A Critical Review of Propulsion Concepts for Modern Airships

Cent. Eur. J. Eng. 2(2) 2012 189-200DOI: 10.2478/s13531-011-0070-1

Central European Journal of Engineering

A critical review of propulsion concepts for modernairships

Review article

Galina Ilieva1, Jos C. Pscoa1 , Antonio Dumas2 , Michele Trancossi21 University of Beira Interior, CAST-Center for Aerospace Science and Technology, Dep. Electromechanical Engineering,

Cal. Fonte do Lameiro, 6201001 Covilh, Portugal

2 Universit degli Studi di Modena e Reggio Emilia, Dipartimento di Scienze e Metodi dellIngegneria,Via Amendola 2, Pad. Morselli, 52122 Reggio-Emilia, Italy

Received 17 December 2011; accepted 18 February 2012

Abstract: After a few decades in which airships have been depromoted to the level of being only considered as a merecuriosity they seem now to reappear. The main reasons for this are related to the recent progress in technology ofmaterials, aerodynamics, energy and propulsion. Airships are also presenting themselves as green friendly airvehicles, in particular if solar powered airships are considered. Their ability to remain aloft for long time periodshave also expanded the range of mission profiles for which they are suited. Herein we have concentrated on acritical overview of propulsion mechanisms for airships. These include a detailed overview of past, present, andfuture enabling technologies for airship propulsion. Diverse concepts are revisited and the link between the airshipgeometry and flight mechanics is made for diverse propulsion system mechanisms.

Keywords: Airship Buoyancy Lift gas Propulsion Review Versita sp. z o.o.

1. Introduction

Airships have the inherent ability of being able to gener-ate lift without the use of aerodynamic flow around wings,as is the case of heavier-than-air vehicles. This makesthem advantageous in cases where we need to hover atlow cost, thus providing long endurance with low energyconsumption. We must make clear that we are not deal-ing with balloons and aerostats, since these do not haveE-mail: [email protected]: [email protected], corresponding author.E-mail: [email protected]: [email protected].

powered means of propulsion. In Europe the revivalismof airships is also being supported by European Union,recently she provided financing for Project MAAT Multi-body Advanced Airship for Transportation. This project,that comprises 12 research institutions, introduces the con-cept of feeder/cruiser airship for transportation of peopleand goods. Along with this renewed interest on airships,there where a number of authors presenting a state-of-the-art update regarding this technology [1]. In particularfocusing in advances on airship dynamics modeling andstructure design. However, there is still a lack of a similarpublication dealing specifically with propulsion concepts,this is the scope of the present work.It is usually stated that the end of the airship era, inlast half of 30s of past century, was mostly due to the189

A critical review of propulsion concepts for modern airships

Hindenburg disaster, but this is not actually the case. Infact, by that time the heavier-than-air airplane was morecompetitive, both in terms of speed, range and productivity.One of the benchmarks that can be considered is theDC-3, that by that time had significantly lower operatingcosts and higher cruising speeds than the Hindenburg,that was considered the most advanced airship of that time.However, airships still continued to provide limited supportfor military activities, during and after the Second WorldWar. Actually, and for certain type of missions these wherestill preferred to airplanes [2].In [3] the authors perform a review of airship structural re-search throughout the years. They also define the severalclass of conventional airship classification. These includenon-rigid, semi-rigid and rigid airships, with this namingconventions being related to the way in which the envelopestructure is conceived. However, unconventional designshave also been proposed, by including spherical, lenticularand winged airship designs. Also deltoid, dart, flat-body,toroid and multi-balloon airship shapes have been pro-posed in a less extent. The most common non-rigid airshipcomprises a flexible envelope made of fabric that is filledwith a lifting gas. This envelope is slightly pressurizedto maintain its shape. It is also equipped with internalballonets, that are internal air compartments filled with air.These can expand and contract to maintain the pressurein the envelope, when the external atmospheric pressureand temperature are changing as a function of height. Al-ternatively, rigid airships include a rigid structure, usuallymade of aluminium, that can support the weight of thecabin, propulsion system and other equipments. Insidethis structure there are several lifting gas cells along thelongitudinal direction, these can expand and contract inorder to accommodate the changes in external pressure andtemperature. At present, airships are mostly constructedas non-rigid or semi-rigid.Even if, as we have stated, unconventional airship shapeshave been proposed over the years, most of the designshave evolved in the so called oblong familiar shape, havinga circular cross section and an elliptical profile. The bestdesign, from an aerodynamic point of view, resulted infineness ratios of around 6 for rigid and 5 for non-rigid,these later are usually smaller and have lower cruise ve-locities than the rigid ones. This classic airship envelopeshad very well designed shapes regarding aerodynamicperformance. Even with present day experimental, andnumerical, methods it is not to expect too much improve-ment in the external aerodynamics of these airships. Inwhat regards to structures and materials the evolution hasbeen very strong in the last 50 years. Also, substantialefforts have been devoted to the modeling and analysisof airship structures [3]. This resulted in analytical basedtools and also in very detailed numerical computations

based on finite element methods (FEM). The advantages ofthe later being that they are able to perform computationsfor static and dynamic behavior of the structure. Thesecomputations where mostly performed for stratospheric andhigh altitude airships, that where mainly developed withmilitary goals [4, 5].The buoyancy force provides an energy-free form of lift,offering a non-traditional approach, to long-duration mis-sions for which traditional aircraft are not well-suited.With the aim of providing lift, balloons full of lifting gascan be included in the main airship envelope, for whicha change in volume and density with the altitude help tomaintain lift force [6].It is well known that the upward buoyancy force, generatedby an airship, is equal to the weight of the displaced air.This force is typically referred to as the gross lift, and isdefined as a function of the net volume of displaced air,density of air for the given altitude and the accelerationdue to gravity. If we subtract the weight of the liftinggas, the net lift can be obtained. The volume of displacedair is equal to the volume occupied by the lifting gas,so the net lift is referred to the net volume and to thedifference between the air and lift gas densities, takinginto account the earth acceleration. The net lift is availableto counteract the net weight of the airship structure withpayload included. At the launching altitude (assumedas sea level), the air density is at its highest value. Theballonets (full of air) are expanded to their maximum volume,and net volume is minimum. As the airship begins torise, the ambient density and pressure both fall, and airis automatically ejected from the ballonets to match thefalling pressure. During the ascent and at a given altitudethe ballonets will become completely empty. No furtherexpansion of the lifting gas volume is possible. Continuedascent causes a reduction in net lift as the densities fallbut the volume remains constant. The pressure differencealso increases, creating a super pressure condition, whichcan result in rupture of the exterior skin if the internalpressure becomes too high. The lifting gas may be ventedto avoid rupture, but this is extremely undesirable in long-endurance applications as it reduces the available lift andshortens the mission life cycle. It is therefore an importantcriteria, particularly in autonomous operations, to keepthe airship below the pressure altitude. The lifting gasballoons are usually located inside the main envelope but,in some special cases, in order to decrease the mass theycan be installed outside.For generating lift force there is also a new experimental,hybrid type, lighter than-air vehicle [7]. These hybridsinclude a mixture of airship technology and conventionalfixed-wing lifting technology. While these use the principleof buoyancy, they also use a lifting body for additionalpayload capacity.190

