The key to Yamaha's MotoGP success

At the final Grand Prix event of the year, Yamaha traditionally holds a press briefing and discloses some information about that year's YZR-M1 machine and its development. This year was no different, and at the recent Valencia round Kouichi Tsuji, MotoGP Group Leader and YZR-M1 Project Leader, gave this year's presentation.

In every year's briefing, it seems the development goals for the M1 are the same as years previously: The focus for the engine in 2012 was "balanced power and fuel;" for the chassis, "maximize corner speed and agility;" and for the EMS (Engine Management System), "direct and natural feeling."

But this year marked a new era for MotoGP - or, perhaps, a return to a previous era - as capacity was increased to 1000cc. Maximum bore size is now limited to 81mm, and minimum weight was slightly increased. Additionally, Bridgestone made changes to the spec tires to cope with the increased power. According to Yamaha, the M1's power increased from "over 200" to "over 240" horsepower, or an increase of approximately 20 percent, and the maximum speed reached at each track increased by an average of 11 km/h.

The presentation notes that faster lap times this year are largely due to the higher speeds, but the improvement in lap time was not as much as expected due to time lost from correspondingly more braking. In an example using data from 2011 and 2012 at the same track, a gain of .3 seconds was seen on one straight from the increased power, but at the end of another straight .2 seconds was lost under braking. In another example, the braking point at the end of the straight moved 17m earlier, a significant change.

To cope with the increased power and changed tire characteristics, the wheelbase was lengthened and the weight bias moved slightly forward. Increasing wheelbase reduces the tendency to wheelie both under acceleration and braking, while moving weight forward reduces wheelies under power but with an increased chance of the rear wheel coming off the ground under braking.

Designers in MotoGP have constantly been moving the centre of gravity around to take advantage of certain characteristics (or minimize difficulties associated with others), and the general trend since MotoGP's inception has been longer, lower motorcycles. This layout helps keep both wheels on the ground under braking and acceleration, an increasing problem as power improves, brakes get stronger, and tires have more grip. The tradeoff with a longer, lower chassis is in cornering performance; the bike is more difficult to turn from side to side, and more lean angle must be used for a given corner radius and speed.

It's interesting to note that rider input - how much of a sacrifice in one area is worth an advantage in another - has directed this trend to a certain extent. Small changes in weight distribution and centre of gravity height can be made with minor adjustments, such as moving the rider around, changing the position of the rear wheel in the swingarm, and so on. But at some point, large objects - such as the engine - must move and new parts must be made. Yamaha arguably made the best transition to 1000cc in 2012, with Jorge Lorenzo winning the championship, both Tech 3 satellite riders reaching the podium during the season and Andrea Dovizioso finishing fourth in the championship.

Thursday, 15 November 2012 07:23

Gearing and Anti-Squat

Quite often when the subject of gearing comes up, many riders and tuners talk about anti-squat at the same time. Gearing and anti-squat are related, and when you change gearing you may also have to make a chassis change to keep anti-squat characteristics the same.

When your bike is accelerating, the weight transfer from front to rear acts to compress the rear suspension, making it "squat." Countering that, the swingarm angle and the relationship between the swingarm and chain run produce a force that acts to extend the rear suspension; this is dubbed "anti-squat." Ideally, these two forces combine to provide just the right amount of suspension compression under acceleration for optimum rear-tire traction without unloading the front tire excessively.

For example, your bike running wide on the exit of a corner when you begin to accelerate is generally a sign of the front tire unloading because of too much squat. But you can make some setup changes to add more anti-squat and offset that; these changes are especially noticeable on a more powerful bike, as there is more squat from weight transfer and more anti-squat produced from the other forces involved.

Here I will skip a lot of math and explanation, but the key is the relationship between the chain run and the swingarm. The closer the top chain run is to the swingarm pivot, and the greater the angle is between the swingarm and the chain, the more anti-squat you will have.

The changes you can make, then, to add anti-squat include: increase swingarm angle by raising rear ride height; increase the chain angle by using a smaller front sprocket or larger rear sprocket; bring the chain closer to the swingarm pivot by raising the pivot location (which also increases the swingarm angle). The opposite applies, and you can make the following changes to reduce the anti-squat effect: decrease swingarm angle by lowering rear ride height; decrease chain angle by using a larger front sprocket or smaller rear sprocket; move the chain away from the swingarm pivot by lowering the pivot location.

