Chassis Alignment Basics

By Dr. Robin Tuluie, Ph.D.

 

As I've set out to cover a large part of motorcycle technology - more or less everything to do with motorcycle chassis technology - and I think it is quite important to clearly define all the terminology and geometry right from the start. As you will see there are not a lot of definitions, but a clear understanding of these basic terms is essential for an understanding of what's to follow.

First there are the geometrical definitions. These dynamical measurements determine how the bike behaves while being ridden.

WHEELBASE:
The distance between the centers of rotation for the front and rear wheels - that is, the distance between the front and rear axle. Typical numbers are 55 to 58 inches for mid-size and larger street sporting motorcycles, 52 to 54 inches for 250 or 500cc Gran Prix bikes and just under 50 inches for 125cc GP bikes. Choppers, as well as the infamous Bohemia motorcycle (a Czech-made, 3-seated behemoth) can easily double these numbers!

WHEEL ALIGNMENT:
(1): In the most basic form it means that the front and rear wheel are in-line. That is, they point in the same direction and are not offset from each other. (2): A more constraining definition of the term also requires that both wheels are in the same plane. That is, one wheel is not vertically tilted with respect to the other. On a motorcycle that is not properly set up or has a bent frame, forks or swingarm (1) can be true without (2) being the case. This is illustrated in figure 1.
Chassis 1

SPROCKET ALIGNMENT:
Exactly the same definitions as for WHEEL ALIGNMENT, but now concerning the gearbox and rear wheel sprockets of the bike.

RAKE:
The angle of inclination with respect to the vertical of the axis of rotation about which the front wheel is turned during the steering process. This is usually the angle with respect to the vertical of the steering head of the frame, but not always. For example, if eccentric bearing cups are used in the steering head to alter the angle between the steering head and the steering stem of the triple clamps, than the rake is the inclination with respect to the vertical of the steering stem. The definition of the rake via the axis of rotation is always true and even applies to non-conventional designs. Typical numbers range from about 20 to a little over 30 degrees, the former for GP-racers and dirt bikes, the latter for cruisers and most vintage bikes. Note however that a relatively well-handling motorcycles have been built (most notably by Toni Foale, a 1970s chassis specialist) with much steeper steering head angles (i.e. a much smaller rake angle).

TRAIL:
The distance, as measured along the ground, of the point at which the front tire contacts the ground and the axis of rotation (see above) would contact the ground. For a conventional design this means visualizing the continuation of an imaginary line through the center of - and parallel to - the steering head tube that extends to contact the ground a few inches in front of the bike. Then, simply measure the distance between this contact point and the center-point of the tire's contact patch (vertically straight down from the front axle). Note that in an accurate measurements of trail, the wheels need to be aligned as defined above (see figure 2 below).

OFFSET:
The perpendicular distance between a line drawn through the centers of the fork tubes and the steering stem center of a triple clamp. The trail is a linear function of the offset of the triple clamps: More offset will yield less trail and vice versa. However, zero offset will not yield zero trail. In that case the trail is a function of the rake and the diameter of the front tire only. The geometry is shown in figure 2.
Chassis 2

Sometimes the top and bottom triple clamps do not have the same offset. In that case the trail and wheelbase (but not the rake!) are altered. Also, if the center of the front wheel axle is not in the center of the fork tube as viewed from the side of the bike, then this is equivalent to a change in the offset (and thereby the trail) of the bike.

CENTER-OF-GRAVITY:
The center of mass of the entire motorcycle, without rider. Usually located somewhat above and behind the crankshaft of the engine. The exact location of the center of mass is an important quantity in the design of top-level racing motorcycles. The vertical projection of the center of mass onto the ground (i.e. the point at which a vertical line drawn through the center of mass hits the ground) solely determines the static weight distribution between front and rear wheels of the bike.

RIDE HEIGHT:
The height of the front of the motorcycle (typically measured from one of the triple clamps) and the height of the rear of the motorcycle. Changes in spring preload, tires, rear shock location or linkage, or changes achieved by moving the forks up or down in the triple clamps all change the respective ride heights. The true purpose of changes in ride height is to affect a change in the location of the center of gravity. Every change in front or rear ride height is primarily a change of the location of the center of gravity of both rider and machine relative to the ground. Other quantities, such as the inclination and length of the swingarm, the location of the swingarm pivot and the geometry of the chain run and sprocket centers are explained as we go along the mysterious road of motorcycle chassis technology.

Now let's get started with our first wheel alignment.

