This section describes the basic function and the most important processes in the 2-stroke cycle a tuner need to consider in general and with the Bimotion 2-Stroke tuning software in particular.

The high performance 2-stroke engine is a pulse resonance engine which means that the operation in general and the scavenging in particular is not only dependent upon the pulses created from piston pumping but also from combustion properties. This is an important process to understand and is the reason why not only the cylinder decides the tuning degree but also the exhaust pipe, cylinder head and ignition. The charging efficiency is dependent on the pulse energy created from the cylinder pressure at exhaust port opening.

Bimotion uses advanced analysis tools as 2-stroke simulation software and finite element modelling for own research and to govern and visualize the theory.

Table of contents: 1.     Basic Function 2.     Combustion            2.1 Burn rate              2.2 Thermal barriers 3.     Cylinder Head 4.     Exhaust Port Opening   5.     Porting and Air Flow 6.     Exhaust Pipe 7.     Case Study

1. Basic function

A combustion engine is basically an air pump which follows the thermodynamic laws, i.e, the relation between pressure, volume and temperature. Air is pumped in, heated to expand and makes a piston (or turbine) to move and create work. The work is preliminary transported as torque and rotational speed, and secondary as kinetic exhaust energy. The residual exhaust energy can be used to run an additional turbine (turbo), expand in an expansion chamber to pump air through the engine (2-strokes) or just dissipate as sound (dragster's). In order to create as much power as possible, the engine has to change the fuel from chemical energy to as much mechanical work as possible. The residuals are unburned fuel and heat loss. A great deal of the piston work is lost by friction from all moving parts on its way to the final destination, the ground. A 2-stroke bike with efficient lubrication and without o-ring chain is usually loosing at least 15% of piston power.  A 4-stroker looses a lot more power with the additional moving parts as oil piston ring, cams, valve train and balance axles.

When evaluation an engine from port timing you can measure the piston displacement from Top Dead Centre (TDC) and calculate the angles in the Bimotion programs, but you need to know rod length to be exact. If this is not known, an approximate value of twice the stroke can be used. You can also print this Degree Wheel and paste on a CD to evaluate the port timing. (Click and print picture in 12x12 cm).
 

2. Combustion

The fuel property will decide the heat release, at which rate the fuel will produce work. A well designed fuel is an important factor to race engines. The heat release time from combustion is mainly decided from the fuel properties, air/fuel ratio and cylinder head design. The fuel ratio of H/C (hydrogen/carbon) in the CnHm molecules varies, n and m takes certain values for different fuels. During the combustion, the molecules break down step by step with different heat release at each step. 50% of the available heat is usually released within 5° -10° ATDC (After Top Dead Center).
 

The picture shows the relationship to crank angle and crank moment. As a schedule example, if fuel blend 1 releases most of the energy in a very short time after TDC, with high peak pressure.
Blend 2 releases heat in a longer period of time with moderate peak pressure, then the moment on the crank will be different over the time for the two cases, thus also power.

2.1 Burn rate
During the pre-state of an combustion, the flame front is expanding at approximately 20 -50 m/s, i.e 70-180 km/h or 45-110 mi/h. Note that this velocity is varying with rpm and mixture which means that we have to design the head to the worst case of our application. This is also dependent on internal flow and heat.

The picture shows three different ignitions at the same condition, a normal operation at which the burn rate varies.

We do not call this an explosion since it is a slow process. As a comparison, explosives as TNT burns at a speed of 6000 m/s. When we get small local detonations which can be defined as explosions, small pieces of the piston and head will be ripped off and deform the metal. The piston may expand at the deck by plasticity and size or blow a hole below the spark plug.
The reason of detonations may be several, but there are dependency to squish band design, temperature, rpm, mixture, compression, ignition setting, fuel atomization, exhaust pipe, etc.
 The right picture is a plot of the gas velocity at the squish band and piston velocity for a 125cc 2-stroke at 14000 rpm as a function of crank degrees. Note that the piston is moving at 20m/s away from the head already at 26 deg. ATDC. At this point, the combustion pressure is still 70% of maximum on the piston.

