POWER SPEED REDUCTION UNITS FOR GENERAL AVIATION PART 5: HOUSING/CASING OPTIMIZED DESIGN FOR PROPELLER-DRIVEN AIRCRAFTS AND HELICOPTERS

Transcription

1 POWER SPEED REDUCTION UNITS FOR GENERAL AVIATION PART 5: HOUSING/CASING OPTIMIZED DESIGN FOR PROPELLER-DRIVEN AIRCRAFTS AND HELICOPTERS Luca Piancastelli 1 an Stefano Cassani 1 Department of Inustrial Engineering, Alma Mater Stuiorum University of Bologna, Viale Risorgimento, Bologna, Italy Multi Projecta, Via Casola Canina, Imola, Italy luca.piancastelli@unibo.it ABSTRACT The purpose of this paper is to focus on the esign of casings for aircrafts an helicopters PSRU (Power Spee Reuction Unit). This paper introuces a rigorous an practical esign proceure for gearboxes. The work starts from the experience of the Authors in Formula 1 an Aircraft gearboxes. For certification, safety an urability reasons, aircraft an helicopter gearboxes i not have the same evelopment rate of the Formula 1 counterparts. A brief history of Formula 1 PRSU/gearboxes forms the first part of this paper. This part inclues also an introuction to material an manufacturing technologies. Then the moal analysis of the gearbox is iscusse, along with the influence of tolerances an operating temperatures. Then cooling is briefly introuce. The gear train is focus of the PSRU. Proper gear meshing in any loa an environmental conition is the main requirement of the PSRU. Unfortunately gears an transmissions are the source of many forcing time-varying forces that act on the housing. This forces not only vary with tolerances, temperatures an loas, but also with wear. Therefore, a comparison of the natural frequency of the housing, the torsional critical spee of rotor system an the flexural critical spees of each of the shafts with the exciting frequency clearly may be use to qualify the gearbox housing. A finite element moelling of the gearbox housing can be carrie out to obtain its natural frequency, stress istribution an force response. Unfortunately, the excitation frequencies vary with tolerances an operating conitions. Furthermore, in aircraft PRSUs, it is common practice to vary the transmission ratio (an the gears) in the same housing. Therefore, the housing shoul ampen a fairly large number of exciting frequencies. This result is obtaine by curve surfaces, ribbing an ouble walling. This approach also reuces the noise prouce by the transmission. In fact, noise raiate by a gearbox is irectly relate to the vibratory level of its housing. Therefore, an aitional aim of this stuy is to analyze the transfer mechanisms between the static transmission error of a gear pair an the ynamic responses of gear an housing of a gearbox. Aerospace an Formula 1 transmissions have many similarities, with Aerospace engineers working on both sies an importing solution. The great avantage of Formula 1 gearboxes was (until the unlucky Regulations of 010) that it was extremely easy to make experiments. This is ue to the fact that all Formula 1 cars are prototypes with test pilots on boar. Therefore, this paper will take avantage of the knowlege achieve in Formula 1 to transfer these ata to aerospace PSRU an transmissions [1-]. Keywors: PSRU, aircraft, helicopter, housing, gear rive, transmission. INTRODUCTION In PSRU (Power Spee Reuction Unit) rive system, gears reuce/increase spee, change the irection of rive, an split/combine torque paths. A gearbox housing has certain functional requirements base on which it was esigne through the years. Historically the gear esign is a typical insie out proceure. Gear transmission is esigne along with the bearings, then the housing is tailore to this assembly. Fixtures, appenixes, accessories, oil tank, cooling, lubrication are also consiere. Aeroynamics may also require a few aaptations. Therefore, the gear housing first esign is base on functional requirements. The moal analysis is then performe. The primary optimization objectives are the natural frequencies of the configuration. It is necessary that they lie sufficiently away from the exciting frequencies that arise from the gearbox; the exciting frequencies are mainly the lowest torsional an flexural of the rotating shafts an the gear meshing frequencies. In this analysis analytical methoology use for bearing ynamic simulation an loa transfer is funamental. However, the first aim of the housing is obviously to house the internals an provie positive locations for the bearings