upright seat infant nacelle

The Two-Wheeler = a full-suspension offroad baby buggy

...sports equipment for the deteriorating father...
...an analysis of my latest practical hack.

(over some photoes with some new friends:)
":-O Your kid is ill? I mean - is he *handicapped*?"
"Oh no, he's perfectly healthy. Just awfully lazy to walk :-)"



As a villager, and a child of technically educated parents, I've always been a DYI tinkerer/gadgeteer, and I've always enjoyed roaming the rural landscape around the house using just my human power. Now that I'm adult, it has just so happend that while I'm a general IT geek by profession (= an urban smartass), I spend a lot of time during weekends in the countryside. Me and my wife both like to walk, and our kids sure don't make that any easier. Especially our first-born son kept providing me with plentiful stimuli to contemplate the design of commercially available baby buggies, and kept reinforcing my focus on this hobby project (first on the sketching, later on construction) - imagine that around 8:30 p.m. you finally manage to drive the defiant junior into bed (after an afternoon spent in turns by towing him sluggishly by the hand, trying to form him, or carrying him on your shoulders) and now that he's finally asleep, the choice is yours: fall asleep yourselfs as well, or try to be creative at the drawing board, or go finish drilling those two holes in Aluminium that you din't manage to complete yesterday before the kid's evening bath time...
In the end it took me two years, from the first scratch of a pencil (okay, it was really a mouse) to the accomplishment of my perpetual motion vehicle. By the time I've finished the "stroller variant", my son has just about grown too old for that (finally started to walk without whining). We went on just a few short rides. So I penciled and built the "infant cab" variant during the following winter, just in time for the second-born baby girl and for the spring weather :-) And fortunately it seems that she enjoys riding "lying on the belly" (supported on her arms), holding her head up high, facing forward - just like her brother a few years earlier.

What it is and what it is not

What it is:What it is not:
  • a freeride two-wheeler
  • fully suspended, with lots of travel
  • for bad roads (and even moderate offroad terrain)
  • a one-off homebrew hack
  • a pretty bulky monster
  • somewhat collapsible
    (wheels, the handle - enough to fit inside the trunk of a Citroen Berlingo)
  • weight slightly over 10 kg
  • an urban buggy
  • a buggy small enough to be stored in a city appartment house
  • a bicycle trailer (could be attached,
    but it doesn't sound very sane...)
  • a commercially produceable model

Advantages and disadvantages:

+++ bobs up and down neatly on the suspension springs
+++ perfectly maneuverable (turning is easy), but it's just as easy to maintain a straight course!
+++ generous wheel spacing => many holes in the road can be "passed between the wheels" (and those that cannot are handled by the suspension).
+++ it allows you to stride swiftly - actually you can even run, but you might kill your baby if you trip and fall
++ a good enough ground clearance / ride height (axle to ground) -> caters for the occasional rougher trails and offroad terrain
+ fits inside the trunk of a car (as long as your car is an Octavia Estate or a Citroen Berlingo, and there's not much else you need to transport in the trunk - tested with 5 passengers)
+/- doesn't have any brakes (doesn't need any - the two-wheeler falls forwards or backwards immediately if you let go of the handle :-P )
- leans a bit sideways on sloped terrain, exagerrating the slope (due to the soft independent suspension)
- doesn't have a bicycle-style prop stand / kickstand - can be parked on the tail or on the nose (it's a good idea to have a piece of string or rubber band in your pocket, to be able to tie the handle to a tree or a fence, if the kid falls asleep and you want to take a break, or you return home from a walk while the kid is still asleep)
- it's not vandal-proof - not very robust against clumsy handling (especially the suspension drawbars, when disengaged from the guides for transportation, are susceptible to accidental bending)
-- impractical for city use (due to its dimensions)

Kid asleep in the cab, leashed to a fence

Analysis of requirements, inspirations, problems solved

The following judgements regarding commercially available baby buggies are admittedly biased (in a negative way) by the motivation of this work, which was to find a solution for offroad / cross-country riding. The professional designer of a mainstream commercial baby buggy faces several obvious limits / requirements:

  1. Collapsibility - the collaped buggy must fit in the trunk of a lower-class hatchback car
  2. Transportability by public transport, storage feasibility, ability to pass through supermarket floor = narrow wheel spacing and short length including the handle
  3. Mass producibility at a reasonable cost
  4. Immunity to careless handling; easy and foolproof manipulation
  5. Intended for riding in mostly urban environment (sidewalks, streets, interior) = limited tolerance to rougher terrain is okay, very basic suspension is acceptable.
  6. Bicycle trailers are intended for riding on asphalt roads or other well-finished surfaces, with a focus on stability (low center of gravity, low clearance, no suspension) and super-light construction.