The outline is as follows: (1) Align wheels. (2) Check whether the wheels are in one plane. (3) Check whether sprockets align (4) Check whether the frame and swingarm are straight. (5) Measure chassis specs. (6) What to do if your motorcycle isn't straight.

Required materials: Don't fret, this really is the poor man's chassis alignment method! You'll need some string, preferable some strong, brightly colored sewing thread. Also, little hand-held lasers (often used as pointers during lectures) can be used by those that think sewing thread is just too crude, but believe me, I have used both and the thread is just as accurate. The reason is that a typical laser aperture is about a millimeter (that is, the 'width' of the beam), not any better than what you get from sewing thread.

Let me just mention right form the beginning this very crucial fact: You do not need to measure any more accurately than about 0.5 mm. The reason is that the wheels on many motorcycles are out of round by this much (or more sometimes!), even brand new ones. Especially once tires are mounted on the wheels, the actual tire surface is never perfectly aligned with the rim even if the rim were perfectly true. Hence this inaccuracy is inherited in *every* motorcycle and any attempt to measure anything to do with the wheels, frame or swingarm more accurately than this is meaningless. I regard claims of measuring machines or frame straighteners that have much higher accuracy than this as misleading, not because the machines don't, but because this superfluous accuracy is wasted and hence meaningless with regard to the errors induced by the wheels and tires.

So, that said, we can all agree that our sewing thread method of alignment will be as meaningful a measurement as any multi-thousand dollar machine can provide. It should also be clear that before you start any of these measurements, check to see if your wheels are true side-to-side. A runout of 1.0 mm is okay, while 2.0 mm is already to inaccurate, besides being a major factor in any possible handling problems you might be trying to solve.

Part (1): Wheel Alignment

Begin by placing the bike on it's center stand, or better yet, support it via wood or stone blocks from both sides as per figure 3. Chassis 4

Put a drop line - which consists of about 2 feet of your brightly colored sewing thread with a weight (a small nut, for example) tied on one end - on the top edge of the rear tire and a little backwards so that it clears the axle and swingarm. Depending on which side of the tire you've put the string, there will be a small gap between the bottom tire edge and the string. For later reference, call this gap `a_r'. For now, adjust the blocks under your bike so that this gap is small, say less than 2 mm. This doesn't have to be accurate yet. Now you've got the bike nearly vertical. Make sure the bike is secure as you'll be doing a lot of measuring and moving around and you don't want the bike to move during this. Turn the steering wheel so it points straight ahead approximately (we'll get it perfectly straight later).

Next string some thread from the back of the rear tire around the front tire and back to the rear tire again (see figure 1 again, posted below). Make sure that the string isn't touching anything else besides the tires! If it's touching the center stand or exhaust you'll need to move the string up or down along the tire. The higher up along the tire you can get the string, the better, but typically the highest one can get is maybe eight inches off the bottom of the tire before the string hits the brake disk or bottom of the engine. At worst, you'll have to remove some parts to gain the necessary clearance. Next straighten the front wheel by adjusting gap `b' in figure 1 to be the same on the left and right side of the bike. Measure the width of your rear tire and cut a piece of wood to this length (I find that a pencil works great for this). Now stick this pencil between the strings, just behind the front tire and perpendicular to the strings. The strings should be nice and tight and hold the pencil in place. If not, tighten the strings and use a piece of duct tape to hold the pencil against the front tire.

Now go back to the rear wheel and look down along the strings: Usually one string will be closer to the front edge of the rear tire than the string on the other side. Adjust the chain adjusters (take the chain off before) to move the rear wheel so that the gap `c' is the same on both sides. If the pencil is exactly the same length as the width of the rear tire, the gap `c' will be zero on both sides. Now go back to the front wheel. Remove the pencil and check whether gap `b' is still the same on both sides. If not, turn the steering wheel ever-so-slightly to make `b' the same on both sides. Stick the pencil back in (make sure it goes in symmetrically, so that it sticks out the same amount on each side), and recheck the alignment for the rear wheel. You may have to adjust the chain tensioners just a tad now, but once you've done this you're done with Part (1). Your wheels are now in-line!

Part (2): Determining whether wheels are in one plane

Now that your wheels are aligned, be very careful not to bump into the bike. Any slight perturbation of the motorcycle can move the handle bars and throw off all your previous work.

Next we'll do an easy, preliminary check to see if the frame and swingarm of the bike are not twisted, which the most common type of damage.

