, cracking, etc. or so on), third-party damage andweather/environment related threats (earthquakes, severe tem-perature conditions, rough seas, etc. or so on). Accidental failure ofoil and gas pipelines, beyond causing significant economic losses,represents a major environmental hazard and a potential threat tolife and must therefore be prevented through effective pipeline

integrity management practices. These include proper pipelinemaintenance and cleaning to avoid obstructions and maintainefficient operative conditions together with periodic non-destructive inspections to assess both internal and external pipe-line status to find possible flaws and damages before they becomecause for concern (Kishawy and Gabbar, 2010; Menon, 2011).

Pipeline inspection gauges (commonly called pigs) play a key rolein pipeline infrastructure management, by both cleaning and carryingout inspection procedures. In the current scenario of pigging proto-cols, the cleaning and inspection purposes are accomplished by twodefinitely different types of pigs: the cleaning pigs and the instru-mented In-line Inspection (ILI) pigs (Kishawy and Gabbar, 2010;Menon, 2011; Tiratsoo, 1999; Quarini and Shire, 2007).

Cleaning pigs are used to remove debris and wax inner pipeaccumulations, and are available in a number of different shapes,materials and densities. Foam or polyurethane pigs are flexibletools and can easily travel through multi-diameter and/or bendedpipelines and are particularly suitable for removing wax (Quariniand Shire, 2007; Hoffmann and Amundsen, 2013). Mandrel pigsare more long-life tools composed of a metal (i.e. or e.g. steel) body

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/petrol

Journal of Petroleum Science and Engineering

http://dx.doi.org/10.1016/j.petrol.2015.01.0080920-4105/& 2015 Elsevier B.V. All rights reserved.

n Corresponding author. Tel.: þ39 110907394; fax: þ39 110907399.E-mail address: giancarlo.canavese@polito.it (G. Canavese).

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎

equipped with easily replaceable elements such as sealing discs,scraper cups, various types of brushes, gauging plates and magnetsto remove any metal object in the pipe (Tiratsoo, 1999; Quariniand Shire, 2007). Due to the relatively high frequency of cleaningruns and the possible harsh conditions of the pipes, cleaning pigsmust be robust and inexpensive to minimize production andoperative costs.

In-line Inspection pigs, instead, are complex instruments, gen-erally heavier than the cleaning ones and primarily made ofmetallic materials, that embed electronics and sensors for datacollection during their run through the pipeline (Rosen Group,2010) and in some cases they may also have an engine (Belleret al., 2006; Zhang and Yan, 2007). ILI technology depends on thekind of the required services. Surface pitting and corrosion, as wellas weld defects are often detected using magnetic flux leakage(MFL) pigs (Kishawy and Gabbar, 2010; Rosen Group, 2010; Afzaland Udpa, 2002). Other tools use electro-magnetic acoustic (ultra-sonic) transducers to detect pipe defects Afzal and Udpa, 2002; Boet al., 2007). Caliper pigs can measure the inner diameter andshape of the pipeline and thus detect deformations (Beller et al.,2006; Zhang and Yan, 2007; Vieira et al., 2008; PipelineInnovations Ltd. (PIL), 2012; Weatherford, 2010).

During the pigging run, the tool is unable to transfer theacquired data to external devices, due to the underground (orunderwater) distance and/or the shielding effect of the pipe itself.For the same reasons, pigs are unable to receive GPS signals, sothey must generally be able to record their own position duringthe trip, using e.g. odometers or gyroscopes, and to detect alimited number of externally installed “markers” signaling specificlocations, in order to locate defects during data processing (RosenGroup, 2010; Beller et al., 2006; Okamoto et al., 1999).

The accuracy of the collected data comes at the price of a moreexpensive and comparatively more fragile nature of ILI pigs, ofsignificant requirements of line preparation and cleaning, of morestringent operating conditions during the pig run and of higherrisks of a stuck-pig event, i.e. when the pig remains blocked in theline without the capability to proceed to the receiving facilities.These factors do not allow for frequent (routine) testing and makeILI a rather rare and special event.