When designing my offroad vehicle, I took pride in (and had the luxury of) ostentatiously ignoring those requirements :-) The resulting design may make you think that I'd been reading Alice by Lewis Caroll (I hate that book) and using hallucinogenic drugs (I refuse any and all addictive substances, except for chocolate).

Today's standard buggies for "varied terrain" have wheels maybe 8-10" in diameter at best, with inflatable or solid tyres maybe 2 inches thick. I have experience with "3-in-1 combi-strollers" by Patron (a popular domestic brand) and Mutsy (the international behemoth). In both cases you can ride plausibly on a good, even forest trail - but holes in the road or trail have to be carried over or traversed slowly and carefully, sometimes you need to prop up the buggy sideways. Pinecones under the wheels on solid ground are pretty annoying, a larger area of coarse gravel is a pain. Many a hefty pothole in the road doesn't fit between the wheels (spacing too narrow). Even the slightest maneuvering (turning) means that you either have to lift the handle a bit (which my wife doesn't like to do), or you can force the handle down with a gut feeling that it must break any second now (which my wife rants about). Mutsy have an interesting invention - a pivoting joint in the middle of the undercarriage, which does make turning somewhat easier, but exhibits affinity to the center position and "swerves its tail like a fork-lift" (and is not fit for rough terrain, because it tends to bend at inappropriate moments on sideways slopes, thus ruining stability).
Suspension on baby buggies, if there is any, tends to be pretty hard and without an option to adjust the spring rate to the current weight of your child => the buggy "rattles" on bumpy roads, which doesn't make the child nor the parent any happier. In terms of suspension, seems to me that even the classic deep-cab prams of the seventies/eighties were better - I'm referring to the classic design where the cab, the handle and the "trestle" were suspended on leather bands against a bare solid-frame undercarriage.

The degree of "joltiness" on bumpy terrain is a result of two basic factors: wheel diameter and suspension thickness (spring rate). And it also has something to do with the overall construction of the undercarriage.
As far as wheel diameter goes, I recall pictures of early prams from the beginning of the 20th century, having wheels of 15 or more inches in diameter, where the front and rear wheels used to overlap (to keep the wheelbase = axle spacing reasonable). Perhaps the cobbled roads of those times were not as smooth as today's asphalt and flat interlocked concrete cobbles. I.e., nothing really new on this front - plenty of prior art.
As far as spring rate goes, this is fairly easy to calculate. The input details into the formula however depend on the overall arrangement of the undercarriage, on the type of the springs used, on the suspension system's geometry etc.

In my early thoughts, inevitably I had to start from the general arrangement of the undecarriage. I did consider a four-wheeler. A two-axle undercarriage with four wheels yields some comfy stability and, with appropriate suspension, perhaps better "jolt balancing" than a two-wheeler - but its maneuverability can be pretty poor. Especially with soft suspension (springs with long travel), maneuverability would converge to zero (with all four wheels on the ground all the time), considering that any design with steerable wheels would be difficult to make in DIY conditions.
I have dismissed the four-wheeler on that basis quite early on, and continued to develop the "two-wheeler idea" only. I was also fairly quick to dismiss a "thick axle" (without independent suspension) - that would spoil all the fun :-) Somehow I had to design independent suspension. Which brings up the question of suspension style / overall geometry. I was wondering about half-axles swinging sideways (along a lengthwise axis/hinge), vs. each wheel on a "dragged arm" hinging along an axis perpendicular to the lengthwise direction (think about the rear axle in a Citroen Berlingo). And fairly quickly I decided for the latter variant - I assumed it would be easier to make in DIY conditions, and expected relatively better behavior with long spring travel over a short distance covered = lesser axial/sideways/torsion strain on the wheel spokes. Another advantage is "inherently convergent geometry" when riding forward.

Next, I had a tiny problem to solve: where do I get the wheels, and how do I mount them to the suspension arms? This was a long-lasting show-stopper.
The one thing I was sure of was: I needed bicycle rims with the common wire spokes, with maximum available diameter, i.e. 28" rims were a clear choice on the market of bicycle components.
Considering my ultra-light passengers, it seemd quite reasonable to use "lefty" hubs (single side mounted) - except that the contemporary double-sided hubs with "quick relases" make single-sided mounting very difficult, and I wasn't happy at all about the other option, which was to use the cheapest / poorest quality / heaviest hubs with the legacy "threaded rod axle". But on the other hand it would be difficult to make a double-sided suspension "fork", suitable for perpendicular mounting... Cannondale has been selling lefty bicycle hubs for several years - for some ridiculous price (considering my needs and budget).