With the wheels aligned, put the drop line on the rear tire just as before, and measure gap `a_r' just as before. Also measure distance `s_r' in figure 3. Chassis 3

If the string touches on one side of the tire on both the top and bottom, move to the other side. There should be at least a small gap now. If not, the bike is perfectly vertical, so record a_r=0. Note that the bike doesn't have to be perfectly vertical! Don't attempt to readjust the bike to get a_r=0, it will only disturb the wheel alignment. Whether a_r=0 or not, it will not affect the accuracy of the measurement we're about to take. Next go to the front wheel and record a_f and s_f there.

It helps if s_r and s_f are the same. If this is not possible for your bike (it should be, though, follow part 2 (b) below). If s_r and s_f are the same, then take the difference and divide by s_f. This gives the angle theta by which the wheels are out of plane:

theta = 57*(a_r -a_f)/s_f

(b) if s_r and s_f are not the same:

theta = 57*(a_r/s_r - a_f/s_f)

Here the coefficient 57 is just the conversion factor form radians to degrees. It also doesn't matter what units you use for a_r, a_f and s_f, as long as they are all the same (i.e. either mm, cm or in, but no mix of them). This formula is the small angle approximation for sin(theta) and tan(theta) and valid here for angles less than approximately 5 degrees, which will always be the case here.

In fact, your value of theta should be between 0.0 and 1.5 degrees. If theta=0, then both wheels are in plane and most likely your chassis is perfectly straight. If theta is less than 1.0 degree, your chassis is not quite straight, or your wheel not properly spaced (in the swingarm), or your forks or swingarm are bent. However, if theta is no more than 1 degree, the "tweak" in your chassis is minor and will most likely be acceptable, even for racebikes. Sandy Kosman once told me that they usually don't bother to straighten a bike unless theta is greater than 1.5 degrees, and my own experience with straightening frames and building chassis has shown this to be true.

Now, if you're a world-class racer and really good at picking up even the slightest chassis imperfections you'll probably notice theta=0.5 degrees. However, as long as theta is about 1.0 degree the chassis will be aligned well enough for most mortals. If theta is considerably larger than 1.5 degrees then either you've goofed somewhere along the line of this measurement or you've got a problem. I suggest re-checking the entire process, including the wheel alignment, as it is easy for the strings to hang up or touch against something and throw off your entire measurement. If you still have theta considerably larger than 1.5 degrees you'll need to find out what's wrong with your bike.

Part 3: Checking the sprocket alignment.

Similar to the method we used previously to align the rear wheel relative to the front wheel, we will now align the sprockets. It should be clear that if the wheels are aligned, the swing arm and frame straight and the sprocket offset and rear wheel spacers correct, then the sprocket alignment will be true. However, for many custom built bikes using different wheels and spacers, relaced spoke wheels or different sprockets this may no longer be the case. I've found that especially on older Vintage machinery often the rear axle adjustment at which the wheels align does not coincide with the adjustment at which the sprockets align. Older Vintage machinery has often been subjected to several previous owners and one is never quite sure what is original and what has been replaced. Even newer machines with cast wheels can suffer from sprocket misalignment if the wheels are not OEM-spec, aftermarket sprockets or wheel spacers are used or a bent swing arm is suspect.

Remove the chain from the motorcycle and chain guard and front sprocket case cover if necessary. Referring to figure 1b below, simply string a line from the front sprocket to the rear one. Align the line at the rear sprocket as shown. If the sprockets are aligned, then this line should just graze the edge of the front sprocket. Make sure that the line is straight.

Chassis 5

If this alignment cannot be achieved, then assuming you have previously aligned the wheels as in Part 1, the logical conclusion is that the rear axle adjustment which yields proper sprocket alignment is not the same as the rear axle adjustment which yields true wheel alignment. Hence something is wrong. First check that the rear sprocket offset is correct. Do this by comparing to the correct OEM wheel and sprocket assembly.

Next check that the rear wheel spacers have the correct length. Also check to make sure the front wheel is installed with the proper spacers, otherwise the wheel alignment would not be correct. If you have OEM wheels, spacers and sprockets and still the sprocket alignment cannot be achieved without disturbing the wheel alignment, the fault lies in the wheel alignment itself or in a bent chassis or bent chassis component (such as the forks or swing arm).

The next step is to look at the chassis itself an determine if the frame, forks or swing arm are straight. This step is also the next step if a proper wheel alignment results in the wheels being out-of-plane as in Fig 2 of part 1 of our article. Just because the wheels and sprockets align doesn't mean that the chassis is straight - the wheels must be in a single plane as well (as described in part 1). Referring to figure 3 of Part 1, this means that the angle

	theta = 57*(a_r -a_f)/s_f

should be close to zero (plus/minus 1 degree). If it is not within this spec, then the chassis needs fixing.