One scenario I often get asked about is as follows: If you change sprockets from, for example, 15-45 to 16-48, the actual ratio is unchanged but the chain is further away from the swingarm pivot; anti-squat should be less. The reality is, however, that chain angle has slightly increased at the same time, in turn increasing anti-squat. In my experience (and any calculations I've seen confirm this), a combination of sprockets that has the same ratio, such as 15-45 and 16-48 or 17-51, will have almost identical anti-squat properties.

One additional aspect to be aware of when making sprocket changes is the clearance between chain and swingarm. If the front sprocket is quite small, or the swingarm pivot excessively high, the chain could actually rub on the swingarm. All that anti-squat geometry then goes out the window and a whole new set of forces comes into play. That scenario is best avoided, and in general larger sprockets are a better choice as they keep the chain well away from the swingarm.

If you make a significant change in gearing, anti-squat characteristics can be affected and you will have to make some chassis changes to suit. And as bikes become increasingly powerful over time, anti-squat plays more of a role in chassis setup; it must be considered even on current middleweight machines, especially in race trim.

Thursday, 01 November 2012 07:25

Using data acquisition to select gearing

As with any setup decisions made at the track, it's nice to have some data to look at to confirm what the rider is saying or even make a change with no input required. A rider may say that he needs taller final gearing after a session, but with some basic data it's easy to actually put a number on how much taller to go - this can help save some significant setup time at a race event.

RPM is the most useful channel to use here, and data acquisition systems usually provide an input that accepts a tach signal. Using rpm alone you can check maximum rpm on the longest straight on the track, minimum rpm in each corner, and see how the rider is using the gearbox over the course of a lap. These are details that I mentioned in my last blog and in my article in the current issue of the magazine, but with actual numbers you can calculate the exact change in gearing required.

Another way to get a nice overview of how the engine is being used is to look at a histogram of the rpm data. A histogram graphically shows the percentage of time that rpm (or any data) falls within a set range; AiM Sports Race Studio 2 Analysis package, as with many acquisition software programs, has a histogram feature and is used here.

Another option is to export the data to Excel and use its histogram feature. Note here that rpm is most often in the range of 13,750 to 15,250, which is ideal for the motorcycle in use (a Yamaha YZF-R6) as its power peak is in that range. Note also that the general shape of the histogram resembles a dyno chart, although turned on its side. Any deviation from this shape - for example, if the most used range of rpm is right at the top of the scale, or if there is a large zone of use below the power peak - can indicate that a gearing change is required.

With an rpm signal alone, it can be difficult to decipher exactly what gear the rider is in at any given time (and sometimes even the rider has trouble answering that question…) and having a "gear" channel that displays this information is very beneficial. Many bikes have gear position sensors that can be tapped into for this use, but another method is to measure countershaft speed - many bikes have sensors on the countershaft for the speedometer, and this also can be tapped into.

If your data acquisition system requires wheel speed as a default input, you can use the countershaft sensor rather than using an additional pickup and sensor on the actual wheel. The ratio of countershaft speed to rpm tells which gear is in use, but this data can be put to further use. The trace comparing the two inputs will also show how smoothly the rider is making upshifts and downshifts, and it will also immediately show any clutch slip. Furthermore, with countershaft speed you can calculate rear wheel speed, and this can be used to look at slip and even more details.

With rear wheel speed and rpm, an x-y chart can be generated that replicates the chart used in the gearing spreadsheet I included in my"> previous blog. In this format, you can see the maximum and minimum rpm used in each gear. There is additionally plenty of information in this chart about how the rider is using the engine, and any clusters of points at a particularly low rpm may need further investigation. The chart show here is also from Aim Sports' Race Studio 2 Analysis program.

The histogram and x-y chart are not essential to determining if a gearing change is necessary, but having the data in these two formats can definitely speed up the process of making a decision and give more insight as to exactly how much of a change to make.


The histogram shown above is part of Aim Sports' Race Studio 2 Analysis package, and shows how much time engine speed spends in each "bin" or range. If engine rpm is most often in a range other than the approximate power peak of the engine, a gearing change may be required.

Also from Aim Sports' Race Studio 2 Analysis package, the x-y chart below plots engine rpm against wheel speed, and shows maximum and minimum engine speed for each gear. The plot density at any given point is an indication of time spent at that combination of engine rpm and wheel speed.


