The Experimental Motorcycle Association is the continuation of “the future of two strokes” page by Tim Hickox. This delves much deeper into the actual design concepts and potential of two stroke engines in relation to dirt bikes and other motorcycles. So sit back, take it in, and open your mind up to the possibilities that are out there. Feel free to share your opinion or thoughts on this subject by using the form at the bottom of this page…
13th July 2009:
I have said that any modern (not traditional) two-stroke must have fuel-injection. I must now explain that ‘fuel-injection’ must have a different meaning for a two-stroke than for a four-stroke. Fuel-injection (FI) on a four-stroke is just an alternative to a carburetor. Both schemes meter a quantity of fuel in relation to some mass of air. Take a short ride on a four-stroke with a carb, then ride the same model with FI, and you might be hard pressed to tell the difference. Ride a two-stroke likewise, and you may not believe that the same basic engine is at work.
To differentiate two-stroke technology, I refer to ‘DFI’, meaning ‘Direct Fuel-Injection’; that is, direct into the cylinder. My definition of DFI results from the need to classify all of the schemes that have proven effective – which covers a surprisingly wide range (including systems with carburetors!). Here it is: Any system that admits fuel independent of the scavenging air. To make that clear, the air that enters the crankcase and passes into the cylinder through the transfer ports ‘scavenges’ the cylinder of the exhaust gases left over from the previous cycle. In the traditional two-stroke, this air carries with it the fuel. This is ‘the problem’, with the traditional two-stroke. So (to oversimplify the solution), don’t do that!
To understand why DFI is such a vast improvement over the traditional scheme, and why the development of the technology has kept a few engineers awake at night, we need to understand that three requirements must be met for an engine to operate efficiently. One: The fuel must be admitted to the cylinder in such a manner that none (or almost none) of the fuel escapes out the exhaust port before it closes. Two: The fuel must be mixed with the air – ‘atomized’ – so that all (or nearly all) of the fuel chemically reacts with the oxygen in the air during combustion. Three: A combustible mixture must be in the vicinity of the spark plug when it fires, or there will be a misfire.
27th July 2009:
Three problems can be more than three times as tough as one problem. Solving the first problem often made the second problem worse. Solving the first two problems could leave the third problem unresolved. A lot of good ideas were not good enough.
The first really effective scheme came from the Orbital Engine group of Australia. Mercury Marine has been using the Orbital system for more than a decade with excellent results. It is certainly the most-proven solution out there. But the Orbital injector must pass fuel and air. Air cannot be forced the way fuel can (because it compresses). They then have been fighting the fourth problem: Their injector has had a rather low frequency-limit. It stops working at about 6000 revs. They may have found another way around this, but the only solution I know of – if the engine is to turn 10k – is to use two injectors per cylinder. This makes it rather more complicated and expensive than the E-TEC option. The Orbital system also requires a small air-pump, which E-TEC doesn’t need. But I don’t want to suggest that these are the only two options, they are just at the head of the pack right now.
I will go into the third problem mentioned in some detail, because it is the most difficult to understand. To reduce power, the throttle restricts the inlet to the crankcase. The amount of air delivered to the cylinder (through the transfer ports) drops in proportion. Remember that the cylinder is filled with exhaust gas from the previous cycle. The fresh air must displace that exhaust. As the throttle is closed, there comes a point where the fresh charge reaching the cylinder is insufficient, the cylinder retains a large percentage of exhaust gas, and the engine misfires (because exhaust gas won’t burn).
One way to deal with this is to create a ‘stratified charge’. A combustible fuel/air mixture is concentrated near the sparkplug, which ensures combustion. The remainder of the cylinder may have exhaust gas or fresh air (no fuel) or a combination of these. With this arrangement, the traditional bad habits of two-strokes are history. Gone is the irregular idle and the surging at light-loads. As a side benefit, it is almost impossible for the engine to not start on the first kick! The exception is that, in the off-road scheme, a capacitor replaces the usual battery. After a few days of not running, this capacitor will run down. The first kick may then only charge the capacitor, and the engine then starts on the second kick.
10th August 2009:
There is another way to deal with over-throttling. Just before that point where misfiring would occur, the DFI computer switches to an alternate-firing (AF) mode. Suddenly, our two-stroke becomes a four-stroke! The power then drops about 40-percent. This feature opens the door to a new option: Traction Control!
We add simple speed sensors to both wheels (much like an electronic speedometer). The outputs from those sensors go to the computer which compares them. The result is a speed differential, which in this control system is called hysteresis.
It works very simply. If the speed of the rear wheel exceeds the speed of the front wheel by a certain amount (the hysteresis), the computer switches to AF mode. But understand that the engine turns much faster than the rear wheel, and the computer can switch modes, back and forth, very quickly, so we can maintain maximum power to the rear wheel with respect to traction.
Traction control was used in Formula One cars, then outlawed. The critics said that it made driving too easy. Some even said that if such technology continued, monkeys could drive the cars. (Given how much F1 drivers are paid…!) For those who say four-strokes are easier to ride, here is the ultimate solution. But more than that, the bike will accelerate faster. Of course, the front wheel can still lift, and this will probably keep out the monkeys.
Because wheelspin is sometimes desirable, we can give the rider options. A button on each handlebar (like a horn button) can vary the hysteresis. Each tap of the right button would give an incremental increase in hysteresis, i.e., more wheelspin. The left button does the opposite. The hysteresis could also vary depending on the gear selected.
I will end this segment with a few numbers from our friends at MotoVerde. Let’s say, after a full season of racing, you want to renew the top-end – replace all the worn parts from the base-gasket up. The parts will cost: For a YZ250, 1172.16; for a YZF450, 3617.53; for a KTM 250 SX, 668.68; for a 450 SXF, 4253.14. All prices are in Euros, in Spain. Too bad those four-strokes don’t last forever.