It could then be said that a missing set of tools in piggingprotocols is a halfway device between cleaning and ILI pig, i.e. apig capable of integrating “smart” data collection features within alow-cost, low-impact and possibly disposable support, whichwould carry no big concern in case of blocking or failure. In fact,such an equipped device, in case of blocking, should simply allow

for another pigging attempt without any serious productionimpact or additional cost (Stephenson, 2004). Within the existingtools, mandrel cleaning pigs can use gauging discs to check thepresence of defects but no information on defect location can beinferred. An efficient integration of a low-cost but “smarter”sensing structure on a cleaning pig instead, may allow for a newlow-risk, cost-saving, and possibly cleaning-concurrent pipelineinspection procedure which can be performed more often than,and partially in replacement of, higher-cost inspection protocolsrelying on ILI pigs.

2. System design, materials and manufacturing process

The goal of the activity described in this paper was to designand develop a prototype of such a “low-cost smart pig” for innershape monitoring, generally composed of a common foam pig as avector, equipped with novel sensing and data storage capabilities.A foam pig has been selected for this scope since, and thanks to itsflexible and shrinkable nature, it complies with the easiest andcheapest removing procedures in case of blocking. Moreover, itallows for the realization of a multi-diameter pig, a highlychallenging and demanding tool in the pigging field (Dawson,2008). To fully exploit the flexibility of the foam pig, the smartsystem to be added must be carefully designed by minimizing andproperly sizing the rigid/metallic parts. The different functional-ities of the system (data acquisition and storage, sensing, energyharvesting, etc. or so on) have been conceived and designed with amodular structure to provide high flexibility. The designed mod-ules can be readily assembled and disassembled, allowing for low-cost, easy replacement of damaged parts and possible mountingon different vector pigs types.

This low-risk pig aims at delivering the information typicallyoffered by caliper pigs, and its “geometric” performances will becompared to those devices. To this end, it must be able to detect,size and locate changes in the internal diameter of the pipe.Moreover, the new tool also attempts to assess and record anestimation of the inner surface roughness, which might help tohighlight corrosion phenomena. Moreover, additional functionalmodules, easily integrated into the pig if necessary, have beendesigned and implemented (Fig. 1a): (i) a module for energyharvesting, through a miniature turbine with plastic scaffoldsmade by rapid prototyping and integrated along a coaxial “bypass”tube inside the foam carrier and (ii) a module for chemical fluidanalysis, also integrated into the same tube, based on sensitive

Fig. 1. Design of a smart foam pig with additional energy harvesting and chemical fluid analysis modules integrated along a coaxial “bypass” tube (a) and “skeleton pig” forgas pipelines (b).

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

films deposited on glass slides. Because these modules were testedin the laboratory, separately from the final prototype, they will notbe included in the subsequent description. Finally, in gas pipelineswith a gas speed higher than about 7 m/s, the tool can be deployedalso without the foam pig carrier, directly pushed by the gas, withthe benefit of preventing the displacement of liquid accumulationspresent in the line. This implementation was called “skeleton pig”(Fig. 1b).

Fig. 2 shows the designed inspection tool. The vector foam pighas been equipped with a metallic disc placed at the rear side ofthe pig from which threaded metallic rods are protruding forconnection of the acquisition and sensing modules through simplelocking nuts. In the realized prototype, a 12″ medium-densitypolyurethane pig, coated with urethane bands has been speciallydesigned and manufactured including a 4 mm-thick aluminumdisc inside, a couple of centimeters below the coating, with6 threaded steel tie rods (7 mm in diameter) evenly deployedaround the center, and at a distance of 50 mm from it. Where nospecial manufacturing is available, as in the case of commercialpigs, the anchoring disc can be applied resorting to differentapproaches. Note that this disc is the only solid metallic compo-nent of the proposed structure, all other parts being made ofplastic material and of reduced dimensions and weight, thereforeit has been dimensioned in order to prevent possible obstructionof the pipe.

