Click on any image to view a larger version.
The HOSSACK front suspension system consists of 2 wish-bones, an up-right and steering linkage. Similar components are found on the front of all racing cars, the only significant difference being in the up-right which has its geometry rearranged.
The wish-bones look and work exactly like their racing car equivalent The up-right performs the same task as its racing car equivalent but has its axle rotated through 90 degrees and over hung. There are few limits on how the up-right is made or what its made from, it could be fabricated, cast or layed-up. The steering link is a little clever though, as it has to pivot on roughly the same axis as the up-right. There is a handle bar pivot but this carries none of the suspension loading and only has to handle the weight of the riders upper half. Norman has run the spring/damper element in several different positions to achieve different conditions as is also common in the racing car world.
In all but HOSSACK 1, the up-right has been made of welded pressed steel profiles making a very strong and light weight shell structure. Detailed stress analysis later showed the up-right could not be made lighter even if it was made in aluminium, indeed an up-right designed for a 125cc racer, made for destruction testing weighed under 1.5 Kg. Further analysis done in Germany for the TUV approval which the HOSSACK BMW was awarded, showed Norman got his numbers a little wrong because the axle structural stiffness was 25 times stiffer than a standard BMW (over kill)!
Before Norman left Mclaren he raced a 350cc Yamsel at club level. He found that when braking very hard into Druids for example, the bike could be made to judder. This he attributed to flex and as it happened the March Atlantic cars of that era did the same thing, leaving the familiar judder marks on the circuit at the braking points. Norman was involved in chassis stiffness tests at Mclaren to investigate the March problem and it was found that increasing the bulkhead stiffness was the answer to the cars problem.
So Norman set out to stiffen up the front end of the motorcycle, his first design (1973) was a braced telly which he never built because it became clear that an additional and real weakness was the steering head structure itself. Just compare: the back end of a bike can carry most of the rider, a passenger and luggage on little bronze bushings less than an inch in diameter.
He knew that there had to be 'triangulation' in the structure, only that way could he achieve the stiffness to weight ratio he wanted. Wish-bones on bikes were GO. Norman drew several different designs but settled on this one. With this design, full triangulation was available therefore low weight as well as a major stiffness increase. Tellies back then were very flexible and no match for todays up-side downs (NOTE first seen on that little BSA Bantam Norman started out on) but even the best telly could never come near the stiffness to weight ratio this design could achieve.
Note: once you have real stiffness inherent in A design you can both soften and direct that stiffness by using rubber bushings; the reverse is not possible.
GEOMETRY: Here the design opened up new opportunities that others could only dream about.
It was possible to juggle with these and even have :- non dive, anti-dive, or pro-dive or all 3 in one set up. It was even possible to imitate the geometry that tellies provide. By choice Norman ran the axle path near vertical to limit dive but here there is an introduction problem :- riders born on tellies can't get used to bikes that don't nod on braking, this is after all how they have learned how hard they are stopping. However after some exposure to this non-dive characteristic the gains become obvious.
The frame design started with some observations then some theory.
Observation 1 - the Mclaren M23 had a front wheel weight of approx. 250lbs on each front wheel and the structure which held that wheel (top and bottom wish-bone) weighed approx. 3lbs and provided huge stiffness. Now the front end of a bike has similar weight but the forks are heavy and worse, they were not very stiff.
Note: Why this preoccupation with stiffness? :- answer the other side of the same equation is low weight! So Norman set out to design a system that could be triangulated and just as importantly symmetrical.
Observation 2 - essence of a good wishbone is that metal goes directly between the load points in straight lines, the curvy devices on the front of the ELF for example are not wishbones. Note: the ELF, the Yamaha and Tony Foales machines were all non symmetrical - this means more metal is required to achieve the same stiffness.
HOSSACK 1's frame started as tetrahedron, pyramid like but on a triangular base :- which is a collection of triangles, the strongest possible geometry. One vertex became the steering axis, and opposite axis at 90 degrees to it became the rear axle pivot. Next these 2 axis were stretched and tilted. Then some engineering compromises were required; at the front some wishbones were needed and at the back a swing arm pivot. The frame as a whole almost obeys Norman's rules of what a wishbone should be; the only compromises were that direct lines between the load points could not be achieved. Still the frame design was good enough to win a championship 10 years later, so though Norman claims he could have made it lighter there was no need. The bike weighed 215lbs in race trim, the engine being close on half the total.
Norman managed to avoid the inventors obsession to rethink everything and stuck to the basics. "Inventors seem to have the compulsion to have a go at everything :- we have seen different handle bars, different steering geometry, different brakes, you name it". Norman kept it simple and used the known numbers and hardware where possible. Norman does not have fixed views on steering head angle; trail; c/g etc, all of these need to be tested and re-evaluated in the light of the extra stiffness the HOSSACK system offers.
Norman is surprized and perplexed to find his design labled as a girder system. "Can it be these people have not bothered to look or do they lack technical savvy?" A simple look at the Hossack design will show that the system does not have a load bearing steering head as does a girder. The front steer axis of a girder system is fixed inexorably to the frame, whereas on the Hossack system this axis moves with the front wheel. The 2 systems do not share the same parts or geometry. In the racing car world where the idea was born, the parts were all familiar and the objectives understood, however this was new to the bike world where people either didn't bother to look at the parts or try to under stand their function.
Norman found enlightenment in Germany though, where people did understand the objectives. Motorrad magazine in 1993 compared Normans BMW K100RS conversion with a standard BMW K100, a Yamaha GTS1000, a TESI, and a BMW Telelever and found it technically superior.
BMW showed enlightenment too with their Telelever which came to market a year or two after Norman's/BMW conversions. BMW themselves tested more than one of Norman's customer bikes thoroughly and they understood the gains that a wish-bone could bring to structural stiffness. It was a step in the right direction though they stopped half way and kept the tellies. They also missed several other opportunities that the HOSSACK system could offer, namely, lower over-all weight, lower steer axis inertia, lower ride friction, lower manufacturing cost (the HOSSACK up-right does not have high percision parts) some additional geometry/ride options and the one sided option to match their rear end.
© 2001 - 2007 Norman Hossack. All rights reserved. The BMW logo is a registered trademark of BMW AG.
This site is best viewed at 800x600 or greater in a browser which supports style sheets (CSS).