This particular technical detail got me stuck for quite a while (before I even started to sketch something). I really had been carrying in my head the idea of an "off-road two-wheeler" since the first time that I pushed a baby buggy with my first-born son = the basic concept dawned on me shortly after his birth. But because at the time I didn't know about suitable lefty hubs, this "project" remained stalled until my son was two years old, which was when I finally stumbled over the website of a local company called Delfi, which sells components and complete wheels both for cycling, and for wheelchairs.
From them, I finally ordered two complete 28" wheels with aluminium cycling rims and wheelchair lefty hubs, along with wheelchair quick-release axles. The quick-releases are swift, simple and reliable - mounting or unmounting a wheel is a matter of 3 seconds, it's a similar system to what Mutsy have on their buggies. I also ordered original housings for the quick-releases, to be embedded in the suspension arms - they're really just a piece of precise pipe, threaded on the outer surface, with a pair of nuts (to be inserted in a hole and tightened). Only after I received the components did I notice, that the neat aluminium hubs carry a logo of "Sachs RehaTechnic" => the hubs seem to come from the traditional German maker of bicycle components, who also makes some wheelchair components... Actually it seems that this business, originally run by Fichtel & Sachs, has probably changed owners several times during the years - among the later owners you can find brands like Mannesmann, SRAM and currently Acelo.
I purchased two extra "quick release housings" and used them to mount the swiveling arms to the brief shafts, coming from the chassis frame (round steel rod, 12mm diameter).
Only after it was clear, that I had the wheels and their lefty mount all together, only then I started to sketch.

As for the spring-loaded element in the "suspension" subsystem, eventually I cobbled that together around compression springs (= the ultra classic spiral type spring). Again I spent quite some time looking for a source = where to buy stock products or have them custom made... until I was delighted to discover a marvellous catalogue on the website of a company called Hennlich = a "local" supplier of various machinery components (as far as I can tell, the group is HQ'ed in Austria, with historical roots in the formerly German fringe of what today is Czech Republic).
I chose springs that are "safe against buckling" (so that they don't need guidance) and I arranged precise parallelism of the push-saucers (spring seats) by mounting them both on a common shaft (drawbar), which itself hinges on two spindles. If I merely mounted the spring seats flat (or even just fixed) against the suspension arm and the buggy's frame, the seats would vary their mutual angle, and would let the spring buckle.
I chose springs made of black steel - these have slightly better mechanical properties than stainless steel springs. Based on the expected weight of the buggy and the weight of the child I estimated the medium force on the spring (for the medium point in the spring's travel) and for this force, I was looking for springs preferably lean, but with maximum possible travel. I ended up choosing two spring models (a pair of each) on the edge of "buckling", for 10 and 20 kg of buggy payload, with reasonable diameter and over 20 cm of travel. I ordered another pair of shorter springs, just in case I ever wanted to use the buggy as a bicycle trailer - although this was never my primary intention.
Within the suspension system, I have let the springs lean straight against the wheel axle (without leverage).
This solution is undoubtedly optimized for DIY building. The suspension subsystem could certainly be made more robust and vandal-proof, with shorter and more stable springs - but that would require more precise metalworking and more complicated bending/welding. That is beyond my capabilities.
It's clear that the suspension is relatively very soft (long travel) compared to the buggy's wheel spacing - the buggy tends to "exagerrate sideways tilt" when riding on a sideways slope. This effect is nevertheless not fatal, and it's an acceptable price for such luxurious handling of rougher terrain. Whenever I run straight through a gaping pothole, and the buggy just lazily bobs down and up a bit, it's difficult for me to suppress a gleeful smile :-) It would be a pity to sacrifice independent suspension in order to solve the sideways tilting, and it would be braindead to carry along some additional hardware for "active tilt compensation" (the buggy should be light and human-powered).
Commercial buggies closely follow the terrain in their tilt, and are quite tall, which means that it's difficult to keep them upright on rough terrain - sometimes you have to "push them upright" from the lower side. Compared to that, my two-wheeler can be persuaded a bit by twisting the handle. Imagine a sloping meadow, heavily ploughed by wild boar, and you need to cross that - if I had the choice between a commercial buggy and my wavering two-wheeler, I'd choose unequivocally the two-wheeler. I've actually tried both :-)