28th August 2009:
Please don’t ask me when you will be able to buy a bike with traction control. I can’t predict the future. More than 30 years ago I wrote an article, “The Source of Power” (published, January 1978). I said then: “Carbureted two-stokes produce high levels of unburned hydrocarbons because of the fuel lost to the exhaust before combustion. Direct-cylinder fuel injection would solve this problem… (In) one test of this principle … the injected two-stroke … showed both a higher thermal efficiency and a lower specific fuel consumption than the four-stroke… Another experimental direct-fuel-injected two-stroke showed full-load specific fuel consumption figures rivaling those of many diesel engines.” … I’m still waiting for that future that I predicted then!
We started this with the intention of talking about two-stroke dirt bikes of the future. What I have been giving you is a lot of history. To clear the air, I had to bring you back to the ‘future’ of about 1995. Do you really want me to predict the future? How long do you intend to live?
With traction control, I did take you where no man has gone before. The technology involved is very simple, but it depends on DFI. Two-strokes with DFI have run an alternate-firing mode, so we know that this works very well. I’m not selling moonbeams. And the control scheme that I laid out is just a variation on the principles used in anti-lock braking systems. Old technology, new applications.
As uncertain as any future may be, I do want to tell you about some possibilities, like traction control, because I think the future of the two-stroke dirt bike could be very exciting. DFI is not the end of two-stroke development, but can be a new beginning. So let’s go on.
If you look at the torque curve of any high-performance two-stroke, you will see a dip, or a valley. In the case of the KTM 105 JMC, this runs from 5500 to 7500 rpm. The engine actually makes less torque at 7500 than at 5500. Immediately after this depression, the torque curve rises very steeply. This is why these engines have that sudden rush, which some riders like, and other riders find difficult to control.
A long time ago, I went looking for a way to fill in this valley, to bring up the bottom end of the torque curve. This result would not only improve performance ‘off the pipe’, but it would make the bike easier to ride. Here then could be the next revolution.
08th September 2009:
It will help us to consider the ’cause’ of this low-torque depression. Some of you already know to blame the expansion chamber. The positive (high pressure) wave travels down the pipe (at the speed of sound) and returns to the port too early, at ‘low’ speeds – when the pipe is tuned for high speed. When the torque falls off, we say that the pressure-wave cycle in the expansion chamber is out of phase with the scavenge cycle.
I know I’m getting very technical, and that I may be pushing what most riders can comprehend, but there is no way to understand engines using kitchen terms. The principle I’m referring to is so crucial that I will explain it by way of a simple analogy:
Most everyone has pushed a child on a swing. We learn very quickly that if we push just after the highest point, the child will go higher and higher. But if we push before that high point, repeatedly, we can bring the swing to a stop. The outcome depends on the timing between the natural period of the swing and the pusher’s effort. We call that the phase relationship. If we push after the top, we put energy into the (swing) system. If we push before the top, we take energy out of the system.
These pressure waves in the exhaust have (are) energy, so they can push (a positive wave) or pull (a negative wave). But whether this pushing and pulling helps us or hurts us depends on whether or not these events are in-phase or out-of-phase with the scavenging of the cylinder. Because the length of the exhaust system is fixed and because the waves travel at the speed of sound (consider it a constant), the systems will be in-phase at the (high) speed that the expansion chamber was designed for, and out-of-phase at low speeds, which is where that valley shows up on the dyno charts.
The ‘elementary solution’ to this problem would seem to be a variable expansion chamber. In most cases today, we have this. The exhaust (power) valve delays and modifies the waves. This works well, but it mitigates, it does not eliminate, the problem.
If you really want to understand two-stroke engines, think of ‘it’ as three ‘pumps’ connected in series – the crankcase (sometimes called the ‘primary pump’), the ‘working cylinder’, and the expansion chamber. The trick then is to get all three of these pumps working together (in-phase) so that each helps (so to speak) and does not conflict with what the others are doing. And making that work over a 10,000-rpm range, really is a trick!
21st September 2009:
To say that the expansion chamber is the ’cause’ of this low-torque region gives the wrong idea. The two-stroke engine cannot be analyzed as ‘parts’ with cause-and-effect relations. The mathematician would say that there are too many nonlinearities. This is why the evolution of the two-stroke has been a process of trial and error.
If we look at what may happen with the piston at BDC (Bottom Dead Center), we see that the exhaust and transfer ports are open and, functionally, we cannot say where one ‘pump’ ends and another begins. (I told you to think of the two-stoke engine as three ‘pumps’. Now, I’m saying that you can’t think of it that way! The two-stroke is a paradox!) So, let’s look at the whole process again and see if we missed something important.
As the piston descends, it compresses the air in the crankcase. The exhaust port opens and the cylinder pressure drops. The transfer ports open and air flows from the crankcase into the cylinder because of the pressure differential. But wait. There is something wrong here. Before the piston reaches BDC, the pressures reach equilibrium. Before the scavenge cycle is even half through, the crankcase stops pumping. What’s worse, as the piston passes BDC, the volume of the crankcase increases and the volume of the cylinder decreases. Simply put, the working-cylinder tries to pump its contents back into the crankcase! What actually happens depends on a lot of variables, but I can now tell you in familiar terms that the phase relationship between the crankcase pump and the scavenge cycle is all wrong. If we say that the expansion chamber has a problem because it is sometimes out of phase, how are we to think about the crankcase pump which is always out of phase? “Well,” a fellow engineer might say, “we don’t think about it at all, because we can’t change it.” Well…what if we could?
An engineer, Manuel Sevilla, asked that question about 25 years ago. And picking up a scheme that DKW engineers had played with in the 1930s, he went to work. The beauty of Sevilla’s approach was that he modified production engines. He was then able to demonstrate the difference before and after the phasing error was corrected. The difference is shocking! The torque characteristic (the shape of the curve) was transformed. Show an engineer the two torque curves and tell him that both engines have the same intake, porting, and exhaust systems. “That’s impossible!”, would be the natural response.
