Versione italiana
Copyright © 2017 Fabio Bizzetti, all rights reserved. NOTE: the disclosure and dissemination of this article is welcome and expressly permitted, provided it is exclusively in the form of a direct link to this web page: any other use is strictly forbidden and will be prosecuted by law.
LAST UPDATED on October 13, 2021
Squish in 2-stroke engines
If you have arrived here you probably already know what squish is, but a brief introduction will do no harm: the "squish crown" or "squish band" or "squish ring" is the outer area of the combustion chamber and, unlike the actual "dome", is shaped so as to be at a short distance (in the order of a millimeter) from the piston to the top dead center. The "squish effect" (which takes its name from the hypothetical noise produced by the crushed mixture between piston and cylinder head) is what we are looking for: the piston, rising to TDC, will compress the fresh mixture against the squish crown, causing it to splash towards the center of the combustion chamber, where we normally find the spark plug.
In this image we see schematically the combustion chamber, seen from below:
The squish crown will rarely be parallel to the piston crown, more frequently inclined upwards by 1 degree or more than the same (clearly, in the case of domed pistons the inclination of the squish crown must be greater than in the case of flat top pistons).
For obvious reasons, a negative squish is never and in no case allowed, i.e. thinner inside than outside: not only would it not have the desired effect, but it would detonate the mixture inside the closed cavity created.
The optimal surface of the squish crown ranges from 45% to 55% of the surface of the piston crown, the most "free" parameter is the height (minimum clearance, at TDC, between cylinder head and piston) of the squish crown, i.e. the “squish clearance”, and is what we will discuss in this article.
What is commonly thought of squish clearance is quite wrong. It is imagined that thinning it serves to make that part of the volume of mixture subtracted from the squish crown, between piston and head, participate in the combustion, as if the combustion took place instantly and completely at TDC, statically. On a 125-144-250-300cc 2-stroke enduro engine, combustion begins when the piston is still almost 2 mm below TDC, and is completed when the piston has already dropped several millimeters, if not centimeters (in the case of of part-throttle operation, i.e. with very slow combustion). So, in the meantime that it is reached by the flame front, the effective "squish" has become very high, in the order of millimeters when not centimeters. Therefore, it is not so much to decrease the volume of squish that leads a greater quantity of mixture to be free to participate in the combustion (even if in fact the boundary layer, therefore gasoline that normally does not participate in combustion, becomes thinner), but the primary function that is actually significant of the squish is to increase the turbulence in the combustion chamber, with the express purpose of speeding up combustion.
Combustion in the absence of turbulence is very slow, in the order of 30-50 cm/s (or up to 2 meters per second in conditions of temperature and pressure found inside the engine around TDC). But it is sufficient to make the calculation (taking into account the bore of the engine, or rather the volume of the combustion chamber since the flame front in the absence of turbulence would develop into a spherical shape) to immediately realize that with a combustion rate of generally 2 m/s it would not be possible to exceed low speeds even using a considerable ignition advance (in itself a bearer of inefficiencies because too early combustion develops pressure when the piston is still rising towards TDC, effectively braking it in that phase).
It is only thanks to turbulence that combustion can be sufficiently speeded up. It's like a fire when a strong wind blows… it propagates much faster than at rest. A good squish makes the peripheral mixture splash away towards the mixture in the center of the chamber at a speed of up to 30 m/s, almost a hundred times higher than the normal speed of flame propagation, and therefore once the spark plug has ignited a first flame front, all those turbulences (vortices and micro-vortices) will spread the flame in a very short time around the main combustion chamber. This very much speeds up the combustion, which is what we want in order to develop pressure and therefore mechanical work in a useful time, i.e. before the piston has already dropped too far, and that consequently the gases escaping from the exhaust port are not too hot (the energy that is not converted into mechanical work is expelled in the form of heat). Also, the engine runs more cleanly (the more complete the combustion, the cleaner the engine runs), and finally, avoid the spontaneous ignition of distant pockets of mixture (detonation). In other words, we want to have the fastest and most complete combustion possible.
