Permanent Revolution

I think of the early years of snowmobile suspension as the era of leaf springs, $2 boat-trailer shocks and bogie wheels. Springs were hard, travel was short, and almost no one considered any alternative necessary. The shocks were sealed and a user had to guess at their condition by how much rust had taken over from their cheesy chrome plating.

When Roger Skime and Arctic Cat introduced the first slide rail rear suspension, it marked one of the most significant advancements in snowmobile technology. Now at least there was some movement, or suspension, at the rear of the sled, but travel numbers remained very slow to improve.
Then came the horsepower revolution, as super-enthusiasts across North America welded, bolted, or cast special crankcases to create three- and four-cylinder engines. The power they made – and the answering multi-cylinder power from the OEMs – made a mockery of boat trailer suspension. It was as though a hundred horsepower had been crammed into a one-horse buggy – something had to be done to control the careening, leaping, and bouncing.
That something was a second revolution – softer, longer-travel suspension, with adoption of effective, controlled damping units previously developed for off-road motorcycles. No more pitiful “rigidimatic” chrome boat-trailer shocks. When this advanced suspension was later applied to lighter, less-power-laden sleds, their human-scale ease of handling and improved grip made the dinosaurs of the “horsepower revolution” obsolete in every respect except straight-line speed.
While many manufacturers dabbled with their own versions of “long travel” suspensions, front and rear, it was Polaris with their Indy IFS that broke loose the front suspensions into the realm of longer travel, and the FAST M-10 that demonstrated what a long travel rear suspension was truly capable of.
Perhaps the largest catalyst for change was the resulting handling behavior from the taller center of gravity brought about by the longer travel suspensions. Riders loved all of the travel and the comfort they afforded, but now the sleds were starting to lift their skis going around corners. Suspension tweaks and all sorts of things were tried in an effort to strike a workable balance between the travel and the handling. And then there was the always-present challenge to reduce the overall vehicle weight. These were over-the-snow vehicles, not through-the-snow machines.
This brought us to the era of cost-cutting large two-stroke twins – and then to the EPA-happy four-stroke power, with its motorboat sound. The extra weight of four-stroke engines was a problem, confining them at first (primarily) to groomed-trail operation. This naturally provoked top-priority weight-saving campaigns to broaden their appeal. Their focus was – you guessed it! – the chassis.
In your imagination, take a vertical slice through a sled chassis at the front of the seat, perpendicular to the direction of motion. What you see is a “hat-section” – the chassis itself. You sit on the crown of the hat, with your feet on its brim. Hat-section stringers are extensively used in aerostructures as a means of stiffening thin shells – such as aircraft fuselages – against buckling. Yet when they are so used, they are riveted to the shell, forming a complete box. Yet the bottom of the hat section snowmobile chassis has to be open, for that is where the track and its mechanism are located. Just for fun, take a cardboard box, close its top, and tape it closed. When you try to twist this structure, it resists strongly because all six of its faces are self-bracing. If you manage to apply enough force to twist the box significantly, it yields by diagonal buckling.
Now cut the tape, open the top of the box, and again twist it. It resists almost not at all, for the open top is now free to deform into a parallelogram. This is the traditional snowmobile chassis – a structure with almost zero torsional stiffness. This is what obliges the experienced rider to use his legs to twist the chassis to “edge in” the track, obtaining grip in the same way skiers do. Riders of the Indy “wedge” chassis would routinely (and often unknowingly) use their feet, pushing on the floorboards, to help “steer” the sled, compensating for the chassis twist and flex when cornering hard.
Now consider the engineer’s dilemma. The Marketing Department orders Engineering to improve the vehicle handling and to dump weight, but snowmobile chassis are already too weak. Taking out weight will be a disaster. What are they to do? Bear in mind that this was not just a four-stroke problem. Two-stroke makers, faced with the prospect of competition from lighter, high-powered four-strokes, fought back with the same weapons. Whatever could be done to a four-stroke chassis to save weight could also be done to a two-stroke chassis.
The first quick-and-easy weight reductions led to some loss of durability (cracking and breakage!), showing that the problem would require serious study, not just scribbling smaller numbers on drawings for the gauges of sheet and tubes materials.
The natural path was obvious – to seek ways to achieve acceptable torsional stiffness (and durability) at the least possible weight. This meant first making measurements to find out what current stiffness is – rather like a tubby person going to the scales as he begins a diet. Where are we starting from?