G. Ilieva, J.C. Pscoa, A. Dumas, M. Trancossi

Table 1. Characteristic parameters of airships over the years.Name of airship (year) Weight [Kg] Volume [m3] Max. altitude [m] N propellers Propulsion power [kW]Graf Zeppelin (1928) 95 000 1 105 1 829 5 1 677Akron ZRS-4 (1932) 180 000 1 105 1 959 8 3 340USS Macon ZRS-5 (1933) 184 000 1 105 1 959 8 3 340Hindenburg (1937) 255 800 2.48 108 1 829 4 2 535GZ-20A Hybrid (2000) 5 824 7.16 106 3 048 2 313VLLA Hybrid (2008) 212 000 2.1 105 3 848 3 SkyCat 20 58 600 3.8 107 2 745 4 1 788HAA Lockeed (2010) 1.4 104 18 300 4 4 000Aeroscraft ML866 (2010) 20 000 3.9 104 3 657 4 2. Earlier propulsion conceptsMost of classical airships relied on internal combustionengines running on Otto or Diesel cycles. The internalcombustion engines are heavier than turbine engines, evenif they present a lower fuel consumption for small sizedengines. Over the years both engines have evolved andnowadays they present very good reliability, in particularthe turbo-shaft engines. For thrust purposes both pro-pellers and ducted fans have been used, with ducted fanspresenting lower noise, and safer handling for ground crews.Airships, by 1930, have incorporated the best that techno-logy could offer in those times. Hindenburgs four engineswere built by Daimler-Benz, these included 16 cylinderengines with a rated maximum output power of 1320 hp at1650 RPM. The engines were started with compressed air,and could be started, stopped and reversed in flight. Byusing a 2:1 reduction gear each engine was connected to afour blade, fixed pitch, propeller [8]. Most of classic designmethodologies for airships can be found in the Burgessreference work [9]. In Tab.1 we present examples of thecharacteristic parameters of airships, including the datafor the most recently proposed.Besides the classical horizontal axis propeller scheme, alsovertical axis propeller configurations where studied, mainlyby NASA [10]. This specific kind of Heavy Lift Airship(HLA) has a classic oblong elliptical shape and is addition-ally supported by four helicopters that can provide extralift, besides the buoyancy lift. In this way the additionallift enables the system to lift off a very gross weight. Theinclusion of the helicopter propulsion system had the ad-ditional advantage of using a well known technology. Thesystem includes also a stern propeller to provide cruisethrust.2.1. Stern propulsionSince the early years of airship development the use ofa stern-mounted propeller was envisaged as a way to im-prove the propeller efficiency. Most of the initial work that

deals with this problem was supported in theoretical analy-sis or in wind tunnel experiments, see [11, 12]. Both designapproaches concluded that a wake propeller would operatemore efficiently than a conventional mounted system. Inreality, a stern-mounted propeller would have an addi-tional advantage, in particular for long endurance flight,that is related to the reduction of noise, since the system ismounted at a greater distance from living quarters. NASAperformed wind tunnel experimental testing and concludedthat a stern mounted propeller would achieve higher effi-ciencies than when using a conventional assembly or evena fin-mounted-propeller. This is of paramount importancein order to ensure an increase of airship range.Still, the use of stern mounted propellers introduces somepenalties on the airship aerodynamics. Due to the suctioneffect, the flow in the aft side of the airship hull undergoesan additional acceleration. This causes a suction forcein the downstream direction that produces pressure drag.Also, the total skin friction of the envelope is increaseddue to the increase in boundary layer edge, although thiscomponent is residual in comparison to pressure drag,see [13]. It is also to be noticed that this effect can alsointroduce a positive outcome, i.e. using the stern-mountedpropeller to prevent boundary layer separation, only atvery low Reynolds numbers [14]. Also, by working in theairship wake, where the axial velocity is decreased, theamount of power needed for the propeller to produce acertain amount of thrust is reduced, with a net reductionof around ten percent. In parallel the propeller efficiency,based on the non-disturbed onset flow velocity, will beincreased. Yet care should be exercised, by matchingthe stern boundary layer profile, for that the propeller isadequately designed in terms of the local inflow velocity atthe blade, in particular regarding the interference effectsthat the hull introduces on the incoming flow. If theseare not adequately taken into account they can causeseparation into the propeller blade [15].A correlated concept was proposed by Goldschmied in[16]. In this case, the integration of hull design, boundarylayer control, and propulsion can produce power gains in191