At this point it should be clear that without turbulence in the combustion chamber, our engines would not even be able to run at speeds above idle. On 4-stroke engines there is already a fair amount of turbulence thanks to the swirl and tumble motions, but a 2-stroke will always have squish as its primary source of turbulence, hence its importance.
Turbulence is that thing that you absolutely do not want either in the intake or exhaust ducts but which, on the contrary, is advantageous in the combustion chamber.
The squish effect can only be achieved effectively with very thin squish gaps. On a 250/300 2T engine, having the squish at 1.80mm or 1.60mm makes little difference (basically there is no important squish effect in any of the 2 cases), while going from 1.40mm to 1.20mm already causes a significant increase in the speed of the mixture that is projected towards the center of the combustion chamber, with the purpose as mentioned above to "mix the flame".
To get all the benefits (which are manifested above all at low revs and partial throttle) it is advisable (on 250 and 300 2T, for convenience we will refer to values suitable for these displacements in our examples) a squish height of no more than 1.20 mm, otherwise all benefits are quickly lost. It is possible to go below this value, up to 1.00mm or even less, but only on engines in order and checked frequently, otherwise in special cases there is a risk of collision between piston and cylinder head, or that the pressure increase of the mixture is too high in the squish band and causes the detonation itself. On the 125/144, on the other hand, a squish height of no more than 0.90mm is recommended. Also in this case it is possible to go down, up to 0.70mm, in engines that are in order and checked frequently, that is, they do not develop excessive play or deposits on the piston crown and on the head such as to further reduce, significantly, the height of the squish.
Regarding the small loss of over-rev that many fear, those who can freely remap the ECU (such as PowerCDI owners) are not subject to it (just retard the ignition the necessary amount at the regimes where the turbulence is now "excessive"), and those who cannot should worry about other things if you want an optimal over-rev. For example, that the motor is not jetted too rich, because this is the main cause of loss of the over-rev, much more than using an optimal squish while maintaining the original mapping. In any case, even on non-remappable control units, the "pros" of having the optimal squish are however on the whole much higher than the "cons", especially for those who mainly use medium-low speeds and part-throttle.
On stock engines a secondary but positive effect is also the increase in the compression ratio that derives from lowering the clearance (thickness) of the squish. Also the raising of the compression ratio has the effect of increasing the performance of the engine (due to the improved thermal efficiency), an effect that is added to that of the squish. Sure, the motor won't double the torque at low end, but the difference going from a squish height of 2.20mm (like the one found in many stock motors) to 1.20mm, feels quite significant. The engine will run cleaner, have more torque, and greater efficiency.
Despite the greater compression, the tendency to detonation in this case does not increase, indeed it decreases, because a good squish effect has a very great anti-knock effect (it was studied and invented in 1919 by the engineer and pioneer Sir Harry Ricardo precisely for this purpose, not for increase of performance, although this side effect is certainly positive) and amply compensates for the pro-detonation effect of the increase in the compression ratio.
If you can get to 1.20mm or less only by putting thinner cylinder-crankcase gaskets (which, however, for enduro bikes at least, it is advisable regardless, because mounting thinner gaskets lowers the port timing and increases the torque at low revs, even if much less than is commonly believed), but it is extremely difficult, especially with stock heads, to reach that squish height only by fitting thinner gaskets. It is therefore necessary to go to the turner and have the head "skimmed" by a few tenths to reach the desired size.
In summary, to increase the pull at low revs and the cleanliness of the engine, especially at part-throttle, the most economical and effective processing that can be carried out is to mount the thinnest cylinder-crankcase gaskets available (this is to lower the port timing and implement at the same time part of the increase in the squish effect) and turning the head (an operation that in a specialized workshop they do for negligible figures) just enough to reach 1.20mm or below. Thinking of succeeding only with gaskets is utopia, because to reach a squish of 1.20mm or below (always referring here, for simplicity, to the values suitable for 250/300 2T engines) you will most likely also need to turn the head.
