human power - velomobiles.co.uk · 2012. 10. 29. · different positions, and dietrich fellenz...

HUMAN POWERVolume 13 number 3 Summer/fall 1998 $5.00/IHPVA members, $3.50

HUMAN POWERis the technical journal of theInternational Human Powered VehicleAssociationVolume 13 number 3, Summer/fall 1998

EditorDavid Gordon Wilson21 Winthrop StreetWinchester, MA 01890-2851 [email protected]

Associate editorsToshio Kataoka, Japan1-7-2-818 Hiranomiya-MachiHirano-ku, Osaka-shi, Japan [email protected]

Theodor Schmidt, EuropeOrtbiihlweg 44CH-3612 Steffisburg, [email protected]

Philip Thiel, watercraft4720 - 7th Avenue, NESeattle, WA 98105 USA

ProductionJS Design

IHPVAPaul MacCready, international president

Theo Schmidt, Switzerland, Chair, 1998

Christian Meyer, Germany, Vice-chair,1998

Jean Seay, USA, Secretary/treasurer

Publisher:HPVAPO Box 1307San Luis Obispo, CA 93406-1307 USA+805-466-8010; [email protected]

Human Power (ISSN 0898-6908) ispublished irregularly, ideally quarterly, forthe International Human Powered VehicleAssociation by the Human Powered VehicleAssociation, a non-profit organization ded-icated to promoting improvement, innova-tion and creativity in the use of humanpower generally, and especially in the designand development of human-powered vehi-cles.

Material in Human Power is copyright-ed by the IHPVA. Unless copyrighted alsoby the author(s), complete articles or repre-sentative excerpts may be published else-where if full credit is given prominently tothe author(s) and the IHPVA.

CONTENTSStability and other factors in the designof displacement boats

Bob Stuart, a well-known Canadianbuilder of beautiful small boats and HPVs,and a previous contributor to HumanPower and HPVNews, reviews the ratherextraordinary range of choices open to thedesigner of human-powered non- hydrofoilboats, giving the advantages anddisadvantages of each approach.

Lower-extremity power output inrecumbent cycling: a literature review

In the last issue of HP (13/2 p. 19) youreditor confessed that he had been remiss innot reviewing such highly relevantpublications as the Journal of Biomechanics,and appealed for more-qualified reviewers.His plea was quickly and thoroughlyanswered. Raoul Reiser and Mick Petersonhave condensed a major study of the wholebiomechanics literature concerned withhuman power into a scholarly paper. This,with its large number of references, islonger than we normally carry. However,the topic is so central to the activities andenthusiasms of readers that we have waivedthe rules and have included it complete.Although Reiser and Peterson show thatthere is still some disagreement amongresearch findings, there are ranges ofrecumbent positions that appear to giveimproved endurance and short-term outputthat can be used for the design of at-leastnear-optimum recumbent HPVs.

earth of authors, muses about the advantagesof using human power for transportation infuture space colonies.

LettersTire differences on vehicles: Charles

Brown suggests that we use different tires indifferent positions, and Dietrich Fellenzresponds.

More on climbing with low bottombrackets: Zach Kaplan and Paul Buttemerconcur that low bottom brackets could givefaster climbing.

A proposed standard for measured dragreductions: Mike Saari would like to see averifiable number that could guide pur-chasers of HPVs.

ReviewsMajor Taylor: The extraordinary career of a

champion bicycle racer, by Andrew Ritchie,reviewed by Wade Nelson.

Trim of aerodynamically faired single-track vehicles in crosswinds, by AndreasFuchs, reviewed by Doug Milliken.

The third European seminar on velomo-bile design, Roskilde, August 1998, reviewedby Dave Wilson.

The proceedings of the eighth interna-tional cycle-history conference, Glasgow,August 1997, reviewed by Dave Wilson.

EditorialMisplaced machismo? by Dave Wilson

Looking ahead-human power in spaceJohn Allen, normally the most down-to-

Volume 13 number 3, summer/fall 1998

CONTRIBUTIONS TO HUMAN POWER

The editor and associate editors (you may choose with whom to correspond) welcomecontributions to Human Power. They should be of long-term technical interest (noticesand reports of meetings, results of races and record attempts, and articles in the style of"The building of my HPV" should be sent to HPVNews). Contributions should also beunderstandable by any English-speaker in any part of the world: units should be in S.I.(with local units optional), and the use of local expressions such as "two-by-fours" shouldbe either avoided or explained. Ask the editor for the contributor's guide. Many contribu-tions are sent out for review by specialists. Alas! We are poor and cannot pay for contribu-tions. They are, however, extremely valuable for the growth of the human-powermovement. Contributions include papers, articles, reviews and letters. We welcome alltypes of contributions, from IHPVA-affiliate members and nonmembers.

2 Human Power

adding small, angled hydrofoils to tiny,raised floats, which help to maintain bal-ance. For him, these have additional merit,because he also adds a sail, and the foils'effect is increased in proportion to need,along with wind speed and boat speed. Hispaddle and sail boat is in production as theTriak. One or two foils mounted to the sidecould also be actively controlled by a surfacefollower, like the front hydrofoil on theFlying Fish HPB, or the side foils on somesailing hydrofoils. One would normally besufficient, but two would be safer for chop-py water. Richard Ehrlich has been experi-menting successfully with a pedal kayakusing amas that get lift from planing as wellas displacement. These are attached to hisspare paddle, which is then clipped to theafter deck.

The author has built a pedal-kayak,Lambordinghy, equipped with a specialpaddle. This normally rests in a pair ofyokes connected to the rudder lines, and sois used for steering. However, the foil-shaped blades are held at a positive angle,and begin work instantly when needed.They plane initially at speed, but may alsobe used for all the usual kayak paddle tricks.It was hoped that the foil-shaped bladeswould provide useful buoyancy at rest, butthey give so little that thin, smooth bladeswould probably serve as well.

Designing for speed usually involves along, slender hull, which is hard to turnquickly. Adding more hulls or fins, especial-ly toward the ends, makes the boat even lessmaneuverable. This effect can be muchreduced with little loss of speed by adding"rocker"; that is bringing the keel line upgradually to near the water surface at stemand stern (fig. 5). Another experiment onLambordinghy had flying amas mounted ona swivel near the stern so that they couldreplace the rudder,-and help

I I I ·I II

rather than hinder the turn-ing. This worked quite wellafter the addition of smallskegs (fins) to the floats (fig.1). This scheme has theadditional merit of allowingdocking without interfer-ence from the outriggers.

CATAMARANS

tatamaran arrangements Iare often used for human-

powered boats (HPBs), and have enoughexcess stability to carry riders easily muchhigher up than do most monohulls. Touristsoften rent paddle-wheel cats, and severalmanufacturers offer much faster propeller-driven versions. Buckminster "Bucky" Fullerwas especially delighted with an inspirationhe called "Rowing Needles."

Cats almost always have enough extrastability to carry sail, though I have heard ofthis being exploited only once, with a pedal-prop supplement to a Hobie Cat. Anotherway to pick up energy to rival what a ped-aler provides is to exploit wave action. Anexperimental craft called the Gausfinachieved 4 to 5 knots in a moderate chop.

One drawback with two full-length hullsis that this shape has the greatest resistanceto turning. Sometimes, two riders (notablyat European HPB contests) each pedal oneof two widely separated props, which pro-vides superb maneuverability. In otherboats, two people would generally use oneprop with the pedals out of phase, since aprop is much more efficient with a constanttorque.

A special case of two hulls is the proa,with one hull much larger than the other.The small hull works as a float or counter-balance as needed, and this combination hasless surface area for the displacement thantwo equal hulls. The Saber Proa is an out-standing pioneer HPB of this type (fig. 2).The large hull could also contain the loadlow down, as does a monohull, thus reduc-ing the need for supplementary stability. Asailing proa usually wants to keep the windon the same side, so instead of tacking nor-mally, it changes direction, sailing "forwardsand backwards." Sailors call a proa that doesnot change direction a "tacking proa". Theborrowed name may need some explanationapplied to an HPB.

Figure 2. Saber Proa (illustration used with permission ofScientific American)

Figure 3. Phil Thiel's Dorycycle

CONVENTIONAL MONOHULLSThe obvious way to improve stability

with a single hull is to make it wider (fig. 3),but there are a couple of tricks that can givegood results with less added drag. Sailboats,which need a lot of stability to counteractthe wind loads, and also have little speed tospare, often use ballast at the bottom of thekeel, the bottom of the hull, or moveableballast on the windward side. Extra weight iscarefully trimmed from other areas, as it isotherwise a liability in fast vehicles. Anotherproblem with ballast is that it is usually verydense, and can cause abrupt sinking if watergets in. Some HPBs also use ballast, notablyGarry Hoyt's Mallard, now renamedEscapade and being produced in Michigan

(fig. 4).Traditional dories have relatively narrow,

efficient bottoms, but widely flaring sides.The total effect is a bit like a trimaran; itstarts to tip easily, but is soon caught by thebuoyancy of the sides. One thorny problemwith adding pedals to a dinghy is the changein trim between having one and two personsaboard, as the pedaler's position is hard tochange.