In my fantasizing about an "offroad two-wheeler", a baby buggy on tall suspended wheels, I really had several "prior art" inspirations:
perhaps the most important inspiration is the two-wheeler gardening cart, on which we used to wheel ourselves (each other) while we were children. Try asking Google Images about "dvoukolak" to get a picture. I'm referring to the big model on 26x2" wheels, big enough for two standard fruit&vegetable crates/caddies (standard here where I live anyway). From that experience, I had a gut feeling for what maneuverability is possible, and also how important it is to precisely balance the center of gravity over the axle, as a way of preventing unnecessary fatigue on part of the riksha driver. At this point, I've also concluded that the handle must be long enough to allow for swift striding, without the driver ever kicking the cart (a common problem with the commercial buggies and the gardening cart). The handle's very grip area must also occur at an appropriate height above ground (i.e. around waist level) for a man around 180 cm tall (I'm 185 cm, my wife is 180 cm). A long handle has another advantage: the force vector, produced by the driver pushing the cart, forms a narrower angle with the ground - as a result, the driver pushes "more forward and less downward" - thus requiring less total force on the handle, for a particular forward push required.
Apart from the gardening two-wheeler, obviously I also recalled the light two-wheeled coach drawn by a horse (called a Cab, abbreviated from Cabriole) and the semi-historical human-drawn Japanese cart called the "rikshaw" = a lightweight two-wheeler, suspended on slim leaf springs.
I have concluded fairly early on that, if I want my buggy to boast suspension with long travel, it won't be much use as a bicycle trailer. After that realization, I didn't bother to optimize my cart in that direction. Pushing some 20-35 kg on a cart through moderate offroad terrain (on foot) is a nice physical exercise and fun. Dragging the same behind a bicycle (+ some aerodynamic resistance) is exhausting, boring and dangerous. I have nevertheless taken some inspiration even from various commercial bicycle trailers - maybe from some cheaper alternatives rather than the holy Croozer himself: I made it very clear in my mind, that my "stroller" certainly will NOT have a simple piece of rug for a seat, without any protection from below (against twigs, rocks and god knows what else). I decided that I'd provide my children with a proper, geometrically comfortable seat with head support, with generous cushioning - even at the expense of a somewhat higher center of mass.
I was also toying with the idea that I could as well build just an undercarriage equipped with "ISOFIX sockets", which could be used to carry any compatible car baby seat. But I quickly concluded that 1) it would not be much simpler than to make a complete seat, 2) car baby seats are not exactly light-weight, and 3) nor are they very comfortable / cushioned (!). My seat (the stroller variant) features an optimum™ fixed comfortable angle between the "bench part" and the "backrest part" (the angle has been measured on my favourite kitchen chair), 3 cm of elastic polyurethane (PUR) foam allover the surface of the seat as a stuffing / cushioning material, an adjustable footboard and headrest, fixed but fairly wide armrests, and this whole seat assembly can be tilted about 45 degrees backwards, to allow a "sitting" child to take a nap in a position that's fairly comfortable for its jugular spine.

The frame of the undercarriage and the two "hockey sticks" that are a key part of the stroller seat involve welded joints. I've had the welding done by a specialist. Many joints however are performed by nuts and bolts, because it would be expensive to have everything welded and again I'd have to have it done by someone else = I wouldn't be able to keep an eye on precision and various "DIY pitfalls" (the aluminium welds could sometimes get in the way of neighboring parts, and thin aluminium material tends to "melt through" when welded.)

The "stroller" seat is mounted above the undercarriage frame using four struts with trapezoidal geometry:
Trapezoidal suspension of the seat
The purpose of this trapezoidal suspension (with variable angles) is to approximate a virtual pivot located in the seat's center of gravity. The "adjustable trapezoid" (and hence the seat's tilt) can be locked in the desired position using two bolts with winged nuts, each holding onto a sideways "cam plate" (a metal plate, 2mm thick steel, with an arc-shaped slot for the tightening bolt). This arrangement allows for continuous adjustment between two extreme positions: "sitting upright" and "bent far backwards / lying half-way down" (= the sleeping position). The whole seat assembly tilts as a whole = the bench + the backrest + the headrest + the footboard (these parts retain their mutual position). The footboard and the headrest are height-adjustable against the seat's basic frame.
In order to establish an approximate center of gravity of the finished seat assembly, I hung the seat with my son sitting in it on nylon rope, and I photoed the seat in two slightly different positions (tilt angles). It took quite some effort to persuade the kid to sit steady for a few seconds in each position, but we've managed that. Afterwards I overlapped the two photoes in software (imported the bitmaps into an Inkscape drawing, set some transparency, rotated and aligned), intersected the "plumb lines" and I was able to come up with a guess at where the center of mass was :-)
Measuring the center of gravity
The precise location of the center of mass is certainly dependent on the child's height and weight, and the "virtual hinge" created by the trapezoidal suspension also tends to deviate from the approximated rotational center in the trapezoid's extreme tilt positions. Nevertheless, even using these approximate techniques, the resulting balance of the two-wheel cart is surprisingly good, quite insensitive to the passenger's particular parameters or the seat's current tilt.