Beware, however, that in this way you will also have a more snappy and reactive engine, "bad" in some respects, so it may not be a modification appreciated by everyone. As we have already explained, when well jetted you have a very reactive engine, therefore "nervous" and someway dangerous to ride, but also much more precise in the throttle response, so if you have sensitivity on the right wrist you will find it preferable to a lazy engine, without much squish, “sluggish", jetted very rich with rough running part-throttle operation, it does not even fourstroke but worse, it fires irregularly, if necessary you give it a little more throttle and the engine responds by drowning even more instead of increasing the speed, with an inconsistent throttle response. There are those who prefer the engine like this, and those who do not, and here we enter the subjective sphere. If you are a PowerCDI owner, however, the choice does not arise: having a powerful and ultra-reactive engine brings only advantages, because the Dynamic Power Control (see the paragraph "Insights on the DPC (DynamicPowerControl)" on the main page of the PowerCDI website) has no problem to make the engine mechanically more powerful and responsive, and at the same time extremely docile, when needed. With the Dynamic Power Control of the PowerCDI you have all the advantages of an optimal squish, without the disadvantages.
FAQ:
Q: How do I measure the clearance of the squish?
A: To measure the clearance of the squish DO NOT use a piece of L-shaped solder by inserting it into the spark plug hole, the measurement would not be accurate at all, with an error even of 2 tenths. It is necessary to measure it simultaneously on 2 opposite sides, to eliminate the lateral play of the connecting rod.
Rather than spreading a straight piece of solder, it is preferable to make a kind of small "omega" in the center, in order to obtain, by compressing it, a slight elastic effect (almost a spring):
Do not use a wire that is too thick compared to the size of the squish (so it is good to make 2 attempts, one to get a rough idea of the clearance, the second to make the precise measurement), because solder wires contain a core of flux and if you squeeze it too much they will have a slight "elastic effect" which will slightly distort the measurement (it will give a bigger result than the real one).
If you do not have solder wire of sufficient diameter, you can twist 2 or more pieces of it or, better still, make a "ball" of molten solder, and then crush it (with pliers or a hammer) to reach the necessary clearance, which must be higher than that of the engine squish by only a few tenths. As a first attempt you can use a clearance of 2.50mm, since in many enduro bikes the squish, as standard, is a little thinner than that value (generally 2.10mm - 2.20mm).
Finally, it is important to know that some heads have the squish parallel to the piston crown, others don’t, so you have to make sure to measure it with a caliper capable of measuring to hundredths of a millimetre, on several points. For example, on many KTM 300 engines equipped with a flat-top piston the squish crown is not parallel to the piston, the outer part is about 0.25mm thinner than the inner one, so you have to carefully measure along the squish radius. On the other hand, on some aftermarket heads the squish band on the head is parallel to the one on the piston, so the measurement is the same, whatever the radius at which we measure.
Obviously the part at the outer extreme, the one in direct contact with the cylinder, will be thicker because it insinuates itself into the gap between piston and cylinder (gap that exists to allow the piston to expand without causing seizure), and should not be considered as a measure of squish, this is an insidious aspect to keep in mind, otherwise if with the caliper we measure the outermost part of the solder we will measure a much higher value than the real one, that is the very small part that has crept between cylinder and piston.
For the reassembly of the cylinder head it is always recommended to use a torque wrench and to tighten the screws with a cross sequence in 2 or 3 stages up to the torque value specified by the manufacturer.
Q: Why do the manufacturers insist on giving us motors with 2mm and more of squish? Are there any contraindications (for example regarding the reliability or the duration of the engine) or are simply "standard" conditions sufficient and enough to the average user?
