Engineering details of the giraffecycle

Frame geometry and access

The compact chassis uses a 3.2 m wheelbase, 3.6 m wheel track and four 1.6 m wheels. The hoof-guide centres are 0.65 and 2.55 m forward of the rear axle line. This centres the animal between the wheels and removes the long empty frame behind it. The two side-truss centrelines sit 1.8 m apart; their 4 panels are 0.80 m long and 1.1 m deep.

The animal walks in from the rear through a nominal 1.8 m corridor and backs out the same way. That dimension is between truss centrelines: the clear width between the 80 mm longitudinal tubes is 1.72 m before sleeves, hinges or other fittings. The completed opening needs its own clearance check. A side entrance between large wheels would be too short on this wheelbase. Rear and centre cross-ties retract, and their flexible bracing cables slacken and stow outside the corridor. The front bulkhead remains fixed. Split rear stub axles avoid a solid shaft across the entrance: a cross-shaft beneath the guarded floor feeds chains rising beside the wheels.

Bij het instappen keek ik minstens zoveel naar Erik als naar de giraf. Als zij een stap achteruit deed, moest hij rustig opzij kunnen gaan, zonder eerst een grendel of een riem te ontwijken.

“During boarding I watched Erik at least as closely as the giraffe. If she took a step backwards, he needed to be able to move calmly aside, without first dodging a latch or a strap.”

The upper and lower side chords are 2.2 and 1.1 m above the road. Sling loads enter reinforced nodes; tension stays support hoof guides at 0.28 m. Cross-frame bulkheads at stations 0.0, 1.6, 3.2 m restrain the longitudinal chords laterally at 1.6 m intervals. The centre bulkhead fits between the front and hind limb sweeps, below the belly. Horizontal X bracing sits in a guarded floor 0.16 m above the road, below all hoof travel.

For loading, parking stands support the frame and locked hoof guides, the rear ramp is deployed and locked, the slings remain slack and the chin boom folds away. Once the animal is in position, the retractable ties close and lock, their cables tension, and only then do the slings take weight. The entry path must remain clear through the complete movement of the animal and the hardware.

A red escape pedal beside a front hoof tray requests the reverse sequence. It is separate from the drive and mechanically locked while moving. At rest, the sequence sets the parking pawl, deploys and locks the stands and rear ramp, lowers the slings, and unloads the structural pins before releasing cable tension and retracting the ties. Pin withdrawal requires all three conditions: locked stands, a deployed and locked ramp, and slack slings. A blocked ramp keeps the exit closed. The sequencing-cam, spring and linkage allocation includes ramp deployment and its interlock; detailed geometry and operating forces still need design and testing.

De rode verf was vooral voor de verzorgers. Voor Truus telde dat het pedaal altijd op dezelfde plek zat, vlak naast haar hoef. Ze moest het kunnen vinden zonder eerst naar beneden te zoeken.

“The red paint was mainly for the keepers. For Truus, what mattered was that the pedal was always in the same place, right beside her hoof. She needed to find it without first looking down.”

Loads used for sizing

The load cases include the article’s 250 kg freight allowance. Animal, vehicle and cargo therefore total 1,330 kg for structural screening. The 900 kg reference animal sets the dimensions; other giraffes require individual fitting.

The two-times-weight case is a chosen screening load. The 60:40 side split allows moderate lateral imbalance. An animal falling against one side, a collision or an arbitrary pothole requires a separate load case. Combined braking and vertical loading is examined below; the complete joint-by-joint combination of loads remains to be solved.

Why a deep truss saves mass

The animal’s load creates a global bending moment across the wheelbase. A connected truss carries most of that moment as a couple: compression in one longitudinal rail and tension in the other. Dividing the moment by truss depth gives the approximate chord force. Treating the entire load as bending in one tube discards this benefit.

For a conservative bending case with a concentrated load at the middle of the heavier side, let W be gross weight in newtons, F the vertical load on that side, L the wheelbase and h the truss depth:

W = (900 + 180 + 250) × 9.81 = 13,047.30 N
F = 2 × 0.6 × W = 15,656.76 N
Mmax = F L / 4 = 12,525.41 N·m
Nchord ≈ Mmax / h = 11,386.73 N
V = F / 2 = 7,828.38 N
Ndiagonal = V / sin(53.97°) = 9,679.77 N

Splitting that side load equally between the quarter points halves the maximum moment to 6.26 kN·m. The harsher central-load case sets the longitudinal tube sizes; shear requires checks at other possible load positions. Carrying 65% of the animal in slings changes where load enters, while the remaining hoof load reaches the frame through the guides. Gross weight includes both contributions.