The data acquisition and storage module (part 4), an ad-hocdeveloped multilayer printed circuit board (PCB), has beenenclosed within a polyoxymethylene (POM or Delrins, Du Pont)disc to maximize its robustness and reliability. It is connected tothe aluminum disc by means of a POM flange (part 3) drilled with6 through holes matching the nuts of the aluminum disc. Outerdiameter of flange and disc together is 140 mm in the prototype. APOM ring spacer has been added in order to improve theassembling flexibility, as it allows for the use of modules withdifferent thickness and/or dimensions.

Polyoxymethylene, a synthetic crystalline polymer (thermo-plastic) made of methylene group-oxygen chains, has been chosenfor its excellent trade-off between performance and cost. In fact, itis a relatively low-cost plastic, with very good tensile strength,stiffness and creep resistance, easy to be machined and withoptimal dimensional stability, as well as exceptional chemicalinertness towards acids, bases and various solvents, even atrelatively high temperatures (Du Pont, 2000; Harper and Petrie,2003). All the POM components have been machined from solid bymeans of a CNC (computer numerical control) drilling machine.

The caliper sensing system (part 5), assembled on the flange bymeans of a threaded hub, is a composite structure (nominal

diameter 5% wider than vector’s) based on harmonic stainlesssteel (AISI 316) stripes buried in polymeric arms (six in therealized tool), able to bend up to 40% from rest position duringthe inspection mission. The bending of the caliper arms is detectedby means of foil strain gauges embedded within the armsthemselves in correspondence of the locations of maximumsupposed strain. This kind of strain gauges, in which conductivepatterns are deposited on a flexible polyimide sheet, provide veryhigh accuracy measurements and are able to work at hightemperatures of up to 180 1C. Within the selected deformationrange the expected relationship between arms bending and straingauges response can be approximated with a linear function. Thepipe diameter variations can thus be obtained by simply multi-plying the acquired signal by the sensitivity coefficient, k. Thiscoefficient depends on the combination of arm geometry, materialflexibility, strain gauges sensitivity, and adopted signal amplifiersgain. The sensitivity coefficient (in the order of 100 LSB/cm) hasbeen chosen in order to achieve a measurement resolution ofabout 10 mm, required to observe the pipe roughness. Fig. 3 reportsthe final realization of the caliper sensing system, together with apicture of the adopted foil strain gauge.

As in the case of commercial, highly expensive caliper pigs(Pipeline Innovations Ltd. (PIL), 2012; Weatherford, 2010), thismulti-channel caliper structure allows for independent arm bend-ing, which is necessary to distinguish the different causes ofdiameter changes via data post-processing, thus discriminatingstructural elements from defects, as well as to determine theo’clock position and extent of asymmetrical dents or bumps. Thepresented choice of sensing structure and materials ensures low-cost and fast production cycle through rapid prototyping methods.The flexible arms have been fabricated by vacuum casting

Fig. 2. Smart Foam pig mechanical design used for the test: (1) foam vector;(2) anchoring disc; (3) connection flange; (4) PCB enclosure; (5) sensing arms.

Fig. 3. Caliper sensing arms (a) and internal strain gauge (b).

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

technique. In particular, a silicone soft mold was used to cast thepolyurethanes resin (PX 223 HT, Axon technologies) together withthe metal stripes. The silicone mold was obtained as negativereplica of a master fabricated ad-hoc by fused deposition modelingtechnique (FDM, Stratasys). Prior to the casting step the straingauge sensors and related wire connectors have been cementedwith a cyanoacrylate glue to the concave surface of each metalstripe and then buried in the polyurethane resin to providemechanical and chemical shielding.

In order to guarantee optimal contact between the caliper armsand the pipe, stainless steel nuts and bolts have been used, whichcan be seen in Fig. 4. This solution allows for easy replacement incase of wear and/or prior to any new mission. In addition, thesmooth and rounded contact surface prevents jamming andrelated risks of “tearing” of the caliper, while still preserving goodsensitivity. Stainless steel is also expected to ensure the necessaryrobustness and chemical stability, at least over the mission time.