an shafts which are capable of maintaining the shaft centers within prescribe tolerances when running uner full loa. In fact, large contact surfaces between teeth are essential with the high tangential loas an their resultant separating forces. The casings must also be stiff enough to contain istortion. In fact, joint faces have fixing centers capable of coping with the loas involve without oil leaks occurring. Finally, as the complete gearbox esign is available, with the stressing checks of the iniviual components finalize, a manufacturing technique for each part shoul be etaile. An important part of this process involves the ecision on the material to be use for each component. Aerospace an Formula 1 transmission have many similarities, with Aerospace engineers working on both sies an importing solution. The great avantage of Formula 1 gearboxes was (until the unlucky Regulations of 010) that it was possible to make experiments. This is ue to the fact that all Formula 1 cars are experimental prototypes with test pilots on boar. Therefore, this paper will take avantage of the knowlege achieve in Formula 60

2 1 to transfer these ata to aerospace PSRU an transmissions [3-6]. Manufacturing technology Formula one an aerospace transmissions are the most avance in the worl, with an unparallel combination of performance, weight an power ensity. A key part of the PSRU is the casing, as it forms not only a housing for the gearbox internals but also acts as an integral part of the vehicle structure. San cast or forge magnesium alloys have long been a favore material for gearbox casings, thanks to its mechanical properties. In fact, magnesium alloys are about 35% lighter than aluminum alloys, an certain alloys can be heat-treate to UTS (Ultimate Tensile Strength) values up to 300 MPa, making them attractive because of their high strength to weight ratio. Unfortunately, the stiffness of magnesium is generally only about 60% of aluminum alloys. Therefore, casings being switche from aluminum to magnesium will nee larger cross-sections an section mouli to achieve the same stiffness as the aluminum part, but can still achieve a weight saving of about 5%. Of course this fact is true for san cast parts. A comparison of san cast magnesium housings an ie cast aluminum ones shows that the latter are slightly lighter. Therefore, in recent years the emergence of new casting techniques for aluminum has topple magnesium from its position as the material of choice. In the past, the major factor governing the weight of aluminum gearbox casings was the casting wall thickness limite to about 6 mm. Traitional san casting methos i not allow thin wall sections to be cast reliably, thus limiting weight savings irrespective of the actual component strength. However, vacuum assiste an investment-rapi-tooling (lost-wax) casting methos that allow for very thin wall sections are now available. Therefore, engineers can now esign casings that are marginally lighter than a magnesium counterpart, while still retaining the same strength characteristics. As a result of these new casting methos, the benefits of aluminum alloy casings now outweigh those of a magnesium unit with comparable prouction costs. As a point of note, a manufacturer of an aluminum alloy gearbox use formula one racing (year 1996) quotes the weight of the unit as being "in the region of 8kg against the 30kg of the same magnesium alloy unit". The evelopment of new concepts for gearbox housings starte on In that perio san cast magnesium case were premium choices in Formula 1. The longituinal gearbox, with the gears behin the final rive, ha problems of porosity an variable wall thickness, ue to problems of positioning the cores accurately uring the casting process. Therefore, the structurally optimum wall thickness was accoringly increase. The situation was worsene by the magnesium alloy well-known reuction in stiffness above 100 C. This technology resulte in a gearbox case mass slightly below 9 kg. A new housing was manufacture with TIG wele.5mm 300M steel plates. These new gearbox, with were CNC-mille own to 1.mm wall thickness an ribs, prove to have approximately the same final mass of the magnesium alloy case. The avantage of the steel solution was a larger stiffness an a greater strength. The manufacturing of the wele gearbox was extremely critical an time-consuming with accurately timing weling steps to avoi istortion an