BOATS LIKE BICYCLESRecently, David Witt and George Tatum

have independently hit on the idea of usinga deep central fin, pivoted like a rudder andunder constant manual control to generate arighting moment for an upright rider on avery narrow hull. David uses his fin to dou-ble as a forward rudder, while George finds

Figure 4 Escapade, a ballasted monohull

4 Volume 13 number 3, summer/fall 1998 Human Power

/.= K f

l


I

Figure 4. Schematic representation of the muthe lower extremity identified as prime moversbiceps femoris long head (BFL), biceps femoriahead (BFS), gastrocnemius (GN), gluteus maxigluteus medius (GM), gluteus minimus (GM), i.(IS), rectus femoris (RF), soleus (SL), semimet

$C,

it

S.

mlicnl

(SM), semitendinosus (ST), tibialis anterior (TA),lateralis (VL), and vastus medialis (VM).

all movements to a smaller degree as joint tistabilizers that may act synergistically or tfantagonistically (Zajac & Gordon, 1989). qlInvestigations have found that the moment Farms of many of these muscles change dra- rematically with joint angle (Hoy et al., 1990; E.Nemeth & Ohlsen, 1985; Spoor & van inLeeuwen, 1992; Spoor et al., 1990). Forexample, Nemeth & Ohlsen (1985) found hthat the moment arm of the gluteus max- irimus decreased 48 mm, the moment arm of ai

the hamstrings decreased 20 mm, and the c)moment arm of the adductor magnus n,increased 46 mm when the hip was flexed in p,the sagittal plane from the anatomical posi- tetion to 90° . This change could have a signif- sticant impact on the power production of rethis important muscle. Si

As mentioned, the non-muscular com- reponents are also affected by alterations inthe geometrical and operational parameters.The gravitational contribution to the pedalforce is altered with variations in hip orien-tation. As the hip orientation is changed,the horizontal distance between the pedaland seat also changes. The gravitationalforce contributed by each segment that isshared between the pedal and seat is there-fore altered. In addition, the torque contri-bution of gravity on each segment will bealtered with a change in hip orientation.Finally, the inertial contribution is depen-dent on pedal cadence. The faster the pedal-ing cadence, the greater the inertial forces(Kautz & Hull, 1993).

I

Since both the muscularand non-muscular contribu-tions to the pedal force aredependent on the geometricand operational parameters, itis hard to predict the best riderposition to produce optimaljoint moments from the hips,knees, and ankles. A numberof investigations have beenperformed to examine theeffects upon recumbentcycling performance by vary-

les of ing one or more of the biome-n cycling: chanical factors. A summary

short of the key results as they relateus (GM), to the hip orientation and)spoas torso angle follows. While the5ranosus other geometrical and opera-vastus tional parameters are impor-

tant, they do not contributelarge changes in the design of

ie vehicle in terms of the size and shape ofhe vehicle nor does space allow for ade-uate discussion of these parameters.[owever, references are provided for the:ader on these parameters in the appendix.xperimental investigationsito recumbent cycling

Too (1991 b) systematically altered theip orientation from -10 ° to 65° in 25°

icrements while maintaining a 90 ° torsongle (fig. 5). Fourteen male recreationalyclists (ages 21-32 yrs) with very little too recumbent cycling experienceerformed the Wingate anaerobic cycling:st with a resistance of 0.83 N/kg of theibject's body mass BM] (5.0 joules/pedal:v/kg BM) in each of these positions.ubjects were strapped to the seat and back:st by means of a lap belt and shoulder

Figure 5. Test position range utilizedby Too (1991b): hip orientation from -10° to 65° in 25 ° increments withconstant 90° torso angle.

harness while pedaling with toe clips.Additionally, hip distance (100% oftrochanteric leg length (distance from floorto greater trochanter when standing erect))and crank-arm length (not specified) werecontrolled for each subject.

Kinematic analysis found that the meanankle and knee angles stayed relatively con-stant, ranging from 90.1 to 92.8 and 98.2to 103.6° , respectively, while the mean hipangle increased from 58.9 to 114.0° fromthe hip orientation of-10 to 65°. Relativepeak power output (highest average powerduring any successive five second intervaldivided by body mass) varied from 10.55+/-1.38 to 11.73+/-1.03 watts/kg BM with the15° hip orientation yielding the greatest val-ues. The 15° hip orientation was significant-ly greater than the -10 and 65 ° hiporientations, but not significantly differentfrom the 40° hip orientation (table 1).Additionally, relative average power (overthe entire thirty-second test) and fatigueindex (percentage of peak power subtractingthe lowest power from the peak power anddividing by the peak power) were calculated.Relative average power values followed asimilar trend to the relative peak powerscores while the fatigue index was similaracross all hip orientations studied.

From this experimental design it couldnot be determined if the changes in poweroutput were attributable to changes in meanhip angle (body configuration), gravityeffects on the lower extremity, or a combina-tion of both. Too (1994) refined the experi-mental design to try to isolate the effects ofgravity on peak power output. A similar set-up and test protocol was utilized while vary-ing the torso angle along with the hiporientation in order to maintain a constantbody configuration of 105 ° . The 105 ° bodyconfiguration was selected because it was themost powerful in the previous study. Sixteenmale recreational cyclists (age 20-36 yrs)performed the Wingate anaerobic testagainst a resistance of 0.83 N/kg body masswith a hip orientation of-15, 15, and 45 °

(torso angle of 60, 90, and 120°, respective-ly) (Figure 6). Kinematic analysis confirmedthat mean hip (80°), knee (100°), and ankle(83 °) angles did not vary significantlybetween these three positions. The relativepeak power was found to be greatest in the15° hip-orientation position, but not signifi-cantly greater than the -15° hip-orientationposition (the total range of relative peak


IA ' '

# S S

I I... I4ii

---; -


In addition to the two sets of experi-ments performed by Too (1989 & 1990)that examined time and total work outputto exhaustion, others have investigated theeffects of body position on long-durationcycling performance (Bevegard et al., 1966;Diaz et al., 1978; Metz et al., 1986; Nadel& Bussolari, 1988; Stenberg etal., 1967;Wescott, 1991). The body's ability to per-form sustained work was found to varywith body position due to alterations in itsability to circulate the blood and exchangegases in the lungs (Bevegard et al., 1966;Stenberg et al., 1967). The blood both pro-vides nutrients to the working muscles andremoves waste products of energy produc-tion. While tasks that are primarily aerobicin nature must consider these physiologicaladaptations that occur with changes inbody position, a primarily anaerobic task,such as a performance for setting thehuman-powered speed record which takesapproximately two minutes to complete,does not rely on optimal blood circulationand ventilation in the lungs during the task(Foster etal., 1995; Too, 1994). However,since a large amount of time might bespent training in the vehicle or in the run-up leading to the sprint, blood circulationshould be considered no matter what the

intended goal of the vehicle.Analytical investigationsinto recumbent cycling

The only analytical approach directed toinvestigate the recumbent position was byLei etal. (1993). This model took intoaccount both the geometry of the rider aswell as the effects of the riding position onthe aerodynamic cross-section of the vehicle.The model did not include any muscles andtherefore the effects of different muscle

lengths in different rider positions. Hip ori-entation, torso angle, hip distance, andcrank-arm length were all varied within con-strained limits which included a minimumtorso angle in order to maintain adequateforward visibility. Both anaerobic and aero-bic performances were evaluated by meansof two different cost functions. For anaero-bic performance the cost function wasdesigned to minimize the moment varia-tions on the hip and knee joints while foraerobic performance the cost function wasto minimize both the average and maximumvariation of the hip- and knee-jointmoments. The model was designed for a tar-get speed of 40.0 km/h with a vehicle

weight of 50.0 kg, air drag coefficient of0.15, rolling friction of 0.01, and pedalingcadence of 60 rpm with a 50 th percentilemale rider.

The optimal aerobic (endurance) perfor-mance was computed to occur with a hiporientation of-5.70 ° and torso angle of26.72 °. The optimal anaerobic (speed) per-formance was calculated with a hip orienta-tion of -25.4 ° and torso angle of 48.1 ° . Both

simulations found the optimal hip distanceto be 0.751 m and crank-arm length to be0.15 m. The model was also run withouttaking aerodynamic drag into account inorder to compare with the results of Too(1991 b). This simulation found the optimalanaerobic position to be with a hip orienta-tion of 22.9 ° and hip distance of 0.662 m ascompared to 15° and 0.666 m, respectively,experimentally established by Too (1991 b).Since the length of muscles crossing the hipsare not considered by the model and aero-dynamics no longer a concern, allowabletorso angles include any angle that wouldmaintain visibility. Crank-arm length of thismodified model was not reported.

Other analytical investigations have beenperformed to investigate the effects of hiporientation on steady-state cycling perfor-mance, as reviewed by Gregor etal. (1991).However, these studies generally constrainedthe hip orientation to stay greater than 70 °

in order to stay within the bounds pre-scribed for the standard bicycling position.The most comprehensive multivariableanalysis was conducted by Hull & Gonzalez(1990). They found that changes in hip ori-entation significantly altered the results ofthe joint-moment-based cost function. Itwas also found that the optimal hip orienta-tion was altered by rider size. In addition tobeing constrained to maintain a standardcycling position, muscles were not includedin the model. Without muscles, the effectson performance by changing their lengthsand moment arms with the changes in posi-tion could not be assessed.

Additional experimental investigationshave looked into various hip orientationsand torso angles while maintaining theupright riding position, such as Umberger etal. (1998). However, the majority of thesestudies have concentrated on the aerody-namic implications of the various positionswith the assumption that time training inthe most aerodynamic position will make ita viable riding position. These studies are

reviewed in Gregor et al., (1991).From the reviewed literature it is clear

that varying one or more of the biomechani-cal parameters may alter the power-produc-tion capabilities of the cyclist. However, theoptimal recumbent riding position is stillnot clear. Nor is it clear whether the stan-dard cycling position is better biomechani-cally and physiologically than recumbentcycling, since none of the reviewed articlestested the subjects in the standard cyclingposition as well as the recumbent positions.Effect of gravity on performance

Brown et al. (1996) showed throughinverse kinematics and EMG analysis thatmuscle-group contributions do change asthe effects of gravity are altered on the lowerextremity. Too (1991a) also noted alter-ations in muscle activity when only theeffects of gravity on the extremities werealtered. However, due to the low power out-put of the cyclists measured relative to theirpeak power in these two studies it cannot beconcluded that peak power output would bealtered by the effects of gravity on the lowerextremities. Too (1990) showed that alteringthe effects of gravity on the lower extremityand blood circulation did not alter the work

output and time to exhaustion in the threepositions studied. In addition, two out ofthe three positions studied were not signifi-cantly different in peak power output (Too,1994). While this limited number of studiesis not conclusive, it appears that the effectsof gravity on performance may be small, atleast in the range of positions studied.Effect of bodyconfiguration on performance

No studies to date have altered the bodyconfiguration while maintaining a constanthip orientation in order to remove theeffects of gravity on the cycling perfor-mance. However, assuming that gravity is asecondary effect based on the previous dis-cussion, body configuration has a majoreffect on power production while cycling.Based on the work of Too (1989 & 1991 b)the 105 ° body configuration may be opti-mal for both peak power production andsustained aerobic performance. However,the differences in the peak power at the130 ° body configuration were not statistical-ly significant. Also, with 25 ° incrementsbetween the body configurations tested, theoptimal configuration may be a position nottested. Gravity also may play a large enoughrole to alter the optimal body configuration



at some hip orientations. More research isneeded into the effects of body configura-tion as well as the effects of gravity to makeany solid conclusions about the optimal rid-ing position.