excess of 50%, as compared to the best streamlined hullwith stern propeller. These gains are larger when thereis an increase in the angle of attack of the airship andalso for higher Reynolds numbers. In this proposed systemit is envisaged to control the boundary layer, around theairship, by sucking the near wall air that is afterwardsexpelled using a reaction propulsive device at airship rearpart. It should be noted that only a combination of thesetwo actions can achieve positive energy in-balance. Since,in alternative, when the fluid has been sucked into thebody it must be pumped out to free-stream static pressureat the flight velocity, so that it leaves the airship ideallyat zero velocity relative to the body. This usually takesmuch power and it is never worthwhile unless it buys asubstantial reduction in propulsion power per unit bodyvolume. But the proposed system introduces also theconcept of sucked air as propulsive effect at airship rearpart. This kind of system was never built, exception madefor wind tunnel experiments, see [17].Stern mounted propellers are usually not acting alone,in particular because they do not offer the flexibility toprovide wider maneuverability. Also, at very low speeds thecontrol surfaces are considered ineffective, since they ceaseto produce dynamic lift to control the attitude of the airship.Most of the time tilting rotors are preferred as a stand aloneor in combination with stern propellers. One of the mostrecent areas of applicability for airships is related to theiruse in the exploration of atmospheric planetary bodies [18].One of the projects that was developed is called AURORA(Autonomous Unmanned Remote Monitoring Airship). Thevehicle is equipped with two vectorizing engines, on thesides of the payload pod, and has four control surfaces atthe stern, arranged in an X configuration, see also [19].3. Solar powered airshipsThe use of solar powered airships is foreseen as easier toachieve than for heavier-than-air aircraft, since there isno need to resort to propulsion power to create lift. Oneof the blocking points related to heavier-than-air solarpowered aircrafts is related to the greater mass/powerratio of electrical powered thrusters in comparison withgas-turbines, the later presenting a more favorable 1/3mass/power ratio [20, 21]. In this way, by using buoyancylift in airships, the power budget is strongly reduced andfeasibility is easily attained.3.1. Some worldwide on-going projectsSeveral countries have decided to investigate solar pow-ered airships. In Russia, the HAA Berkut solar poweredairship was designed with three different envelopes of di-

(a)

(b)Figure 1. External auxiliary balloons for HAA Berkut airship: at sea

level (a) and at pressure altitude (b) [22]. Reproducedunder permission of AUGUR-RosAeroSystems.

verse length and volume, each targeted to a specific missionprofile, but having the same maximum diameter, see fig. 1.An alternative innovative concept is proposed for the AT-LANT airship, [22]. It is claimed that it involves a combi-nation of the best features of dirigible, aircraft, helicopterand vehicle on air-cushion. This concept uses the positivefeatures of the dirigibles wide range of traveling, and of-fering the ability to carry high loads, coupled to economicefficiency and being ecological friendly.AEREON was proposed as a tri-lobed balloon that gen-erates a large degree of dynamic lift [23]. For efficientflight this includes the idea of using a gas-filled aerofoilcomprising 3 streamlined rigid hulls connected in parallelby a structure of airfoil section. The helium was to beheated by propane burners under each gas cell so that itcould fly without releasing gas or ballast.In [24] it is proposed an airship using the geostrophic flightconcept. The planning of the trajectory for the airship isproposed to be made using two alternative possibilities,depending on how are taken into account the air variation oftemperature and pressure, and also relying on the behaviorof the lifting gas over space and time. The two consideredpossibilities are: a) the airship volume of lifting gas isvaried, by inflating or deflating the ballonets, so thatthe pressure of the lifting gas is the same as that of thesurrounding air at any time. In this case there is nopressure differential across the envelope. The ballonetsmust be large enough to accommodate the required changesin volume, so that the net static lift remains constant overthe trajectory. b) the volume of lifting gas remains constantduring the flight. As a result, the density of the lifting192


gas remains constant, so there is a pressure differentialacross the envelope of the airship. It is assumed that theenvelope is able to withstand the pressure difference, sincethe net static lift changes over the trajectory.The two previous referred methods are the most widely usedto generate lift. However, there have been also proposalsrelated to use of steam as a lifting gas [25]. It is claimedthat the main advantages are that steam is extremely cheapand easy to produce in the field. For control purposesit can be vented freely as required, and can be replacedduring flight by boiling stored water ballast. A minorbenefit is that icing upon the envelope ceases to be aproblem, and abundant heat is available for de-icing thecontrol surfaces. In a steam supported airship it would beeasy to vary the volume of lifting gas and its correspondinglift.The disadvantages of a steam airship, as compared with anhydrogen or helium airship, are that the lift specific poweris lower due to his higher density. Also, the condensingwater from the lift producer steam needs to be continu-ally re-boiled. However, the waste heat of the internalcombustion engine used to provide propulsion could beused for generating steam for the envelope. Yet, it is to beexpected that this concept is only feasible for very largecraft. However, the main drawback of this approach isrelated to the not so efficient power-to-weight ratio. Thisidea of hot air, or thermal airship, was also considered fora blimp-like shape, with fins and propellers.Besides airships based on lifting water steam there wherealso proposals to use heated helium. In this concept theidea is to use it in the inner envelop to augment and/orcontrol the gross lift [26].3.2. Propulsion systems for solar powered air-shipsLighter-than-air (LTA) vehicles represent a unique andpromising platform for many applications that involve along-duration airborne presence. Because they achievetheir lift through buoyancy, these vehicles require muchless power than traditional aircraft. This opens the doorto renewable sources of energy, which would enable con-trolled flight to be maintained almost indefinitely. TheSun is the main power source for the airship and solararrays are the means of converting this incoming energyinto an usable form [27].As an example we can cite a balloon made of Mylarscrim material, whose concept is that of an airship whosegondola is suspended from a classic balloon, [28] thegondola and its connected propulsion system are freeswiveling, permitting rotation with respect to the balloon.Therefore, the propulsion device can be oriented into the