oil leaking. The passage to titanium alloy wele gearbox (1995) obtaine a reuction in mass of aroun 40% (17kg). It was not until the Minari team began experimenting with a cast titanium unit in 000 that a viable option for titanium alloy casing construction was foun. The use of a rapi tooling casting process, with the extensive use of rapi prototype wax patterns (create using a SLS Selective Laser Sintering process), meant that not only i the prouction cost of the transmission fall consierably, but the entire esign coul be optimize to a greater egree. This is ue to the complex patterns that coul be manufacture through SLS, allowing for internal structures that were not possible with traitional casting methos. Process time is slightly longer however, an cannot compete with traitional casting rates. The cast titanium housing was aroun 0 kg with consierably improve stiffness an reliability when compare with the titanium alloy wele solution. The titanium alloy increase hot stiffness, also meant better gear reliability an greater efficiency. As the gears o not isplace significantly uner loa, they can be groun to truer profiles for greater efficiency an urability. With a completely ifferent approach, in 1995, Ferrari manufacture a titanium gear case bolte to a CFRP bell-housing/oil tank, onto which the suspension units were mounte an onto which acte suspension loas. In 1998 Arrows manufacture a complete CFRP gearbox with bone titanium-alloy transverse bulkheas to carry bearings. The CFRP main case weighe about 9kg. A welcome avantage of the CFRP gearbox was that the oil ran cooler C instea of Figure-1. Mass of F1 gearbox with ifferent manufacturing technologies. the more usual 130 C. This is ue to the insulating properties of the carbon/bismalamie casing that reuce the heat transfer from the extremely hot rear part of the Formula one car to the insie of the housing. 603

3 Figure-1 summarize the gearbox mass with the ifferent manufacturing technology [7-9]. Casing esign The WWII German School use 3D curve surfaces for castings. The DB 605 engine is a goo example of this approach (Figure-). Figure-. 3D surfaces (WWII DB 605). The curve surfaces are a goo way to compromise between manufacturing simplicity an goo overall stiffness an amping. Another classical approach is external ribbing. This approach has two variants: large ribs (Figure-3) or small ribs (Figure-4). The small rib approach often combines the large ribs with the curve surfaces. The ouble walling is typical of liqui coole integral crankcases but is also use in gearboxes to reuce noise emissions an to use two casings of ifferent material bolte together (Figure-5). Usually, the housing is tailore on the transmission with a minimal gap (usually 5mm) to reuce air pumping. A breathing system an the cooling arrangement are ae. Mounting points an accessories are inclue an the resulting shape is reinforce with ribs or ouble walls to obtain the necessary stiffness an natural frequencies. In more etails, the esign of housing continues with the following stages: esign of a single/multiple stages spee reuction transmission; etermination of the forcing (exciting) frequencies; computation of the critical shaft spees of the shafts; static analysis of the housing; etermination of the natural frequency of the configuration to make sure that the exciting frequencies are sufficiently away from the housing natural frequencies. Finally, a simulation of the manufacturing process is performe an the moul is esigne [10-13]. Figure-3. Housing with large ribbing. Lubrication methos The preliminary esign an layout for the gearbox internals inclues the internal gear pack with the location bearings an shafts. At this point the remaining vital internal part of the gearbox is the lubrication system. This system shoul always start from the beginning of the esign, as this one area is traitionally the cause of many problems both at the esign stages an when the gearbox is in use. A lubrication system shoul be effective an efficient at the high spees an heavy gear tooth loaings that can be experience at take-off. The lubrication system can be esigne in varying ways. Most low power (<00HP@6500rpm) manual automotive transmissions an aircraft PRSUs rely on splash lubrication. The problem with splash lubrication is that it is spee epenent. There are centrifugal effects, hyroynamic effects an effects from the gears working as pumps. As the