It is interesting to note that the 105 °

body configuration falls in the range used inthe standard cycling position. Cavanagh &Sanderson (1986) reported that elite pursuitcyclists rode with an average torso angle of145°. Considering that the standard cyclingposition general uses a hip orientation (seattube angle) from 70 to 90 ° (Burke & Pruitt,1996), the cycling position for these elitepursuit riders has a body configuration from105 to 125 °. The 125 ° body configurationis within 5° of the position that Too(1991b) found to be not statistically differ-ent from the 105 ° body configuration. Thisknowledge lends further support to theassumption that the musculoskeletal biome-chanics at the hip may be of more concernthan the effects of gravity on the lowerextremity for power output. These conclu-sions may be a bit premature, however, sinceToo (1989 & 1991b) did not test his sub-jects in the standard cycling position andthey would not be considered 'elite pursuitriders' who may have trained into their rid-ing position. The human body is highlytrainable and may be able to adapt to aseemingly non-optimal position and make itoptimal, as long as the cycling position isnot too drastically different from a goodpower-producing position.

Additionally, care must be taken whenimplementing the results from Too andBrown. These studies are based on resultsfrom recreational cyclists with little to norecumbent cycling experience. Also, theseresults are based on studies where the sub-jects used toe-clips rather than clipless ped-als which are more common for highperformances. Clipless pedals may make alarge difference in recumbent cycling sincegravity is acting differently on the legs thanin the standard cycling position.

While analytical models can providegreat insight into what position may beoptimal and why, the results from thosereviewed in this article were not used here tojustify one position over another. Thesemodels did not incorporate the effects ofchanging the muscle lengths and momentarms across the hips that appear to be amajor determinant of performance based onthe experimental results. Without the inclu-

sion of these musculoskeletal effects the ana-lytical results are difficult to use, even if theyappear to be consistent with the experimen-tal results.

CONCLUSIONSAlong with the power-production capa-

bilities of the rider and the aerodynamics ofthe vehicle, additional performance factorsand practical constraints must be consideredwhen designing a successful human-pow-ered vehicle. Performance factors include thepower-train efficiency, vehicle dynamics,and road friction for a land vehicle. Thefairing must also be designed to allow ade-quate air flow in order to keep the ridercooled. Practical constraints include the visi-

bility, controllability, structural stability,safety, and comfort of the vehicle. Many ofthese factors are not independent. For exam-ple, the rider position, in addition to affect-ing the rider's ability to produce power andthe aerodynamic design of the vehicle, mayalso affect the drive-train construction andthus its ability to transfer the energy fromthe pedals to the wheels efficiently. For thesereasons, the overall design may not includethe optimal riding position for peak-powerproduction/sustained performance or thedesign with the lowest aerodynamic cost.However, a global optimal for all designconstraints should be selected.

More research is needed into the effects

of body configuration and gravity acting onthe lower extremity, as well as the effects oftraining on cycling position. None-the-less,an aerodynamic position with a low hip-ori-entation angle (either positive or negative)combined with a body configuration from105 to 130 ° appears based on current litera-ture to be justifiable for a high-performancehuman-powered vehicle.

APPENDIXThe following are references that discuss

geometric and operational parametersbeyond hip orientation and torso angle.References followed by an 'R' have informa-tion directly related to recumbent cycling.Hip distance: Gregor et al. (1991), Too

(1993b) R, Too (1990b) RCrank-arm length: Gregor et al. (1991),

Inbar et al. (1983), Too (1990b) R, Too(1996) R, Too (1998) R

Foot position on pedal: Burke & Pruitt(1996), Hull and Gonzalez (1990)

Foot-to-pedal interface: Broker & Gregor

(1996), Gregor etal. (1991), Moran(1990), Moran & McGlinn (1995),Wheeler et al. (1995)

Pedaling cadence and power output: Coast(1996), Gregor et al. (1991), Too(1990b) R, Urlocker & Prassas (1996),Whitt & Wilson (1982)

AUTHORSRaoul F. Reiser II, M.A., C.S.C.S.

is a doctoral candidate in MechanicalEngineering at Colorado State University.His dissertation is investigating some of thebiomechanical questions still remaining asthey pertain to recumbent cycling.Previously, he worked for the United StatesOlympic Committee where he worked withmany sports, including cycling.

M.L. "Mick" Peterson, Ph.D. is an assis-tant professor in the mechanical engineeringdepartment at Colorado State University.His life changed when he read RichardsBicycle Book as a kid growing up in themountains of Colorado. Since then he hasbeen an avid bicycle commuter and is now afaculty advisor for the CSU HPV team.Authors' address: Department ofMechanical Engineering, Colorado StateUniversity, Fort Collins, CO 80523 USA

REFERENCESAlexander, RM & Ker, RF (1990). The

architecture of leg muscles. In MultipleMuscle Systems: Biomechanics andMovement Organization (Edited byWinters, JM & Woo, SL-Y), pp568-577.Springer-Verlag, New York.

Bevegard, S, Freyschuss, U, & Strandell, T(1966). Circulatory adaption to arm andleg exercise in supine and sitting position.JApplied Physiology 21 (2), 37-46.

Broker, JP & Gregor, RJ (1996). Cycling

biomechanics. In High-Tech Cycling(Edited by Burke, ER), ppl 4 5-1 6 6 .Human Kinetics Publishers, Champaign.

Brown, DA, Kautz, SA, & Dairaghi, CA(1996). Muscle activity patterns alteredduring pedaling at different body orienta-tions. JBiomechanics 29(10), 1349-1356.

Burke, ER & Pruitt, AL (1996). Body posi-tioning for cycling. In High-Tech Cycling(Edited by Burke, ER), pp79-99.Human Kinetics Publishers, Champaign.

Cavanagh, PR &Sanderson, DJ (1986). Thebiomechanics of cycling: studies of thepedaling mechanics of elite pursuit riders.In Science of Cycling (Edited by Burke,

Volume 13 number 3, summer/fall 1998Human Power 11

ER), pp91-122. Human KineticsPublishers, Champaign.

Coast, JR (1996). Optimal pedalingcadence. In High-Tech Cycling (Edited byBurke, ER), pplOl-116. HumanKinetics Publishers, Champaign.

Diaz, FJ, Hagan, RD, Wright, JE, &Horvath, SM (1978). Maximal and sub-maximal exercise in different positions.MSS 10(3), 214-217.

Foster, C, Hector, LL, McDonald, KS, &Snyder, AC (1995). Measurement ofAnaerobic Power and Capacity. InPhysiological Assessment of Human Fitness(Edited by Maud, PJ & Foster, C),pp73- 8 5. Human Kinetics Publishers,Champaign.

Gregor, RJ, Broker, JP, & Ryan, MM(1991). The biomechanics of cycling. InExercise and Sport Science Reviews (Editedby Holloszy, JO), Vol. 19, pp 12 7-1 69 .Williams & Wilkens, Baltimore.

Gregor, RJ & Rugg, SG (1986). Effects ofsaddle height and pedaling cadence onpower output and efficiency. In Science ofCycling (Edited by Burke, ER), pp6 9-90.Human Kinetics Publishers, Champaign.

Gross, AC, Kyle, CR, & Malewicki, DJ(1983). The aerodynamics of human-powered land vehicles. Scientific American249(6), 142-152.

Hoy, MG, Zajac, FE, & Gordon, ME(1990). A musculoskeletal model of thehuman lower extremity: the effect ofmuscle, tendon, and moment arm on themoment-angle relationship of musculo-tendon actuators at the hip, knee, andankle. J. Biomechanics 23, 157-169.

Hull, ML & Gonzalez, H (1990).Multivariate optimization of cycling bio-mechanics. In Biomechanics of Sports VI(Edited by Kreighbaum, E & McNeill,A), pp 15-41. Montana State University,Boseman.

Hull, ML & Hawkins, DA (1990). Analysisof muscular work in multisegmentalmovements: application to cycling. InMultiple Muscle Systems: Biomechanics andMovement Organization (Edited byWinters, JM & Woo, SL-Y), pp 62 1-638.Springer-Verlag, New York.

Hull, ML & Jorge, M (1985). A method forbiomechanical analysis of bicycle ped-alling. J Biomechanics 18(9), 631-644.

Ice, R & Waite, J (in preparation). Overuseinjuries in cycling. InternationalJ SportsMedicine.

Inbar, O, Dotan, R, Trousil, T, & Dvir, Z(1983). The effect of bicycle crank-lengthvariation upon power performance.Ergonomics 26(12), 1139-1146.

Kautz, SA & Hull, ML (1993). A theoreti-cal basis for interpreting the force appliedto the pedal in cycling. J Biomechanics26(2), 155-165.

Kita, J (1997). The unseen danger (Specialreport: impotency and cycling). BicyclingAugust, 68-73.

Kroemer, KHE (1972). Pedal operation bythe seated operator. SAE Paper 72004.Society of Automotive Engineers, NewYork.

Kulig, K, Andrews, JG, & Hay, JG (1984).Human Strength Curves. In Exercise andSport Science Reviews (Edited by ?), Vol.12, pp4 17 -46 6 . Williams & Wilkens,Baltimore.

Kyle, CR & Caiozzo, VJ (1986).Experiments in human ergometry asapplied to the design of human poweredvehicles. Int. J Sports Biomechanics 2,6-19.