A linear, pin-jointed side-truss calculation separately assembles each member’s axial stiffness EA/L and solves Ku = F. With the central upper node loaded, the model predicts 2.62 mm of vertical deflection. It also identifies a centre upright carrying the full 15.66 kN. That upright is therefore 70 × 2 mm, larger than the other uprights. The calculation fixes the support locations and treats the joints as rigid against slip; it is a two-dimensional check of this topology.

Tube sizes, buckling and imperfections

The principal tubes are aluminium 6061-T6 with mechanically sleeved or gusseted joints. The calculation uses density 2,700 kg/m³, elastic modulus 70 GPa and parent-metal proof stress 240 MPa. A welded joint requires a strength allowance for the loss of T6 properties near the weld. Steel pins and tension rods require isolation from the aluminium. Hydro, Alloy 6061 data sheet; Hydro, Marine Standard Items.

For outer diameter D, wall thickness t, inner diameter d, area A, second moment of area I and section modulus Z, use metres, newtons and pascals throughout:

d = D − 2t
A = π(D² − d²) / 4
I = π(D⁴ − d⁴) / 64
Z = I / (D/2)
mass per metre = ρA
PE = π²EI / (KL)²       [K = 1, pin-ended member]

Euler’s P_E is the buckling load of an ideal straight column. The preliminary screen takes the smaller of 0.4 P_E and A × 70 MPa. It then adds a beam-column estimate for 3 mm of joint eccentricity and initial bow L/500. The 70 MPa limit and Euler reduction are explicit preliminary design assumptions. Code compliance and fatigue life require separate checks. MIT, Solid Mechanics Laboratory: Elastic Buckling.

Pscreen = min(0.4 PE, A × 70,000,000 Pa)
e0 = 0.003 + L/500
σcombined ≈ N/A + [N e0 / (1 − N/PE)] / Z
Check N ≤ Pscreen and σcombined ≤ 70 MPa

For the transverse bracing, a 6 mm steel tension rod with a specified minimum 4.8 mm thread-root diameter takes 2.29 kN, giving 126.7 MPa against a separate assumed 150.0 MPa rod screen, requiring a steel grade with at least 300 MPa proof stress. Crossed pairs take opposite directions of shear; chord restraint requires tensioned rods with firm anchors. Rod ends and their load-spreading attachments need their own checks.

The rear and centre bulkheads use retractable 6 mm steel wire cables. The selected supplier example has 19.60 kN minimum breaking force. Applying a chosen factor of 5 and an assumed 80% termination efficiency gives a 3.14 kN screen. Lateral load plus 0.30 kN preload demands 2.59 kN. Termination durability and bending life require separate evaluation. CERTEX, 6×19M-FC steel wire rope.

The 80 × 1.5 mm chord weighs 0.999 kg/m. A 40 × 3 mm tube weighs 0.942 kg/m but has only 21% of its bending stiffness. In this case the smaller tube reaches 111.7 MPa under the same force and fails the chosen screen. Larger diameter supplies useful stiffness without requiring a thicker wall.

Bracing is equally consequential. Removing the intermediate restraints makes the chord 3.2 m long; its ideal Euler load falls to 19.23 kN and the reduced screening load to 7.69 kN, below the 11.39 kN demand. Effective restraint requires diagonals connected through load-carrying joints. These checks also leave local wall buckling, dents, cut-outs and connection flexibility unresolved.

Hoof guides, chin support and wheels

L. Wieman described the automatic transmission and head controls in a 1967 technical report. Their supports carry concentrated loads as well as transmitting commands.

The frame’s axial forces occur alongside local bending: hoof carriages apply moving, offset loads to guide rails, and the chin boom bends. These need local checks in addition to the frame calculation. The 500 N downward mast load is a structural sizing case; the brake command is an upward chin movement.

We keken vooral naar wat er gebeurde als ze even aarzelde. Een gelijkmatige tred zag er prachtig uit, maar ze moest ook midden in een stap kunnen stoppen zonder dat de andere voeten werden meegetrokken.

“We mostly watched what happened when she hesitated for a moment. A steady gait looked beautiful, but she also needed to be able to stop halfway through a step without the other feet being dragged along.”

Guide rail: M = F L/4; σ = M/Z; δ = F L³/(48 E I)
Mast: M = Fside L; σ = Fvertical/A + M/Z
Mast deflection: δ = Fside L³/(3 E I)

Each guide has two tension stays. A 30 × 1.5 mm stay taking the complete 3.00 kN endpoint load carries 4.03 kN tension at 30.0 MPa. Its sideways component puts 2.20 kN into the floor tie; the associated column screen includes 24.5 MPa combined stress. The floor members and these stays are included in the mass schedule.