Thursday, 18 October 2012 11:25

Track Time with John Sharrard - 2012 BMW S1000RR

This summer, our editor asked me if I'd be willing to put the 2012 BMW S1000RR through its paces on the race track.  I argued for awhile, then reluctantly agreed… Yeah right!

Shortly after, I was at Calabogie Motorsports Park with the sleek BMW for a Pro 6 Cycle-hosted  track day.  Pro 6 always provides a safe and well-run track day, at what just may be the nicest race track in the country.  Calabogie offers superb tire grip and smooth, consistent pavement, allowing traction no matter which line you select.

Monday, 22 October 2012 12:24


In the latest issue of Inside Motorcycles, my article discusses some of the advantages of changing your motorcycle's final-drive gearing. This can make a significant difference, especially on late-model sportbikes that are equipped with tall gearing and very close-ratio gearboxes. For track-day riders and racers, there are some additional subtleties to consider when changing gearing to suit a particular track.

In general, final gearing should be selected so that your bike just reaches maximum rpm in sixth (or top) gear at the end of the longest straight on the track. This gives you full use of the gearbox without having to worry about hitting the rev limiter in top gear on the straight; each individual gear ratio will be as short as possible, for maximum acceleration and a good launch at the start. From this starting point, looking at performance in various parts of the track can guide you to make a gearing change.

If the gearing in more than one corner is not ideal, you can change the final-drive gearing to a higher or lower ratio to better suit those particular corners - although there may be a trade-off on other parts of the track. For example, when I raced my TZ250 at Shannonville, I started with final gearing to reach maximum rpm in sixth at the end of the back straight. But in the tighter turns on the front section, the engine was either revving too high in first gear or bogging in second. I switched to a taller final ratio to make first gear more useful in the tighter turns. The tradeoff was that I ended up not using sixth gear at all on the back straight, and first gear was more difficult to use at the start of the race.

One other aspect to consider is that the spacing between gears (in terms of ratio) is not the same between each gear. Almost every gearbox has a fairly large gap between first and second, and the ratios get successively closer with each higher gear. This means that if you can use the higher gears more than the lower gears, engine revs will drop less with each shift and your bike will accelerate better. Continuing with my TZ250 example above, I could have selected shorter overall gearing to not use first gear at all, and dealt with the engine over-revving a bit on the back straight. This would have also given me the benefit of using the higher, closer ratios, and also decreased the chances of missing the first-second shift. Overall, however, using first through fifth was a better combination for that particular track. Everything is a trade-off, and it's a matter of finding what works best for the greatest portion of the track.

Attached below is an Excel gearing chart that I use. You can input all the information for your bike - maximum rpm, the internal gear ratios, final-drive ratio and tire circumference - and the chart creates several tables. Top speed in each gear for a selection of sprockets is shown; this data can help you better make a decision when it comes time to change your gearing. Also included is a table showing how swingarm length changes with each sprocket selection. If the wheel has to be moved significantly forward or rearward with a gearing change, you can consider another combination that is close to what you need, or go with a shorter or longer chain.

Gearing is one setup variable that is easy to change and can make a huge difference to your bike's performance on the track. Most racers have a large selection of sprockets on hand and constantly change gearing, but even track-day riders can take advantage of a change to better suit a particular track.

Download the Gearing Excel file here

Thursday, 04 October 2012 10:13

Ride better using data acquisition: Braking

Another data channel that shows a significant amount of information about the rider's habits - and is easily generated in almost any data acquisition system - is braking G. This channel shows actual deceleration in units of g, as opposed to brake pressure, which shows what's happening at the lever but requires an additional sensor.

The braking G channel can be calculated from a number of sources. Taking the derivative of the speed channel (yes, calculus does come in handy sometimes) is one method. Another is to use the data provided by the internal accelerometer that most data acquisition systems have. Or GPS longitudinal acceleration can be used, a channel found on many of the GPS-enabled lap timers currently available. The data shown here is from an AiM Solo GPS-based lap timer, with the braking G channel created using AiM's Race Studio software. The raw GPS longitudinal acceleration data does show braking force, but manipulating the data to show just deceleration and make braking a positive value makes it much easier to visualize what's actually happening on the track.