gearbox is run through its revolution range, the oil jets will switch over an move aroun. This is particularly critical for roller an ball bearings with small projections that woul intercept the main oil jet. The main problem of this gearbox is the energy issipate by ipping an the necessity to keep the oil level uner strict tolerances. For this reason a tight visual-an-automatic control of oil level is strictly necessary. A proper shape oil tank that allows significant oil consumption without altering significantly the oil level is also manatory. Another approach is to install a simple recirculating system within a wet sump gearbox. This system has no external fittings, except for the filler plug, the oil breather an the rain. The most common system incorporates an external oil cooler an oil collector tank. Therefore the gearbox has a ry sump, a scavenge an a pressure oil pipe, an oil breather an a rain plug. The oil filler is installe onto the oil tank, which coul also incorporate the oil breather/pressurizer. The lubricating oil is circulate aroun the system by means of a positive isplacement pump that operates with reasonable efficiency. The level of this efficiency is extremely epenent upon the esign of the pump inlet an outlet. Therefore, it is essential to provie a smooth-flowing system along with the pump boy. In a simple recirculating system, the pump retrieves the oil from a sump in the bottom of the gearbox casing. The sump shoul be esigne so that it collects the oil as it rains from the internal running gear to the bottom of the casing. It is also essential that the rotating gears in the gearbox o not ip into the collecte oil, as this woul create heat in the oil with the relevant reuction in efficiency. The oil from the sump is scavenge through a gallery, cast an machine in the gearbox casing. At some point in this gallery, between the sump an the oil pump, a filter shoul be inclue. This is a cylinrical frame covere with fine mesh gauze, with a ro-type magnet in the center of the tube. The oil from the pump is elivere through cast an machine galleries in the gearbox casings. The metho usually aopte to lubricate the bearings is an oil jet machine into one of the rille oil galleries. Oil jets are also use to lubricate an cool the rive gears. Research has shown that, oil spraye onto the 604

4 rotating gears before the point of mesh, creates wear, heat an lowers efficiency. The gear tooth is worn by the wege of oil spraye into the mesh point as the gears rotate, having a hyraulic effect on the tooth surfaces in the meshing zone. The continual spray action as the gears rotate eroes the tooth surfaces. This phenomenon can be easily be confuse with the initiation of pitting an can ultimately lea to tooth failure. Wege spray erosion generates heat, with a resultant loss of efficiency. On the contrary, as the lubricating oil is spraye onto the rotating gears after the point of mesh, tests showe that a film of oil is retaine on the gear tooth face which is thick enough to prevent metal-to-metal contact. The majority of the oil is use as a coolant to remove the heat create by the gear meshing uner loa. A quick removal of heat reuces the amount of wear on the tooth surfaces reuce with no erosion effect. The system that inclues the oil tank an raiator works in a similar way except that it is more complex. In this case the oil pump has two stages, the first stage is a scavenge pump, generally or 3 times larger than the secon stage, which scavenges the mixture oil an air from the gearbox sump, through the filter to an collector, from where it is pipe into the raiator an then into the oil tank. The secon stage of the pump raws the oil from the oil reservoir into the oil gallery system via another external fitting, the gallery system being ientical to that of the recirculating system. The choice of the lubrication system epens mainly on cooling requirements. Vibrations The main source vibratory an acoustical behavior of gearboxes is the Static Transmission Error uner loa (STE). STE comes from the meshing frequency ue to variable elastic eflections of gear teeth meshing uner loa, teeth-gears manufacturing errors an shaft misalignments. Uner operating conitions, STE generates ynamic forces an moments transmitte through bearings to the housing. High ynamic mesh loas may correspon to the excitation of some critical (natural) housing moes amplifying the high potential