The piston is not moving symmetrically during the stroke due to the crank and connecting rod geometry.  It moves faster at the top and slower at the bottom. By using a short rod, this will give a 2-stroke more time for scavenging which is critical to high speed engines. The drawback is a high piston velocity at combustion and decreased torque, but the benefits usually wins.
 

2.2 Thermal barriers, coatings.

A thermal barrier is a surface treatment that is preventing the heat from warming the metal, thus reducing thermal expansion, metal softening and improving combustion. The piston crown is sprayed with a ceramic layer which mirrors the heat from the metal and instead of being transferred to raised metal temperature and finally more cooling need, is preserved and dissipates into piston work.
The technique of thermal spraying was developed in the space-, aircraft-, and power plant industry at turbine blades as a shelter to heat shocks. The combustion is as mentioned above not static, the magnitude and location of the heat release is changing all the time. The process can be resembled to a finger glove hitting the surfaces.

Thermal coatings to the crown of the piston are popular, typically, we only see a benefit on crown coatings when the piston is made too thin or too fragile for the application. Then the crown coating becomes a kind of band-Aid for a part that might prematurely fail in an uncoated situation. Increased power always increase the need of more cooling, not only from combustion, but also from friction. Protection from heat and friction is a necessity in modern race engines in order to prevent metal melting and/or to minimize the wear.

Skirt coatings minimize friction and allows pistons to run at slightly tighter clearances than normal. This is especially important on forged pistons that have a higher expansion rate than cast pistons. There is no downside to a properly applied piston skirt coating.

The inside of the cylinder ports are also treated to decrease heat transfer between the metal and the in going air flow. Coatings inside exhaust pipes are also common to prevent heat from being transferred to the engine and cooling system by heat radiation and convection.
 

3. Cylinder Head

High efficient squish bands speeds up the combustion by adding more kinetic (velocity dependent) energy to the gas mixture. They will also transfer more heat into the head wall surface, which will be a measured as a power loss through the cooling system. Why?
If you move your hand quickly in warm water or blow on your skin in a hot sauna you will feel that the skin gets warmer due to heat transfer from velocity. In the Bimotion Advanced Head program, this factor is called the Area/Volume factor. If a geometry change would give a higher factor with the same squish velocity, then we could expect more heat loss due to the increased area. The actual value don't need to be focused, but the changes to different head shapes are interesting to observe, especially if cooling is a critical factor in the original configuration. 

With high performance engines kinetic energy is needed and with short compression times (due to high rpm and short connecting rods) the heat transfer will not necessarily be too high to the cooling system. (Adiabatic compression). A head designed to a high tuning degree will work best with a high speed engine, and a moderate tuned engine will feel a great improvement with a moderate tuned head instead of an head without a tuned squish band at all.
As you probably already noticed, the head geometry will be somewhat dependent on the fuel type for a critical tuned GP engine and require a lot of testing. As mention before, it's not easy to simulate a head combustion by a software since it is irregular. But, there are certain rules that can be used to predict the function by software's. The Bimotion Advanced Head program uses a lot of well calibrated methods from physical tests and experience to design efficient 2-stroke heads.

The mechanism of a squish band is to push the gas as close to the spark as possible at the ignition phase. During the squish, the gas will also increase the vaporizing of the fuel and add kinetic energy which increases burn efficiency. This is one of the reasons why the ignition timing needs to be reduced with higher crank speed in a 2-stroke engine. The charging from the exhaust pipe is another. 
Usually, squish velocities of 25-50 m/s are the upper limit dependent on design, materials, cooling, fuel, etc. Too high squish velocity will transfer too much heat from the gas to the surrounding metal and will make the gas self detonate due to the energy increase.

The fuel usually burn with this mentioned speed and it will serve as a good limit for the squish velocity. A bad designed squish band will cause detonations which will destroy the surrounding metal and sometimes hammer the piston, making it plastic deformed over a big area and size due to the expansion.  When the gas at the squish band is moving into the centre, it will have to increase its velocity due to the fact that the area is decreasing. The red length is shorter than the blue length in the picture and the gas must pass these gates.  3.1 Head optimization

To optimize the squish behavior we need to have a constant squish velocity over the squish band. This is achieved by tapering the squish band height with the corresponding area ratio, so that A is the Squish Gap and B is the reduced height found as Y(C4) in the coordinate table of the program. This height reduction also reduces the inefficient burned volume. The blue line shows the mathematical correct squish band shape. We can see that a strait line will approximate the shape perfect over the squish band width.  