5th October 2009:
Sevilla’s experiment was a great proof-of-concept, but his engines were bigger, heavier, less efficient (they needed DFI) and more expensive – or would have been if they had gone into production. He showed us what we needed to do, but not how to do it.
So we come to what I call AST – Assisted Scavenging Technology. To our traditional two-stroke – with DFI – we add a plenum chamber. This is nothing more than an empty space in the crankcase/gear casing. The only requirements are that it be air-tight and that the volume be at least equal to the piston displacement. (There are also what we call ‘packaging’ considerations.) This chamber is connected to the crankcase through a reed valve. If the pressure is higher in the crankcase, air will flow into and charge the plenum. When the pressure in the crankcase falls, the higher pressure remains in the plenum. There are other changes required, but these are mostly modifications to existing parts (like cast-in ports that permit transportation of the air), and not additional parts. It sounds too simple, but let’s run through it.
As the piston compresses the air in the crankcase, the plenum chamber is also charged. The exhaust port opens, then the transfer ports. The pressure quickly drops in the crankcase until it is equal to the pressure in the cylinder. So far, everything is normal, except that the crankcase pressure, when the ports open, is lower; this is because we have lowered the primary compression ratio through the additional volume of the plenum. This is not a problem. This lower pressure, at this point in the cycle, is consistent with Sevilla’s engine. At low speeds – where we want to pick things up – lower pressure means lower velocities through the ports, less mixing of fresh charge with exhaust residuals, and less charge loss out the exhaust. In technical terms, the ‘trapping efficiency’ is improved. Of course, if we only lowered the primary compression ratio, we would see a power loss at high speeds.
Now, as the piston nears BDC, a valve opens and the compressed air in the plenum chamber discharges into the cylinder. The pressure at this time is actually higher than when the transfers first open, because the pressure in the cylinder is lower. The only ‘pressure’ that means anything is the differential.
A simple way to think of this is that we have added a fourth ‘pump’. The traditional crankcase works through the transfer ports, very much like always. Call this ‘phase one’. Then, near BDC, when that system has run out of steam, the air in the plenum is released and AST keeps air moving into the cylinder until the AST ports close. Call this, ‘phase two’. From here, it gets interesting!
20th October 2009:
Is anybody still with me? AST is very simple mechanically, but the thermodynamics are very sophisticated. In the traditional two-stroke, when the transfers open, there is a great inrush of air, then… casi nada. This pattern is dictated by the mechanical constraints of the simple engine. From the standpoint of thermodynamic efficiency, it would be much better if the pressure started low, when the ports open, and then the pressure increased until the ports closed. AST doesn’t quite do that, but it is a step in the right direction, and it is simple. But what I have described is only the beginning. AST changes all the rules.
The AST ports are independent of the normal transfer ports (which work the same way as always). This means that phase-one can be optimized for scavenging – displacing the residual exhaust gas. Phase-two can be optimized so as to retain as much fresh air as possible in the cylinder – what we mean by trapping-efficiency. These are really two different problems, but until now we had to accept a single compromised solution.
A very neat trick is now possible. The AST ports can be as high as the exhaust port. Phase-one requires a ‘blowdown’ period, the time between the exhaust opening and the transfers opening. But the AST valve is closed until nearly BDC, so nothing can blow-back through those ports. Phase-one can then be about 60-degrees, and phase-two can be, maybe, 80-degrees. In effect: asymmetric timing!
I should make it clear that the AST valve is not something that can go BANG! We get the necessary timing function by modifying parts that we already (in any simple engine) must have. AST requires no new moving parts except the reed valve between the crankcase and plenum. And so the stage is set.
At the same time that I was working on AST, I was putting together a simple DFI system. But I was aiming at a different target from the computer nerds. I wanted engines for any application. It has been estimated that there are about 100-million small, two-stroke engines operating in Southeast Asia today. Most of those are in small motorcycles. People bought these things because they were cheap (but also because of performance, ease of repair, etc.). If you are trying to live on $1000 – $1500/year, cheap is good! These folks aren’t going to buy any fuel-cells, or hybrids, or computers. Of course, Honda, et al., would like to sell them four-strokes, but they rejected that option last time – they bought two-strokes!
So we know what they don’t want! What’s left? How about more simple, cheap two-strokes with lower fuel consumption and emissions than the four-strokes? As it turned out, AST was the secret to making a really effective, carburetor-based DFI system.
6th November 2009:
If AST is given, we get DFI very simply. The traditional carburetor, of course, is replaced with a throttle valve at the inlet to the crankcase. (Reed valves are retained.) A much smaller carburetor is then connected, through another reed valve, to the crankcase via a long manifold. This carburetor functions as always, but the (rich) air-fuel mixture never quite gets to the crankcase. On each inlet cycle, the manifold becomes charged with mixture, while the crankcase remains fuel-free.
Here is the trick: This ‘manifold’ is also an AST port. So when the AST valve opens, the pressure in the plenum chamber ‘blows’ the fuel mixture into the cylinder. (This clears the manifold/port for the next cycle.) Presto! ‘Fuel injection’! But now it gets good. The AST fuel port directs the mixture to the sparkplug – it’s a straight shot. At the same time, other AST ports open. These air-only ports create streams that converge just above the exhaust port. Some of the air will be lost out of the port, but it’s only air. Mostly, this air prevents the fuel mixture from migrating toward the exhaust port. It also prevents the fuel (generally speaking) from contacting the cylinder wall, which adversely affects lubrication and emissions. Finally, it helps direct the mixture to the sparkplug and confine it to the combustion chamber.
