RIBLETS FOR TURBULENT DRAG REDUCTION

Riblets are low fins approximately aligned withthe airflow. Studies at NASA and elsewhere have shownthat a drag reduction of around 7.5 percent could beachieved on a surface having a turbulent boundarylayer. As Bruce Holmes of NASA Langley pointed out (HPvol.5 no. 1 p 7), the fairing of an HPV might have justover half of its surface in natural laminar flow,giving a low drag. The turbulent drag on 45 percent ofthe surface would contribute about 65 percent of thetotal drag, so that a reduction of 7.5 percent issignificant. Holmes was advocating a larger reductionin drag by using suction at the surface (eg through aporous skin) to maintain laminar flow. Using ribletswould be a less-complex but partial solution.

The optimum shape of the riblets was found to beof a vee-form, having a height, h, given by:

13(U.5*skin-fr.coeff)**O.5

The optimum spacing, s, is 15/13 times the height. Ipresume that the skin-friction coefficient is that forturbulent flow before the application of the riblets,and could be estimated from a plot of flat-plate datagiven in most fluid-dynamics texts, against a Reynoldsnumber based on the length of the surface from thevehicle nose.

For more information, see "Optimization andapplication of riblets for turbulent drag reduction",by M. J. Walsh and A. M. Lindemann, NASA Langley LAR13286. Dive i lson

GOSSAMER CONDOR AND ALBATROSSCASE STUDY

Paul MacCready has provided the IHPVA with copies of anexcellent 60 page monograph on the design and execution·of the Gossamer Condor and Albatross Human-PoweredAircraft. The document, authored by James D Burke, isone of a Professional Study Series published by theAIAA (American Institute of Aeronautics andAstronautics). Included are 18 large photographs ordiagrams along with 30 pages of textual material. As abonus, the paper includes 13 pages of detailed drawingsby Pat Lloyd -- which originally appeared in thebritish Aeromodeller Magazine. IHPVA is offering thismonograph to members for $US 10, postage included. Anyprofits, after reimbursing Paul MacCready for hiscosts, will go to the IHPVA treasury.

Anyone who is thinking of designing an HPA, individualswho are facinated by the Gossamer saga, and those of uswho just like to read about examples of innovativeengineering should find this paper well worth itsprice. Send your orders to:

Si ncerel y,

IHPVA, Dept AIAAPO Box 2068Seal Beach, CA 90740

Paul R DesrardinsExecutive VP, IHPVA

HUMAN-POWERED AIRCRAFT BOOK NOTE

There is uch material in the following book that Ithink would be of interest to HPA designers andbu I ders:

Unconventional Aircraftby Peter W. BowersTAB Books, IncBlue Ridge Summit, PA 172141st edition, Ist Printing (c) 1984

PEDAL HEIGHT AND CROSSWIND EFFECTCHARLES BROWN -

I've been working towards developing a fully-fairedvehicle that could be used for commuting, and I'd liketo share some o the things I've found out in person.

I become more and more convinced that placing the bot-tom bracket much higher than the seat results in a lossof power. It certainly gets uncomfortable. I have builttwo bicycles in this manner and both we slower thanthey should have been. This opinion is seconded, in away, by relatively tall but fast vehicles, such as 6oldRush and Bluebell, which had their sewats higher up andthus should have had more air drag. Yet they were ableto beat some highly respected lower-slung designs, suchas Vector, which had their cranks uch higher thantheir seat. Studies of the Vector showed that riderDave Gryllsw was producing 70 watts when he shouldhave been able to manage 750 watts.

No matter how uch I ride and try to get used to aseat-scraper my legs always feel a bit tired I coastdown hills rather than pedal. The feeling is like thatexperienced when one works for some time with one'shands above one's head, and they get tired. I believeit is due to a build-up of lactic acid in thebloodstream. I built one bike with a sewat adjustablefor height. To my tastes, I would not tour with a bikethat had a bottom bracket much higher than 3 inches (80mm) above the seat. One way to avoid this and stillhave a low seat would be to build a linear-drivevehicle. I tried twisting the chain in a figure- anthaving the pedals go backward so the legs pushed whewthey were in a lower position, but this felt unnaturaland was extremely uncomfortable to pedal.