A: There are substantially no contraindications on the reliability and durability of the engine, since the increase in performance occurs mainly at reduced throttle and low revs, and therefore if the driver in a certain condition needed a certain amount of torque that he will still be looking for from the engine, but it will do so by (unconsciously) decreasing the amount of throttle, therefore from a mechanical stress point of view in normal riding, at part-throttle, there are no big differences, it is the efficiency that increases (less fuel consumption for the same mechanical work produced) and this is certainly not a negative aspect. Apart from what could also be a choice to make the engine intentionally more “sluggish” and therefore more manageable by an amateur (since there are no dynamic engine power management technologies present on the standard ECU), the production processes of the manufacturers have manufacturing tolerances and cannot devote too much time to the care and tuning of the engine. They have to automate the production process as much as possible, and disassemble and reassemble the engine to find the best gaskets, turn the cylinder head, etc... to cancel these tolerances, is not in the order of things that motorcycle manufacturers feel compelled to do (nor can you blame them for that). Among other things, it would not even make sense to deliver the bikes with optimal squish, and then jet them as the stock jetting (even there, for specific reasons, but this does not mean that the user must then maintain the stock jetting, which often is so rich as to be unacceptable, especially by those enduro riders who use constant and partial throttle a lot).
The tuning is the task of the owner or the tuner, not the manufacturers, which only have to give a base to work on, perhaps with wide tolerances and poorly finished, but which conforms to certain standards. After all, as we have seen above, wanting an optimal squish is not a universal thing, and some riders, not equipped with electronic engine management (such as the Dynamic Power Control of the PowerCDI) may also prefer a high squish clearance with all that goes with it. In this case, if the bike had been delivered from the factory with an already optimal squish, it would be much more complicated to bring it to the greater clearance than it is, starting from excessive clearance, to bring it to the optimal one from the point of view of combustion and pure performance.
Q: How much does the compression ratio increase on my 300 2T if I take 6 tenths off the squish?
A: The calculation is very simple: we start from the bore, in this case of 72mm, divided by 2 to obtain the radius, then squared and multiplied Pi, we have 40.715 cm2 of piston crown area, multiplied by 6 tenths (0.06cm) we find 2.4429 cm3 (or cc, which is a different way to express the same thing).
Now, if the combustion chamber volume (complete with squish) was 27cc before, and the displacement was 293cc, we had a geometric compression ratio, calculated as (293cc + 27cc) / 27cc, of 11.85:1. Switching to 24.56cc (27cc - 2.44cc) of combustion chamber volume, we obtain a new compression ratio of (293 + 24.56) / 24.56 or 12.93:1.
So one more point of CR, but as already explained it will not increase the tendency to detonate, on the contrary, especially if the new squish is effective, i.e. no higher than 1.20mm, the engine will still be less likely to detonate than before, when it was 1 CR point less compressed but with the squish at 1.80mm.
Q: I put the thinnest gasket between crankcase and cylinder, lowering the diagram, but now the piston at bottom dead center does not fully reveal the transfer ports, is that a problem? Do I have to change the gasket to one that is thick enough so that the piston also completely reveals the lower part of the transfer ports?
A: The fact that, by fitting a thinner cylinder-crankcase gasket, the piston occludes the lower part of the transfer ports has little or no effect on performance, and in any case there is little to do, not being able to "file" the piston crown, but in any case the negative effects are much smaller than the advantage obtained by lowering the timing (provided that the premise was to increase the pull at low revs).
The piston, going down, at a certain point begins to uncover the transfer ports: it is there that most of the gas transfer takes place (thanks to the primary compression of the crankcase but above all to the vacuum in the cylinder created by the expansion chamber exhaust), then the difference of pressure between transfer and cylinder decreases and although the transfer ports remain open, their surface, however very large at the BDC, becomes less influential on pressure drops. So the fact that the underside of the transfer ports is not completely uncovered at BDC is of little influence and can be overlooked when looking for the benefits of simply lowering the cylinder through the use of a thin gasket, in order to lower the timing without modifying the cylinder and, more importantly, to approach the optimal squish value without turning too much material from the base of the head.
Copyright © 2017 Fabio Bizzetti, all rights reserved. NOTE: the disclosure and dissemination of this article is welcome and expressly permitted, provided it is exclusively in the form of a direct link to this web page: any other use is strictly forbidden and will be prosecuted by law.
Reach the main PowerCDI website.