energy store by the meshing stiffness. Also critical shaft spees can be excite by the time-average meshing stiffness an bearings stiffnesses. Finally, a ispersion of critical spees takes place ue to the variability of STE, resulting from geometry an assembly errors authorize by esigners (tolerances) an operating conitions. In this context, the Taguchi's metho is valuable to estimate in a very simple way the statistical moments of a function of multiple ranom variables of known probability ensity. In this case, statistical moments are estimate from Gauss-quarature numerical integration. Figure-4. Micro-ribbing on the casting of an automotive engine. The Taguchi s methos allow calculating the response function for the samples that are necessary to take into account the eventual nonlinearity of the response function. The number of samples is equal to the prouct of the number of levels chosen for each factor. The basic assumption is that the intercorrelation between variables is low. Usually 3 levels for each factor are neee to treat the three retaine ranom parameters. Therefore, the principal avantage of this metho is the ease of its numerical implementation an its short computing time. Figure-5. Double walle automotive transmission. Probability ensity function cannot be calculate by the moifie Taguchi s metho. However, from the estimation of the influence of the uncertain parameters compute samples is achieve by using variance analysis. The STE is obtaine by a set of successive rotational positions of the riving wheel. The ieal tooth contact lines containe in the action plane are iscretize in a some number of slices. The STE an the meshing stiffness are perioic functions in relation with the cyclic variation of the number of in contact teeth pair. The evaluation of these functions is time-consuming an requires many calculation steps. Therefore, the final equation has the general form (1). K k( t DX E MX CX ) (1) In this equation, X is the vibratory response of the meshe housing noes, M an K are the mass an stiffness matrices provie by the finite element metho with the aition of the bearings stiffness matrices, k(t) is the perioic meshing stiffness, D is a coupling matrix of the two toothe wheels an E is the generalize force 605

5 vector which results from the STE. Matrix C represents amping which is usually very small an can be neglecte in most cases. The vibratory response is then compute with stanar numerical time integration. Generally, K an D are non-linear ue to bearings. In fact the ynamic properties of any rolling bearing varies uring operation, especially the stiffness which mostly epens on rotating spee an loas applie. Therefore, the rolling bearing varying stiffness is moelle by bearing ynamic matrixes. For example, the raial stiffness of single-row ball bearing with raial contact an without raial clearance has the form of equation () K r 1.19z D F () Where K r is the raial stiffness [N/mm], D is ball iameter [mm], F is the external raial loa [N] an z is the number of rolling elements. Equation () is very simplifie since it oes not inclue clearance an rotational spee. For oblique contact ball an roller bearing stiffness epens mainly of rotational spee an play (wear). Play epens also on assembly an assembly tolerances. Therefore, even with the Taguchi metho vibration calculations are complicate. Even natural moes analysis (3) that comes from the eigenvalues of equation (1) is time consuming an shoul be repeate for various spees, an loa conitions. K kd X 0 M (3) Therefore, the esign metho practically use is very simplifie. In this simplifie methos the esigner checks that forcing frequencies are well away from housing natural ones. Forcing frequencies etermination The vibration analysis of housing is base on the fact that only very specific elements in the rotating parts of any machine will prouce forces in the casings that will cause vibration at specific frequencies. One of the most important of the forcing frequencies is the rotational velocity of the shafts, an it arises from the fact that any rotor will always have a certain amount of resiual imbalance. This imparts a raial centripetal force on the bearings, causing the structure to vibrate at the funamental frequency an to its harmonics; usually the secon an fourth orers are significant. Another forcing frequency comes from the bearing tones, which are characteristic of each bearing geometry, are