I wish someone had warned me how severe the crosswinteffect is on a fully-faired two-wheeler! One of thehighlights of my bicycle studies was speaking face-to-face with Gardner Martin. I was surprised to learn thathe hadn't been able to do uch about the crosswincproblem either. I resolved to study the problem orclosely. Crude tests showed the crosswind effect variesroughly with the square of distance above the ground;two designs can be compared for crosswind effect b)cutting out their silhouttes in a irror-image patterr( bottom to bottom ) from a sheet of cardboard. Balancethis along the ground line like a seesaw and the shapethat rises has the least total crosswind effect.The center of crosswind forces is calld the cnter ofpressure. Location of the center of pressure relativeto the center of gravity is extremely important.Ideally, the center of pressure should be behind thecenter of gravity, so that crosswinds would steer thevehicle into the wind like a weathervane. This isunlikely to happen in the real world because the orestreamlined a vehicle, the farther forward the centerof pressure tends to be. On the really streamlinedvehicles we build, the center of pressure can actuallybe ahead of the vehicle itself C[ow is this possible -Editor]. Correcting this by putting a fin on back,like the Red Shift vehicle, would require a large tailto do much good, increasing the side area and overallcrosswind effect. The fin would add a lot skin drag,partially offsetting the advantage of the fairing inthe first place. About the best we can do is to trimdown the nose of the vehicle as much as possible. TheGold Rush has remarkably little fairing projectingahead of the front wheel. It also uses a small frontwheel of about 16 inches (400 ) diameter. StephenDelaire's Rotator/Cargo Carrier has part of its fairingturn along with the front wheel. Eric Andbergen'sUelerique (Second Scientific Symposium, pages 104 106) has the front wheel between the rider's knees. Thecrosswind problem on fully-faired two-wheelers is sosevere that, if you are going to build one, I recommendbuilding it as low as you can, and keeping side area,especially in front, very low.

Charles Brown22928 OxfordDearborn, MI 48124

Chapter 6, "Other Wing Shapes", might be especiallyuseful if adapted to various HPA design function needs.

Sincerely,

Edwin G Sward215 Cambridge StreetWorcester, MA 01603

(h~free-stream velocityflidkineatic viscosity

Power requiredtWat

60C

,51

3

2

00-HP

00.

00-

0

0.

1

I

MITMonarch

I I

Mr readv I I I-Gossamer Rochelt _ALbatross MusculairI I I Z I I I I~~i I

Speed PlaneMusculair 2

,/l

LI41

I 7-

I1

I

I /

A/

1AFi'l

I

, X

I |, I ,.'Lj_.'rNZ 2m Wing Span .Long-is -t I (Low Power Flight)Plane 20m Wing Span i

|I I |||(Speed Flight)

0 10 15Flying Speed[m/s]

Fig. 2: Power RequirementofMan powered Planes

AIRCRAFT STRUCTURE.

The main structural elements of the fuselagethe vertical main tube and the stabiliser strulwhich the propeller shaft rotates. All component!made of carbon-fibre-reinforced composites.relatively large cabin is faired to conform torelatively insensitive NrACA 64021 profile.

The self-centering ailerons and rudderrlylar-covered and have a surface area conside:larger than required to achieve a high f:stability. A pair of springs keeps the rudderneutral position and thus eases the control protfor the pilot.

The pedal-power train, which weighs only 450 racing bicycles 1.2 kg is normal), transmits the Ithrough a fine chain to the carbon-fibre propishaft, supported in four bearings, and back to2.72-m-diameter pusher propeller. At barely 100 rlthe pedals the propeller runs at 230 rpm. The pipropeller, developed in 1980 for a solar aircraft,been modified for the special conditionshuman-powered flight, but still has over 86 peefficiency - see figure 9.

CONTROLS.

FIRST GOAL ACHIEVED: THE FIGURE-OF-EIGHT PRIZE.

During the three-month period of construction ofthe aircraft, the pilot completed a training programset up by the Sports College of unich. The first hopwas accomplished at the unich Mlilitary Airport atNJeubiberg at the end of May, 1984. It was done withoutfairing on the pilot cabin. At the end of only twoweeks of training, on June 18, 1984, the flight over aone-mile figure-of-eight course was achieved in 4 min.5s, almost twice as fast as Bryan Allen's flight inMacCready's Gossamer Condor in 1977.

THE SECOND GOAL: THE KREMER SPEED PRIZE.