The squish  taper angle is not constant, it increases with increased squish gap A. The taper angle is tangent with the piston edge at B.

The Head geometry gets more important with high squish velocities, the surrounding surface needs to be protected from the heat transfer. This can be achieved by different thermal barriers (surface treatment) and with the gas itself. The secret key is to avoid the hot gas to transfer heat. This is why we like sharp inside corners from multistage heads. The two pictures below shows the difference.

In the right picture above there is no barrier to the head surface, more heat transfer to expect due to high gas velocity next to the head surface. In the right picture beside, the small gas pocket above the red line forms a natural barrier to heat transfer with low gas velocity next to the surface.

4.  Exhaust port opening

Many people talk about a port area as a target value and indirect refers to a header area, with an area factor for the port. That approach is a completely waist since port area don't give any information about port shape. Time area is still the standard unit for 2-stroke ports since at least 1971. An area change far up on the port will give a completely different change than on the bottom of the port. It would also affect the pipe differently. The time area distribution is simply different over the port height because the port is open different periods of time during the stroke.
In the Bimotion Advanced Port & Pipe program we can see this distribution in a chart. (Se pictures below.) We should be more careful with the machining precision at the upper part of the port since that area is opened for a longer time.

A square port gives an increasing time area distribution with port height.

The distribution curve will also show the effect of auxiliary exhaust ports which adds blowdown time-area. We need a certain blowdown time-area to equalize the cylinder pressure to the crank case when the transfer ports opens. If this not works, we get blow back to the crank case and that will drop the power rapidly.

 

Auxiliary exhaust ports adds time area which can be seen in the distribution chart.

   

The same port but without auxiliary ports.

   
Now, the blowdown timearea is not just a figure to match, it is dependent on how strong pulses the pipe deliver. If the pipes tuning degree is high (strong pulses), then we get away with less blowdown time area !
And the transfer ports don't need so much time for scavenging since the pipe suction wave is strong, pulling out the gas from the crank case. We could then have wide and low transfer ports. With low ports we can go for higher rpm without getting blow back into the transfers.
Why do we need to take all this in consideration for the exhaust port design? The answer is that it is not only a time area target value, the whole system needs to be investigated since there is an interaction, a balance between pressures as the pipe is affecting even the reed valve and the carburettor at BDC!!!
The normal procedure for race engine design would be to keep the exhaust port as low and wide as possible and to match the blowdown target (depending on pipe). The exhaust blowdown target is a statistical value for the tuning degree decided by the bmep target (braked mean efficient pressure). The port duration should not exceed the recommended value.
 

5. Porting and air flow

When using the Bimotion Advanced Port&Pipe, you should be aware of the validity of the calculations.
The time area values are only valid for the port apertures, not the port channels. That is, it's a measure of what the port can flow not what is really does flow. Bad shaped port channels may need more port- and blowdown time-area than recommended, or more true, the port channels may need to be reworked to a state of the art shape with smooth surfaces. The recommendations for the higher bmep's assumes this since it's based on race engines.

Now to a practically application, let us assume that you want more power not only the first 10 minutes of a race but also all way into the finish (less heat problems) and better drive, faster throttle response, faster starts, less clutch slip and a higher appreciation of enjoy. Perhaps you change parts as exhaust pipe, head, ignition, etc. and convince yourself that it became an improvement. Or just different ?
Let us then tell you a "secret".  There is a reason why the fastest riders use special tuned engines, tuned as a handcraft, not bought as single parts from the shelf. An engine which is correct adjusted or tuned manually is not believed to go faster. The difference should be so obvious that you get astonished and needs some time to adapt as a driver !

In order to make a qualified port work you either need long experience and a flow bench and/or an efficient software to calculate the correct relationships between port shapes and port areas. Bigger holes doesn't necessarily flow more air during a whole cycle since the flow is pressure dependent as well. There need to be a balance between kinetic energy (air speed) and static pressure. Big holes = high static pressure and low kinetic energy, and visa versa. The flow must be maintained until the ports are closed, otherwise backflow will occur with power loss, especially below peak power. But high speed power=big holes (?)... Yes, but this is why it is very important to locate the port area exact where it is needed, at the correct time of the cycle where it will meet the pressure pulses from the exhaust pipe most efficiently.
By other words, we need to work very carefully with port shapes and port timing. Not port area only. The Bimotion Advanced Port & Pipe program has been especially designed to take most of these parameters into account. It's not possible to tune a single port without having the whole picture.
 