relative wind. Control surfaces connected to the propulsionsystem enforce an orientation towards the wind stream.The balloon altitude control is made by venting gas and/ordropping ballast.Nevertheless, for solar airships, the electric power sub-system is still a significant part of the total weight of theoverall propulsive system. It is therefore highly desirableto minimize propulsion power requirements, among otherreasons to reduce battery and solar cell weights. It is alsodesirable to use low disk loading propellers (i.e. largediameter, slowly rotating propellers similar to hoveringhelicopter rotors) that provide better efficiency throughthe minimization of the power required to develop a giventhrust. Usually we can devise, in a preliminary assessment,the thrust requirements for classical shaped balloons onthe basis of known experimental drag coefficients obtainedfrom tabulated data.Another kind of solar powered airships are based on anon-rigid inflating oblong elliptic shape. The more classicelliptic shape is in this case only visible when it is fullyinflated, at cruise altitude. At lower altitude it is deflatedand only a spherical portion of the nose is identified. Thepropulsion system comprises, in some concepts, a stern-mounted propeller coupled to an electric motor.Inflation and launching of this airship is accomplished ina manner analogous to the launch of a classical balloon.During vertical ascent, the center of buoyancy is abovethe center of gravity providing a stable arrangement. Inthe flight altitude, the center of gravity is also locatedbelow the center of buoyancy. This location of the centerof gravity causes the vehicle to rotate from the vertical tothe horizontal position as the float altitude is approached.This movement is due to the center of buoyancy beingshifted from the nose to the center of the main axis of theairship, this happens when the airship achieves the cruisealtitude. All this can be clearly seen in Fig. 2.Whatever may be the solar powered airship concept, itneeds to operate for extended periods of time and performtasks throughout the day and in night time periods. So,the power system must be capable of providing sufficientenergy to meet all these needs. This means that enoughenergy must be collected during daylight hours to meetthe power requirements of the complete day and nightperiod. Since the weight is one of the most importantfactors affecting airship power requirements then, insteadof carrying energy in batteries, other sources of renewableenergy can be used. Thus, solar photovoltaic arrays forenergy production can be used coupled to regenerative fuelcell energy storage systems, [27, 29]. A detailed parametriccomputation regarding power needs, and power harvestingfor Sun can be seen in [3032].Accordingly, a permanent supply of electric power for theairship needs must be guaranteed. The electric power193


(a)

(b)Figure 2. Unconventional non-rigid inflatable airship: a) Ascent se-

quence after launch without being fully inflated; b)Fullyinflated airship at cruise altitude after ascent.

is needed not only for propulsion but also for other com-ponents, such as rudder machines, ballonet fan, ballonetvalves, payload electric equipment, and also for the controlof the airship.However, due to the variable angle of incidence betweenthe Sun radiation and the solar cells during flight, thesolar power supply varies permanently and can even bezero. Therefore, the power supply has to be buffered, as inthe case of LOTTE airship, with flight accumulators servingas an intermediate reservoir of energy [33, 34].The airship power systems should be constructed in de-centralized and redundant fashion, as in the power systemof the P500 [35]. In this case there are two power systems a smaller power system and a single larger one. Yet, moreredundancy can be introduced by using distributed powersystems connected through a power ring or net.These systems can thus function as back up of each other.The engines and control systems will continue to functioneven in the event of loss of one or more (in case of severalpropulsion systems) of the independent power modules.This allows the airship to be under control even in theevent of a significant failure.Thus, as we have discussed, most of the power drives usednowadays are of electrical kind, obtaining energy fromsolar cells or from batteries. In early days, piston enginesdriving propellers were used but this means that fuel isneeded it was not an ecological and economical approach[36]. The use of liquid fuel introduces an additional problem,that is related to the buoyancy compensation as the weight

of liquid is reduced on course of traveling, since this mustbe compensated by adding a ballast load. In order toeliminate this problem gas bags of inflammable fuel whereproposed and used to sustain power needs of airships.Nowadays, for high-altitude airship propulsion, electricmotors which exhibit an higher torque-to-weight ratio thanpiston engines are the almost unique way to provide shaftpower.Let us now consider the following: an airship hull in for-ward motion is normally unstable. It need to be feededwith a continuous provision of corrections from deviationsof a fixed direction of flight. This is the purpose of fins orof vector control thrusters. In most of classical airships theconventional control technology is based on fixed fins andmovable control surfaces, attached to the hull, in order toaccomplish flight direction corrections. One main disad-vantage of such control elements is their relatively highweight and also their drag in calm weather. Alternatively,vector control thrusters are located in the hull at forwardand aft stations to replace these fins and surfaces. In orderto provide control moments, two thrusters are normallyactivated, one in the bow and one in the stern.The incorporation of vector thrusters into the design ofairships serves two particular purposes. The first is toprovide direction controls in an efficient and effective way,that can completely replace the fins and movable parts ofthe conventional airships. Since we already noticed thatthese later ones are inefficient at low speeds, as well as toeliminate the weight and drag of such control surfaces. Thesecond is to provide an increase in static lift whenever itis required, if they are operated with a vertical component.These lift thrusters are capable of long periods of operationand are driven by electric motors which may be energizedfrom a central or distributed power plant. These systemsare not limited to be used with electric power, since theycan also be used with gas turbine power systems.The use of this type of propulsion can be seen per exampleon the airships Polar 6000 and Polar 3000 [37]. Polar6000is powered by twin 300 hp vectorized direct-drive LycomingIO-540 engines (STC SS05290AT). The low-speed maneu-verability is enhanced by the Polar-family of hydraulicpositioning lateral thruster system, which increase theloiter stability and reduce the ground crew requirements.Polar 3000 is powered by 3 300 hp Ford Diesel enginesand 2 1, 000 hp vectorable PT6 turboprops.3.2.1. Propellers for solar airshipsPropellers are in wide spread use as a thrust devices forairships and aircrafts, they present very high efficiencyat low to moderate velocities. They can present them-selfs with a fixed or controllable pitch. The fixed-pitchpropeller is used when low weight, simplicity, and lowcost are needed. There are mainly two types of fixed-pitch194