This is a familiar progress. Back in the late 1940s a British national racing car was created – the BRM. Its very ambitious V-16 engine eventually developed over 500-hp, and was carried in a traditional ladder chassis – of almost zero torsional stiffness. A perceptive young engineer, the late Tony Rudd, noticed that this car lost grip in corners on rough pavement, as its wheels pattered rapidly up and down, spending most of their time out of contact with the road. As a result, drivers could not hold line unless they slowed down. Changes to suspension springs and dampers had almost no effect – a classic syndrome. Rudd realized that most of the chattering motion was not the fault of the suspension – it was coming from chassis flexure, which has almost no damping to dissipate it. As he found ways to stiffen the chassis, changes to suspension variables became progressively more effective. Grip increased, and laptimes improved.
This makes perfect sense. If the suspension is attached to a structure that is itself very flexible, flexure will occur in both. Because the springiness of the chassis has no damping, it continues to oscillate rapidly after a bump is hit. This oscillation rapidly varies the load on the tire, destroying grip. Anyone who has ridden in buses has noted that when a wheel hits a pothole, there is an impact followed by a period of rapid shaking – a kind of “wubba-wubba” motion that takes time to die away. This is the flexing of the corner of the chassis over the wheel that hit the bump.
The bottom line here is that for suspension to do its job properly (including keeping the skis on the ground), chassis flex must be reduced to a small minimum. Why had this not been addressed before in snowmobiles? Because we were used to it, and accepted it as normal.
How can added stiffness be applied to a snowmobile’s chassis? Is everything different because a snowmobile makes only 3-point contact with the snow? If so, torsional rigidity might be irrelevant. But what about the case in which the sled is making a hard turn and hits something with either the outside ski or the track? This applies a sudden torque to the chassis, which now oscillates in the classic wubba-wubba fashion. Skis pop up into the air, losing steering contact with the snow, and track grip takes a nosedive.
How can you stiffen a box that’s open on the bottom? The classic case is the fuselage of a bomber aircraft, which must have large bomb bay openings in its underside. In this case, torsional stiffness is restored by running closed-section torque “boxes” along the sides of the bomb bay, formed of sheet metal. This is what you will see if you stand under the open bomb bays of a B-29. But a snowmobile’s hat-section rear chassis has no room for such boxes.
Ski-Doo found another way in 2003. Engineers know that the stiffness of a structure is related to the volume it encloses. A solid aluminum rod, weighing the same per foot as a 747’s thin, circular fuselage, has much, much less torsional stiffness because its material is too close to the center to have any “leverage” over applied stress. But rolling that rod into thin sheet and forming it into a large, thin-walled tube moves all the material far from the center, allowing it to resist stress more advantageously. The sled engineers saw that they could transmit stress from the forward bulkhead (to which the ski suspension attaches), up to the top steering bearing mount, and down via a pair of splayed-out struts to the sides of the hat-section rear chassis. These two triangles enclose a lot of space, and their members are straight rather than curved. This gives a great increase in torsional stiffness with minimum metal, thereby making suspension more effective and banishing the wubba-wubbas.
Now we hear that chassis assembly is switching from welding to self-piercing rivets. Hmm, weldable alloys like 6061 have only moderate strengths, and the welds themselves are soft in many cases. But if your structure is riveted, you can use higher-strength but non- weldable alloys like 7075. When the B-29 bomber’s wing material was changed from 2024 to 7075 it was possible to make the new wing over 700-lb lighter, yet handle an increase of more than 25% in aircraft gross weight.
A weld begins with the hot, liquid puddle. As the electrode moves on, cooling begins from the edges of the weld inward. Any materials having low solubility in the liquid metal are driven toward the center. This leaves a zone of impurities down the center of the weld as it solidifies. These impurities, being poorly integrated into the aluminum, can easily produce tiny cracks under the action of repeated stress. The weld may be very strong, in the sense of a one-time tensile test, but it is very susceptible to fatigue cracking. This has been a durability problem in snowmobile chassis. Pop the belly pan off of most any big-bore twin from a few years ago with several thousand miles on it and take a close look at the chassis welds; you might be shocked at what you find.
In some cases welded structures are post-weld heat treated in an oven to re-solution these precipitates and restore metal properties, but size and sagging make this difficult for large or thin structures. In any case, it adds expense and takes time. These problems illustrate another important reason why aircraft structures employ riveting, which is now a thoroughly understood technology.
Yamaha were first to move clearly toward the goal of higher chassis stiffness with their 1997 SX 700 and its Pro Action Plus chassis that was both lightweight and more torsionally rigid than the models they replaced. The chassis evolution continued with their Deltabox chassis in the 2003 RX-1, as being the first with a heavy four-stroke power unit in a high-performance sled also gave them the most pressing need for solutions. Any time vehicle structures are made lighter, higher-quality materials are required – materials which can tolerate thousands of cycles of increased stress without cracking. The new Hitachi-Yamaha thinwall die casting process achieves this by maintaining a strong vacuum over the melt and in the mold, thereby preventing the formation of a surface slush of aluminum oxide. Oxide inclusions in the metal act as crack nucleation sites. Casting under vacuum also continuously pumps away gaseous material that would otherwise generate traditional casting porosity – the prime source of cracking in castings. The result is complex cast shapes with fatigue properties close to those of the ultimate – a forging.