To also win a Kremer Speed Prize, the first ofwhich meanwhile had been won by the Monarch studentgroup from FlIT, it was necessary to improve upon theprevious speed by more than the required five percent,and the plane had to be aerodynamically refined andoptimized. As a result of the test flights the pilotlearned to fly the plane perfectly. Meanwhile,MacCready with the Bionic Bat had won the second Kremerspeed prize by improving on the MIT speed by more thanfive percent. Both teams used energy storage and hencehad approximately twice the peak power available(figure 2).

On August 21, 1984, pilot Holger Rochelt, inoptimum conditions, flew the speed course in2 min 31.38 s, improved MacCready's speed by sevenpercent, and for the first time established a speed

are record for human-powered flight without energy storage.t ins areThethe

arerablylightin ablems

g (inpowerllerthe

pm ofusherhasof

rcent

The control problem has been solved veryelegantly, economically and ergonomically.

While a road racer forms a fixed unit with hisbicycle and force is transmitted between the hands andthe handlebars as well as between the feet and thepedals, the HPA pilot must keep his/her body almostimmobile above the hips to allow the controls to behandled sensitively. Precise control is more importantthan the absolute maximum in power output. That apilot experienced only in flying hang-gliders was ableto control the craft at the first attempt can beattributed to the ergonomically designed joystick,which actuates all three control surfaces. Whensteering, the pilot has only to envision that he holdsthe wingtips with his hands, and twisting of thecontrol surfaces will cause the plane to perform thedesired manouevers. Sideways tilting of the controlstick operates the ailerons, rotation about a verticalaxis acts upon the main rudder, and rotation of thehandgrips in the same way as opening the throttle of amotor-cycle acts on the elevators. A co-workerexperienced only in model-plane flying achieved a500-m-long clean flight on his first attempt.

a

i1

I0. )

'0 .4.

6)0.3.0

0 00. 0

L-

.

a:d._ 1

a

I n

0 T 42 ", .to El

MN

e0 t 4

4)a.

.,, e 4C 1 V4)4) 4) 00 M:4 0 4" C ,k * C0 3A

4N

11

Nu

4.

W aU

4. 4. C.4 C ,Cup0 *EDC a U

4. I 0 6)

0 .. .7 r a: a.- X: o

-:

_

.U.

d·I

L - !.

I /tMUL _- .- . 1--C ---I r=-.~,F�1 I..-/, , I·

_ I 7 I,,, ,,

I

Mc rea dy-Bionic: Bat a, :w

aF II I i

I ~ , !

1.II,

1C

Iel

t

THE FIRST HPA PASSENJGER FLIGHT!

Musculair 1 became an attraction at a few airshows and surprised everyone by demonstrating theastonishing manoueverability of such a large aircraft.To test the available reserves of the pilot andaircraft and to close off the 1964 flying season,Holger Rochelt on the last flight took along as apassenger his sister Katrin who, at 28kg, weighedexactly the same as the bare airplane. So on October1, 1984 the first human-powered passenger flight in thehistory of aviation took place covering a distance of500m at 5m altitude!

THE END OF USCULAIR 1 AND THE BIRTH OF A HIGH-SPEEDbUCUL5bUR.

When in the spring of 1985 Musculair 1 wasinvolved in a traffic accident on the road and washeavily damaged, the idea arose of building an aircraftpurely for high-speed flight. The large reservecapability and the good-natured flight characteristicsof the all-round usculair 1 led to the expectation ofa significant increase in performance. The author'scalculations showed that designing purely for a fastplane, a time of two minutes, which is 45 km/h (12.5m/s) for the first 1500-m course, would be achievablewithout energy storage. This speed is significantlyhigher than the new lacCready speed of 37.7 km/h (10.5m/s) of Dec 2, 1984. Based on the knowledge of thesuccessful Mlusculair 1, the construction of usculair 2(figure 4) was relatively simple. Since provenconcepts had only to be adapted for fast flight, wemerely had to re-optimize the aerodynamics, mechanics

cr

._TL;.

T

5

A7

N EI1_ _

C4

I-

ir X0T

and construction methods for the new conditions. Theaerodynamicist Dieter Althaus of the University ofStuttgart modified and optimized the successfulWortmann profile FX 76 MP precisely for the high-speedconditions (lift coefficient 0.8 at Reynolds numbers of600,000 inside and 400,000 outside) without reducingits good characteristics. To avoid torsional problemsas were experienced in rusculair 1 and to achieve theaccuracy and surface finish required for the laminarprofile, the wings were covered with a 3-mmfoam/fiberglass sandwich, and then covered with Mylarfilm (figure 7). The main carbon-fibre-reinforced sarwas made in four pieces, and designed for three g"weighed only 4 kg. Through the special design of thewing tips the intensity of the wing-tip vortex, andhence the induced resistance, was slightly reduced.