propellers: climb and cruise. The climb propeller has alower pitch, therefore it introduces less drag. This lessdrag results in higher RPM and more horsepower capa-bility, this helps to increase performance during take-offand climb, that is why they are called climb propellers.However, they present lower performance during cruiseflight, the opposite holds true for cruise propellers.The first practical controllable pitch propeller for aircraftwas introduced in 1932 [38]. These propellers are adaptableto different thrust values and air speeds as a function ofthe altitude, so that the propeller blades dont stall, hencedegrading the propulsion system efficiency.The reduction of weight of the propulsion system is impor-tant when we are in the design process of a stratosphericairship, a particular attention to this topic must be payed.It is also very important to increase the efficiency of thesystem, since the available energy generated by solar cellsdistributed over the hull is limited. One solution to increasethe efficiency of the propulsion system is to use a sternpropeller, as its was already mentioned in Section 2.1.A new concept for propeller efficiency increase was alsoproposed in [39], the easy-to-mount propulsion systemconcept. This author has shown that the total weightof the propellers is proportional to N12 , where N is thenumber of propellers. Therefore, the greater the number ofpropellers the lighter their total weight. The design of thisDistributed Multi-Propulsion Units System is performedby considering the benefits of boundary layer flow in theincrease of efficiency of the distributed propellers.The system comprises a certain number of small propulsionunits, each of which consists of a small propeller, a smallmotor and a small battery. They are all distributed on thesurface of the hull. The main advantages of this systemare: a reduction of total weight and size of propulsionunits and of propellers and also a reduction on energylosses caused by Joule effect in power lines.There is also a possibility that performance improvementsbe based on propeller aerodynamics. In [40] an increaseon propeller efficiency was sought by using a relativelythin airfoil section for the blades, but this concept is onlyacceptable if the stall performance of the blade is not verymuch compromised. For the proposed, and designed, profileit is stated that to obtain high efficiency the propeller tipMach number should be no higher than 0.950.97. Also, toreduce the structure weight, the chord length is properlyreduced to form a pointed-tip, sharp root and large pressureside shape. A propeller having similar shape must beredesigned to achieve diverse thrust needs.In the AEREON III airship thrust needs were coveredby a 6.4 m in diameter, 2 blades, helicopter rotor, thatis being used as propeller and is powered by an 80 hpSolar Gas turbine [23]. The use of propellers with such ahigh diameters, however, is not desirable because of the

limitations in Mach number at the tip radius, as previouslystated. Their use can be made at the cost of a reductionin performance.Propellers for Vertical Take-off and Landing (VTOL), andcruise operation, are also used in some other projects [37].Among those we can cite the Titan airship project, thataims to use an airship to explore the Titan moon of Saturn.An alternative propulsion unit is proposed in [41]. It isa propeller mounted in a duct with a row of ports inthe wall of the duct, the ports are put downwards of thepropeller. These side ports allow an air flow in the directiontransverse to the axis of the duct. The duct has a closurethat stops outflow from the rear of the duct, and there areadditional regulated closure valves to control flow througheach port. The selective operation of each of the portsprovides radial thrust, allowing a control in the directionor altitude of the airship.For reducing propeller blade tip losses and to help indirecting its thrust, the ducted fan is more efficient thana conventional propeller, especially at higher rotationalspeeds, [42]. By appropriately sizing the duct, the airvelocity can be adapted through the fan and allowed tooperate more efficiently at higher air speeds than a pro-peller would. For the same static thrust, a ducted fanhas a smaller diameter than a free propeller. Ducted fansare also noiseless since they reduce the tip speed andintensity of the tip vortices, that both contribute to noiseproduction. They also allow for a limited amount of thrustvectoring, something normal propellers are not well suitedfor. This allows them to be used instead of the tilt-rotorsin some applications. However, very small clearances arerequired between the blade tips and the duct. The designis complex due to the required high RPM and minimalvibrations, [43]. Yet, they offer an additional advantagerelated to enhanced safety on the ground.3.2.2. Hybrid propulsion systems for solar powered air-shipsAn airship with hybrid propulsion incorporates the staticlift, from the lifting gas, along with some form of dynamiclift, such as helicopter-style rotors or airplane-like wings.A critical issue on airship flight mechanics is related onhow to compensate the sudden increase in the airshipsstatic lift when heavy payload is unloaded. The moststraightforward approach is to add into the airship theamount of weight which equals the payload that is removed.Another alternative is to use a slightly negative buoyancylift, and so the airship is much less buoyant during cargounloading, reducing instabilities. In this case, to achievetake-off, an additional dynamic lift must be provided inorder to compensate the added payload weight. Severalexamples can be found in the literature related to thatconcept.195


The Lockheeds Skunk Works task force took its hybridairship P-791 to first test flight in 2006 [44]. The airshiphas two propulsion motors on the exterior of its envelopeand another two attached to its tail. These generate about20 percent of the dynamic lift when the vehicle is flyingforward, the remaining dynamic lift is provided by the staticlifting surfaces.The hybrid airships present however some technical chal-lenges. For one, the addition of dynamic lift increasesaerodynamic drag. Second, to help with generation ofdynamic lift, they also typically have a flatter profile thanconventional airships, but this shape gives them a higherratio of envelope area to gas volume, increasing the air-ships empty weight. Higher weight and drag, of course,mean that more propulsive power and more fuel/energyare needed, both of which make the ship more heavier.Some hybrids employ multiple lobes in their design, whichcan create problems as the gases inside heat up from theSuns rays. Helium conducts heat six times more efficientlyas air, so a multi-lobed hybrid may tend to be inclinedtowards the side that is not exposed to the sun. Perhapsthe biggest issue, though, is thus the hybrids potentialto pitch nose up or down and to roll from side to side.Conventional airships, of single lobe, avoid this problembecause the majority of its gas volume is positioned wellabove its center of gravity, imparting whats known aspendulum stability. The higher up the center of lift is, themore stable the airship is; conversely, the closer the centerof lift is to the center of gravity, the greater the tendencyof pitching from wind gusts.To overcome all the previously referred problems, Boeingand the Canadian company SkyHook International haveproposed a different approach: the rotary-airship hybrid,[44]. It combines a conventional ellipsoidal envelope withfour powerful helicopter rotor units, which are installedbelow the helium envelope. The helium is sufficient tosupport the weight of the vehicle itself, leaving the fullpower of the rotors to lift a 36-metric-ton payload. One ofthe first applications of the SkyHook is moving equipmentand supplies for oil-drilling operations in northern Canada.When the airship is airborne a propulsion system is usedto control vehicle attitude and provide thrust.The propulsion system of this airship comprises a numberof maneuver thrusters mounted outside of the envelope,[39]. Each maneuver thruster includes a propeller, whichis powered by a motor, and a cowl which encases andprotects the propeller. In order to balance the pressureinside the envelope, while in flight, air may be periodicallypumped into and vented out of the ballonets to keep thevehicle neutrally buoyant, in response to pressure andaltitude changes. To descend, the ballonets are filled withair via the ballonet fans to increase the density of thevehicle. During descent, the ambient air pressure once