Lei, Y, Trabia, MB, & Too, D (1993).Optimization of the seating position in ahuman-powered vehicle. In Biomechanicsin Sports Xl (Edited by Hamill, J,Derrick, TR, & Elliott, EH),pp 1 15-1 19. University of Massachusettsat Amherst.

Martin, G (1984). Notes on human pow-ered practicality. In Proceedings of theSecond International Human PoweredVehicle Scientific Symposium, pp1 1 5-117;103. Long Beach, CA: IHPVA, Box2068, Seal Beach, CA.

Metz, LD, Moeinzadeh, MH, White, LR,& Groppel, JL (1986). In Biomechanics:The 1984 Olympic Scientific CongressProceedings (Edited by Adrian &Deutsch), pp 289-295. MicrofilmPublishers. University of Oregon,Eugene.

Moran, GT (1990). Biomechanics ofcycling-the role of the foot pedal inter-face. In Biomechanics in Sports 17i (Editedby Kreighbaum, E & McNeill),pp43- 4 9. Montana State University,Boseman.

Moran, GT & McGlinn GH (1995). Theeffect of variations in the foot pedal inter-face on the efficiency of cycling as mea-sured by aerobic energy cost andanaerobic power. In Biomechanics ofSports XII (Edited by Barabis, A &

Fabian, G), pp105-109. Budapest,Hungary.

Nadel, ER & Bussolari, SR (1988). TheDaedalus project: physiological problemsand solutions. American Scientist July-August, 350-360.

Nemeth, G & Ohlsen, H (1985). In vivomoment arm lengths for hip extensormuscles at different angles of hip flexion.JBiomechanics 18(2), 129-140.

Seirig, A & Arvikar, RJ (1989). Modeling ofthe musculoskeletal system for the upperand lower extremities. In BiomechanicalAnalysis of the Musculoskeletal Structure forMedicine and Sports, pp99-154.Hemisphere Publishing Corp., New York.

Spoor, CW & van Leeuwen, JL (1992).Knee muscle moment arms from MRIand from tendon travel. J Biomechanics25(2), 201-206.

Spoor, CW, van Leeuwen, JL, Meskers,CGM, Titulaer, AF, & Huson, A (1990).Estimation of instantaneous momentarms of lower-leg muscles. J Biomechanics23(12), 1247-1259.

Stenberg, J, Astrand, P-O, Ekblom, B,Royce, J, & Saltin, B (1967).Hemodynamic response to work withdifferent muscle groups, sitting andsupine. JApplied Physiology 22(1), 61-70.

Too, D. (1998). Comparisons betweenupright and recumbent cycle ergometrywith changes in crankarm length(abstract). MSSE Supplement 30 (5),s81.

Too, D (1996). The effect of pedalcrankarm length on power production inrecumbent cycle ergometry. InBiomechanics in Sports XIII (Edited byBauer, T). Lakehead University, ThunderBay, Ontario, pp350-353.

Too, D (1994). The effect of trunk angle onpower production in cycling. ResearchQuarterly for Exercise and Sport 65(4),308-315.

Too, D (1993a). The effect of hipposition/configuration on EMG patternsin cycling. In Biomechanics in Sports XI(Edited by Hamill, J, Derrick, TR, &Elliott, EH). University of Massachusettsat Amherst, pp12 6 -131.

Too, D (1993b). The effect of seat-to-pedaldistance on anaerobic power and capacityin recumbent cycling (Abstract). MSSESupplement 25(5), s68.

Too, D (1991 a). The effect of body orienta-tion on EMG patterns in cycling. In



Biomechanics in Sports IX (Edited by Taut,CL, Patterson, PE, & York, SL). IowaState University, Ames, Iowa,pp109-115.

Too, D (1991b). The effect of hip posi-tion/configuration on anaerobic powerand capacity in cycling. Int. J SportBiomechanics 7, 359-370.

Too, D (1990a). The effect of body configu-ration on cycling performance. InBiomechanics of Sports VI (Edited byKreighbaum, E & McNeill, A). MontanaState University. Boseman, Montana,pp 5 1-58.

Too, D (1990b). Biomechanics of cyclingand factors affecting performance. SportsMedicine 10 (5), 286-302.

Too, D (1989). The effect of body orienta-tion on cycling performance. In VI/IInternational Symposium ofBiomechanicsin Sports (Morrison, WE, Ed.).Melbourne, Victoria, Australia, pp53-60.

Umberger, B.A., Scheuchenzuber, H. J.,and Manos, T. M. (1998). Differences inpower output during cycling at differentseat-tube angles (abstract). MSSESupplement 30 (5), s81.

Urlocker, JA & Prassas, SG (1996). Phasicmuscle activity of the lower extremity atdifferent powers and pedaling cadences incycle ergometry. In Biomechanics in SportsXIII (Edited by Bauer, T). LakeheadUniversity, Thunder Bay, Ontario,pp354 -357.

Wescott, WL (1991). Comparison ofupright and recumbent cycling exercise.American Fitness Quarterly, October1991.

Wheeler, JB, Gregor, RJ, & Broker, JP(1995). The effect of clipless float designon shoe/pedal interface kinetics andoveruse knee injuries during cycling.JAppliedBiomechanics 11, 119-141.

Whitt, FR & Wilson, DG (1982). Humanpower generation. In Bicycling Science(Second Edition), pp2 9-70. MIT Press,Cambridge.

Wilson, DG, Forrestall, R, & Hendon, D(1984). Evolution of recumbent bicyclesand the design of the avatar bluebell. InProceedings of the Second InternationalHuman Powered Vehicle ScientificSymposium, pp92-103. Long Beach, CA:IHPVA, Box 2068, Seal Beach, CA.

Winter, DA (1990). Biomechanics and MotorControl of Human Movement (SecondEdition). John Wiley & Sons, Inc.,

New York.Winters, JM & Stark, L (1988). Estimated

mechanical properties of synergistic mus-cles involved in movements of a variety ofhuman joints. J Biomechanics 21 (12),1027-1041.

Zajac, FE (1989). Muscle and tendon: prop-erties, models, scaling, and application tobiomechanics and motor control. In CRCCrit. Rev. Biomed. Engng (Edited byBourne, JR), Vol. 17, pp359-411. CRCPress, Boca Raton.

Zajac, FE & Gordon, ME (1989).Determining muscle's force and action inmulti-articular movement. In Exercise and

Sport Science Reviews (Edited by Pandolf,K), Vol. 17, pp187-2 30. Williams &Wilkens, Baltimore.

LOOKING AHEAD:HUMAN POWER INSPACEJohn S. Allen

The IHPVA has competitions forsurface, air, and waterborne vehicles. It isonly a matter of time until there are alsocompetitions in extraterrestrial environ-ments. This article will look into some ofthe challenges and possibilities of human-powered travel on other moons and planetsand in space stations. The insights to begained may be useful in planning spacemissions; but also can lead to some usefulEarthbound applications.

This article is not the first discussion ofhuman power in extraterrestrial environ-ments. In his early story "Pokanyne ofMars," science fiction author RobertHeinlein described human-powered aircraftthat flew in a sealed cavern used for a moon

colony's air storage. Conditions in the cav-ern were much more favorable for human-powered flight than on earth, due to themoon's low gravity and to an atmosphericpressure 3-4 times that on earth. Heinlein'swere birdlike wing-flapping craft, not pro-peller-driven ones like those that have nowactually been built and flown. In his juvenilestory "The Rolling Stones," Heinlein de-scribed bicycling on the surface of Mars (1).

Human Power editor David GordonWilson suggested a human-powered lunarvehicle to the National Aeronautics andSpace Administration (NASA) some 15

years before the Apollo lunar missions.NASA rejected Wilson's idea and sent anelectrically-powered cart, the Lunar Rover,with the astronauts on the Apollo mission.The astronauts brought back movies of theLunar Rover throwing rooster tails of dust,skidding and bouncing like a dune buggy.This was possible because of the moon's lowgravity, and despite the Rover's low power.Wilson described the human-powered lunarsurface vehicle in an article which appearedin Galileo magazine in 1979 (2).

In his article, Wilson argued that ahuman-powered off-road lunar surface vehi-cle would effect impressive savings in launchweight for an Apollo mission, and could goabout 18 mph (30 km/h) with a rider'spower input of 75 watts (1/10 horsepower)-even faster than the electrically-poweredvehicle that actually went to the moon. In asurprising turn, Wilson showed that humanpower would actually be too high for safepersonal transportation on paved surfaceson the moon. The low gravity would bothincrease speed and decrease the ability tocorner and brake. As an alternative, Wilsonpointed out that the power-to-speed rela-tionship for a human-powered aircraft in alunar colony would be ideal, and deter-mined that flying such aircraft would begreat fun - not the hard work it is with thehigher gravity on Earth. Wilson also advo-cated low-speed human-powered transportvehicles for use in lunar colonies.

Let's look into some of the science thatunderlies Wilson's observations about sur-face travel, and see where that leads us.

Gravitation has many important effectson human-powered travel. Yet, if we do notventure beyond Earth, Earth's gravity is easyto take for granted. When we look intoenvironments beyond earth, we can begin tounderstand just how drastically the amountof gravitational force affects the operation ofhuman-powered vehicles.

This starts with the human body itself.Humans, like other life forms on Earth,have evolved in response to Earth's gravity. Itis well-known that larger creatures mustwork harder against gravity than smallerones; the heavy bones and strong muscles oflarge dinosaurs, bears and elephants are nec-essary to maintain the ratio of bone andmuscle cross-section to body volume.

Since aviators frequently experiencehigh-gravity conditions, the effect of higher-than normal gravitation in the short term is

Volume 13 number 3, summer/fall 1998 13Human Power

well-known. Several times Earth's gravitycan be tolerated in a seated or recumbentposition, but mobility is seriously impaired.

Since it is unlikely that anyone willsubject experimental subjects to life in acentrifuge for weeks at a time, the effect ofhigher-than normal gravitation in the longterm can only be imagined, and this will bethe case for a long time, since all planets inour Solar System that have higher gravitythan Earth are gaseous: they have nosurface to stand on. Even a hypotheticalnuclear-powered space station forinterstellar travel would accelerate verynicely at a comfortable 1 g.