SEMI-RECUriEFrT PILOT POSITION AND ELLIPTICALLHAi'JUHLLL.

The semi-recumbent position of the pilot wasexpected to result in an improved energy balance. Thepilot cabin could be made smaller and held to a truerprofile through the use of a superlightfiberglass-sandwich fairing. We have not yet es-tablished the optimum sitting position of the pilot atwhich he could deliver high power and yet is able topilot accurately. The pedal power output was improvedby about 5 percent through the use of an ellipticalchainwheel. Proven components like the controls, therudder configuration, and the pusher propeller wereused without modification.

With a multitude of clever solutions, GunterRochelt was able to realize a very simple cleanconstruction, functional down to the last detail, andhighly efficient aerodynamically. With the mosteconomical use of materials, and an almost stingyapplication of epoxy resin, the Musculair 2 weighed,ready to fly, 24 kg (figure 4). At the very beginningof the flight tests in September 1984, it was apparent(not really unexpectedly) that the airplane, incontrast to the good-natured rusculair 1, could beflown safely only with a fast powerful flight of about250 watts pedal input, and then was very sensitive tocontrol inputs. Shortly afterwards Flusculair 2 washeavily damaged in a crash landing at an air show, butcould be rebuilt in a little over a week.

MUSCULAIR TECHNICAL DATA.

PLANE MUSCULAIR 1 MUSCULAIR 2Type HP all-purpose HP speed planeBuilder Gunter Rochelt, Munchen, W. GermanyConstruction High-wing monoplane with rear prop.Span 22m (20m for speed) 19.5mLength 7.1m 6.OmFuselage height 2.12m 1.5mWing area 16.5 sq.m. 11.7 sq.m.Aspect ratio 29.3 32.5Airfoil Wortmann FX76 MP FX76 MP

root 16% thick modified bytip 14% thick Dieter Althaus

Empty weight 28 kg 25 kgFlying weight 82 kg 78 kg(with passenger 110 kg)

Wing pressure 49 N/sq.m. 65.4 N/sq.mMin.flying speed 7.5 m/s 10.0 m/sMin.power at s eed 200 W 8.5 m/s 250 W 10 m/!Full" " ' 265 W a 11 m/s 315 W 12 m/Min. sink rate 0.22 m/s 0.27 m/sMax. glide ratio 1:38 1:37Propeller Solair 1 mod. 2.72m dia. 2.68m dia.Materials for both: "Sigri" carbon fibre

"Rohacell" foam"Styrodur" foam"Conticell" foam"Bakelite L20" epoxy resin"Mylar" film.

E~~

Ss

FX 76- MP14D

/1 ; -03'

\\ \

X. Di I

20

C.

52/ ' ~\9

0.4 k

5 LO A4 8 12 16 20 24

101t, a

2~~~~~~~~~~~.0 ~~20bL;U.C' FX76-MP160 C.

2 (/ ( ·- 0.3 52

. . +

5 12 I6 2Q-' 4 4 12 16 20 24

Fig.5:Polars of theWingAirfoil I .1. .

A NJEL' KRE1ER SPED PRIZE.

Unusually beautiful fall weather allowed continua-tion of the test flights. On October 1, 1985, Rocheltachieved a new speed world record of 2 minutes 21s atthe airport of Oberschleissheim, near unich, but couldnot better MacCready's speed by the required fivepercent. On the following day, a bicycle racer startedworking on the pilot two hours before the start to gethim physically and psychologically ready for the toughjob ahead, and brought him into super form. The courseselected was a long loop over the runuay such as tomake best use of the minute early evening thermaluplift. Hence Holger Rochelt was able to increase,under the most favorable conditions, the world speedrecord and the Kremer speed prize to two minutes andtwo seconds, or 44.26 km/h (12.3 m/s).