There are several areas of interest in the braking G channel. Foremost is the maximum value, which can range anywhere from .8 g (a street bike with street tires on moderately abrasive pavement) to 1.3 g (a race bike on race tires at the track). At the beginning of the braking zone, the data trace should ramp up smoothly to its maximum, at a rate of about 1 g per second. This value represents the time it takes for the bike's weight to fully transfer to the front tire for maximum braking; a slower rate indicates the rider could be getting on the brakes quicker. The point of maximum braking should be early in the braking zone, as shown in the graph by the red triangle at approximately 7500 feet. Braking at higher speed is much more effective than at lower speed, so the earlier you can get to maximum deceleration in a braking zone, the better.

Note that if the braking zone is short enough, the rider may not be able to reach the maximum value seen in a longer braking zone before having to release the brakes to enter the corner. For example, at the end of a short straight, braking G may increase to just .4 or .5 g before decreasing to zero at the corner entry - there is just not enough time/distance to get to maximum braking. Through the middle of the braking zone, the top of the trace should be smooth and remain close to the maximum value.  In the attached graph, the rider holds a steady .85 g of braking for 250 feet before beginning to release the brakes.

When the brakes are released near the entry to the corner, the braking trace should smoothly reduce to zero. The rate of release depends on the type of corner and the rider's style, but the trailing edge should be smooth with no bumps or dips. If desired, further math channels can be generated to graphically display how quickly the brakes are applied and how smoothly the rider holds the maximum and releases the brakes.

Deviation from this ideal shape for the braking G data trace - a smooth increase to maximum, constant in the middle of the braking zone and a gradual taper to zero at the corner entry - requires further investigation to find whether the cause is a particular characteristic of the track, something involving bike setup, or an aspect of the rider's style.

Thursday, 20 September 2012 09:40

Ride better using data acquisition: Lean angle

One of the more useful channels to look at when using data to improve your riding is lean angle. This channel can show some interesting detail about what’s happening on the track, and it’s easy to visualize what is actually going on from looking at the data in graphical format. While many riders think absolute maximum lean angle is the most important number, there is far more information that is of value and can be utilized.

It’s not entirely necessary to have a lean angle sensor to obtain the required data, although a proper sensor is ideal. If you have a GPS system, or even one of the newer GPS lap timers such as the AiM Solo, lean angle can be calculated from lateral acceleration in a software math channel. Calculated lean angle is far from exact and does not take into account aspects such as the rider hanging off the motorcycle, but even still can provide plenty of information.

The attached chart shows several corners of Chuckwalla Valley Raceway in Desert Center, California. The top blue trace shows speed in mph (note the scale on the left) while the bottom red trace shows lean angle with its scale on the right. Here, lean angle is calculated from GPS-based lateral acceleration. Distance in feet is labelled across the bottom of the chart, while the individual corners and segments are numbered at the top. Turns 7 through 9 make up a right-left-right chicane; here you can see lean angle going to near 50 degrees in each turn, then returning to zero as the rider straightens the bike to vertical (zero degrees).

In each turn of a chicane or cluster of corners close together, the rider would ideally reach the same maximum lean angle. In turn 7, the rider reaches 50 degrees of lean, but somewhat less in turns 8 and 9. The cause may be a characteristic of the track, such as camber or available traction, but it is somewhat (and surprisingly) common that even expert-level riders will not quite reach maximum lean in the second turn of a chicane; the fix is simply a short discussion with the rider. In most cases, however, some further examination is required to determine the cause and any possible action.

How quickly the lean angle trace changes from maximum to zero and back to maximum shows how quickly the rider is transitioning from side to side between the corners. This value, usually referred to as roll rate, can be visually ascertained by looking at the data in time (rather than distance), or a math channel can be used to actually calculate a number. Here we see a roll rate of approximately 45 degrees per second between each corner. Note, however, that roll rate is dependent on speed; the faster the motorcycle is travelling, the more difficult it is to transition from side to side and the maximum possible roll rate correspondingly decreases.

Later in the lap, turns 10 and 11 are two long sweeping right-handers. Here, the lean angle trace would ideally ramp up smoothly and quickly to its maximum and remain at that maximum value for the length each corner. In turn 10, we see that the rider has accelerated through the corner (note speed is increasing) and lean angle must therefore decrease. In turn 11, there is a portion early in the turn where the lean angle decreases, then momentarily increases before returning to a fairly constant value. Again, these anomalies may be due to the track and not something the rider is necessarily doing wrong. In turn 11, for example, a look at the track map shows that section is actually two corners with a small straight in between, and the rider accounts for the slight straight with these changes. But perhaps a different approach would result in a faster lap time?