The effects of zero gravity on the humanbody are well-known, thanks to long toursof duty on the Russian Mir space station. Inthe long term, muscles and bones degener-ate in zero-gravity or low-gravity space envi-ronments where they are not subject to theloading for which they are designed. A pro-gram of exercise is necessary to minimizethese effects. In the future, space stationsmay be placed into rotation to simulategravity and avoid these issues.

As Wilson has pointed out, human-pow-ered vehicles can provide some of the exer-cise necessary to counteract the ill effects oflow gravity. But the vehicles, too, wouldhave to adapt. Let's examine how travel inwheeled HPVs would be different on theMoon, with 1/6 the gravity of Earth.

The coefficient of friction () betweentire and road surface does not vary greatlywith loading. Therefore, a bicycle's maxi-mum lean angle on an unbanked turnwould be the same on the moon, since bothweight and cornering force are proportionalto gravity. However, with less gravitationalforce, any given curve radius (r) wouldrequire a lower speed (v). You would leanway over in a turn, without being able toturn very sharply.

With one-sixth Earth's gravity,the achievable cornering radius

would berm (Vm2( ge (bere Ve) . gm) [rm)

so that if ge =6and if de = 1,

the cornering radius on the moon, rm,would be six times that on earth for thesame speed, v. For the same corneringradius, rm = rc, the speed would have

to be reduced by X = 2.45

The "instant turn" emergency maneuver,

pulling the handlebars in the directionopposite the turn, would still be the fastestway to begin a turn, but it would take sec-onds, not a half-second.

If the front wheel were to skid fromsteering too quickly or leaning too far, theresulting fall will be so slow that it would bepossible to recover control. The techniquewould be to yank the bike upright toincrease traction momentarily, at the sametime steering into the turn. Skilled BMXriders recover from low-speed front-wheelskids on earth, but anyone could do it onany bike on the moon.

Falls will be in slow motion, 1/6 earthspeed, so there will be little risk from fallingoff the bike onto the pavement. Most seri-ous accidents will be collisions, with impactdue to forward motion. And it will be hardto stop that forward motion. In lunar gravi-ty, braking would suffer as much as corner-ing, since the achievable braking force isdirectly proportional to weight. On a con-ventional bicycle, excessive front-wheelbraking causes unweighting of the rearwheel, and in the extreme case, the rider ispitched forward off the bicycle. On theMoon, pitchover would occur at 1/6 thebraking force on earth. It would take only avery weak front-brake application on a dia-mond-framed bike to pitch a rider forward,arcing slowly upwards two to three metres(ten or fifteen feet) while moving forward atfull speed.

Speed on upgrades and downgrades willvary much less from that on the level; sixtimes the slope would be needed to achievethe same acceleration or deceleration as onearth. Hills and bumps also would pose aserious hazard. A bicyclist could ride veryquickly up steep hills, and would easily leavethe surface at a hillcrest or at any largebump in the road.

Acceleration would be limited by front-wheel lift, up to quite high speeds. It wouldalso be very easy to skid when accelerating("burn rubber") with a bicycle; special carewill be necessary to avoid it when starting.Careful control of motions of the center ofbody mass would be necessary to keep fromlifting the tires off the pavement, and if theylifted, they would bounce-again. ..and...again-as in a slow-motion movie, beforethey finally settled down. At low speeds, itwould easily be possible to yank the frontwheel up into a "wheelie" and even to flipover backwards.

These conditions would also tend towardhigh speeds. Yet it would be much easier tolearn to balance and steer a bicycle on theMoon than on Earth. Since the bicyclewould fall over much more slowly, slowsteering corrections would keep it upright.On the moon, a conventional bicycle wouldbe a fool's thrill machine the way a fastmotorcycle or Jet-Ski is on earth.

Given all the problems, a recumbentwould have a major safety advantage underconditions of low gravity. And, given theprobable small size of lunar colonies, slow,three-or four-wheeled transport vehicleswould be most practical for use inside thecolonies. Such vehicles could transport loadsmany times the rider's mass. A vacuum ormagnetic device to increase friction wouldbe helpful for emergency braking, and bemore practical the larger the vehicle.

The advantage of the slow rate of fallingdoes point toward a device which might beuseful on Earth: a bicycle simulator thatwould slow down the rate of falling to allowpeople to learn to balance and steer easily. (Iunderstand that a virtual-reality company inCambridge, MA has built a bicycle simula-tor, but I don't know whether this has beenused to teach balancing and steering).Traffic behavior

Due to the low traction, the trafficcapacity of roads in moon colonies would belower than on earth. This leads to anotherpoint in favor of human power: the higherspeed of motorized vehicles would not bedesirable. Greater following distances wouldbe required, and it would be necessary toslow down more when approaching inter-sections-just as when riding or driving onpacked snow on earth.

The limitation on acceleration, alongwith the low rolling resistance and air drag(assuming low pressure of the artificialatmosphere) would afford very little advan-tage for a motorized vehicle over a human-powered vehicle for personal transportationin a lunar colony.

It is conceivable the bicyclists might rideon steeply-banked tracks in a lunar colony,for exercise and for sport. A bicycle trackwith high bank angles could approximatethe conditions of cycling on earth, and pro-vide lunar cyclists with healthy exercise,while the artificial gravity due to the cen-trifuge effect of the track prevented bonedegeneration.

Since air supplies in a lunar colony have

:l�4 Volume 13 number 3, summer/fall 1998 Human Power


to be renewed artificially, combustionengines would be out of the question.Vehicles would have to use either humanpower, or electrical power with batteriescharged from a solar or nuclear power plant.Distances in lunar colonies would be small,typically like those in a large warehouse orfactory complex on earth. Weather would becontrolled, and lighting would be availableat any time-it would have to be, since themoon's night lasts fourteen earth days.

The environmental conditions, like theoperating conditions, point to the usefulnessof human power for goods transport-andperhaps also as a stationary power source,rather than primarily for personal trans-portation.

One possibility for higher-speed travelwould be a pressurized tunnel in whichcyclists would travel in a helical path. Thiswould allow travel between colonies.Pedaling technique

The moon's gravity would have a verydramatic effect on pedaling technique. Toeclips would be essential, and cleats or clip-inpedals highly recommended, because gravitywould be little help in keeping feet on thepedals and an excessive pedal thrust couldloft a rider right off the bicycle. Saddles andbicycling shorts might have mating Velcropatches to keep the rider in place.

The most effective pedaling technique byfar will be a light spin. Standing up on thepedals as we know it will not be effective,since the rider's weight will be so small. Astanding rider would have to push with oneleg and pull with the other. Pulling hard isn'tgood for the knees, so lunar cyclists willprobably experience a new knee ailment:"moon knee." As a stress injury, this wouldbe unusual on the Moon.

On the moon, the recumbent positionwill have a much smaller effect on the wayleg weight influences pedaling. The advan-tage of being to push against the seat backwill be more important.

1) Thanks to cyclist and science-fiction fanSheldon Brown for the Heinlein refer-ences.

2) Wilson, David Gordon (1979)."Human-powered space transportation."Galileo magazine nos. 11 & 12 (doubleissue) Boston, MA 1979. A brief articleabout the pedal-powered lunar roveralso appeared in Time magazine ofJanuary 8, 1979.

John Allen is an engineer who has devotedmuch of his life to the improvement of bicyclingconditions, working with state and local agen-cies, the League ofAmerican Bicyclists andmany others. He has been a contributing editorat "Bicycling', his book, The Complete Book ofBicycle Commuting, is a [email protected]

LETTERSTIRE DIFFERENCES ON VEHICLES

There are two points on which I disagreewith Dietrich Fellenz, in "Tip over and skidlimits of three- and four-wheeled vehicles"HP (13/2/p.8 ). He assumes that all tires invehicles are identical. On many recumbentsthis is not the case. From my experiments, ifall tires have the same tread and tire pres-sure, the product of tire width and diametershould be proportional to tire loading.

If the designer tries to equalize tire-con-tact areas by combining a small-diameterwide wheel with a large-diameter skinnywheel, as is often done on short-wheelbasebicycles, it does give the desired skid perfor-mance, but also gives strange and unpleasantcharacteristics to the ride and to the steer-ing. On rough surfaces, one end feels as if itis dancing to a waltz, which the other end isdoing a jig. Also, spreading the wheels fur-ther apart seems to slow down the transitioninto a skid, so that a person with normalreflexes has a chance of recovering. Bikeswith two wheels close together can be 'outfrom under you' before you know it.Presumably, three- and four-wheelers withclose-set wheels also break free quickly.

In his discussion, Fellenz does not gointo detail on the importance of front-to-rear position of the center of gravity on skidperformance. As Huston, Graves & Johnsonmake clear in "Three-wheeled-vehicle dy-namics" (IHPVA Second Scientific Sympo-sium), the CoG should be located forwardof the midpoint of the wheels. (They alsoassume that all wheels are identical-what isit with these multi-track people?) If thefront wheels do the steering, one wants therear wheels to start skidding a little beforethe front. Thus the operator has a chance tocorrect the skid, given this remaining abilityto control the vehicle. This condition isgiven with about 55% of the weight on thefront wheels for two or four identicalwheels.

Automakers use this forward weight bias,

combined with softer suspension in thefront than in the rear, not only to give goodhandling properties at the limit, but also toimprove the ride (the "flat" or "boulevard"ride, first used in 1934 in the ChryslerAirflow). On unsuspended bicycles, howev-er, the ride qualities improve with a morerearward weight distribution (within reason-able limits). I set up my unsuspended bicy-cles with identical wheels to have about45% of the weight on the front wheel forthe ride improvement. I lean forward whenriding and encountering a patch of gravel,for instance. This setup can be modified forvehicles with non-identical wheels asdescribed above.

