Some Thoughts on Squish Action
During the last 20 years, but more so during the last 18 months I have been
taking an intensive look at the effects of squish. This is part of an attempt to
include a combustion model into simulation software where the model
calculates its parameters rather than having it prescribed by the user.
Determining the values to prescribe combustion accurately one requires a
fairly sophisticated engine test cell with the capability to record combustion
pressure traces. As part of this exercise I have changed my views on squish
quite a bit. Some of what I found will be described next.
Discussing squish without first taking a brief look at turbulence is a waste of
time as the main function of squish is to increase the turbulent intensity in the
combustion chamber. Without turbulence in the combustion chamber we
would burn the mixture at the laminar burning rate which is ten to twenty
times slower than the turbulent rate. This would make practical engines that
rev higher than about 1500rpm an impossibility.
Turbulence in a 2stroke engine is caused by three mechanisms. The first and
most important of these is the scavenging process. The kinetic energy of the
inflowing scavenge streams is converted at a specific rate to turbulent kinetic
energy which leads to an increase in turbulent intensity. At the same time this
turbulent intensity is being damped out by the viscosity of the fluid (air/fuel
mixture) and converted to heat. From this there follows at least the following
• The turbulence generated by the scavenging process influences the
combustion rate so it is to be expected that a 2port system will have
different combustion characteristics from a 7port system.
• The faster an engine turns the less time there will be for the turbulence
to be damped out by the viscosity thus increasing rpm will, inside
some limits, increase the burn rate provided that the turbulence created
by the scavenging process does not decrease faster with the increase in
The next generator of turbulence is the piston movement. As the piston
displaces fluid during the up stoke it imparts kinetic energy to the fluid which
gets converted at a certain rate to turbulence. Obviously this increases with
rpm so higher rpm will have a higher turbulent intensity and thus faster burn
rate unless the dissipation rate is greater than the generation rate.
And then, the third generator of turbulence is the squish action. The squish
action causes gas to flow towards the center of the cylinder. The speed of this
flow is what we calculate with various pieces of software and is known as
MSV or maximum squish velocity. Now to repeat what has been said in the
previous two paragraphs; this kinetic energy is converted to turbulence at a
rate depending on the conditions inside the cylinder at that time. Also, MSV is
a function of squish band geometry and rpm. This leads to the following
• As rpm increases so does MSV, and so does the available kinetic
energy that can be converted to turbulent intensity. If this does get
converted the burn rate will increase.
• As the squish area ratio increases and or the squish clearance decreases
the MSV also increases with the same results as for the previous point.
The picture that emerges from this is that depending on the residual turbulence
in the cylinder at the time of the squish action, the squish action can either
increase the turbulent intensity or have no effect on the turbulent intensity.
This can translate into some of the following:
• On an old 2stroke with two transfer ports and high primary
compression the squish action can increase the burn rate so much that
maximum pressure occurs so early that a large amount of negative
work is done, a lot of heat is generated and power is lost unless the
timing can be adjusted. On older engines that can sometimes not be
done so a squish head is detrimental.
• On some engines the increased turbulent intensity can shorten the burn
times after max power so much that it kills over-rev. On others where
the burn time was too long it can improve over-rev.
• If the port and pipe combination is a bad mismatch that looses a lot of
fresh charge (and thus kinetic energy) during the open cycle increasing
the MSV will usually help such an engine and vice versa.
There are many more such examples but the general trend is clear: Without
knowledge of the turbulent intensity inside the combustion chamber at the
time of combustion it is not possible to say whether the engine needs more or
3. Squish Action
There are unfortunately not many papers published on squish action, mostly
because in most 4strokes it is of lesser value. They have to make space for
valves which is more important. They do however generate turbulence
through tumble and swirl which is not really practical for loop scavenged
2stroke engines. There is one class of 4stroke engines that use squish though
and that is the diesel engines with the bowl in piston type combustion
chamber. Measurements on these engines have shown that the calculated
squish velocity using the MSV equations corresponds well with the measured
results. The conclusion is that the MSV calculation is acceptable, the problem
is with its worth.
Another side effect of the increased squish velocity is that the convection heat
transfer between the gas and the piston and head is increased. This is often
claimed as the primary reason to use squish action to stop detonation.
Preliminary calculations have shown that this effect, especially in high
revving engines, is almost negligible.
4. Combustion Picture
To help with forming a better mind picture of the combustion process a
simulation was done on an YZ250U engine and the dimensions of the burnt
sphere of gas calculated for each time step. This, together with the piston
position from TDC was written to file and a series of seven pictures generated
A squish clearance of 1.2mm and a squish area ratio of 50% were used. This
resulted in a MSV of 28m/s at 10deg BTDC.
In the model the following assumptions were made:
• Combustion proceeds in a spherical shape. This is not far from reality
but without a detailed computational fluid dynamic simulation not
possible to predict.
• The ignition timing is set at 20deg BTDC. For the tested engine it was
19deg but as the pictures were generated at 10deg intervals this was
• The measured delay period was 9deg but this was changed to 10deg to
get the combustion to start at the measured position and to correspond
to a 10deg interval.
• The actual combustion period was 51deg but was changed to 50deg
once again to correspond to the 10degree intervals.
The first picture shows to scale where the piston is relative to the combustion chamber at the point of ignition.
The second picture shows the piston position at the end of the delay period and thus the start of turbulent combustion (10deg
BTDC). This is also the point where the squish velocity is at a maximum (MSV). The flame kernel is now bigger than a turbulent
The next picture shows the piston at TDC. As can be seen the flame sphere has grown substantially in size and is just beginning
to touch the piston crown. If a plug with a protruding tip was used the flame would have touched the piston a few degrees
At 10 degrees after TDC the hemisphere of the chamber is almost fully enflamed and the flame front is on the point of moving
into the squish area. As can be seen the squish area is rapidly opening up and will be experiencing reverse flow
At 20deg ATDC the flame front is about halfway into the squish area and the squish clearance is quite big already. No chance
of the flame in the squish area getting quenched.
At 30deg ATDC the squish area is now fully enflamed and only a small part of the piston crown and of the cylinder bore is
not yet enflamed. At this point even the reverse squish flow is negligible.
The final picture is at 40deg ATDC and shows the piston position with the complete chamber enflamed
This is still way before the exhaust port opens which points to the fact that for burning mixture to exit the exhaust port
something serious has to be wrong with the combustion process.
So what have we learned:
• As far as squish is concerned whether it helps or not depends on the
residual turbulence at the time of combustion.
• As for squish quenching the flame at the edge of the squish band to
stop detonation is not looking very likely.
This is not the final word on the subject. As I learn more I keep changing my
mind and I am sure it will continue to happen for a while yet. Hopefully
somebody reading this will be able to help me right where I am still
misunderstanding squish and definitely turbulence!
©Copyright 2005, Neels van Niekerk, published with permission.