The "lying rider cab" can be used for a ride on the back or on the belly (if the kid is old enough to support itself on its arms, which happens to be a recommended exercise at this age.) I made the cab deliberately somewhat wide, so that the child can spread the arms wider when riding "belly down", for better stability. And a younger infant can have some toys in the cab. The windscreen is especially useful when riding "belly down" and is deliberately wide, to give the child a good view (wide viewing angle).
Kid on its side and looking around
It's a great sight to watch your kid, who doesn't want to exercise "belly down" at home and is normally bored all day long, to watch him or her insist on "belly down" when riding the two-wheeler, and to watch him observe what's ahead and comment on the scene - I recommend that you take a walk in a gang, and maybe take a dog along. The windscreen was inspired by baby buggies that feature a removable "overcoat" on the cab, with a front window underneath, filled with "nest fabric". Our three-in-one Patron had that - and for a period of time, our first-born "heir to the throne" refused to ride otherwise than "belly down".
The description above sounds rather optimistic, and indeed at this point I should offer a grain of salt. Riding belly-down is certainly tiresome for the kid, and obviously the buggy won't repel hunger, growing teeth or flatulence. A longer ride on rougher roads and trails (let alone off-road terrain) in itself is tiresome for the small child - and especially during first rides you need to understand that the kid hasn't quite gotten used to the buggy yet, and can have a problem falling asleep in the new buggy (especially if it's used to taking its after-lunch nap in its own bed, which doesn't wobble about and shake). I would say that the infant's preference regarding its favourite sleeping place tends to develop over time. For instance our baby girl used to fall asleep immediately as soon as we laid her down in a (commercial) buggy. Later there was a time when she preferred to sleep in her own bed with her favourite hairy blanket (rug), which she would rub on her face lengthily and happily before she'd finally fall asleep...
=> it takes just a bit of common sense and sensibility to the kid.

The kid enjoys riding a buggy when it's tired. It may even take a nap. But if it's relaxed, a lengthy ride quickly becomes boring. This is not my own discovery - I've heard the same thing from parents who go cycling a lot, with a baby seat or with a trailer (Croozer et al.).
Be cautious about how you dress your child for the ride. A "fast" ride in an open stroller means airflow (draft), which can be uncomfortable to the child (especially in the face). You don't even have to run, even a swift stride can cause "too much wind". Pay attention to how you dress your child - similar to a cycling baby seat. A sitting child doesn't move, isn't physically active during the ride, doesn't develop heat, can easily feel cold, has a completely different perception of temperature from the parent pushing the riksha.
Don't make your walks/trips too lengthy. Don't exagerrate the distance nor the time your child is strapped to the buggy. Do care about your child. Don't try to shake it out of its skin. If needed, my two-wheeler buggy can cope with various sorts of nasty surface: a meadow ploughed by wild boar, a decaying pile of twigs left behind by loggers on a forest trail, a heavily washed forest road with free rocks as big as a fist, across a roadside ditch and up the following six-foot slope... but that doesn't mean the baby enjoys such stunts.
Choose roads that you'd happily ride on a bicycle yourselfs.


The Sketchup model contains both the "stroller" seat and the infant cab (which was created later). Each of these two variant has two or three specific layers - which you have to disable/enable by hand, depending on what you want to see.
Infant cab

The Sketchup model ultimately isn't perfectly real. Some tiny parts are missing in the model, because I was too lazy to draw them - such as two pieces of a double housing for the ball bearings holding the swing-arm pivots, lathed from 40mm Aluminium rod. And e.g. the pipe material of the handle ultimately is not a 40x2 mm pipe - instead, I chose 25x3 mm (I took what my supplier had in stock).

Work started (sketching): May/June 2009
Buggy basically operational (stroller): August 2011
Infant cab finished: February 2012
This page displayed on the web: April 2012

Home | Contact | Česká verze