On a related but unconnected matter,one would think that the ride qualities ofunsuspended bicycles could be pretty wellpredicted by the wheelbase, the CoG height,the wheel diameters, the vertical flexibilityof the frame, the tire shock absorption, andthe weight distribution. I rated (subjective-ly) all of the thirty-odd bicycles I've builtover the years for ride quality, and matchedmy ratings agains my experience. The corre-lation was pretty good except for the fivebicycles that had the most-similar tires onboth ends seemed to ride about 20% betterthan expected. They had a general similarityto the Rotator 'Tiger': three had all-20"tires, and two had 16" front and 20" rear.The 20" models had the nicest steering ofany of my bikes. Is there something specialabout this layout?

-Charles Brown1875 Sunset Point, #206

Clearwater, FL 33765

Dietrich Fellenz respondsI read Charles Brown's letter with interest

and have the following comments.I am delighted to see such outpouring of

empirical and anecdotal information derivedfrom actual riding experience. I believe anydisagreement is with what I didn't write,rather than with what I wrote.

My paper was restricted to first princi-ples, and obviously one can think of a lot offine tuning for specific applications.

I agree with the observation that themixing of unequal wheel types and sizesmay introduce strange dynamic ride effects.I also agree that a short wheel base wouldcontribute to a hair-trigger skid behavior.As far as the lateral holding ability in a turnis concerned, Charles Brown seems to say:

Human Power Volume 113 number 3, summer/fall 11998 115


when you exchange a large-diameter/skinnywheel with a small-diameter/wide wheelthat the product of tire width and diametershould be held constant, all other parame-ters being the same. That is nearly the sameas requiring that the contact area shouldremain constant, which I would agree with.

I am not sure to which extent wheel loadand contact area contribute to the lateralholding ability in a turn. I only know thatboth are involved, so that I feel justified instating that a lightly loaded wheel with thesame tire would have more lateral holdingability than would be expected if it wasdetermined only by its proportionality tothe wheel load. From this would follow: ifyou want a safe rear-first skid setup (becauseonly so can you steer your way out of a skid)you should unload the front wheels slightly.

I don't think the 1934 Chrysler is a goodexample of a safely skidding vehicle. Front-loaded vehicles have a way of skidding infront first if you attempt to make a fast turn,meaning that at that point there is no moresteering authority available.

A slightly rear-loaded vehicle would per-haps skid at the rear, but you can steer yourway out of the skid if you do it soonenough. A small amount of rear loadingwould be desirable. over any front loading.

I am still hoping that someone will stepforward with a realistic tire-friction model,where the effects of tire dimensions, contactarea and wheel load, and perhaps others likerubber characteristics are properly account-ed for. That would go a long way towardsputting this issue to rest.

Charles Brown states that he can predictthe ride qualities of unsuspended bicyclesfrom design parameters. I would be interest-ed to see how this process is quantified.

-Dietrich W. Fellenz

5191 Devon Park Court

San Jose, CA 95136-2825 USA

MORE ON CLIMBINGWITH LOW BOTTOM BRACKETSZach Kaplan

I normally cruise at a cadence of about90 RPM. I go up to 120 RPM only insprints or if the particular HPV I am ridingdoesn't have a high enough top gear tomaintain a lower cadence on a particulardownhill. However I don't do much sprint-ing and I usually coast on descents unless Iwant to get a running start for a climb that

immediately follows the descent. On longrides (over 12 hours) my cruising cadenceoften drops to 80 RPM. I have found thatwith the better circulation I get with thelow-bottom-bracket position I am able topedal at slightly lower RPM in general. Ifind a cadence of 90 RPM+ more essentialon a high-bottom-bracket bike to reducefatigue in the muscles of the legs though myknees will protest if I "lug the engine" at toolow a cadence regardless of bottom-bracketheight.

In the course of extensive riding of bothhigh- and low-bottom-bracket recumbents Ihave found in general the steeper the climbthe greater an advantage the low-bottom-bracket position has over a high-bottom-bracket position. This is because the steeperthe climb the closer to pedalling upsidedown is the rider of a bike with a high bot-tom bracket. As far as I know the muscles in

the legs of the human body were designedto work most efficiently with the best bloodcirculation when low in relation to the levelof the heart such as when walking or run-ning. In this respect the typical upright dia-mond-frame bicycle has a more ergonomicposition than any of the currently producedrecumbents. This is perhaps why many peo-ple find they climb faster and easier on adiamond-frame bike than on a recumbent.The pedalling dynamics of a low-bottom-bracket recumbent are closer to that of adiamond-frame bike. Perhaps pedalling inthe low-bottom-bracket position also feelsmore natural because one does not have to

support the weight of one's leg as much dur-ing the power stroke on a low-bottom-bracket bike. Gravity can be used more toget the crankset past the dead spot at thebottom of the power stroke. By lifting theleg on the return stroke energy can be storedand used to help push the leg back downduring the power stroke. This may helpspread the total energy expenditure of theleg muscles out over a greater duration ofeach revolution of the bottom bracket thus

decreasing peak stresses and fatigue. In anycase I believe the low-bottom-bracket posi-tion enables better climbing both because ofthe way the muscles operate in relation tothe effects of gravity and through improvedblood circulation for similar reasons. I admitthis is all speculation on my part and moreresearch needs to be done in this area, per-haps some tests using heart-rate monitorsand watt meters that can measure power

applied to the pedals during the various por-tions of each revolution of the cranks.

It remains clear to me that I climb fasterwith less apparent effort on a low-bottom-bracket recumbent vs. a high-bottom-brack-et recumbent. Contrary to Tim Brummer'sspeculations, in my case the steeper theclimb the greater the advantage the low bot-tom bracket has. I would think most "nor-mal humans" with "normal" circulation andmuscle development would have similarresults if they were accustomed to riding inboth positions and climbed a steep hill withhigh- and low-bottom-bracket recumbentsof similar weight.

In Northern California we have an annu-al 320-km (200-mile) ride called TheTerrible Two. It is known as one of the mostdifficult double centuries in the US as it has4880 m (16,000 feet) of climbing some-times in the 15-20% range. Only a smallnumber of recumbents have completed thisride within the 16.5h time limit. I havecompleted The Terrible Two three times. Allthree times I was in very similar physicalcondition and the weather conditions weresimilar. In 1995 on a Lightning R-84 withstretch-fabric F-86 full fairing my time was14:47. In 1996 on a Lightning R-84 with amore aerodynamic sailcloth F-86 full fairingmy time was 14:46. In 1998 on an EasyRacers Gold Rush with stretch fabric bodystocking my time was 14:38. The ride doeshave some level portions and on levelground the Gold Rush with body stockingis about 20% slower than the F-86. TheGold Rush also weighed slightly more thanthe F-86 as it has wide tires running at 4.5bar (60-70 psi) rather than narrow tiresrunning at 8 bar (110-120 psi). I attributethe 8-minute time savings riding the lessaerodynamic Gold Rush entirely to theclimbing superiority of its low-bottom-bracket position.

I am tall for my weight, so that thismakes me more sensitive to cross windswhen riding a fully faired Lightning.However even if I gained a lot of weight Iwould still be relatively sensitive to sidewinds when riding a two-wheeled vehiclewith an aerofoil shape and lots of side area. Iknow of 90-kg (200+ lb.) riders who havebeen blown off the road riding LightningF-40s. One of my scariest cycling momentswas when I was riding a Lightning F-40with a full touring load down the shoulderof US 101. A strong gust of wind from the

Volume 13 number 3, summer/fall 199816 Human Power

right side blew me into the traffic lane.Luckily the vehicle in that lane had justpassed me.

-Zach Kaplan 235 Pacific Way

Muir Beach, California 94965 USA

Paul ButtemerStarting in 1990, I have had the

pleasure of riding and competing in HPVsbuilt by George Georgiev (the Torso andVarna streamliners). In conjunction, wehave built several unfaired recumbents forthe purposes of training. These unfairedrecumbents were designed to put the riderin the same position as is required to ridethe fully faired vehicles. The bottombracket is between 250-280 mm (10 and11 inches) above the seat, and the rider isreclined such that the shoulders are slightlybelow the highest point attained by theknees during the pedal stroke.

Since starting to ride recumbents, myriding has been split roughly 50/50 betweenthe recumbent and the road bike. My localclub has weekly time trials, and I regularlyride these TTs on either type of bike.Initially, it was obvious to me that climbingon the recumbent was noticeably slower. Myhope was that this was due to lack of train-ing in the recumbent position, so I purpose-ly did lots of hill training on the recumbent.

In order to properly evaluate the twopositions in relation to climbing, it isimportant to equalize both the weight andthe aerodynamic properties of the bicycles.The increased attack speed and the preserva-tion of momentum, both due to streamlin-ing, make it possible to "roll over" somequite substantial hills on a recumbent, espe-cially a faired one. I will cite two, althoughadmittedly extreme, examples. The roadrace at the 1991 IHPVA Championships, inMilwaukee, took place on a ring roadaround a park.

When Matt Weaver was checking outthe course, he noticed a hill and was wor-ried that he didn't have the right gearing inhis Cutting Edge to climb the hill. Afterthe race, he didn't remember the hill at all,and never had to shift out of his top gear.We had a similar experience when first test-ing Georgiev's Varna at speed. We found agood road to ride on, except that it hadone worrisome hill. This road is actuallypart of my local club's 20- and 40-kmtime-trial courses, so I knew this hill well.

On a road bike, I have to shift to the smallchainring and climb out of the saddle toget over this hill. On my first run withVarna, I gave my best sprint just before thehill. To my great surprise, I crested the hillat over 79 km/h, 22 m/s, 49 mile/h, andactually had to put on the brakes for fear ofgoing airborne off the lip!

When testing hill-climbing potential, Ihave attempted to equalize my road bikeand my recumbent with respect to the prop-erties noted above. My road bike weighs inat 10 kg (22 lb.) and my recumbent at justover 12 kg (27 lb.). When testing the roadbike, I always rode carrying a bag of toolsand a full water bottle (bringing its weightto approximately 10 kg. without drawingthe attention of my fellow cyclists), with aset of Specialized composite wheels, andwith time-trial bars set for my best aeroposition. The aero bars were not used whenthe speed dropped below about 7 m/s,25 km/h. The road bike gearing is 42-52front and 12-21 rear, with 175-mm cranks.When testing the recumbent it was alwaysequipped with spoked wheels front andback, a mirror and a small pannier bag. Therecumbent gearing is 34-44-54 front and12-21 rear, with 175-mm cranks. Coast-down tests, where the speeds were in thevicinity of 9 m/s, 32 km/h, gave a veryslight advantage of 0.1 m/s to the recum-bent. This difference increases to 0.8 m/s inthe vicinity of 14-m/s, 50 km/h, coast-downspeeds.