E

a1A.0

a

l OA.0a

Ia

30

0.2000-

1500

1000

500

0

--rFe

'50

t40

30

'20

'10

o

15- -- -

Z- ….,Thrust

1

' __.._. ,Torque_ -_ -__Power obsorbed - ff ciency

50

U 5 10 1 rlying peeu Zv

Fig.9 :Propeller Operation Graphs at 360 rpm Wind Tunnel-measured]

r. ... -

* ̂.&&.<

C.

a'

.2

0a

C

I

I

4 i 12 1 2 -4

C

C

0

Fig.8 :PropellerType SOLAIR I mod.Computation and design E.Schcberl

Minimum induced loss design andoperation in turbulence as rear prop.

Design data for Solair I Modification for Musculair 1(measured values) Diaeter 2.72 mDiameter 2.65 m Thrust at 21 N

Pitch appr. 2.5 a Flying speed 8 m/s

Thrust at 120 R Power absorbed 195 wattFlying speed 11.7 m/s Efficiency 8Power absorbed 1700 watt Pitch-angle -1.5 0

Efficiency 82 S adjustment

2.x o2 r5ton at>;2.7 ,

Prop, .',fi g/ / /t/ i-

Prop. airfoil Gttingen 795/796; 0 thick

Cf-ud- upper chord Styrodur ribRohacell core / Cf-ud- reinforcement

Styropor-sheet covering D/ iagonal weave Cf / rlar foil coveri

// / ohacell/Cf--/ ./ trailing edge

Cf-ud- tension chord Cf-rovings diagonally braced

Fig.6; Wing structure of USCULAIR 1 withWYRTrLN' - airfoil FX 76 rP (man powered)

Gf-Conticell-Gf Cf-ud-upper chordsandwich -covering / G-Conticell-Gf

' Gntfi ch

Styrodur rib / Cf-ud-tension chord Cf-reinforcedtrailing edge

Mylar foil covering

Fig.7: Wing structure of MUSCULAIR 2 with

WORTMANN- airfoil FX 76 A, modified by D. Althaus

7

._

f

--

-

GETTING PRACTICAL -Recumbents on dirt roadsJIM ROBERTS -

The jury is stilt out on the viability of current recutmbents ondirt roads. Ruts, washboard, gravel, the usual brain-dead car/trucks,dog packs and the unique characteristics of the bikes make for acourageous or foolhardy tight-rope ride.

One real killer is resonance, say twenty cycles per second. Ilive a mile off a paved road. Most of this mile is maintained by thecounty: at best, gravelled maybe once a year and graded every monthor so, according to desperation of need. It is heavily travelled inall weathers.

I built a good (read strong) Easy Racer recumbent out of a tallFuji frame, opting for the parallel brace rather than the diagonal(see diagram). I wanted to build in the vertical flex I thought nec-essary for survival on these roads. It works to the imited extentof taking most of the grief out of rocks and holes. Performance onwashboard is less predictable and you must choose your speed - thatis frequency - with care. The resonance is induced by the gravel andbuilds as a harmonic in the bike frame. A fat twenty-six-inch(660-mm) tire and a suspension fabric seat might dampen thesefrequencies enough to prevent the collapse of my spinal discs. WhatI have at the moment doesn't do that job.

My experimental electrical-conduit (kink'n go) seat is a one-piece affair that cantilevers the bearing portion of the seat andarches the back to the rear axle, rather than making a column loadout of the seat back; the back is thus preserved from direct kicksfrom the rear by structurally dissociated rear stays. The load-bearing sections are made from 1/8" (3-mm) plywood and plenty ofEnsolite and foam. One important virtue is that the coccyx is com-pletely free of seat burdens and well aerated. The umbar arch isnot vague, it helps avoid spinal compression and it allows imited'back riding' in rough conditions. The design logic is good thoughunsightly and heavy in its present incarnation. It would be betteras a composite structure. I'd also ike to try the cushions used byair-freight shippers; these would certainly act to dampen the killerfrequencies.

I was pleased to read Charles Brown's suggestions (HP, Fall 85)on frame design and his espousal of vertical flexibility. I agreethat vertical flex is highly desirable and don't doubt that he caneliminate torque and lateral flexing from his minimalist frame, buthe won't eliminate the vertical flexing that comes with high pedalloads. While he clearly regards frame flex as energy loss, I have mydoubts. In my experience, a small amount of winding (torsionalspringiness) is advantageous. It does two things: first it takes theshock load out of the top end of the stroke and, second, it seems toextend the drive period at the bottom of the stroke (presumably asthe frame and drive unwind). Winding experienced at the pedal actu-ally comes from a number and combination of sources: verticalflex/suspension wind, drive-chain stretch, and torsional flex. Tomost designers these are all regarded as apparent losses. For thetime being, I'd as soon see these frame molecules agitated. grantthat we are talking of no more than an inch (25 mm) of pedal travelin a ong-crank, flexible Easy ike mine. Brown will experience morewinding, but he may also realize a net power gain to the wheel.

