125x125 banner
Stylish Motorcycle Riding Gears

Motorcycle Engine Balancing Act

Kevin Cameron has been writing about motorcycles for nearly 50 years, first for <em>Cycle magazine</em> and, since 1992, for <em>Cycle World</em>.

Kevin Cameron has been writing about motorcycles for nearly 50 years, first for <em>Cycle magazine</em> and, since 1992, for <em>Cycle World</em>. (Robert Martin/)

Let’s say we’ve built our own single-cylinder engine as a shop project. All is in readiness: We have spark, fuel, compression, and everything is correctly timed. The engine starts and runs—hooray!

But it vibrates quite badly, and the vibration intensifies as it revs up. What to do?

Addressing the situation methodically, we see that this vibration has two possible sources: linear and rotational. Linear, meaning “in a straight line,” describes the up-and-down reciprocating motion of the piston. Rotational means it’s coming from a rotating imbalance, such as the crankshaft itself.

So far, so good. But to which of these two categories does the connecting rod belong? What we do is weigh the two ends of the rod in a horizontal position. We assign the weight of its big end (the end that attaches to the crankpin) to rotating mass, while we add the weight of the small (piston) end to the reciprocating parts such as piston, rings, wrist pin, and wrist-pin clips.

Our first job is to achieve rotating balance. To do this we make up a balance weight to attach to the crankpin, equal in mass to the rotating imbalance we have just measured (the big end of the rod and its bearing). Ideally this would take the form of a ring or any geometry that does not have a heavy side.

With this weight attached to the crankpin, we can test for static balance by placing our crank on horizontal knife edges to see if imbalance rotates it. If it does, very likely the heavy point is the crankpin and the weight we’ve attached to it.

Our goal now is to either make the crankpin side of the crank lighter, or to add extra material to the crank at 180 degrees to the crankpin, until the crank is in static balance; that is, it doesn’t roll when placed on the knife edges.

Another step, more important in longer cranks, is to achieve dynamic balance as well. To understand this, imagine a dumbbell consisting of two wheels joined by a shaft. If one of the wheels has a heavy side but we balance it by removing mass from the other wheel, we can achieve static balance. But if we spin the dumbbell, it will wobble. In order to achieve dynamic balance (no wobble), the original heavy spot and any metal we remove to balance it must act in the same plane, perpendicular to the axis of rotation.

One of the two counterbalancers found in KTM’s LC4 single. The other is at the end of the camshaft to cancel fore and aft movement.

One of the two counterbalancers found in KTM’s LC4 single. The other is at the end of the camshaft to cancel fore and aft movement. (KTM/)

Once we have both static and dynamic rotary balance of the crank, we can move on to address the linear (straight-line) imbalance caused by the piston’s motion. To do this we remove the crankpin weight we used to achieve dynamic balance and build up our engine again. It is normal in crank balancing to balance 100 percent of any rotating imbalance.

When we start it, as we’d expect it still shakes terribly. This is because so far we have done nothing to balance the up-and-down shaking force of the reciprocating parts—the piston, rings, wrist pin, clips, and the small end of the con-rod.

We know where to add mass to balance them: at 180 degrees to the crankpin. Sometimes, as in engines with full-circle flywheels, this added extra mass takes the form of cylindrical slugs of heavy metal, such as lead, tungsten, or even depleted uranium, pressed into holes bored parallel with the crank axis and out near the flywheel rim. Or the flywheels may be made with lightening “pork chop” cutaways near the crankpin.

When we start adding mass at 180 degrees to the crankpin, what do we find? If, for example, we balance 25 percent of the reciprocating mass (a “25 percent balance factor”), we find that the up-and-down shaking has also decreased by 25 percent. Progress?

Fore-and-Aft Shaking

This results because when the piston is at TDC, the counterweight is at 180 degrees to it, and when the piston is down, the counterweight is up. The 25 percent counterweight cancels 25 percent of the vertical shaking.

But what happens when the crank has rotated to either 90 degrees or 270 degrees after top dead center (ATDC)? At the 90-degree position, our counterweight is trying to pull the crank forward, but there is no piston force opposing it. Same at the 270 degree position: now the counterweight’s inertia is trying to drag the crank back, but there is little piston force because it is once again in mid-stroke.

What we have achieved is a 25 percent reduction in up-and-down shaking, but we have also created a new fore-and-aft shaking force that is 25 percent as great as that of the piston in the up-and-down direction.

Try as we may, putting counterweights in every imaginable place, what we learn is that no counterweight added to the rotating crankshaft can ever completely balance the up-and-down motion of the piston. If, for example, we add enough counterweight to cancel 100 percent of the up-and-down piston shaking force, we have created a fore-and-aft shaking that is of the same magnitude as the vertical force we are seeking to cancel. Rather than doing away with the piston’s shaking force, all we’ve achieved is a change in its direction.

Subjective Balance

This is why the old-timers gave up on seeking perfect balance and instead tried to achieve a kind of “subjective balance.” Motorcycle frames and other parts are flexible, and when you bolt a vibrating engine to them, there will be ranges of rpm at which the bar vibrates badly, or the seat subframe fatigue cracks, or the riders complain that their feet or butts go numb after an hour. The engineers took note of what counterweight—measured as a percentage of the engine’s reciprocating weight—felt least bad to riders, a trial-and-error process. What they found was that the best balance factors (percent of reciprocating weight) were high—typically in the range of 65 to 85 percent.