Such complex shapes used to be fabricated – usually by multiple pressing and welding operations. Cast bulkheads and other parts thus cut assembly time and increase fatigue strength. The die-cast front bulkhead on the ’97 SX 700 demonstrated the importance of having a “strong backbone” to mount suspensions to, as this model railed around the corners flat and firm. Suspension designs were only part of the equation; chassis design had just revealed its importance.
When Yamaha’s four-stroke RX-1 appeared, such complex castings enabled the engine to be surrounded (as it is in Yamaha’s Deltabox motorcycle chassis) and integrated into the front bay structure at low cost, greatly stiffening it. Yamaha claimed at the time that the RX-1 chassis, although lighter than that of the previous SRX, had twice the torsional stiffness.
In interviews, Yamaha engineers said that their chassis torsional stiffness development has been aimed at maintaining skis and track against the snow at angles optimum for grip.
Polaris, for the most part, are watching and waiting, as Arctic Cat has been up until this year. What if a really light, fast-accelerating sled – even with a somewhat floppy chassis – outsells a somewhat heavier sled with enhanced stiffness? Let’s find out – sometimes you can make an advantage out of what others consider a liability. When in 1903 Napier’s first in-line six auto engine displayed audible crankshaft torsional vibration, their sales manager S.F. Edge cheerfully called it “power rattle”.
Arctic Cat’s new-for-2007 Twin Spar Chassis resembles what Ski-Doo have done with their pyramidal structure, for it not only ties front and rear bulkheads together, it also passes twisting stress from the tunnel forward through two parallel members to the rear bulkhead. The twin spars referred to are those joining forward and rear bulkheads.
Ski-Doo’s pyramidal design, because it employs the natural stiffness of triangular bracing, achieves maximum stiffness at minimum weight. Arctic Cat employ a forged front bulkhead, which extracts maximum fatigue strength from the material.
Ski-Doo introduced the use of self-piercing rivet assembly, which has since spread to other makers. This not only allows use of higher-strength non-weldable aluminum alloys, it also enables a higher degree of assembly automation (robotics). This makes clear the fact that lower weight and chassis stiffness aren’t the only goals of all this work; production costs (and profitability) remain a huge catalyst, as well. A casting of complex shape provides what might otherwise have to be fabricated – machined, welded, bolted, &c. – thereby saving on manufacturing cost.
One unlooked-for outcome of the weight reduction campaign was reduced durability. When you simply take metal away, everything bends more, and by bending more, fails sooner. But when you employ less metal in a more intelligent way, the result can be lighter weight, increased stiffness, and greater durability. This is what Arctic Cat claim for their Twin Spar Chassis, saying, “The Twin Spar Chassis outlasted the most demanding test cycle ever employed by Arctic Cat”.
Manufacturer promotional materials give the impression that design just naturally goes from good to better, as a river flows serenely downstream. This paints a picture of engineers in a conference room, discussing how to spend their overflowing R & D budget. Nothing can be further from the truth! With snowmobile sales sliding and budgets tightening, no one has enough R & D money for anything but absolute essentials. The fact is that design is driven by forces too powerful to ignore. When sleds break, unhappy customers call, citing their warranties and talking about lawyers. This turns on the R & D tap but quick! Being able to meet EPA regulations has brought about a host of four-stroke offerings, but if four-stroke sleds are too heavy, mountain and other deep-snow users will leave them in the showroom. To help make them sell, weight reduction becomes urgent. If that weight-reduction threatens to make parts fail early, flex has to be designed out by rational process, and materials have to be upgraded. These are the real reasons we are seeing these changes right now.
When all aspects of a system are in relative balance, there is no reason for change. But when something upsets that balance, as home-made triples and fours did in the Horsepower Revolution, suddenly the suspension is hopelessly inadequate. It is fascinating to see the impact of upsets like this, and the strong engineering responses they provoke.
Another force for change is the specialist constructors such as FAST, Inc. Here, experienced and far-thinking persons try to imagine the future of the snowmobile, unimpeded by boring stuff like production cost or tradition. They build what they see as the future, and we are often amazed. Sometimes the result is a new level of performance which drags the mainstream manufacturers along in its wake. Sadly, the real money to be made from innovation seldom goes to the inventor, whose advanced machines are too expensive – or even too different – for most of us. That money is earned when the mainstream absorbs the new ideas into high-production sleds. This leaves the creative types suffering for their art, wondering if they’ll be able to meet next month’s payroll, and having to be satisfied with having changed the world a little bit.
This constant interplay of ideas, crisis necessity, and economic realities brings us some pretty nice sleds.