The on-board electronics, battery powered, has been limited tothe essential signal conditioning, analog-to-digital (A/D) conver-sion and data storage, while more advanced data processing isexpected to be performed by dedicated software running on anexternal computer. The input signals to be acquired originate fromthe six strain gauges embedded in the caliper arms and from athree-axial SMD accelerometer integrated on the PCB and used todetermine the pig position and orientation along the pipe. Toallow data downloading and post-processing and battery rechar-ging, the board is equipped with a USB (mini-USB/USB) computerinterface protected by an airtight POM plug. For further protection,once definitively enclosed in the POM disc, the PCB has been fullycoated with a specifically selected epoxy resin, in order toguarantee perfect isolation from the external environment. Fig. 4shows the realized PCB: it includes a high-gain wide-dynamic-range asymmetrical bridge for each strain gauge, the acceler-ometer (ADXL330, Analog Devices), a 2 GB mini-SD card memoryslot for data storage, an auxiliary SDRAM for temporary datastorage, a main microcontroller for measurement and systemmanagement, a dedicated microcontroller for USB interface con-trol, a real-time clock and the battery charger circuits (twodifferent batteries are used for analog and digital circuits to

improve immunity). The acquisition scheme and the storageprocedure implemented in the main microcontroller firmwarehave been specifically designed to maximize the sampling fre-quency and simultaneously provide minimum power consump-tion. Considering that common flow velocities during piggingmissions are in the range of 0.2 to 3 m/s (Pipeline InnovationsLtd. (PIL), 2012; Weatherford, 2010), a reasonable minimummission duration of 20 h has been assumed, which enablesinspection of 20 to 100 km of pipelines.

The system automatically alternates two acquisition modes: anormal mode where all inputs are acquired at 1 kSa/s and a fast modewhere only a pair of strain gauges, placed on opposite arms, areacquired at 16 kSa/s, while acquisition from other inputs is sus-pended. The time sharing scheme is approximately 99% normalmode and 1% fast mode per second of acquisition and the arm pairselected for fast acquisition changes clockwise continuously. Thissolution provides periodic windows with measurements of highspectral quality, while still assuring an acceptable battery life. Thesehigher-accuracy measurements provide sampling steps in the orderof tens of micrometers (at the typical pig speed), which are suitablefor fine defects detection and surface roughness estimation.

A magnetic switch allows to completely shut-down the on-board electronics during long-term inactivity, to achieve a “deepsleep” battery life of up to 5 years.

A crucial choice from the assembly standpoint was the elec-trical connections between the contacts of the strain gauges,which emerge from the stop plate of the caliper, and the PCB,since operation under harsh environments and modularity of thestructure must be both ensured. The high pressures and tempera-tures that can be reached in the pipeline make commercial male-female connectors unreliable, therefore direct welding and sealingof the wires has been adopted. Even if this requires longerassembly times, it has the advantage of being extremely low-cost (as no expensive connectors are needed) and space-saving.

The PCB input wires protrude from the epoxy resin and passthrough pin-holes made in the bottom face of the POM disc. Oncethey are welded to the strain gauges contacts, they are tightenedby rotation thanks to the coupling between the threaded flangeand caliper, and kept in place by grooves milled on the back of thePOM disc. The sealing at the interface between the two moduleshas been obtained by suitable o-rings.

For the field test, two identical prototypes of caliper foam pighave been fabricated. Fig. 5 shows both the final assembly and theseparate functional blocks of the two tools.

3. Mechanical characterization results

After its realization, the caliper module has been subjected tomechanical and functional characterization in the laboratory, inorder to evaluate functionality and figure of merits of the designed

Fig. 4. Ad-hoc electronic system.

Fig. 5. Pictures of the two realized prototypes.

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

sensing structure. As anticipated, two prototypes have beenfabricated and tested showing very similar results.

The test set up is shown in Fig. 6: Fatigue structural test andfunctional stability tests have been performed simultaneously bymeans of a fatigue testing machine (Bose ElectroForce 3200)equipped with 225 N range load cell and operating in

displacement control. The caliper module is fasten to the fixedclamp of the testing machine in such a way that the arm under testconnected to the moving clamp results subjected to a periodicbending strain radial-oriented with respect to the caliper axis.