Of course, all of the air delivered by AST is in addition to the air that passes through the four, main transfer ports; but the AST airflow is against the ‘loop’ pattern that those ports establish. The air from the transfers must displace the residual exhaust gas; but all gases have mass, inertia, and once we get the whole mess going out the exhaust port, the good (air) naturally follows the bad (exhaust). In the worst case, with a traditional engine, 40% of the fresh charge will be lost. If this contains fuel… well, there is the problem. So AST not only stops the fuel mixture from going out the exhaust port, it stops much of the air. That is, the air from the AST ports that is lost out the exhaust, is less than the air that would be lost without AST. And this improvement in trapping efficiency is most beneficial at low speeds, for there is then more time for air to circulate and escape out the port.
That we can gain so much with so little additional complication and expense is exciting. If we aim for low fuel consumption (rather than maximum power), we can get diesel-engine efficiency from a very simple, light, cheap (relative to any diesel) two-stroke, gasoline engine. Is this what all the big guns are afraid of?
19th November 2009:
When we add “Fuel” to AST, we get: Fuel/ Assisted Scavenging Technology, or FAST. Yes I know, it’s too cute; but it’s descriptive and easy to remember. And I insist, a FAST engine is so different from anything that has come before that it constitutes a new class and we need a new-class name.
Actually, there is more to FAST – it gets better – but its limitations must also be appreciated. It is very simple, meaning that very few new parts are needed (i.e., things you’ve never seen before). But many of the old parts must be different. Consequently, I have not found an acceptable way to modify an existing engine. (This is relatively easy with E-TEC.) New castings are required. The whole thing needs to start with a clean sheet of paper. And that simplicity only holds for single – and some twin-cylinder engines. FAST is not an alternative to E-TEC, et al. Those systems won’t work on a lawnmower (because the cost would be too great). FAST won’t work on a six-cylinder outboard. And that traction-control scheme that I laid out requires computer control. But altogether we have very nice solutions for any application that I can think of.
So if all the ‘problems’ that were supposed to kill the two-stroke have solutions, what else can we do? I pointed to the KTM 105 JMC as an example of what we can expect from a very small, light engine. I then said that most riders would perform best with something around 175-cc. But there are giants among us, and events like the Dakar, so there is a demand for largerengines. Over about 200-cc, I would abandon the simple single. The opposed twin has all the advantages except cost. Of course, the crankshaft axis must be transverse, like the old Douglas, and not like BMW’s famous boxer.
Consider this: We take two of those 105 cylinders and connect them to a common crankcase. We now have more than 60-hp, from 210-cc. But the advantages over a 60-hp single are great. For one, we have that super-wide powerband. We have a very smooth engine that will save the rider a lot of fatigue on long rides. (I don’t think we really appreciate how much vibration takes it out of us.) The thermal and mechanical stresses are so low with these little pistons that the engine is nearly unbreakable; 100-cc kart engines typically turn over 16,000-rpm. The opposed cylinders mean that both cylinders fire together, so rear-wheel traction is the same as it would be for a single-cylinder engine at the same rpm. This also means that only one expansion chamber is required. Because the forces on the crankshaft are balanced, it can be lighter than a single-cylinder crankshaft, if both engines produce the same power. And those balanced forces so reduce the loads on the whole engine structure that magnesium can be used for the crankcase. In short, this twin will be no heavier nor wider than a single with the same peak power. Here is a map of the road not traveled.
3rd December 2009:
There is another side to this two-stroke vs four-stroke debate. I have been telling you about a lot of two-stroke research and development that few people hear about – and there is much more to tell. But I could also paint a bleak picture of the four-stroke engine by telling you, once again, what the manufacturers of these engines don’t want you to know. Of course, some people have seen the light. KTM, Gas Gas, TM, Husky, et al., have been selling lots of two-strokes. And the AMA and FIM are under a lot of pressure to ‘fix’ the race rules, to support the riders, rather than the Japanese manufacturers.
However all this may play out, I want to wrap up this question as to the viability, practicality, and advisability of using two-stroke engines in dirt bikes, road bikes, cars, or what-have-you. I can do no better than to repeat what the engineers at GM said: “A state-of-the-art two-stroke engine is simply the best thing that we know how to make.” After about 130 years of work on internal combustion engines (and alternatives), this is where we are. If this surprises, if the common man cannot understand it, if corporate giants wish it were not so, the facts remain. That no motorcycles are being produced with these engines is THE problem! However imperfect the latest two-strokes may be, everything else that has been tried is worse. If, as it turns out, all the big players invested in the wrong technology, well, that’s the way it goes. As Kierkegaard said: “He who fights the future has a dangerous enemy. The future is not; it borrows its strength from the man himself and when it has tricked him out of this, then it appears outside of him as the enemy he must meet.”
As for knocking four-strokes, I don’t think they need any help. The Yamaha WR 250 R made 27-hp on the MotoVerde dyno. That’s not the problem. With a full tank of gas, it is one cantaloupe short of 300 pounds. And what do they say is the “worst” attribute? “High price.”
So let us move on. We don’t want to lose sight of the motorcycle, the dirt bike as a whole. We want a two-stroke engine because it will permit the best bike overall. That is, a future of two-strokes must take into account the whole machine, where the engine will influence the design of so much more. Let me take you a little farther down that road not traveled…
15th December 2009:
We might begin by thinking of something conventional – an RM125 will do. We are going to rearrange and substitute parts until we arrive at something very un-conventional. Will this be the dirt bike of the future? Well, let’s see. (If you have trouble following the words, draw a picture.)
First, we bore and stroke the engine to 175-cc. We rotate the cylinder forward until it is just 15-degrees above the horizontal. Another rotation and the exhaust comes out the top. The reed valve still feeds the crankcase, but the entry angle is much improved. The gearbox moves from behind the crankcase to below it. The arrangement now is much like the new Husaberg, but the engine is much smaller and as far forward as the front wheel will permit.