The guide rail needs supports at the stated 0.55 m spacing throughout the stroke. The mast result shows why braces are necessary for control feel even when stress is acceptable: an unbraced tip moves about 26 mm. The budget includes mast braces; the deflection calculation uses the unbraced mast stiffness.

Truus hield Erik heel goed in de gaten. Zelfs als hij maar een stap opzij deed, kon haar hoofd al meedraaien. Zijn looproute hoorde daarom bij het afstellen van de besturing.

“Truus kept a very close eye on Erik. Even if he only took a step to one side, her head could turn with him. Where he walked was therefore part of setting up the steering.”

Each wheel bracket spans 0.9 m sideways. Its upper 60 × 1.5 mm strut rises 0.30 m from the hub to the lower side chord; its lower 50 × 1.5 mm stay falls 0.64 m to the floor. Solving the triangle under the 10.45 kN wheel reaction gives 10.54 kN compression above and 12.27 kN tension below. The upper column reaches 53.1 MPa combined stress; the lower stay reaches 53.7 MPa in tension. The 10.00 kN horizontal force requires reinforced 60 × 2 mm lower end-bulkhead ties, which reach 46.0 MPa; floor ties take 55.1 MPa tension. The mass schedule includes these reinforced ties.

Wheel bracket: H = Rwheel × lateral reach / (upper rise + lower drop)
Nupper = H × upper member length / lateral reach
Nlower = H × lower member length / lateral reach

The wheel allocation is 9 kg per complete wheel, including tyre and hub. At full gross weight and 0.3 g braking, each front wheel carries 5.12 kN before road impacts. Combining twice the vertical weight with the same braking transfer raises this to 8.70 kN for an equal left/right split, or 10.45 kN at the front wheel on a side carrying 60% of the load. These are requirements for wheel, tyre, hub and bearing selection. The 9 kg allocation still requires component selection and testing against these loads. Using 14 kg wheels instead raises total mass to 199.5 kg.

Connections and lateral loads

The side chords remain continuous. As a conservative example for a critical frame clevis, the following joint carries the 11.39 kN maximum chord force. A 12 mm steel pin works in double shear, through a 10 mm central steel lug and two 5 mm cheeks; the central lug is 40 mm wide. The example follows the longitudinal chord’s load path; removable rear ties carry transverse loads.

Sleeves and gussets must spread the concentrated loads into the 1.5 mm aluminium wall. Edge tear-out, pin bending, fastener preload, lug fatigue, removable-tie stiffness and the sleeve-to-tube load path still need detailed sizing. Positive pins and retained locks must keep the retractable ties engaged in travel. Sleeved load spreading is necessary wherever these joints meet thin aluminium walls.

With a centre of mass 2.5 m above ground, 0.15 g cornering transfers 1.36 kN from the inner side to the outer side. The outer side carries 7.88 kN, roughly 60.4% of total weight. At 20 mph, that lateral acceleration needs a turn radius of 54.3 m. The rigid-body tip threshold of 0.72 g is only a geometry result: animal motion, tyre grip and bumps can impose earlier limits.

At 0.3 g, front-axle load rises to 10.23 kN and rear-axle load falls to 2.81 kN. Total required brake torque at the wheels is 3,131 N·m. Appropriate front/rear brake distribution and a separate parking system still need a design. The simultaneous torsional case of one wheel hitting a bump requires a three-dimensional analysis alongside the side-truss model.

Longitudinal load transfer = (M a_brake) hCOM / wheelbase
Lateral load transfer = (M a_lateral) hCOM / track
Ideal rigid-body tip acceleration / g = track / (2 hCOM)

Complete mass budget

The calculated structure includes the less visible transverse ties, tension rods, rail supports, local sleeves and retractable-tie fittings. Most of the rest of the vehicle remains an explicit allocation, included alongside the structural schedule in the complete-machine budget.

Structural schedule

The 179.51 kg result gives an about 180 kg design description. Structural loads use 180 kg. Both are rounded from the component sum. The 6 kg contingency is small relative to the unresolved wheels, joints and mechanisms. The finished mass remains dependent on component selection and detailed design.