The maximum amplitude (peak to peak) of the applied dis-placement has been varied in the range 0.04 to 12 mm, theminimum (720 μm) corresponding to the expected pipeline sur-face roughness amplitude and the maximum (76 mm) to theexpected maximum restriction (720% of pipeline diameter). Toevaluate the dynamic response of the module under differentconditions, the actuating signal frequency has been varied from1 to 187 Hz, corresponding to 1000 to 5 sample per period.

The graphs reported below show some of the obtained mea-surement results under different conditions. They report, as afunction of time, the displacement signal measured by the calipermodule, in combination with the designed electronics (blue plot)and compare it to the displacement (green plot) and force (redplot) signals measured by the testing machine.

Figs. 7 and 8 show the response to sinusoidal time-varyingdisplacements at different frequencies, respectively 5 Hz (200 pointsper period) and 100 Hz (10 points per period). In both cases thedisplacement measured with the caliper module is clearly synchro-nous with the applied strain and has a very similar shape. Themaximum discrepancy is less than 1% in case of low frequencyvariations, that is well in the range of the measurement error. In caseof high frequency variations, the system dynamic is still able tofollow with reasonable accuracy the variations at fundamentalFig. 6. Laboratory test set up for caliper module mechanical characterization.

Fig. 7. Strain measurement results: 76 mm strain, frequency 5 Hz.

Fig. 8. Strain measurement results: 71 mm strain, frequency 100 Hz.

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

frequency. However, in this case, measurements are affected byhigher order harmonics generated by the non-linear mechanicalresponse of the arm.

The results reported in Fig. 9 refer to a 720 μm high frequency(100 Hz) sinusoidal applied strain (green plot) emulating the expectedsurface roughness of the pipeline. The acquired signal (blue plot) is

rather close to the minimum detectable signal, i.e. of the order of theLSB (Least Significant Bit). However, the sinusoidal trend is still visibleand good results can be obtained by simple signal averaging(black plot).

In order to evaluate the response linearity of the caliper module, alow frequency triangular waveform (12 mm peak-to-peak strain

Fig. 9. Strain measurement results: 720 mm strain, fequency 100 Hz. (For interpretation of the references to color in this figure legend, the reader is referred to the webversion of this article.)

Fig. 10. Strain measurement results: 76 mm strain, triangular wave, period 10 s.

Fig. 11. Strain measurement results: 73 mm strain, frequency 20 Hz, 80 min fatigue test.

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎6

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i

amplitude, 10 s period) has been adopted. The measurement results,reported in Fig. 10, show segments with fairly constant slopeindicating a good linearity of the proposed system.

Finally, to estimate the device durability, one of the caliperarms has been subjected to a long-term, worst-case fatigue testconsisting in the application of 10 Mcycles of bending strain(6 mm peak-to-peak sinusoidal oscillation) at 20 Hz frequency(about 80 min total test duration) together with a pre-compression force of 6 N corresponding to a displacement offsetof 25 mm.

Fig. 11 shows four sections of data acquired in four differenttime intervals during this test. In particular the first and lastintervals correspond to the beginning and end of the test, whilethe second and third ones correspond to respectively 20 and60 min from test-time beginning. The measured displacementsreveal a homogeneous and constant behavior of the sensor undertest, which shows no sign of drift or degradation during the 10Mcycles, together with an excellent mechanical fatigue resistanceof the arm under test. Indeed, during the entire test the measuredsignal does not show remarkable variations meaning that thepossible hysteresis or creeping behaviors related to the polymericpart of the arm are well compensated by the presence of theharmonic steel core.

The results obtained during the laboratory characterizationcampaign can be considered satisfactory and prove that therealized caliper system is suitable for the intended application.An on-the-field test characterization of the two prototypes hasbeen therefore planned and executed, and they will be presentedin Part B of the present paper.

4. Conclusions

In this paper the development, fabrication and testing of a new,low-cost and low-risk plastic caliper pig has been presented. The newtool is able to detect, locate and size inner diameter changes androughness changes thanks to the special strain gauge assembly. Theuse of a foam vector together with the proper design enables the pigto negotiate significant restrictions and to prevent operational pro-blems in case of getting stuck. The low-cost nature of the caliper foampig allows for frequent use, and its modularity provides cheap andeasy replacement of damaged parts. Two prototypes have been

produced and tested in the laboratory, with good results; robustnessand potential capability of application to surface corrosion measure-ment were proved. Field test results, carried out in a 7 km eni pipelinein Italy, will be presented in Part B.