Now the fun starts. Reverse the gear shafts so that the countershaft is forward of the mainshaft (not behind), and directly below the crankshaft. We then place the swingarm pivot in front of the countershaft. The swingarm is longer than normal and the rear shock can be moved right up against the back of the engine. You should notice that all these masses are moving forward. The shock doesn’t need linkage (for the reason I will explain), so it is something like the KTM.
What we call ‘the frame’ has become a new proposition. The problem has always been to connect the front and rear wheels (steering head to swingarm axis) with the lightest, stiffest structure possible. But the traditional layout puts the engine between those points, which is why we have these massive frames. Look at the new picture: A cast aluminum beam (no welding) drops from the steering head down to the new swingarm pivot. It splits to form a yoke that goes around the cylinder. The swingarm axle also passes through the front of the gearcase, so everything ties together in a rigid unit. That’s it. That’s the ‘frame’! Of course we need a rear subframe to carry the seat, fender, etc.; but that is not really ‘structural’. If the rider could do without a bunch of doodads, the bike could be ridden with just that short, front frame.
With the swingarm pivot ahead of the countershaft, a funny thing has happened to the rear suspension. When you land from a jump, you open the throttle and the top-run of the chain pulls at the rear wheel. If you analyze the forces involved in this process, you see that some fraction of that drive force acts to compress the rear suspension. The farther the suspension is compressed, the greater the fraction of chain-pull acting to compress the rear suspension. Of course, we must make the rear spring stiffer (than would otherwise be the case) in order to deal with this force. All this happens when the swingarm axis is behind the countershaft, which is what everybody takes for granted. With the axis ahead of the countershaft, the force of chain-pull acting on the suspension is less, but more important, it acts like an additional ‘spring’ that acts to prevent bottoming of the suspension. And the more the suspension is compressed, the stronger is this new ‘spring’. It’s a funny thing.
29th December 2009:
A forward thinker is going to put one and one together; with the cylinder horizontal and the gearbox below, the back door is open to add another cylinder. Indeed, I began with the twin (about 25 years ago). The single is a twin with a forgotten cylinder. The intent was to cover a whole range of engines from a 100-cc single to a 450-cc twin. That’s not to say that the 100 would share any parts with the 450, but from a design standpoint, it is mostly a matter of scaling things up or down.
Clearly, with this layout, the swingarm must go around the gearbox. To keep the width at the footpegs within the normal range, the gearbox should be as narrow as possible. (The clutch is on the crankshaft, so it is out of the way; the drive chain must be on the right, not the left, to avoid some conflicts.) This objective is met by using a special gearbox that Sachs produced for their 175/250 ISDT (ISDE) bikes. It gives seven speeds, but it is only four gears wide (ignoring dogs and shift forks). The trick of turning a four speed into a seven speed is accomplished through two small gears on a third shaft.
Let’s move forward. I said the frame drops down from the steering head. Forget that – the ‘steering head’ is gone. The bearings for the steering axis have moved to the fork. The triple-clamps are also history, and so are the fork tubes. I begin with Valentino Ribi’s Nuova Sospensione. When Roger DeCoster was with Honda, he played with Ribi’s innovation for a couple of years and regarded it as a superior design. It is not a telescopic fork. It looks something like the old Greeves leading-link fork, but the geometry is quite different. It has two main advantages: very little sticktion and very little dive on braking.
I have changed Ribi’s fork in two ways. First, that short tube with a bearing at each end – the old ‘steering head’ – is cast in one-piece with the fork legs. The legs curve in at the top to meet that tube; at that point the fork is only as wide as the handlebar clamps. This structure is very rigid – there is nothing to tweak. The second change is the elimination of the two spring/shock units that give it that Greeves-look.
Change of subject: A few odd notes. The best enduro rider recently has been Ivan Cervantes. His 300 KTM gave him seven wins in ten events. So eyes were wide open when he rolled out a four-stroke in Slovakia. After the event he was asked to comment. He spoke of the weather and the special tests. What about the four-stroke? He said, “…las 4T …son muchos mas torpes…”. My dictionary translates “torpe” as “stupid; dull; clumsy; slow.” Second in World Enduro (and first in Slovakia) was Christophe Nambotin. You can tell how he likes it – look at that shiny expansion chamber on his Gas Gas. And, Suzuki has taken out full-page adds to tell everyone that Carlos Pando is the Spanish Enduro Champion for 2009, and they tell us that we can buy an RM250 E (or an RM125 E) two-stroke just like his. But somebody should wonder why, after so much money has been spent on four-stroke design and development for the last 10 years (and NOT spent on two-strokes) that they are no better than “torpes”.
12th January 2010:
In place of Ribi’s suspension units, I use a yoke that goes from the fork’s lower link – which carries the axle – up to a ball-joint above the tire. This pivot is on the steering axis, so turning the front wheel does not affect the suspension, or vice-versa. The yoke transfers wheel motion to a rocker-arm that pivots on the main frame. A single spring/shock unit sits beside the rear shock and is fixed at the bottom. As the front wheel goes up, the inboard end of the rocker-arm compresses the spring. The fork, now free of springing/damping functions, is exceptionally light; and the mass has moved to a more central location, reducing the polar moment of inertia of the whole machine.
There are some details: The air filter is where you would expect to find the gas tank. Air enters about where a gas cap would be, goes down into an airbox, then up through a flat filter element. Gravity and vibration shake dirt off the filter rather than forcing the particles through the filter and into the engine.
Where two radiators might be, there are two small gas tanks. The radiator is under the seat – out of harms way. Below that is another gas tank with a built-in fuel pump. The front tanks stay full until the rear tank is empty. In this way, as much as four gallons can be carried without the rider feeling that he is doing something unmentionable to a watermelon.
If you have been able to follow this, you might be surprised. I have told you how to build a motorcycle quite unlike any of the current manufacturers have even suggested might be possible. All the Me-Too dirt bikes – that haven’t changed significantly in 25 years – are not the only way to go. The question now must be: How do we get turned around and moving in a new direction?