The earlier 80 kg figure depended on aggressive wheel and mechanism allocations without corresponding component designs. Deep, properly connected trusses carry the global bending moment efficiently within the present mass budget. The explicit cross-frame members and guarded floor raise the estimate above the initial 170 kg aim. Reaching that earlier target would require finding savings in identified components. Heavier wheels alone move the present budget to about 200 kg, with other increases possible as details are resolved.

Effect on energy and braking

The energy comparison belongs to the later 1994 reconstruction described in the Giraffecycle article. J. Borst studied the fairings and coauthored the 1967 road-load report; A. van Loon's 1971 study examined surface limits and progressive braking. The unladen performance comparison retains the rounded 180 kg vehicle and 900 kg animal. It uses partial-fairing drag area 1.2 m², 95% drivetrain efficiency, 25% incremental muscle efficiency and 1.1 kW of resting, movement and posture cost. These physiological and aerodynamic values are assumptions. The running metabolism reference comes from a comparative mammalian equation extrapolated to this animal. Taylor, Heglund and Maloiy (1982), Energetics and mechanics of terrestrial locomotion.

The unchanged running reference is 10.725 kW. Raising vehicle mass from 80 to 180 kg adds 251 W in the mild-road case; the model retains an estimated energy saving. The article rounds the smooth and mild cases to about 45% savings. The 20 mph performance estimates apply to the unladen vehicle; loaded cruising speed remains unestablished.

At 20 mph, unladen translational kinetic energy is 43.17 kJ; the fully loaded sizing case has 53.16 kJ at that hypothetical speed. Wheel rotation adds further energy. At the article’s gentler 0.12 g deceleration, the ideal braking distance remains 34.0 m because mass cancels from v²/(2a). A 1.5 s brake-pressure ramp raises that to 40.5 m, before recognition time. Repeated stops and descent heating need brake-temperature calculations.

Giraffe proportions in the illustrations

The rendering target uses a compact trunk and long legs: about 1.8 m from shoulder to rump, with about 2.6 m from the chest indentation to the forehoof. Their ratio, 0.69, is close to the adult wild-female Masai giraffe mean reported by Cavener and colleagues. Their wild-male and wild-female trunk/foreleg ratios were 0.649 and 0.692; “foreleg” here begins at the chest indentation. Their withers-to-croup height ratios were 1.221 and 1.176. These measurements support a back that slopes down towards the rump. Cavener and colleagues (2024), Sexual dimorphisms in body proportions of Masai giraffes.

The illustration uses an approximately 2 m neck rising about 40° above horizontal, with the head independently facing forward. These selected proportions describe the reference animal. The vehicle must fit around that animal; lengthening the torso to fill the chassis gives the wrong silhouette. The scale-model giraffes follow the same proportions.

What remains to be established

The present work establishes reproducible section properties, a connected side-truss solution, selected buckling and bending checks, load-transfer estimates and a parts budget. It also specifies a clear rear entrance, with axles and raised crossbars outside the animal’s path. Passing those checks establishes a coherent concept whose construction and animal handling still require detailed design and testing.

The main unresolved work is a complete three-dimensional node model, combined torsion and impact cases, all joint details and fatigue life, wheel and tyre selection, rolling resistance, brake heat, and the mechanical transmission’s peak loads. Loading stands must support an unevenly moving animal with the rear and centre bracing retracted. Slings, head support and hoof guidance need contact-pressure, release and biological evaluation. Exact clearances should be checked through the full motion of the retractable ties and all four legs.

The proposed Tandem Giraffecycle would need a separate frame and braking analysis: two independently stepping giraffes, one steering position and two brake commands change both the loads and the control system. The wild-giraffe self-service station proposal adds a different problem: an unattended machine would have to select and check the fit of its supports for an unfamiliar animal. Each proposal requires calculations specific to its layout and operation.

The numerical tables are generated from the engineering calculation model; energy values use the vehicle power model. Dimensional records and physical tests establish tube-wall thickness, joint engagement and structural strength.

See also

• Dromedary Express — four sliding foot trays with steering and braking assigned to a seated human attendant.
• Struthio courier — two alternating foot platforms and beak-pad steering for an ostrich carrying eggs.
• Capricorn tandem — independent treadles for two goats, with the front goat steering and either animal able to brake.

Sources

1. CERTEX, 6×19M-FC steel wire rope.
2. Hydro, Alloy 6061 data sheet.
3. Hydro, Marine Standard Items.
4. MIT, Solid Mechanics Laboratory: Elastic Buckling.
5. Cavener and colleagues (2024), Sexual dimorphisms in body proportions of Masai giraffes. Table 3.
6. Taylor, Heglund and Maloiy (1982), Energetics and mechanics of terrestrial locomotion.