again increases and additional air may be blown intothe ballonets, thus providing the required pressure withinthe envelope. As the airship increases or decreases inaltitude, it is important to maintain balance between theambient air pressure and the pressure of the lighter-than-air gas inside the envelope. If the appropriate air pressurewithin the envelope is not maintained, a catastrophe canresult. For instance, if the vehicle experiences some typeof mechanical failure and begins to descend rapidly, theambient air pressure will increase too quickly causingthe envelope to collapse, which may result in structuraldamage. If this occurs, the airship will essentially fall fromthe sky.There are many specific requirements which address thistype of emergency descent situation. In order to maintainthe structural integrity of the envelope in this rare situa-tion, it is mandatory that every airship or hybrid aircrafthas the ability to pump air into the ballonets at an higherflow rate. In an effort to meet this requirement, currentdesigns for airships or hybrid aircrafts include a numberof emergency ballonet fans (in addition to the ballonetfans that operate under normal conditions). These arecapable of pumping air into the ballonets very quickly.Thus, the emergency ballonet fans serve the sole purposeof rapidly inflating the ballonets in the event of an emer-gency descent situation, and are several times larger andheavier than the ballonet fans needed for normal opera-tion. The need to have these additional ballonet fans foruse only in an emergency is costly, and also results inan increase in the total weight of the vehicle. Therefore,it would be useful to provide a propulsion system thatcontrols vehicle attitude and provides forward thrust undernormal conditions, and is capable of performing the func-tion of emergency ballonet fans in an emergency descentsituation. Alternatively dynamic lift can be used to solvethese emergency situations. This can be seen in the hybridZPG-2 (non-rigid hybrid, semi-buoyant, airship), that ispowered with four 5,500 hp Rolls Royce Tyne Turbo-Propengines [23]. The engines were mounted along the rearwing edge of this delta shaped airship. The helium liftinggas would be carried internally in large individual cells,and the design was intended to operate just like a slightlyheavier-than-air airship.As previously discussed airships can operate in verticaltake-off and landing conditions (VTOL). Powerful enginesthat use thrust vectoring technology for change in directionare included on most current airships, in order to overcomeboth low speed maneuverability and changes in incomingwind flow. VTOL is realized mainly by vectorized thrusters[45]. Since airships can be very large, commercial airshipsrequire relatively large storage facilities and open fieldareas for both take off and landing. However, they haveless requirements than traditional aircrafts.196


In most cases, as we have mentioned, more than two en-gines driving propellers, or ducted fans, are located onthe hull in order to provide thrust for takeoff (VTOL mode)and forward airspeed (cruise mode), and also to providecontrollability to the airship.The Very Large Luxury Airship project uses a stern engine[46]. This drives two propellers to provide yaw and pitchcontrol at slow forward speed. The rear engine configura-tion includes one fixed propeller, so that the plane of theblades is vertical and parallel to the longitudinal axis ofthe airship, to provide yaw control. A separately controlpropeller is mounted on the rear engine assembly, so thatthe propeller plane axis can pivot 90 degrees. This engineprovides pitch control for the airship at low airspeed withthe plane axis pointed downward, and thrust in forwardflight with the plane axis pointed aft.4. Unconventional propulsion in air-shipsAmong one of the less conventional means of propulsion,for airships, we can cite the cycloidal rotor. This consistsof several blades rotating around an horizontal axis, per-pendicular to the direction of normal flight. The angle ofthe individual blades, to the tangent of the circle of theblades path, is varied by a pitch control mechanism. Thisis designed so that the periodic oscillation of the blades,about their span axis, may be changed both in amplitudeand in phase angle. The net force vector acting on theaxis of the cycloidal rotor may be varied in magnitude anddirection by the movement of the eccentricity point thatcontrols the blades pitch.The special performance characteristics of cycloidal rotorscan be seen in Fig. 3. The blade on the upper and lowerpositions produces upward force with a certain angle ofattack, but the blades on the left and right positions makea very small amount of force, because the blade has verysmall angle of attack [4749]. This is only an example ofone of the possible working positions.The cycloid propeller can provide 360 of vector thrust ina very easy way by simply moving the eccentric positionthat controls the blade pitch. So, it can provide thrustforce in any direction and also to help increase hovercapability of the airship. Thus, the vehicle has very goodmaneuverability and can obtain very fast response on thepilot control command. There is no strong blade tip vortex,since all sections of the cycloidal propeller travel at thesame speed. Therefore, the noise of cycloidal propellerswill be much lower than for screw propellers. Cycloidalpropulsion was proposed for VTOL Vertical Airships [50].A nuclear-powered rigid airship of 298.7 m length and52.4 m in diameter, with a volume of around 353 960 m3

(a)

(b)Figure 3. Cycloidal rotor for propulsion: a) section of the cycloidal

rotor; b) 3D view of the rotor.

of helium, was also proposed and designed [23]. This hasbeen a topic of controversy due to security reasons, albeitproposed and discussed in many occasions now and in thepast.Another alternative concept for airship propulsion is relatedto the so called fish-like movement in air, propelling anairship by undulating its hull and a caudal fin, see Fig. 4.The option of this fish-like body in air is realized bydielectric elastomers [51, 52].The bending of the fish-like body increases the velocitydrastically and introduces a relevant contribution to thepropulsion. Besides the aft hinge, another hinge or bendingpoint is applied towards the head (or fore position) of theairship, that might decrease the drag further and leadto a better efficiency. An active fin ray, with suitable197


Figure 4. Operation of a fish-like airship using dielectric elastomer actuators.