My recumbent could be made faster byreclining the rider further, and with theremoval of the mirror and pannier bag.(The pannier bag fits only partially withinthe profile of the rider, and adds about 40square inches or 0.026 square meters to thefrontal area. The mirror adds 9 squareinches or 0.006 square meters.)

The data below were collected in 1992,after three seasons of recumbent riding, andat a point that represents my best level of fit-ness between 1990 and the date of writing(1998). I have roughly verified these data in1998, with performances on both types ofbicycles at levels slightly below the 1992results. The data represent climbing at atime-trial level of exertion with mid-seasonfitness. I have experimented with cadencesfrom 50-90 when climbing, but I get mybest results on both bikes when pedaling ata cadence in the range of 60-75, the steeperthe grade, the lower the cadence. However,

the ideal cadence was not possible to main-tain in some of the situations describedbelow, especially on the recumbent.

1. Our 16-km course rises at an averageof about a 2.5% grade for the first 6 km. Todate, I have ridden this time trial 26 timeson the recumbent and 25 times on the roadbike. I have various checkpoints along thiscourse, but a major one is the end of thisrising grade. Here, on my road bike, I gen-erally show an average speed of about9.7 m/s, with my best being 10.0 m/s(36 km/h). On the recumbent, I generallyaverage 9.2 m/s, with my best being 9.4 m/s(34 km/h). Just for interest's sake, the returntrip on this stretch generally goes as follows:12.2 m/s average on the road bike and13.2 m/s average on the recumbent. I amusually faster over the complete 16-kmcourse on the recumbent, even though I getbehind by about 40 seconds by the end ofthe 2.5% grade.

2. Close by, we have a hill which is two-km long and rises mostly at a steady 6%grade, with the last 200 meters at 8% grade.Some of the local club members like to trainon this hill, but we have never held a raceon it because there is too much traffic for anorganized event. I have probably climbedthis hill over a hundred times on each typeof bike, and almost always go hard. On theroad bike I can climb this hill at an averagespeed of 7.8 m/s (28 km/h), but on therecumbent my very best is 6.7 m/s.

3. The last kilometer before the turn-around of our 20-km TT course rises atabout an 8% grade with the final100 meters a nasty 12% grade. When ridingthe recumbent, I am always pushing as hardas I can in my 34 x 21, but it is all I can doto maintain 3.1 m/s (11 km/h) on the 12%grade. While on the road bike, my speednever drops below 5.3 m/s (19 km/h).

4. One of our hill-climb TTs starts at the20-km TT course turn-around. This is aroad up to a ski resort, and goes skyward ata 14-16% grade for the first 2.5 kilometers.I have tried to ride up this hill on myrecumbent many times, but this is a sprinteffort for me, and I can ride only about 200meters before falling over. Perhaps a lowergear than 34 x 21 would help, but I can rideit on my road bike in a 42 x 21 gear, albeitout of the saddle at a snail-like 3.3 m/s bythe end, legs searing.

Despite my best training efforts, and mywishes otherwise, I have to concede that I


cannot climb as well on a high-bottom-bracket recumbent as I can on a road bike.It seems that the steeper the grade, the morethis is so. I have read that this could be dueto increased lactic-acid buildup, but I don'tbelieve this, because I can make my legsburn much more when climbing on theroad bike. To climb well, you have to beable to push hard on the pedals, and I feel asif I can push harder while on the road bike.I can also accelerate faster from a standstillon the road bike (even without rising fromthe saddle), and before any lactic acid couldhave built up. Perhaps the upright positionallows the legs some mechanical advantageover the recumbent position when pushinghard. On level or downhill road, wherecadences are higher and pressure on the ped-als is lower, I don't perceive a pedaling dif-ference between the two bikes, and therecumbent realizes a speed gain consistentwith its measured aerodynamic advantage.

Of course, one rider's experience doesnot allow us to draw any definitive conclu-sions, and, relatively speaking, I am not agood climber. Any mountain goats out therethat ride both road bikes and high-bottom-bracket recumbents?

-Paul Buttemer 905 Sandpines Drive

Comox, BC, Canada V9M 3V3

A PROPOSED STANDARD FORMEASURED DRAG REDUCTIONSMike Saari

Performance of any human-poweredvehicle is determined by a variety of factorsincluding weight, inherent drag (especiallyaero drag and rolling resistance) overall riderstrength and other ergonomic factors.While most bike manufacturers publish"vehicle weight", they generally do not pub-lish hard data regarding the inherent drag or"overall speed" of their vehicle. The reason issimple: there is no established standardmethod for doing so.

(Bicycle "performance" can consist ofseveral parameters including hill-climbingability, starting acceleration, level-groundsustainable cruising speed, plus variousother more subjective individualistic aspectssuch as rider comfort, etc. Here I amfocussing solely on the measurable issue oflevel-ground cruising speed.)

"Drag coefficients" (Cd or CdA) comeclose to solving this need but are not quitesufficient. For one thing, the total drag

depends on the individual rider. For anoth-er, CdA measurements do not includerolling-resistance drag elements. As a con-sumer trying to choose between an uprightand various recumbents with or withoutaerodynamic fairings or other enhance-ments, at present all that we really have togo on are various anecdotal reports, e.g.,"Such-and-such recumbent is really fast."This is not enough to convince a consumerto change her/his riding position. We needto do better.

One might object that the goal is impos-sible, given that "every rider is different".But I believe we can define a single useful,measurable criterion that any rider can use.Here is a straw-man proposal. Most likely itcan be improved upon or replaced with amuch better version, but this could help toget the ball rolling."Inherent Speed Factor" (ISF)

We would start by specifying a "baselineupright bicycle" which is defined to have an"inherent speed factor" of Ix. (This baselineconfiguration would need to be specifiedprecisely for reference, e.g., "Model XXXwith 5-bar (75-psi) tires, no add-on compo-nents, rider in upright position.") It shouldnot be necessary to specify the rider heightand weight, since we are looking to define arelative criterion not an absolute one.

Now if a given recumbent bike and riderexhibits a level-ground cruising speed of 22km/h versus only 20 km/h for the baselineupright, then that recumbent bike couldclaim "Inherent Speed Factor (ISF) of 22/20or 1.lx".

Thus a given manufacturer might pub-lish something like the following."Measured ISF as follows: 1.lx for baselinemodel; 1.2x with front fairing; 1.33x withfront fairing and body stocking; 1.35x withfront fairing, body stocking and front/rearwheel covers." (Please note: I have no idea ifthese values are realistic.)

How would a consumer use these values?The consumer simply begins with her/hisexisting flat-ground speed on an uprightbike, then multiplies by the ISF to deter-mine the approximate improved speedassuming a similar level of effort. For exam-ple, consider an average rider who can sus-tain 20 km/h on an upright bike. Asemi-streamlined recumbent with a pub-lished Inherent Speed Factor of 1.3x wouldresult in a predicted sustainable speedincrease of 6 km/h (or 26 km/h total).

Manufacturers could measure their ISFby a variety of means. We would do well topublish some precise methods, but the sim-plest would have a rider simply cruise on theupright and then on the test recumbent,striving to maintain a similar level of effort,then compare the speed difference. (Othermethods might be more precise.)

A slightly different alternative systemcould measure "Inherent Drag Reduction"(IDR). For example, a bike with half thetotal drag at a given speed (compared to theupright) would have an IDR of 2x. A simplecube-root function converts an IDR to anISE (A bike with an Inherent DragReduction of 2x would have an expectedInherent Speed Factor of 1.26x.

IDR (Inherent Drag Reduction) valuesare a bit more precise and meaningful (andalso larger so they look better), but requirethe consumer to perform a cube-root opera-tion to calculate a speed increase-messy.But in either case, with such a systemrecumbent-bike makers would have some-thing tangible to crow about and consumerswould have a realistic basis for comparison.

I propose that IHPVA define and specifyan industry standard for specifying theinherent drag reduction (IDR) or inherentspeed increase (ISF) for various human-powered vehicles. I am willing to organizesuch a committee if asked (I know sometechniques whereby any committee canreach good, fast decisions-without requir-ing a strong autocratic chairperson andwithout falling into the morass of "consen-sus" nor the mayhem of "majority rule").

We all know that a streamlined recum-bent bike is "significantly faster" than anupright on flat terrain. With a well-definedstandard we can better quantify such gains(if any), better serve the bicycle consumer,and possibly do a better job in teaching oneof the tangible benefits of recumbents.

-Mike Saari

TECHNICAL PAPER REVIEW

"Trim of aerodynamically faired single-track vehicles in crosswinds", AndreasFuchs. Presented at the 3rd EuropeanSeminar on Velomobiles, 5 August 1998,Roskilde, Denmark.

One of the fundamental problems limit-ing the acceptance of fully faired bicycles istheir sensitivity to crosswinds. While theeffect varies with vehicle design and the

18 Volume 13 number 3, summer/fall 1998 Human Power~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~18 Volume 13 number 3, summerlfall 1998 Human Power

strength of the wind, this problem is notgoing to go away. One of my favorite say-ings on crosswind handling is, "Small-boatwarnings apply!"

Andreas Fuchs has studied the literatureon this topic and then made a majoradvance-he has taken a moderately com-plex math model of bicycle handling andadded in the forces and moments due tocrosswinds. With this tool, he then analyzesa number of typical HPV configurations.

The title points out that this is a "trim"analysis, which implies that the solutions arestatic, not fully dynamic. Static analysis andtest (i.e., most wind-tunnel testing) is stillused extensively in aircraft design. Properlyinterpreted, statics can be a very powerfultool for understanding the nature of stabilityand control problems.