forces generate by efects in the races an rolling elements of the bearing itself. Roller an ball bearings inner race rotating an our race stationary the natural frequencies can be calculate with equations (4), (5), (6) an (7). S B 1 FTF cos (4) P zs B 1 BPFI cos (5) P z B 1 BPFO cos (6) P P S B BSF 1 cos (7) B P Where S is the ifference of rotational spee between inner an outer ring (RPS). FTF is the funamental train frequency. BPFI is the funamental frequency roll frequency pass of the inner ring. BPFO is the funamental frequency roll frequency pass of the outer ring. BSF is rolling element spin frequency. B is the rolling element nominal iameter. P is the pitch iameter an φ is the true contact angle for oblique contact bearing. φ epens on velocity, preloa an loa. All the calculate frequencies [Hz] have a tolerance epening on manufacturing quality. Usually gears are AGMA quality 1 an bearings have tolerances between H/h5 an H/h6. Housing natural frequency evaluation It is then check that the natural frequencies of the housing are away from the forcing ones. In this case errors are common in the FEA (Finite Element Analysis) evaluation. Usually FEA overestimates frequencies. The normal FEA calculation errors [%] for an aluminum alloy ie casting are summarize in Table-1. Table-1. FEA aluminum alloy ie casting errors. Calculate frequency % 100 True frequency % 9 Stanar eviation 5% Table-1 shows that normally true frequencies lie within 97= (9+5) % an 87=(9-5)% of the calculate values. This is ue to the material moel, the calculation metho an the manufacturing technology. It is necessary to avoi that natural housing frequencies lie close to the forcing one. It is also goo rule that the gearbox casing o not have important natural moes in the most important frequency range for the human ear [50-000Hz]. Since noise amplitue epens on the vibrating surface extension, ribs shoul contain this imension. That is a reason for micro-ribbing. This technique is often applie on surfaces that prove to be critical from acoustic measurements. Static evaluation an bearing misalignments Figure-6 shows a atypical arrangement of a large PRSU with titanium alloy micro-ribbe casings. Large 606

6 PRSUs are to be avoie at all costs since imensions increase isplacements with a cubic law. The same happens for the mass. In this case the requirement of propeller shaft position was manatory. Figure-7. Static SF for a PSRU. Figure-7 shows the typical problem that large parts of the housing are unloae (blue color) with SF (Safety Factor) larger than 15. Unfortunately these parts are resonating, so thickness cannot be reuce for ynamic (vibrations) an manufacturing reasons. Misalignment on the propeller shaft bearings are controlle by small isplacements on the bearing supports. Figure-6. PRSU example. Figure-6 shows a PSRU with a 3-shafts arrangement. In this solution the intermeiate shaft hols the propeller governor pump. The lubrication is force an uses the pump of the piston engine. An oil filter an a collector is positione at the bottom of the housing. Oil is taken from the engine sump an the engine pump irects a jet onto the upper gear mesh at exit station. The spur gear solution avois axial loas. Therefore the engine pinion is supporte by the crankshaft journal bearings. The spur gear also compensates the axial isplacement of the crankshaft. The casing is arrange in two halves with a vertical separation (not shown in Figure-6). In orer to save weight the propeller shaft has journal bearings. For this reason a maximum static isplacement (misalignment) of 4μm is allowe between journal bearing centers at full loa. This result is obtaine through a titanium alloy casting. Microgrooves are present in the structure but not moele in this CAD moel that is use for FEA. In the FEA moel the microgrooves are taken into account by moifying the material ata. Walls are 3 mm thick (overall) with ribs of 3mm. The static simulation results are epicte in Figure-7. CONCLUSIONS This paper introuces a practical esign proceure for gearboxes. For certification, safety an urability reasons, aircraft an helicopter gearboxes i not have the same evelopment rate of the Formula 1 counterparts. Aerospace an Formula 1 transmissions have many similarities, with Aerospace engineers working on both sies an importing solution. The great avantage of Formula 1 gearboxes was (until the unlucky Regulations of 010) that it