27th January 2010:
All this future-talk has been about engineering that I and others have done in the past. What could happen in the future would be the implementation of such work. This is where the breakdown occurs. GM engineers show the world a (two-stroke powered) 100-mpg car; then GM’s management said, “But we can’t build that, so we will just have to keep selling more Chevy Suburbans”. You see, we engineers do little experiments to prove what is possible. What happens thereafter – if anything – is outside our realm of responsibility. The real problem is that a very small group of persons sitting behind very big desks decide what shall be produced, and consequently, what we can buy. It is not for me to impugn such decisions – in business, somebody has to watch the bottom line. If there is anything to lament, it is that we do not have other options.
But wait! There is a glimmer of light at the end of this dark corridor. I want to suggest a radical idea… no, a revolutionary idea. We are not the first to fall under the judgments of an elite. Herschel Smith, in ‘Aircraft Piston Engines’ (1981), said: “Whether improvement will take place is doubtful today, because benevolent governments have raised the cost of certificating new designs to such impossible heights that a manufacturer experiments at considerable risk.” That statement would be equally true in the case of motorcycles, cars, and many other products. But even as Smith was making that bleak prediction, the future of light and ultralight aircraft was being silently stolen from corporate and political interests by a band of backyard empiricists. These guys flew the banner of the Experimental Aircraft Association. Without any investment in yesterday and without any desire to ‘control’ the future (with standards, regulations, and certificates of compliance), they raced ahead willy-nilly. They ignored what Piper, Cessna, and Beechcraft were making and they operated just beyond reach of that ‘long arm of the law’.
In retrospect, what the EAA accomplished was extraordinary. On the surface, they designed, built, and flew airplanes. It’s how they did this that was revolutionary. It’s time for those of us with helmets to leave behind those chained to their desks. If you can hold the throttle open, welcome to the Experimental Motorcycle Association!
09th February 2010:
EMA (pronounced ‘Emma’, like the woman’s name) can serve several important functions. First, it can pull together an international cadre of like-minded individuals into a community. The model here is the ‘scientific community’, where the free exchange of information, peer review, and experimentation replace secrecy (telling us only what we ‘need to know’), manipulation (i.e., marketing), and coercion (rules that force this – four-strokes – and prevent that). Obviously, science has been successful. What has been given too little attention is that the ‘scientific community’ represents a social order in contradistinction to all the business/political schemes. (Read Michael Polanyi’s essays: Mutual Authority and The Free Society).
The second aim is education. Nobody has all the answers. But some of us have learned a few things, and where one person lacks knowledge, another can serve as mentor.
Third, to accomplish anything, we need resources. These are ‘out there’, but they vary greatly in accessability, quality, and cost. Some of us will find such resources and tell the rest of us where we might go when we need something. As EMA grows, some persons will offer services specifically aimed at the creation of experimental motorcycles.
Fourth, we don’t need any more UJMs (Universal Japanese Motorcycles). EMA must encourage new designs – pushing those limits that the manufacturers cannot justify. We need designs that are easier to fabricate in small numbers, cheaper, lighter, and filling those niche interests that ‘aren’t worth the trouble’.
I am going to expand on the possibilities that are open to EMA, but this may sound like a lot of pie-in-the-sky speculation. Just remember that we are not breaking new ground; the trail has already been cleared. Try to imagine more than 10,000 home-built aircraft coming together at the EAA’s annual meeting in Oshkosh. Do you think that building airplanes is easier than building motorcycles? Dirtbikers are used to taking risks. So let’s get going and aim for our own EMA “Oshkosh”!
27th February 2010:
I’m sure there are some who will say that building our own motorcycles is not revolutionary, it’s ridiculous! So I must explain that, since WWII, two revolutions have slipped by almost unnoticed. I might ask first: Why do we make things in big factories rather than little workshops? The answer should be: Economies of scale, and the division of labour, as Adam Smith said in ‘The Wealth of Nations’. But Smith didn’t say that if one man spends all day putting points on pins he will soon go crazy. He didn’t say that the greatest improvement in the productive powers of labour was the institutionalization of inhuman practices. (Read Elton Mayo: The Social Problems of an Industrial Civilization.)
Of course, Smith’s pin factory was a trivial example. More than 100 years passed before the problems of manufacturing complex machines, like cars, were worked out by the two Henrys, Leland and Ford. But the world has changed since Fords ‘Model T’. Those economies of scale depended on treating a person as if he were ‘a cog on a wheel’. A multitude of ‘social problems’ brought governments into partnerships with business in the production of contradictions. Henry Ford didn’t have to deal with the EPA, CARB, NHTSA, CPSC, OSHA, etc. Newt Gingrich recently asked: “Is there anything in your life you think would be better if it were run by government bureaucrats?” Well, better or worse, manufacturers have to deal with all that and when you buy their products, YOU have to pay for it all!
Back in 2006, U.S. News reported estimates of monies U.S. companies spend annually on research and development ($194-billion) and on tort litigation ($205-billion). When there is more money going to lawyers than to engineers, you should not have to wonder why GM has been going out of business!
I don’t have solutions to their problems, and I am not making a political statement; what I am saying is that those old economies of scale, that made Ford and GM the largest companies in the world, no longer give the mega-corporations the advantage. Government intervention – the supposed ‘solution’ to all those ‘social problems’ – has created so many complications and additional expenses that both the products and the customers have been left in the dust.
And this brings us to the key concept, the next revolution, that the EAA, et al., have been pulling together: The backyard builder is not any kind of ‘business’ at all, and so he is freed of all those expensive parasites and the costs of feeding them. The money he spends goes into the product – the product that he wants. You might have to think about this for a while. It really is a new idea!