stiffness, may improve the thrust drastically. But due to theadditional weight, and to the large horizontal distance fromthe center of lift, this option may be difficult to implement.In particular because it can compromise the stability ofthe airship. Additional active fins would lead to bettermaneuverability and may improve propulsive efficiency.Another alternative propulsion system is based on air-jets[53]. This system was recently developed at UniversityModena Reggio-Emilia and is based on a novel control-lable Coanda effect nozzle which has been designed tooptimize the dynamic angular deflection of a synthetic jetformed by two or more primitive jets.5. ConclusionsAirships are nowadays regaining visibility due to theirfriendly relation to green air transportation systems, inparticular due to their use of static buoyancy lift insteadof the large energy consuming dynamic lift. In this contextEuropean Union have recently funded a cooperative re-search project called MAAT Multibody Advanced Airshipfor Transport. The main is idea to use the cruise/feederconcept to allow for the transport of people and goods.For this novel airship concept new envelope designs andpropulsion concepts are envisaged.In the framework of airship propulsion systems there hasbeen a large diversity of concepts proposed along theyears, as was evident from the work herein presented. Oneprimary conclusion is that propulsion is strongly relatedto power availability. Also, the energy source can be ofdiverse nature, from gas turbines to solar power.In present times the most promising concept for energyproduction is that of solar powered airships. This presentsthe advantage of being ecologically friendly, and energy iseasily transmitted by means of electric cabling to the sev-eral places where power is needed. This can be then feedto electric drives, using either open or ducted propellersor even more unconventional propulsion mechanisms, suchas air-jets or cycloidal thrusters.It is nowadays possible to obtain very efficient thrusters towork at, or near, sea level conditions. If one decides to useonly a large propeller, then the efficiency at high altitude

would not be so difficult to achieve. The problem is thatis is required to consider distributed propellers havinglower dimensions, working at low-Re and high tip Machnumbers, implies that the design of thrusters for these HighAltitude Airships is still under development. Yet, propellersand ducted fans have been designed, showing reasonableefficiencies, in particular for applications related to high-altitude UAVs.However, we propose that alternative propulsion mecha-nisms be investigated, capable of achieving better efficiencyand dynamic performance gains. This alternative thrusters,however, have not reach yet the level of development ofclassic thrusters, so we are not able to say that they willbehave better than conventional ones. But, it would bemost important to develop a program of research on thisnovel propulsion systems. In particular because they canbe more adapted to be used as small, distributed sys-tems. Since we have demonstrated that these distributedpropulsion systems reduce the weight/power ration.Whatever may be the propulsion concept it must be fitted,optimally, to the airship envelope shape. It can benefit, forits performance, from external aerodynamic flow features,such as a friendly boundary layer behavior. But it can alsoon its own contribute to the overall aerodynamic efficiencyof the airship, in case it is adequately distributed aroundthe hull.The use of distributed propellers is advantageous, since itreduces considerably the weight of the propulsion system.Also, it is more convenient to use this distributed propellerswith decentralized power unit, that can provide redundancyin case of failure. They can also contribute to a reductionin power cabling weight.In constant cruise flight the propulsion thrust must beable to compensate for the drag force, and eventuallyfor the needed dynamic lift. However, when accelerat-ing lighter than air vehicles its very important pay atten-tion to the added masses and ground effects [54]. Theyexert acceleration in take off conditions and, in case ofMAAT cruiser/feeder concept, also when the feeder ap-proaches the cruiser airship. This is another area, relatedto unsteady propulsion design conditions, where it is ofparamount importance to perform additional research work.198


Acknowledgments

The present work was performed as part of Project MAAT Multibody Advanced Airship for Transport with ref.285602, supported by European Union through the 7thFramework Programme.References

[1] Li Y., Nahon M., Sharf I., Airship dynamics modeling:A literature review, Progr. Aero. Sci., 2011, 47, 217239[2] Ardema M.D., Young A.D., Missions and vehicle con-cepts for modern, propelled, lighter-than-air vehicles,AGARD Report No 724, Advisory Group for AerospaceResearch and Development, 1985[3] Liao L., Pasternak I., A review of airship structuralresearch and development, Progr. Aero. Sci., 2009, 45,8396[4] Wang X., Shan X., Shape optimization of stratosphereairship, J. Aircraft, 2006, 43 (1), 283286[5] Nejati V., Matsuuchi K., Aerodynamics design andgenetic algorithms for optimization of airship bodies,JSME International Journal, Series B: Fluids andThermal Engineering, 2003, 46(4), 610617[6] Mueller J.B., Paluszek M.A., Zhao Y., Development ofan aerodynamic model and control law design for ahigh altitude airship, the AIAA Unmanned UnlimitedConference in Chicago, IL, 2004, 117[7] Gammon S., Frye M., Trevino R., Qian C., The develop-ment of the tri-turbofan airship model for autonomousflight control research, AIAA modeling and simulationtechnologies conference and exhibit, AIAA-2006-6620,2006[8] Grossman D., 2011, http://www.airships.net/hindenburg/design-technology[9] Burgess C.P., Airship design, The Ronald Press Com-pany, 1927[10] Goodyear Aerospace, Feasibility study of modern air-ships, phase II executive summary, NASA ContractorReport 2922, NASA, 1977[11] McLemore C., Wind-tunnel tests of a 1/20-scale airshipmodel with stern propellers, Tech. Rep. TN D-1026,NASA, 1962[12] Cornish J.J., Boatwright D.W., Application of full scaleboundary layer measurements to drag reduction ofairships, Tech. Rep. Report No. 28, Mississippi StateUniversity, Aerophysics Department, 1960[13] Lutz T., Leinhos D., Wagner S., Theoretical investiga-tions of the flowfield of airships with a stern propeller,Proceedings International Airship Convention and Ex-hibition, 1996, pp. 112