The lean angle channel is just one way to easily and quickly find areas of the track that need attention. Quite often more investigation into other data channels and some consultation with the rider is required, but the lean angle channel provides some easy reference and is usually a good place to start.

Thursday, 06 September 2012 09:10

Colorado to Toronto with Larry Tate: A Daily Ride Journal - Day Nine: Home At Last! (updated daily)

Napanee, Ontario – So I’m home safe and sound, as is the Multistrada, with 5,688 miles on the clock (never did flip it back into metric, that’d be 9,152 km if I had). Since when I picked it up there were 2,886 miles showing, my little jaunt from Denver covered 2,802 miles, or 4,508 km in a week of riding. Hardly Iron Butt territory (done that, got the T-shirt and plaque, never again), but not bad mileage, either, for all the sightseeing I did.

Thursday, 13 September 2012 17:05

The Contact Patch

Over the last few weeks, several people have asked about the tire's contact patch, and how it plays a role in trail - especially given that the tire is constantly deforming and the contact patch is very much a dynamic feature itself. As most riders know, tires play a huge part in how a motorcycle works; steering, comfort, stability, cornering…almost every aspect of performance is highly dependent on the tires used.

My last three blogs have shown how trail changes as the motorcycle moves on its suspension and leans into a corner. We know that trail is the distance from the steering axis to the contact patch, measured either perpendicular to the steering axis (normal trail) or parallel to the ground (mechanical trail). But the big assumption is that the contact patch can be represented by an easily determined single point; the equations I used in my last blog are based on the tire having a perfectly circular profile with no distortion from weight or other outside forces. In reality, the contact patch is a large area affected by an almost infinite number of variables, and from that it's difficult to equate to a single point – and hence define trail.

The picture from my previous blog shows how the contact patch changes with lean angle. When the bike is upright, the patch is a nicely shaped oval and it’s fairly obvious that an equivalent point can be placed in the center of the oval. When the motorcycle leans over, a tire with a round profile retains that contact patch shape right to the very edge of the tread. Most tire manufacturers, however, design front tires with at least some triangular shape. This promotes quick steering and puts more rubber on the ground when the bike is leaned over - the contact patch widens and begins to lose that nice oval shape. Where to place a single point to define the center is less obvious, and it correspondingly becomes more difficult to put an actual value on trail. At extreme lean angles, the patch becomes smaller and crescent shaped as the very edge of the tread is used.

The tire’s profile can significantly alter the shape of the contact patch over the range of lean angles, in turn affecting trail and how the motorcycle behaves. Even the subtle change in profile due to wear can make a difference; most riders can feel that their bike handles differently when the tires are replaced, even if the exact same brand and model of tire is used.

The distortion of the tire due to weight transfer, cornering forces and acceleration/braking forces must also be considered, as these forces can further contort the contact patch, making it even more difficult to identify a central point. This brings the compound and construction into play, as those properties determine how the tire distorts under load. The contact patch loses almost any resemblance to an oval, and it’s almost impossible to determine its shape mathematically based on the properties of the tire.

We normally think of trail as a very definable and easily measured dimension that helps define the handling characteristics of a motorcycle. But there are so many factors that depend on the tire itself that the measurement given in a specification sheet, and even the values for dynamic trail calculated using approximations and assumptions, tell only part of the story. This at least partly explains why simply changing tires can transform a bike’s handling (in either a good or bad way!), and why some riders and bikes have difficulty adapting to certain brands or models of tires. In the pre-spec-tire era, riders could pick and choose a profile, construction or compound of tire that worked more favourably with their style or masked particular handling deficiencies. With that option largely gone, it’s up to the rider or the bike to adapt to the tire – sometimes a difficult proposition.

Friday, 24 August 2012 11:03

Vintage Sunbeam S7: An Inside Glimpse at the Gentleman's Machine

Catching a glimpse of a Sunbeam motorcycle moving down the asphalt is a rare experience, especially in Canada.

Friday, 24 August 2012 10:39

Page 12 of 19