Part 4: Checking The Chassis

So let's suppose that somehow your chassis alignment and sprocket alignment don't agree and/or the front and rear wheels are out of plane. After all, if they do agree (to within the limits of accuracy stated above), then you're home free, your wheels and sprockets aligned and both wheels in the same plane. At that point, there is little reason to suspect a bent chassis unless several parts are bent and conspire to give aligned (which also means "in plane") wheels and sprockets - a somewhat unlikely scenario!

Chassis 6 The first check is to see if any of the axles are bent: either roll them on a flat surface or hold a straight edge along them, turning the axle in the process to at least two different spots. The next check is to see if the swing arm is bent. The most common way for the swing arm to be bent is a twist in the cross tube - imagine standing behind the bike, taking hold of the swing arm legs and pushing one down and the other up. This would result in a twist of the cross tube. An easy way to discern this without dismantling anything is as follows: Referring to figure 2, and with the sprockets and wheels aligned (or as close to it as possible with the suspect bent chassis), get the rear wheel vertical to ground, using a carpenters level or the string method. Next see if the swing arm axle is horizontal - you can do this by either feeding a string or straight rod through the swing arm axle (if hollow) and holding a level along the string, or by measuring the distance d1 and d2 of figure 2 from the axle center to either a flat and level ground or to a carpenter's level that is set up on the ground in a horizontal position, perpendicular to a line connecting the front and rear tires.

Your swing arm is straight enough if the twist angle is less than 1 degree, that is if the difference in the distance d1 and d2 of figure 2 from the axle center to horizontal is less than

	(d1 - d2) < w * sin(1 degree), or            (d1 - d2) < (w / 57.0)

If it turns out that (d1 - d2) exceeds the above limit, I recommend removing the swing arm and checking it again on a flat work bench, using the same principles as above.

It is also possible for the swing arm to be bent sideways, such as would occur if the rear wheel were impacted from the side. Any sideways bend would manifest itself in an offset of front sprocket to rear wheel sprocket and would be easily detectable during a sprocket alignment.

If the swing arm is indeed bent it can usually be straightened and braced by a decent frame shop, or you could do it yourself if familiar with such operations.

CHECKING THE FRONT END

The first thing is a quick preliminary check which consists of sighting along the plane in which the forks lie. Gauge one fork tube against the other by making sure they are parallel. If not, try loosening the bottom triple clamp bolts and front axle and, if applicable, any fork brace, and twisting the front wheel back and forth slightly. Bounce the front end up and down a few times (make sure the top triple clamp bolts are tight!) to center the wheel, then snug the bottom triple clamp bolts and and then the front axle. If the fork tubes are still not parallel, then either they are bent or the triple clamps are bent. To find out which one is at fault, disassemble the forks first and rotate the fork tube in either a lathe or on V-blocks using a dial indicator, or roll it on a flat surface. If the forks are straight, put them back into the triple clamps and, if the triple clamps are indeed bent, lying a flat plate across the fork tubes will immediately show this. Both fork tubes and triple clamps can be straighten if not bent too severely. However, as soon as any crease or chipping of the chrome is evident on the fork tubes they are history.

CHECKING THE FRAME

Chassis 7

Checking any frame is not a conceptually difficult part in so far as there are only two important dimensions involved (apart from making sure that none of the motor mounts, seat mounts, etc. are bent). A frame's function is basically to connect the steering head and swing arm pivot points in as rigid and economical a manner as possible and also to provide mounting points for various ancillaries. For our purposes, the question of whether the chassis is aligned or not condenses now to whether the frame's steering head is perpendicular to the swing arm axle. The inclination of the steering head with respect to the ground (the steering head angle) is also important for the handling of the machine but will not affect anything we've discussed above as far as wheel alignments and such are concerned. We will return to the question of measuring the steering head angle in part 3 of this series of articles. We therefore represent the frame as in figure 3.

Figure 3 shows that the steering head must lie in a plane normal -- that is perpendicular -- to the swing arm axle. It is not generally required that this plane intersect the swing arm axle in the center of the axle, 1/2 way between the frame spars. Some machines have asymmetrical frames with respect to the center of the swing arm axle. However, in all cases, the steering head MUST lie in a plane normal to the swing arm axle. In addition, for a perfectly aligned chassis, the steering head must lie directly vertically above a line connecting the center of the tire contact patches when the wheels are aligned and vertical.