This may sound nuts to designers better educated than myself.But they will agree that slaving the human egs to rotary motioninvolves osses and system strains, especially in recumbent postures.I further believe that analysis of the running/climbing man' showsthat the maximum thrust comes in about two thirds of the way throughthe power stroke and that this corresponds favorably with the stresslines and structure of legs and pelvis. I suggest that knee strainin recuimbents results from working too hard past the flat spots atthe top and bottom of the pedalling circle. That is where the unnat-ural oading occurs: too early a oading on the top eg and negativeloading (read joint separation) on the bottom. Flexibility, iftuned, might deliver a winding process tailored to how the eg worksbest. It may well be, as far as I am concerned, that the principalcontenders for the Dupont prize are employing 'winding' to advantage,knowing or unknowing.

In any event, the problem of 'compliance' with recutmbents isreal, if not for comfort and physical survival, then for efficiency.This point was welt illuminated by Ray Wijewardine (HP, Spring 85):a recumbent bike with independent suspension allows for a twenty-percent reduction in energy requirement over that without on roughroads (saving energy otherwise ost in vertical acceleration). Norhas Ray complained of osses thrugh suspension or drive winding. Iam bothered by the complexity and suggestion of dead weight when con-sidering articulated suspension systems. I feel that the need forcompliance and the damping of harmonics can be met by frames withgraduated flexibility.

Conventional bikes perform well in the dirtperfectly as reflexive springs and dampers.who makes up 80% of the "sprung weight" of theinto "dead weight" when the rider is supine.

because the etogs workSince it is the ridervehicle, this turns

I sense a ot of anxiety about flexibility as t pertains to on-trot. "Aren't we bound to suffer oscillation at speed with tong, flitframes?" This would certainy be a egitimate concern for a 'highrider, or a conventional bike where the center of gravity is somedistance from the drive/frame center ine. But in 'low riders' wehave a close proximity of forces and masses close to a common center;that makes it a different battlgame. Couldn't flexibility just asreadily enhance control as destroy it? My point is that there may beopportunities here that have been viewed in the past as iabilities.

On the other hand, 'tow riders' are definitely hard to control onless-than-optimal surfaces for different reasons. The unique charac-

teristics of ong-wheelbase, ow-center-of-gravity control were wellreviewed by Des Messenger (HP Wnter 85): greater steering anglesand greater ean, for a given turn. Very serious trouble on a dirtroad. Add to this the short footprint (contact patch) of the 20"wheel, steep steering-head angle (Easy), ight steering oads and youcan just forget avoidance maneuvers; it wants to dig a hole much ikea dull spade.

Sometimes I have the feeling that some of us are coming from wayout in eft field. We might doubt, for instance, that there will everbe a Dupont Prize for a Practical HPV (poorly cloaked challenge).You have to believe that there is a cultural entrenchment here thatmakes such a thing unlikely.

The 65-mph prize is going to be * taken with current technology,endowed with enough money, drive and peak conditions. The greaterchallenge, certainy deeper, ies with the practical HPV. Here wegrapple with ambient physical and political realities. Practicallyspeaking, I can't hack riding a bike while pumping dirty air and/oraccounting for kiler drivers of cars, uniformly blind drunk on dis-embodied power.

Let's face it, t's a dog's life for practical bikers in thiscountry. In other parts of the world - et's say most of it -thingsare different. There, bicycling represents basic mobility, the primeobject of iberation - through technology. In spite of universallyinferior status, the bicyclist does maintain his right-of-way, ascertainy as pedestrians do here (no snickering). The practicalproblem ultimately includes the where of it.

Maybe we should ook to India or China to lay down the appropri-ate design parameters. You might believe a supine bike could be madeentirety (hardware excepted) of bamboo, epoxy and silk.

Jim RobertsThe Wave ProjectChimayo, Box 408New Mexico 87522

* has beenl Ed.

Roberts' Flexible Easy Showisng Extreme Vertical Flex'Dissociated' Seat and Parallel Frame Base. weRigid Factory Type Diagonal Brace Shown as DashedLines.