[14] Lutz T., Funk P., Jakobi A., Wagner S., Calculationof the propulsive efficiency for airships with sternthruster, 14th AIAA Lighter-Than-Air Technical Com-mittee Convention and Exhibition, 2001[15] Hirner A., Dorn F., Lutz T., Krmer E., Improvementof propulsive efficiency by dedicated stern thruster adesign, 7th AIAA Aviation Technology, Integration andOperations Conference (ATIO), No. AIAA 2007-7702,2007, pp. 18[16] Goldschmied F.R., Integrated hull design, boundary-layer control, and propulsion of submerged bodies,Journal of Hydronautics, 1967, Vol. 1, No. 1, pp. 111[17] Goldschmied F.R., Wind tunnel demonstration ofan optimized LTA system with 65 power reductionand neutral static stability, in AIAA Paper 831981,1983[18] Elfes A., Bueno S., Bergerman M., Paiva E., RamosJ., Robotic airships for exploration of planetary bodieswith an atmosphere:autonomy challenges, AutonomousRobots, 2003, Vol. 14, pp. 147164[19] Dorrington G.E., Concept options for the aerial surveyof titan, Adv. Space Res., 2011, Vol. 47, No. 1, pp. 119[20] Pascoa J., Xisto C., Goettlich E., Performance assess-ment limits in transonic 3d turbine stage blade rowsusing a mixing-plane approach, J. Mech. Sci. Tech.,2011, Vol. 24, pp. 20352042[21] Pascoa J., Mendes A., Gato L., A fast iterative inversemethod for turbomachinery blade design, MechanicalResearch Communications, 2009, Vol. 36, pp. 630637[22] AUGURRosAeroSystems, 2010,http://rosaerosystems.com/projects/obj687[23] Kaley N., The modern airship: A review of 40 yearsof airship golfier, Technical-Scientific Journal onModern Aerostatic Problems, development, 2003,Vol. 2, pp. 112, http://www.agaeroplast.com/new/engl/Nigele.pdf[24] Martnez F., Study of a zero-emission airship transportsystem based on the geostrophic fligth concept, Tech.Rep., Universitat Politcnica de Catalunya, 2011[25] Goodey T.J., Steam LTA past, present and future, in4th International Airship Convention and Exhibition,Cambridge, England, July 2002, pp. 117[26] Rapert R.M., A heat transfer model for a heated heliumairship, Masters thesis, Naval Postgraduate School,1987[27] Colozza A., Initial feasibility assessment of a highaltitude long endurance airship, Tech. Rep. NASA/CR-2003-212724 36, 2003[28] Vorachek J.J., A comparison of several very high al-titude station keeping balloon concepts, in Proc. 6thAFCRL Scientific Balloon Symposium, AFCRL 70-0543, 1970, pp. 283286[29] Hugh B.-S., Tom swift and his electric airship, IEEE199


Aerospace and Electronic Systems Magazine, 2011,Vol. 26, No. 10, pp. 411[30] Dumas A., Trancossi M., Madonia M., Giuliani I., Multi-body advanced airship for transport, SAE International,2011, DOI: 10.4271/2011-01-2786[31] Dumas A., Pancaldi F., Anzillotti F., Trancossi M.,High altitude platforms for telecommunications: de-sign methodology, SAE International, No. 09ATC-0010,2009[32] Dumas A., Anzillotti S., Madonia M., Trancossi M.,Effects of altitude on photovoltaic production of hydro-gen, Proceedings of the 5th International Conference onEnergy Sustainability, ESFuelCell2011-54624, 2011[33] Lutz T., Funk P., Jakobi A., Wagner S., Summary ofaerodynamic studies on the Lotte airship, Presentedat the 4th International Airship Convention and Exhi-bition, July 2831, 2002, Cambridge, England, 2002[34] Krplin B., Solar airship Lotte, 2011,http://www.isd.uni-stuttgart.de/lotte/

infodownload/engl_v02.pdf[35] Summers M., Lessons from tragedy: a review of thehelios difference, 2011,http://heliosairships.blogspot.com/2011/

06/lessons-from-tragedy-review-of-helios.

html[36] Lowry J., Fixed-pitch propeller/piston aircraft opera-tions at partial throttle, J. Propul. Power, 1999, Vol. 15,pp. 497503[37] Buerge B., Polar family airships for surveying andcargo, 2011, http://www.airshipstothearctic.com/docs/pr/14_Brandon_Buerge.pdf[38] Kinney J., Frank W. Caldwell and variable-pitchpropeller development, 1918-1938, J. Aircraft, 2001,Vol. 38, pp. 967976[39] M-Harada, Distributed multi-propulsion units system,JAXA, NAL RP-2002003, 2003, pp. 1013[40] Liu P., Duan Z., Ma L., Ma R., Aerodynamics propertiesand design method of high efficiency-light propeller ofstratospheric airships, 2011 International Conferenceon Remote Sensing, Environment and TransportationEngineering (RSETE), 2011, pp. 80418044[41] Pavlecka V., Thrusters for airship control, US patent4402475, 1983

[42] Turner R.C., Notes on ducted fan desing, AeronauticalResearch Council Current Papers, 1966, No. 895, p. 44[43] Piolenc F., Wright G., Ducted fan design, 2001,http://massflow.archivale.com/ductbook.htm[44] Hochstetler R., Airships for the 21st century, 2010,http://spectrum.ieee.org/aerospace/

aviation/airships-for-the-21st-century[45] Kothmann, The kothmann multi-use airship, US patent6648272b1, 2003[46] Rademacher A.T., Very large luxury airship (VLLA),http://www.engr.sjsu.edu/nikos/pdf/

VLLAirship[47] Wheatley J.B., Simplified aerodynamic analysis of thecyclogiro rotating-wing system, Tech. Rep. 467, NACA,1933[48] Wheatley J.B., Windler R., Windtunnel tests of a cy-clogiro rotor, Tech. Rep. 528, NACA, 1935[49] Kim S.J., Yun C.Y., Kim D., Yoon Y., Park I., Design andperformance tests of cycloidal propulsion systems, in44th AIAA/ASME/ASCE/AHS/ASC Structures, Struc-tural Dynamics, and Materials Conference, No. AIAA-2003-1786, 2003[50] Engstrom J., 2011,www.airship.org/verticalairships[51] Jordiand C., Michel S., Fink E., Fish-like propulsion ofan airship with planar membrane dielectric elastomeractuators, Bioinspiration Biomimetics, 2010, Vol. 5,DOI:10.1088/1748-3182/5/2/026007[52] Michel S., Bormann A., Jordi C., Fink E., Feasibilitystudies for a bionic propulsion system of a blimp basedon dielectric elastomers, Proc. of SPIE, 2008, Vol. 6927,pp. 115[53] Trancossi M., Dumas A., CFD based design of a noz-zle able to control angular deflection, in ASME 2011International Mechanical Engineering Congress & Ex-position, 2011, Vol. IMECE2011-65440[54] Mueller J., Paluszek M., Development of an aerody-namic model and control law design for a high altitudeairship, American Institute of Aeronautics and Astro-nautics, No. ADA451761, 2004, p. 17

200

A Critical Review of Propulsion Concepts for Modern Airships

Documents

Transcript of A Critical Review of Propulsion Concepts for Modern Airships