References

Afzal, M., Udpa, S., 2002. Advanced signal processing of magnetic flux leakage dataobtained from seamless gas pipeline. NDT E Int. 35 (7), 449–457.

Beller M., Barbian A., Strack D., 2006. Combined in-line inspection of pipelines formetal loss and cracks. In: Proc. ECNDT, Mo.2.5.4, pp. 1-13.

Dai Bo et al., 2007. An ultrasonic in-line inspection system on crude oil pipelines.In: Proc. Chinese Control Conference, 26–31 July 2007, pp. 199–203.

Dawson,k., 2008. Multi-diameter pigging—factors affecting the design and selectionof pigging tools for multi-diameter pipelines. In: Proc. PPSA Aberdeen Seminar.⟨http://ppsa-online.com/papers/08-Aberdeen/2008-02-PE.pdf⟩.

Du Pont, 2000. Delrins Acetal Resin Datasheet. ⟨http://www.dupont.com/products-and-services/plastics-polymers-resins/thermoplastics/brands/delrin-acetal-resin.html⟩.

Harper, C.A., Petrie, E.M., 2003. Plastics Materials and Processes: A Concise Encyclopedia.Wiley-Interscience.

Hoffmann, R., Amundsen, L., 2013. Influence of wax inhibitor on fluid and depositproperties. J. Petrol. Sci. Eng. 107, 12–17.

Kishawy, H.A., Gabbar, H.A., 2010. Review of pipeline integrity managementpractices. J. Pressure Vessel Technol. 87 (7), 373–380.

Menon, S.E., 2011. Pipeline Planning and Construction Field Manual. Gulf Profes-sional Publishing, Boston.

Okamoto Jr., J., et al., 1999. Autonomous system for oil pipelines inspection.Mechatronics 9 (7), 731–743.

Pipeline Innovations Ltd. (PIL), 2012. PIL Caliper Datasheet. ⟨http://pipeline-innovations.com/downloads/PIL_Caliper_Data_Sheet_010612.pdf⟩.

Quarini, J., Shire, S.A., 2007. Review of fluid-driven pipeline pigs and theirapplications. Proc. IME Part E 221 (1), 1–10.

Rosen Group, 2010. Review of advanced in-line inspection solutions for gaspipelines. In: Proc. PPSA Aberdeen Seminar. ⟨http://ppsa-online.com/papers/10-Aberdeen/2010-07-Rosen-slides.pdf⟩.

Stephenson, N., 2004. Emerging issues of the next 10–20 years. PPSA AberdeenSeminar.

Tiratsoo, J.N.H., 1999. Pipeline Pigging and Inspection Technology. Gulf ProfessionalPublishing, Boston (second ed.).

Trench, C.J., 2001. How pipelines make the oil market work—their networks,operation and regulation. Allegro Energy Group.

Vieira, R., Wakamoto, E., Maruyama, N., Massatoshi, C., 2008. Digital ultrasonicsystem for internal corrosion assessment on oil pipelines. In: Proc. ABCM Symp.Series Mechatronics 3 (2008): 543-551.

Weatherford, 2010. MultiCal™ 360 High-Resolution, In: Multi-Channel Caliper ToolDatasheet. ⟨http://www.weatherford.com/dn/WFT085744⟩. ⟨http://www.weatherford.com/dn/WWW018938⟩.

Zhang, Y., Yan, G., 2007. In-pipe inspection robot with active pipe-diameter adapt-ability and automatic tractive force adjusting. Mech. Mach. Theory 42 (12),1618–1631.

G. Canavese et al. / Journal of Petroleum Science and Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Please cite this article as: Canavese, G., et al., A novel smart caliper foam pig for low-cost pipeline inspection—Part A: Design andlaboratory characterization. J. Petrol. Sci. Eng. (2015), http://dx.doi.org/10.1016/j.petrol.2015.01.008i