8th March 2010:
I said there were two revolutions. At the same time that manufacturers were sinking into a mire of uncontrollable costs – which must, of course, be passed on to the consumer – another revolution was quietly growing at the opposite end of the economic spectrum. About 25 years ago we engineers started talking about “desktop manufacturing” and “rapid prototyping”. When advanced machine tools met computers, Adam Smith’s ‘division of labour’ became meaningless. Individual parts – a motorcycle frame, crankcase, etc. – could be made by machines with almost no labor involved (after the computer is programmed).
Consider the building of a motorcycle as a two-stage problem. First, all the individual parts must be made. Then, all those parts must be assembled. I think a lot of you would say that, if you were given all the parts, you could probably put a motorcycle together in a couple of weekends. But where do you get the parts? You might be surprised to learn that even a big company, Chrysler for instance, makes less than half the parts that go into their cars. Most things are ‘farmed out’.
When Gas Gas started making motorcycles, they hardly made anything. They had suppliers make the parts and Gas Gas was mostly an assembly and marketing operation. The idea is that the company would buy a widget made in Poland, a gizmo made in Italy, and a whatsit made in Ireland. Wherever the quality and price were right, there was the source of parts.
Okay. So why don’t those suppliers send all those parts to ME, and I will put them together? We can cut out a whole bunch of middlemen, I save a bunch of money, and I end up with the same machine. But let me suggest something a little more down-to-earth. Let’s say that I buy a CRF250 with a blown engine – cheap. I throw away everything that says “four stroke”, but I keep the clutch, gearbox, waterpump, etc. Next, I buy two KTM 105 cylinders, two pistons, two conrods, etc. Okay, now I have almost everything I need to make a very special motorcycle. You see, I don’t need 1000 unique parts, I don’t need 100 unique parts, but I might need a dozen. And this is where rapid prototyping comes to the rescue. I’m sure that everyone knows that there are manufacturing processes that are only practical and economical for large-scale production. What might not be so well known is that there are a lot of processes that are only practical small scale. In fact, the manufacturers are always building and showing off really trick stuff, and then they say, “But we can’t produce this, so forget about it!” But they did ‘produce’ one!
23rd March 2010:
When I referred to a unique motorcycle, I do not mean ‘one-off’, I mean: not like anything produced for the market. If everybody did their own thing, there would be no need for EMA. We need to bring people together for the reasons already given, and one more. Making almost anything involves unit-costs and fixed-costs. A unit cost is, for example, the material that goes into the part (plus waste). A fixed-cost, for example, might be the writing of the computer program that guides the machine that cuts the metal that produces the part. The machine itself is, of course, another fixed cost.
Let us say that for some part the unit costs are $100 and the fixed costs are $1000. If I make one part, that part must cost $1100. If I make 100 parts, each part costs $110. If I make 1000 parts, the unit-cost will be $101. So you can see that making one part is too expensive, and making 1000 parts is probably not worth the trouble – that would make me a manufacturer and I would then have to assume all those fixed-costs that society has deemed ‘necessary’.
Depending on what we are talking about, 100 parts may be too few or too many, but I hope I have made the point. And we must always choose the method of production and taylor the design for our small scale. What this leads to is a certain approach to deciding what is practical, or cost-effective, and what is not. The way to keep down costs is to keep things simple. Hello two-stroke!
If you look at the Model ‘T’, Henry Ford’s formula was endless simplification. “You can have any color you want, as long as it’s black!” Making cars different colors was a complication, and cost money, so Ford said, “Forget about it”. And simplification is just what manufacturers today cannot do. They must continually push up prices to cover the costs of all the parasites. They do this by selling you all sorts of junk that you don’t need and telling you that it’s “better, improved, high technology”. Eventually, they all end up making toys for the rich, which is exactly the opposite of what old Henry was trying to do.
This leads us to EMA’s guiding principle, which we have taken from Thomas Edison. In his laboratory, he got people going with this order: “There’s a better way. Find it!”
If I seem to harp too much about costs, it’s because this is the argument most often posed against doing things small-scale. The economist, E.F. Schumacher, said that, “…a considerable number of design studies and costings… have universally demonstrated that the products of an intelligently chosen intermediate technology [rapid prototyping] could actually be cheaper than those of modern factories.” And the problem is that people have not been taught to think this way. We were taught to work, make money, and then buy everything we need. To move in this new direction, small groups have to communicate, get organized, and pool their resources. This is what EMA is all about.
For anyone who wants to get into the economics and practical implementation of small-scale manufacturing, I suggest reading Schumacher’s book, ‘Small is Beautiful-Economics as if People Mattered’; and Roy Morrison’s book: ‘We Build the Road as We Travel’.
28th March 2010:
I have been dancing around with this, but some of you may have been paying too much attention to the music, so I want to make it explicit. This is the principle that the EAA introduced: If I make my own motorcycle (airplane, car, etc.), I must take responsibility for my creation and my use of that creation. Any question of liability vanishes. I cannot sue myself, nor can a government protect me from me (but they will keep trying!). With such a simple act, most of the parasites are lost – think: lawyers, tort litigation, insurance companies, government regulators, et al. And, of course, if you don’t need them, you don’t have to pay for them!
It must follow that your use of your creation cannot endanger others. This is why only the constructor of an ‘experimental aircraft’ is allowed to fly it. For motorcycles used off-road, this should not be a problem. EMA is not limited to dirt bikes, but operation on public roads involves many complications. (However, these vary from country to country and there are certainly many interesting possibilities for on-road applications if one can thread the legal gantlet.
Although the EAA is the model that best proves what can be done when a group of persons serve a common interest, it is not the only fraternity that may guide us. Years ago, Robert Washburn discovered that quite a surprising number of people were interested in building their own I.C. (Internal Combustion) engines in home workshops. Whenever he could talk to one of these persons, he didn’t seem to know what other engine builders were doing. Seeing a need, he started ‘Strictly I.C.’ magazine. He thought he might be able to pull these people together into a self-supporting community. It took a while, but that’s just what he did.
