Posted - 05/21/2012 : 10:23 AM
The story told in tip 256 http://www.msgroup.org/Tip.aspx?Num=256&Set= is not complete. One might say that the tip contains "nothing but the truth" but not "the whole truth", leading to not quite accurate conclusions. I will not repeat the content of the tip, so please read and understand it before reading this post.
I will be talking about the "neutral steering" - the absence of torque at the handlebars needed to keep the motorcycle in a turn at a chosen radius and speed. The tip leads one to believe that, in order to achieve neutral steering, you need to ride either very quickly in open turns or ride at a lower speed but through very tight turns.
This is true only if you assume that the motorcycle and rider are aligned, i.e., that the motorcycle's and the rider's centres of gravity (COG) are aligned. Such a situation is shown on the left in the picture above (borrowed from elsewhere on this site). If a rider leans to the inside of the turn, that is, if he displaces his COG to the inside of the turn, as depicted on the right side of the picture, the situation changes. Here's a graph from Cossalter. The blue solid curves are similar to the ones in tip 256. If the rider is aligned with the motorcycle, the inside handle needs to be pushed for the majority of radius/speed combinations.
The red dashed curves show steering torques when a rider moves his COG just 5cm (2in) to the inside of the turn. And they paint a completely different story! Only when in a tight curve (left of point A on the graph) with 25m radius and at low speed of 10m/s (meaning it takes 3.9 seconds to make a 90 degree turn) does the rider push on the inside handle. For every other turn with any radius and at any speed you need to push on the outside handle to keep a steady yaw rate.
A slight detour: Although the motorbike is unstable in this regime (see the figure in tip 256), the instability is generally not very serious. The typical times of the capsize mode are on the order of a few seconds, so the motorbike does not instantly fall over and the rider has plenty of time to correct this instability. Racing riders actually use the unstable regime of the capsize mode to help them turn motorcycles in a turn.
Now compare situations at points B and C. At point B the rider is riding in a curve with a radius of 50m with a speed of 15m/s and needs to push on the inside handle with a torque of magnitude around 0.4. At point C the rider has moved his COG 5cm to the inside of the turn. Now he is riding in a curve with the same radius and same speed, but now he needs to push on the outside handle with a torque of magnitude 0.4.
If the torque is -0.4 at 0cm displacement and +0.4 at 5cm, it is necessary that the torque is 0 somewhere between the two displacements (not necessarily at 2.5cm displacement - the relations are not linear). In other words, by moving the COG towards the inside of the curve, the rider easily achieves neutral steering.
What is more: if almost all torques need to be negative for a non-displaced rider and if almost all torques need to be positive for a displaced rider, then a rider can achieve neutral steering under almost any circumstances.
The theory and conclusions were clear to me for a long time, but before posting I wanted to confirm it with experiments. This weekend I got a chance. We went riding in relatively twisty roads, for example sections http://maps.google.com/maps?saddr=V...9.313,-0.004 and http://maps.google.com/maps?saddr=R...16.669,0.564.
My motorcycle is a Honda CB900F Hornet (AKA Honda 919 in the States) http://en.wikipedia.org/wiki/Honda_CB900F, a naked motorcycle with an almost upright seating position - not a sports motorcycle. The experiments were performed on downhill sections for a simple reason: by choosing the correct gear and using engine braking it is possible to have an almost constant speed without needing your hands on the throttle. So if neutral steering has been achieved, you would be able to simply take your hands off the handlebars and the motorcycle should continue by itself.
And this is exactly what happens. By leaning into turns and setting the correct COG offset to the inside of the turn, I was able to let go and hover my hands just above the handlebars while the motorbike kept turning at constant radius and speed.
To prevent any unnecessary discussion: I am NOT claiming that I can body-steer the motorcycle. I am fully aware of the fact that counter-steering is the only way to effectively turn a motorcycle.
The procedure in each turn was to choose, set and stabilise the speed for the turn, use counter-steering to enter the turn and simultaneously lean beyond the motorcycle's lean angle. Once the transitional phase was completed and the motorcycle had settled down, I took the hands off for the remainder of the turn. Finally I used hands-on counter-steering to straighten the bike at the end of the turn. I was able to repeat this sequence for many of the turns on different sections of the road with different speeds, road inclinations and turn radii, so it was not just a fluke. The times were short enough that I didn't notice any capsize instability. The speeds were of course deep in the counter-steering regime.
I do not have access to other motorcycles. I also don't have the time to implement equations in computer code to be able to reproduce the steering torque graphs for other geometries. So it is quite possible that for some class of motorcycles (e.g., cruisers with a large trails) the geometry always forces the required steering torque to be negative. On the other hand, I don't think my motorcycle is anything special, yet it easily exhibits neutral steering, so I would tentatively conclude that neutral steering is achievable on many motorcycles.
There's a simple way you can test at least a part of what's been discussed here. (I have just thought of this, but will not be able to test today or tomorrow, because it is raining cats and dogs here). If you look at curves for turns of radius 100m and higher, you will notice that none of the torques fall below 0.5. In other words even in a very long turn you need to press on the outside handle at any speed. It seems logical to be able to extrapolate this to curves with an infinite radius (i.e., straight lines).
So if you move your COG to the right of the symmetry line while riding in a straight line, the motorcycle will lean to the left and the combined COG should remain vertically above the wheels. If the ideas in this discussion apply to your bike, you should have then have to press the left handle to keep the motorcycle going in a straight line. This is consistent with the need to counteract the leftwards-directed camber thrust (see tip 48 http://www.msgroup.org/Tip.aspx?Num=048&Set=) which is the result of the left lean of the motorcycle.
Posted - 05/21/2012 : 6:12 PM
| Thank you for this detail.
I was taught the technique in a "Pro Rider" course a few years ago. On my Ultra Classic I could move my butt either side of the seat and partially onto the saddlebag. In this position I could make tighter curves/circles without scraping a peg.
I like knowing how to give myself extra cushion if I missread a curve.