was extremely easy to make experiments. This is ue to the fact that all Formula 1 cars are prototypes with test pilots on boar. Therefore, this paper will take avantage of the knowlege achieve in Formula 1 to transfer these ata to aerospace PSRU an transmissions. Therefore, a brief history of Formula 1 PRSU/gearboxes forms the first part of this paper. This part inclues also an introuction to material an manufacturing technologies. Then cooling is briefly introuce. Unfortunately gears an transmissions are the source of many forcing time-varying forces that act on the housing. This forces not only vary with tolerances, temperatures an loas, but also with wear. Therefore, a comparison of the natural frequency of the housing an the forcing components of the transmission is the most feasible approach. A finite element moelling of the gearbox housing can be carrie out to obtain its natural frequency. FEA (Finite Element Analysis) errors an the variation of the excitation frequencies with tolerances an operating conitions are also iscusse. Furthermore, in aircraft PRSUs, it is common practice to vary the transmission ratio (an the gears) in the same housing. Therefore, the housing shoul ampen a fairly large number of exciting frequencies. This result is obtaine by curve surfaces, ribbing an ouble walling. This approach also reuces the noise prouce by the transmission. In fact, noise raiate by a gearbox is irectly relate to the vibratory level of its housing. Therefore, an aitional aim of this approach is to ampen the frequencies in the most sensible part of the spectrum 607

7 (50-000Hz). Finally, a simulation example is briefly introuce. REFERENCES [1] Wolfer H. Ignition Lag in Diesel Engines VDI- Forchungsheft 39, 1938, (English Translation, RAE Farnborough, Lib. No 359, UDC , Ig5g). [] Marine Propulsion System Reliability Research Base on Fault Tree Analysis, Conglin Dong, Chengqing Yuan, Zhenglin Liu, Xinping Yan, Reliability Engineering Institute, School of Energy an Power Engineering; Key Laboratory of Marine Power Engineering an Technology (Ministry of Communications), Wuhan University of Technology, Wuhan , China. [3] Steven Lewis. A short explanation of the moifications mae in a poor quality ECU remaps. HDI-Tuning Limite [4] L. Piancastelli, L. Frizziero, E. Pezzuti Aircraft iesel engines controlle by fuzzy logic. Asian Research Publishing Network (ARPN). Journal of Engineering an Applie Sciences, ISSN , 9(1): 30-34, EBSCO Publishing, 10 Estes Street, P.O. Box 68, Ipswich, MA 01938, USA. [5] L. Piancastelli, L. Frizziero, E. Morganti, A. Canaparo. 01. Fuzzy control system for aircraft iesel engines, International Journal of Heat an Technology, ISSN , 30(1): [9] Flui Dynamic Drag: Practical Information on Aeroynamic Drag an Hyroynamic Resistance Sighar, F. Hoerner Sighar, F. Hoerner. Flui Dynamic Drag: Practical Information on Aeroynamic Drag an Hyroynamic Resistance. Publishe by the Author, English ISBN: [10] L. Piancastelli, L. Frizziero, I. Rocchi. 01. An innovative metho to spee up the finite element analysis of critical engines components, International Journal of Heat an Technology, ISSN , 30(): [11] L. Piancastelli, L. Frizziero, G. Zanuccoli, N.E. Daizic, I. Rocchi A comparison between CFRP an 195-FSW for aircraft structural esigns. International Journal of Heat an Technology, ISSN , 31(1): [1] L. Piancastelli, L. Frizziero, E. Morganti, E. Pezzuti. 01. Metho for evaluating the urability of aircraft piston engines. Walailak Journal of Science an Technology, ISSN , 9(4): [13] L. Piancastelli, L. Frizziero, G. Zanuccoli, N.E. Daizic, I. Rocchi The "C-Triplex" approach to esign of CFRP transport-category airplane structures. International Journal of Heat an Technology, ISSN , 31(): [6] L. Piancastelli, L. Frizziero an I. Rocchi. 01. Feasible optimum esign of a turbocompoun Diesel Brayton cycle for iesel-turbo-fan aircraft propulsion. International Journal of Heat an Technology. 30(): [7] L. Piancastelli, L. Frizziero, N.E. Daizic, I. Rocchi Analysis of automotive iesel conversions with KERS for future aerospace applications. International Journal of Heat an Technology, ISSN , 31(1). [8] L. Piancastelli, L. Frizziero, E. Pezzuti Kers applications to aerospace iesel propulsion. Asian Research Publishing Network (ARPN). Journal of Engineering an Applie Sciences. ISSN: , 9(5): , EBSCO Publishing, 10 Estes Street, P.O. Box 68, Ipswich, MA 01938, USA. 608