This tells us that we and the motorcycle itself are more sensitive to up-and-down shaking than we are to fore-and-aft shaking. This also explains why in general riders have preferred the large 65–85 percent balance factors that cancel most of the vertical shaking but trades that for a large fore-and-aft shaking that annoys us less. The footpegs scrub fore-and-aft under our feet rather than giving us an up-and-down buzz. Same with the seat. Because the bar is flexible, the maker will sometimes install weights inside the bar ends to kill the high frequencies that quickly put our hands to sleep.

That was considered enough through the 1960s, but modern riders, given a ride on an old single or 360-firing parallel twin (which vibrates like a single because its pistons move up and down together) usually ask in dismay, “Are they all this bad?” Modern riders expect a higher standard of engine smoothness.

Learning to Love Vibration?

Life would be simple if the rider were the only part of the package to be fatigued or annoyed by vibration. Many have been told “Real bikes vibrate—get used to it!” But we have no way to tell the seat frame, footpegs, or the foaming fuel in the carburetor float bowls to man up. There is no way to dress up malfunction or broken parts. Maybe a bit of vibration is acceptable to remind us of the romantic past, but the basic truth is we’re better off without most of it.

A Better Vibration Solution

Excellent balance can be achieved in single-cylinder engines, but only by adding extra parts.

Example 1: If we do as the auto industry does, and add counterweight equal to 50 percent of the reciprocating weight, we have reduced peak main-bearing loads by half, which is good. Let’s look at what those loads are as the crank rotates. At TDC, with a 50 percent balance factor, as the piston stops and reverses direction its inertia yanks upward at 100 percent, but is opposed by the 50 percent balance weight 180 degrees away from the crankpin. The net upward force is 100 minus 50, or 50 percent of peak shaking force.

As the crank rotates to 90 ATDC, there is almost no piston force because it is in mid-stroke, but the 50 percent balance weight, located at 180 degrees to the crankpin, is now trying to pull the engine backward at 50 percent of peak shaking force.

Keep rotating, and when we get to BDC, the piston is at the bottom of its stroke and the inertia of its stopping and changing direction is trying to push the engine downward with 100 percent of peak shaking force. But the crank counterweight, pulling in the opposite direction (straight upward) at 50 percent, reduces this to a net 50 percent.

The crank continues its rotation, and when it gets to the 270 degrees ATDC, the piston is once again in mid-stroke creating little inertia force, but the crank counterweight’s force now tries to yank the engine forward with a force of 50 percent of peak shaking.

How can we sum this up? What we see first is that the net force in these four positions is constant at 50 percent of peak shaking force. And that constant force is rotating, but when we look closely, we see that it is rotating backward, opposite to the crank!

We can roughly cancel this by adding a gear-driven crank-speed balancer, also rotating opposite to the crank, but arranged to cancel the crank’s net backward-rotating imbalance force.

Example 2: During World War II, the aircraft-engine industry tested prototype engine cylinders on a standard Universal Test Engine. In order to provide balance that would prevent such test units from constantly breaking things (total recip weight was roughly 8 pounds) they were equipped with two crank-speed balance shafts geared together. Their eccentric weights were phased to arrive together at TDC and BDC, but to be 180 degrees to each other at the 90 and 270 degree positions. Thus, the forces generated by the two balancers added to zero at those 90 and 270 positions, meaning they generated no net horizontal force; but at TDC and BDC they could be sized to add up to values equal and opposite to the piston’s shaking force. The result was smooth operation. If a heavier or lighter piston was to be tested, it was a simple matter to change the masses on the balance shafts to cancel any desired piston weight.

Example 3: Rather than provide extra parts just to achieve balance, why not add a second power cylinder and find a way to make the shaking forces of the two pistons cancel each other? Some options:

  1. Do it as <a href=”https://www.cycleworld.com/story/blogs/ask-kevin/the-advantages-of-flat-motorcycle-engines/”>BMW’s boxer engines</a> do, by building a flat twin whose pistons move in opposite directions, with crankpins 180 degrees apart, thus canceling each piston’s shaking forces, though there is always some “crank wobble” caused by the two pistons not sliding along the same axis.
  2. Build your engine as a 90-degree V-twin with a 100 percent balance factor crank counterweight, or as Massimo Bordi did with <a href=”https://www.cycleworld.com/2007/09/17/cw-classics-ducati-supermono-first-look/”>Ducati’s Supermono single</a>, replace one of the two pistons with a sliding weight, resulting in a smooth single-cylinder engine. When you sketch the positions of the pistons and balance weight on the crank at the four positions—TDC, 90, BDC, and 270—you find that everything cancels.
  3. Do it as Kawasaki did with its <a href=”https://www.cycleworld.com/kawasaki-kr250-road-racer-history-classics-remembered/”>Gen 2 KR250 tandem twin GP engine</a> of the late 1970s/early ‘80s. It had two cranks geared together, one ahead of the other, such that its pistons came to TDC and BDC simultaneously. They then provided 100 percent balance-factor counterweights on each crank to cancel TDC and BDC piston shaking forces, while the counterweights on the cranks canceled each other at the 90 and 270 crank positions. The late great <a href=”https://www.cycleworld.com/tags/dan-gurney/”>Dan Gurney</a> built his very smooth big four-stroke tandem twin using this same scheme.

That’s enough for today. I’m going to get up and go see if there are any dinner plans.

View full post on Cycle World