How strong pulses can the pipe cope with ?
How much responsibility is laid on the pipe dependent on the exhaust port shape ?
Does the transfer ports match the pipe pulses ?
Does the transfer flow match the intake capability ?
etc, etc.

By using time-area (flow capability) as a design target for a certain tuning degree and a lot of recommended combinations we can by 40 years of experience say what will be a success combination and what will not. In fact, simulated and tested designs usually shows that it is not worth the extra work to try and find a more efficient combination or shape unless there is a large budget for test, simulation and manufacturing. But even if there is, this simplified method is needed in order to get the basic configuration correct as a pre-stage to more advanced analysis.

When 2-stroke cylinders are manufactured, different manufacturing- and production  limitations needs to be accounted for. It could be cast draft angles, tooling split lines, tooling wear,  production economy, etc. By other words, it's not possible to cast all shapes and desired surfaces from manufacturing. These built in restrictions must be adjusted by hand if maximum performance is wanted. Apart from the surface finish, the port shape can  also be adjusted to fit the system. Any adjustment done should be possible to motivate by the tuner, otherwise it shouldn't be done.  The picture shows the difference between raw manufactured surfaces and adjusted shape and area. The increased exhaust flange area and port time-area must be correct matched to the exhaust pipe, .i.e. the standard pipe may need to be changed to fit the new port properties.

6. Expansion chamber

When the exhaust port shape is decided from time area targets etc. we can dimension the exhaust pipe header diameter. This is critical to the pipes blowdown efficiency and the pressure build up. We need to pressurize the pipe with a strong and not too short pulse and the length will interact with the pulse resistance to create the necessary pulse shape. This means that it is not always better to use a diverging header (L1), sometimes a diverging header may result in that the cylinder exhaust evacuation is too rapid and causes the pipe to 'loose the breath' during the stroke. It will gain power at high rpm but lose in the lower range, mostly okay for race engines though. The phenomena needs to be viewed in a simulator to be fully understood.

To simplify some we say that the first diffusor (L2) acts in the lower rpm range and the last diffusor (L4) in the upper range. The rear baffle (L6,L7) angle decides the top end rpm range and power 'hit'. Steeper angles will increase the pulse strength and decrease the pulse length, i.e. shorten the rpm range at which the usable power is produced. The internal length between the diffusors will also decide the power production characteristics, so if the first diffusor is relative long then it will gain power in the lower rpm range etc. A pipe that is pressurized with an early opened port (long duration) will be able to retain the pressure through the steep angles and deliver a strong suction pulse back at BDC. With strong pulses engaged we don't need too large/high transfer ports. The pipe will then help and pull out the gas from the crankcase, even manage to open the reed valve and pull more air through the engine. When the transfer ports are closed, a second returning wave in the pipe pushes back the fresh gas that was spilled out into the header, in the remaining blowdown window that is. The charging pressure is often in the region of two atmospheres (bar) and way over that.

This is why the pressurization of the pipe needs to be well investigated together with the exhaust port, we simply cannot fit any pipe to an engine. It have to fit the exhaust port and even the transfer ports too!!

Finally, the stinger need to be large enough to let the engine breathe. A small stinger diameter (and long stingers) will increase the internal pressure and power but also make the engine run hotter.  The length becomes critical to pulse resonance over about 9000 rpm. At that point the plunging effect at the stinger end will interact with the pipes internal pressure and help to lower the evacuating pressure at piston BDC. However, for high rpm engines, the stinger in general needs to be smaller than on low rpm engines. High frequency pulses have shorter wave length and will fit a smaller pipe better. But since the high rpm engine also needs to breathe more this can often even out.

 

7. Case Study Download the Bimotion case study of Kawasaki's legendary KX500 2004 motocross engine ! A step by step guide about how to calculate a 2-stroke. Click the picture ! [2.6Mb]