I think that almost anyone has to be amazed at what these guys do. There are DOHC fours, V8s, nine-cylinder radials – almost anything that you can think of, and some things that have never been seen before. And they make everything: pistons and rings, valves, crankshafts, camshafts, etc. About the only things they don’t make are screws, springs, and ball bearings. These engines, of course, are operational, but many of them are also works of art.
Another group that you may not be aware of are those craftsmen who make parts for vintage cars, motorcycles, and aircraft. If you have a car that’s worth half a million dollars, you don’t send it to a junkyard because it needs a part, and you won’t get a new one at Pep Boys. Somebody goes to work and makes a set of pistons, a crankshaft, or whatever it takes.
To repeat what I said earlier, the road has already been built. There are people out there making all sorts of things. The question is not: “Can we build our own motorcycles?” The question is: “Why aren’t we doing it?” The purpose of EMA is to bring to motorcycling what has been going on in other fields for a long time. The big problem is that we need to work together. So, to start with, EMA needs a home, that is, a home-page, a web site where we can toss around ideas: what we want; what we don’t want; who’s doing what; where can we get what we need; etc. Then, people like myself will have a way to pass on to others what we have learned, and together, we can push motorcycles into that ‘future’ where manufacturers are afraid to go. Of course, you can just forget about EMA, keep buying what is offered for sale, and that will tell the manufacturers that they are giving everyone what they want.
It’s your play.
12th April 2010:
I will give you now an example of what a few guys can accomplish in a backyard workshop. The quote comes from Nick Ienatsch. The encounter took place at Daytona, 1994.
“I exited the international horseshoe on the ultra-fast factory Bimota Tesi, tucked in and wound out, riding like the veritable wind at gale force. Suddenly a blue and pink streak flashed past on the rear wheel: Andrew Stroud on the Britten V-1000, leaving me and the rest of the pack in its roaring wake on the way to a complete and resounding Daytona victory. I’d never seen anything like it – this New Zealand-made, handcrafted work of art…”. There was nothing like it. The Britten had no frame – in the conventional sense. The steering head was part of a stiff brace that tied the two cylinder heads together. The swingarm pivoted off of the gearbox. The rear shock was attached to the front of the engine. There was no telescopic fork. The radiator was under the seat. All this strangeness might have been open to criticism, but it worked!
Some time later, at Road Atlanta Raceway, Ienatsch had a chance to race “John Britten’s backyard miracle”. His conclusion: “John Britten’s V-1000 is the neatest bike in the world. That’s a strong statement considering the bikes I’ve seen and ridden in the past decade, but I’ll stand by it. Understand this: Besides the Ohlins suspension pieces, the Brembo brakes and a handful of scattered parts, nothing on this bike can be traced to anywhere but the Britten workshop. … John Britten and his team imagined, designed, built and developed everything from the carbon-fiber wheels to the nasty bodywork to the fuel-injection system…”.
So, after blowing away all their machines, did Yamasakihonsuki rush to copy the Britten? Noooo. Will nothing change (for the better!) until another John Britten comes along? What if, instead of making everything, all in one grand effort, as the Britten team did, we make the “neatest bike in the world” in a series of small steps? We might concentrate first on the engine, then add the frame, then develop the suspension. What matters is that we leave the status quo behind and march, step by step, into the future.
24th April 2010:
With the Britten story, I have really finished. There is so much more that I could include, but if you are only going to buy what comes to market, you might be happier not knowing how good things could be.
To tie things up, let me describe the Sulzer RTA 84 two-stroke marine engine: bore – 31.5 inches; stroke – 94.5 inches. In 12-cylinder form, they make 48,360-hp at 87-rpm. “These giant engines (34-ft high) are designed for maximum fuel economy in long distance supertankers.” You probably didn’t know that almost all large engines (over 12-inches bore) are two-strokes. Why? Look at the formula for horsepower. If the displacement and engine speed (rpm) are the same, a four-stroke must have twice the cylinder pressure of a two-stroke to make the same horsepower. If you hit the nail half as often, you have to hit it twice as hard. To deal with the higher pressures, the conrod, crankshaft, all the bearings, the whole engine structure has to be made much stronger, i.e., heavier. As engines get larger, the cost (in size, weight, and money) grows proportionately. It makes no sense at all to build four-strokes, because there is no payback – the four-stroke doesn’t have any advantage to offset all the disadvantages. What is true for those large engines is also true for smaller engines. (Please don’t miss the story of the new Skidoo 800, which was recently tacked on to the end of part one – The Future of Two Strokes.)
“In 1936, work started on the development of a very highly rated, two-cycle gasoline-injected, sleeve-valve petrol engine for aircraft propulsion… . Toward the end of the programme , short-duration tests were successfully carried out at 354 BMEP at 4000 rpm…”. That was a single-cylinder engine of 1639-cc (5-inch bore) and those numbers give 358-hp. (Quote from the autobiography of Sir Harry Ricardo.) Try to imagine something like a Dodge Ram pickup powered by a neat, little twin-cylinder, two-stroke engine, conservatively rated at 400-hp. I’m referring to an engine that was running in the 1930s! That engine, built today with modern materials, lubricants, and electronic controls, would be a killer – of four-strokes. Ricardo Consulting Engineers tried many times in the next half-century to get somebody to adopt this engine. There were no takers.
Ricardo’s E65 was very different from the Sulzer, and both of those very different from a KTM 105. There are a lot of different ways to make two-stroke engines, depending on the application. So I will end this with a statement: From tiny model airplane engines to supertankers, NO power source has proven to be so flexible and efficient as the two-stroke cycle internal combustion engine. And there is still great potential to be explored.
What Are Your Thoughts On The Future Of Two Strokes and The Experimental Motorcycle Association?
Share your opinions and ideas with everyone. Where do you see engine design and dirt bike production heading? How do you feel about the information Tim has presented?