The paper opens with a through outlineof the literature (both bicycle and motorcy-cle) and then presents a good summary(with figures) of the bicycle model devel-oped at the Cornell University BicycleResearch Project. Next is a complete deriva-tion of the addition of the aero forces andmoments to the model. To quote directlyfrom the paper, "The resulting equationallows the designer to trim a single-trackvehicle so that it keeps its course in a steadyfield of crosswind."

The next sections describe various waysto develop the aerodynamic data required bythe model. This has not been measured toany great extent-most bicycle wind-tunneltests report the drag (in one form or anoth-er) but not the side force and yawing/rollingmoments required to locate the center ofpressure. In recent personal correspondence,your reviewer was able to supply a sample ofthis type of data, taken from tests sponsoredby Moulton and published (in part) in theIHPVA's Second Scientific Symposium vol-ume.

Final sections look at the consequencesof different design decisions, as expressedby the various terms of the trim equation.All the primary variables are included anddifferent trade-offs between fairing andchassis design are discussed. This is perhapsthe most interesting section for vehicledesigners. One notable conclusion (whichagrees with practice) is that in most casesthe center of pressure should be in front ofthe center of gravity. The paper ends withsuggestions for further research (graduatestudents take note!) and an excellent

reference list.I would like to personally thank Andreas

for taking the time to research and write thisexcellent paper. It should be "required read-ing" for all designers of faired bicycles.

-Doug Milliken

245 Brompton Rd. Buffalo, NY 14221 USA

Doug Milliken has been an IHPVA membersince the early 1980s, serving as the director ofthe DuPont Watercraft Speed Prizes and VP-waterforfive years. Starting with his wind-tunnel test workfor Alex Moulton in 1980, hehas been interested in the effects of crosswindsonfaired bicycles. He wrote "Stability? orControl?"for Human Power in 1989 (refer-enced in Andreas'paper) and has been waitingsince then for a detailed analysis of the effect ofcrosswind-which Andreas has now done.

REVIEWS

Major Taylor: the extraordinarycareer of a champion bicycle racerby Andrew RitchieISBN 0-8018-5303-6 $15.95Reviewed by Wade Nelsonwadenelson @frontier. net

Imagine cycling in 1898. Bicycles greatlyoutnumber cars. Paved roads are few and far

apart. Bicycles, or "wheels" as they arecalled, had undergone a remarkable trans-formation: The high-wheelers of yesterdayhave been replaced by all new, chain-drivensafety bikes. Saturday-night bike races heldon high-banked wooden tracks thrill largecrowds long before sports like baseball orfootball captured America's attention. Fromamong the riders of the day emerges anunlikely hero, Major Taylor, a young blackman. At a time when many white membersof society are still arguing whether blacks are"fully" human, Taylor stuns everyone bybecoming the world's fastest bicycle rider.More amazing still is the number of peopletoday who have never heard of Taylor, or hisrise to the pinnacle of bicycle racing.Fortunately, a recent book by AndrewRitchie describes this man's courage in theface of obstacles probably greater than thosefaced by any athlete at any time in history. Ifyou love cycling or just want to read anaccount of a true, untarnished Americanhero, this book is for you.

As a child, Taylor became acquainted

with a young white boy, Dan Southard. Thetwo were inseparable, and when Dan andhis mates began riding "wheels," Southard'swealthy father ensured that one was avail-able for Taylor to ride. At the time, the exor-bitant cost of a bicycle, nearly $100, wassomething few blacks could afford. Unableto participate in many other sports domi-nated by whites Taylor got on his bike androde, rode, and rode. At 13, Taylor took hisbike to the Indianapolis firm of Hay &Willits for repair. Afterwards, he performeda piece of trick riding which led to an offerof employment as a greeter for the shop.Hay soon pressed Taylor, nicknamed"Major" for the snappy military uniform hewore each day, into participating in a ten-mile road race. Taylor broke down in tearsand begged Hay not to make him ride. Toeveryone's surprise, he won. Taylor's win andassociation with the shop led to his meeting"Birdie" Munger, a bicycle racing hero ofdays gone by who was ultimately to becomehis coach and mentor.

Ritchie's book covers Taylor's meteoricrise into the world of bicycle racing, theincredible discrimination he fought, and the"world" of bicycle racing during that era.Every issue one sees in cycling today seemsto be present in 1895, whether battlesbetween sanctioning organizations (LAW vs.ACRA), issues over technology (chains vs.shaft-drive bicycles), personality battlesbetween rival superstars, even argumentsover the best possible diet for cycling.Discrimination against blacks, even in manynorthern towns, was the order of the day.Bicycle racing was "big business" rather thanthe obscure, niche sport of today. Owners ofbicycle tracks were mixed in their decisionsto allow Taylor to race, knowing that hecould be extremely good for the purse, whileat the same time under intense pressure toexclude blacks. Taylor's wins and losses ledto tremendous excitement in the crowd.

Writing this book, Ritchie turned toTaylor's daughter who maintained a com-plete scrapbook of newspaper articles abouthis career. Cycling was, in that era, the lead-ing spectator sport and reporting of cyclingevents was far more comprehensive than it istoday. The Worcester, MassachusettsTelegram provided nearly race-by-race cov-erage of Taylor's career. Interestinglyenough, an e-mail bicycling newsletter by awriter for this same paper is how I firstlearned of Major Taylor and Ritchie's book.


Taylor wrote his own biography, entitled"The fastest bicycle rider in the world"which Ritchie also used as a resource. AsTaylor's racing career took him first to Eur-ope, and later to Australia, foreign paperscoverage of "The Black Whirlwind" wasextensive. Ritchie's book takes advantage ofthat excellent coverage and truly gives thereader a taste of the flavor of the era.

Cash prizes meant that a good ridercould earn a fair income from racing, andindeed, Taylor did. Yet there was one thinghe would not do; race on Sundays. After hismother's death Taylor made a commitmentto Christianity he never relinquished. Timeafter time race promoters offered Taylorhuge sums of money to race on Sundays. Hedeclined these offers till the very end of hiscareer. It is Taylor's sticking to his beliefs,and his turning the other cheek to numer-ous fist-fights and other racetrack battlesdesired by his white opponents that lead thereader to respect him so immensely. Ononly one occasion is Taylor described asphysically defending himself, when heswings a piece of lumber at a group of whiteriders coming to his locker room ostensiblyto kill him. Perhaps by the grace of God, hemisses, and escapes unharmed. Were theremore such circumstances? We don't know,but from Ritchie's book we get the flavor ofa man whose beliefs shaped his life and hiscycling success into something all mightseek to emulate, regardless of the color ofour skin.

The end of Taylor's life is a sad story. Thedepression and a series of bad investmentsrobbed him of his considerable wealth,eventually leaving him penniless. He diedalone in a charity hospital, and was buriedin a pauper's grave. In 1948 we see membersof a bicycle club approaching FrankSchwinn, of the then mighty SchwinnBicycle Company, and obtaining the fundsto re-inter Taylor in a more suitable grave inthe Mount Glenwood Cemetery, south ofChicago, a place this reviewer hopes someday to visit. A small bronze plaque was pur-chased which perhaps says it all:

Dedicated to the Memory of Marshall"Major" Taylor 1878-1932 World'schampion bicycle racer who came up thehard way without hatred in his heart, anhonest, courageous, and God-fearing,clean-living gentlemanly athlete. A creditto his race who always gave out his best.Gone but not forgotten.

If you love bicycling, or just want in thisday of school-yard massacres, philanderingpresidents and other bad news to read thestory of a man who lived life in a way wewould all do well to emulate, Ritchie's bookis for you.

THIRD EUROPEAN SEMINAR ONVELOMOBILE DESIGNRoskilde, Denmark, August 5, 1998Reviewed by Dave Wilson

"Velomobile" is European for "HPV". Iwas asked to be the North American repre-sentative on the committee. I did not do agreat deal of the grunt work. That fell main-ly to Carl Georg Rasmussen, who producesthe famous three-wheeled "Leitra" nearby,and to Andreas Fuchs, a powerhouse fromthe Bern, Switzerland, Engineering School(university). The seminar was similar toHPV symposia in the US in that there were,at first, no papers offered and little apparentinterest. However, the snowball startedrolling and in the last few months severalauthors had to be turned away. The seminarwas different from US HPV symposia inthree respects:

(1) there was a considerable charge forattendance; (2) would-be members of theaudience had to be turned away by the pre-vious week; and (3) audience members eachreceived a bound copy of the papers uponentry to the seminar room.

I would like to add a fourth: English wasspoken! But that would be unkind. WeEnglish-speakers were very favored in thatall papers with one exception were given inEnglish.

Doug Milliken has reviewed one paper(by Andreas Fuchs) elsewhere in this issue. Ihope that we will have more in-depthreviews of other papers. My purpose here isto give brief comments on the papers andon any trends that might be of general inter-est. One observation may be made at thestart: human power is a respectable academ-ic pursuit in Europe, because a large propor-tion of the speakers had university careers orbackgrounds. That included the moderator,Carl Georg Rasmussen, a physicist who wassecretary of the technical university until hedecided to devote full time to velomobiles.There is also great enthusiasm for humanpower and the movement is very much aliveand well. This symposium had particularlystrong representation from Germany.

The first speaker was Vytautas

Dovydenas, a pioneer in "biotransport",who published his "Velomobiles" (and thusis the originator of the word) in Lithuania in1979 (subsequently translated and pub-lished in St. Petersburg (1986) and Berlin(1990)). He is a passionate advocate forhuman power, pointing out that a commonfeature of almost all cities, rich and poor,developed and developing, across the worldis traffic jams and fumes, noise and anger.

Carl Etnier, "kinetic Yankee in KingHarald's port", gave a light-hearted but valu-able paper co-authored by John Snyder onissues in pedicab design. Carl operates apedicab in Oslo, John one in NorthAmerica, each as a combinat

human power - velomobiles.co.uk · 2012. 10. 29. · different positions, and dietrich fellenz...

Documents