[GP125] All that you wanted to know on Aprilia RSA 125, and more, by Mr Jan Thiel and Mr Frits Overmars (PART 1) (Locked)

With 'theoretical cycles' you mean the Carnot diagram 8)? Yes, it would be nice to have a really fast combustion; then you could ignite later and the heat losses would be smaller. But there is a big gap between theory and practice.
In a 125 cc race engine the spark-plug sparks at about 14° BTDC @ 13000 rpm. Then combustion starts really slow: after the spark another 10° go by before you can detect the first signs of combustion. Then combustion continues for a little over 40° (this is at full throttle; with less cylinder filling, combustion is a lot slower).

Is it useful to slow down combustion on some operating phases? Yes and no. I will try to explain this in a little physics lesson.
No matter which way the piston is moving, as long as there is combustion, the pressure in the cylinder keeps rising. Only after combustion is finished the expansion of the burnt gas can begin.
Slow combustion means that the piston is already well on its way down before expansion can begin; it means less expansion for the burnt gas in the cylinder before the exhaust ports open.
Less expansion means less cooling down of the gas in the cylinder: it is still hotter when it enters the expansion chamber. In hotter gas the speed of sound is higher and that means a higher resonance frequency for the cylinder-pipe system, so it works better at high rpms.

But how do you slow the combustion speed down? Less squish? Mixture too rich? You do not want to do that...
Fortunately there is a simpler solution. We do not slow combustion down; we just start it later: we retard the ignition timing. As far as the exhaust gas temperature in the expansion chamber is concerned, the effect is the same: the engine runs better at high revs.
That is the reason for programmable ignition systems.
Below the power band the ignition advance can be more than 30° so that there is a whole lot of expansion; the burnt gas contains hardly any energy when it enters the exhaust pipe, so the exhaust pulses that arrive at the wrong moments at low rpm, are weak and will not disturb the scavenging too much.
At the rpm of maximum torque the ignition advance is about 14° (careful, a litte too much advance here and you get detonation), and at maximum revs the advance can be 10° or less (Honda even went to 3° after TDC ).

yes, and we are to use more advance when the engine isn't "on pipe" (low-mid rpms) also because of fresh charge pollution by the unburnt gases.

Frits, on the simmetric two stroke design you posted some pages ago, aren't the pulses from the exhaust pipes on the way to distort the central fresh mixture flow, in such a way that they are not able to scavenge well the upper part of the combustion chamber?
on the classic schnuerle design the exhaust poses a great deviation on the flow coming out from the ports.
also on the new design there is no coanda effect that keeps the gases attached to the wall so it should be even more sensitive to exhaust pulses.

huh... and sorry for the intrusion

Welcome Nasone32

a quick presentation would be great

Thanks

Marc a écrit:

nasone32 a écrit:

Frits, on the simmetric two stroke design you posted some pages ago, aren't the pulses from the exhaust pipes on the way to distort the central fresh mixture flow, in such a way that they are not able to scavenge well the upper part of the combustion chamber?
on the classic schnuerle design the exhaust poses a great deviation on the flow coming out from the ports.
also on the new design there is no coanda effect that keeps the gases attached to the wall so it should be even more sensitive to exhaust pulses. huh... and sorry for the intrusion.

Ciao, Nasone32, benvenuto quoi. Ma qual'intrusione? Non te preoccupare . And now, to maximize the joy for all francophones I will continue in english (I just wish my french was a bit better....).
The pulses from the exhaust pipes return to the cylinder after the transfer ports have closed, so the scavenging flow from the transfers is not disturbed. The mixture that is pushed back from the exhaust ducts, enters the cylinder in the same direction as the scavenging streams, just a little later, so the central flow is enhanced, not hindered.
Both in the classic Schnürle design as well as in my symmetrical system the change of direction from the upward scavenging angle to the cylinder bore is so abrupt that the coanda effect is not capable of clinging the scavenging flow onto the cylinder bore above the ports (except when the port has a very big upward angle, which is bad for the effective flow area).
Instead the coanda effect causes the incoming flow to cling to the piston crown, which is very desirable because it is the main cooling factor for the piston.
This is also why I do not like flat pistons; the coanda effect is much more effective when the piston is domed.

Thank you Frits, understood.

Sorry but this morning I was a bit in a hurry so i'll do a quick presentation
I'm an Italian two-stroke-lover, 21 years old i work mostly on scooter engines and differently from the rest of the scooter world I work with theory, flowbench, simulations etc.

what has all this to do with Aprilia? well now i'm working on a new scooter cylinder which comes directly from an old Aprilia design, take a look if you want:
this is the 90cc scooter cylinder (50mm bore 46mm stroke)
[Vous devez être inscrit et connecté pour voir cette image]
looks much like the aprilia one doesn't it?
[Vous devez être inscrit et connecté pour voir cette image]

i'm also developing a new 100cc (50mm x 50mm) with bridged exhaust intead of boosted exhaust
here it is, freshly casted (on the right) [Vous devez être inscrit et connecté pour voir cette image]
and here after some porting
[Vous devez être inscrit et connecté pour voir cette image]
the 50mm stroker doesn't fit in the original scooter crankcase so i have to cast also new crankcase, it's all a work in progress... little time to work on engines unfortunately

well just for your knowingness here is another photo of the aprilia one
[Vous devez être inscrit et connecté pour voir cette image]

and here is mine (still got to work a lot on it)
[Vous devez être inscrit et connecté pour voir cette image]

since i can cast my own cylinders it would be nice to try your new design (but I don't have time at the moment).

Frits,

In fact you use the "flat" part of the MBT (maximum brake torque) curve?
that allow you to modify ignition time without change the torque a lot.

I knew this system was used in GP500 and I know it is used for cold start calibration in automotive(to reduce pollution)

But do this system allow you a big range of "tuning" for the pipe?

nasone32 a écrit:

i'm working on a new scooter cylinder which comes directly from an old Aprilia design, take a look if you want:
this is the 90cc scooter cylinder (50mm bore 46mm stroke)
[Vous devez être inscrit et connecté pour voir cette image]
looks much like the aprilia one doesn't it?

Looks more like a Sprocatti to me

Citation :

since i can cast my own cylinders it would be nice to try your new design (but I don't have time at the moment).

That might not be such a bad idea.... I am developing the cylinder as a 125, but I know of several people who would love to have a 50 cc version.

Yes i am working closely with Franco Sprocatti , if I remember well he made some castings for Aprilia in 1998 too. the first castings are from him, the 50mm stroker is a prototype.
I am casting my own next days.

Camus14 a écrit:

Frits, In fact you use the "flat" part of the MBT (maximum brake torque) curve?
that allow you to modify ignition time without change the torque a lot.
I knew this system was used in GP500 and I know it is used for cold start calibration in automotive(to reduce pollution). But do this system allow you a big range of "tuning" for the pipe?

In a racing two-stroke any ignition change directly influences the torque curve. I do not know about automotive applications, but yes, variation of the ignition can considerably change the behaviour of the engine. In GP500 this was used to make the engine character less violent in the low gears and in the wet.

@ nasone32: I noticed that the inner radius of your transfer ducts ends in a sharp edge at the cylinder bore.
Most cylinders are like that. But not the Aprilia RSA...
Below I have tried to show this, using a picture of my FOS cylinder (which has radii identical to the RSA).
There is some good horsepower in that modification 8)

[Vous devez être inscrit et connecté pour voir cette image]

thank you very much Frits!
didn't notice that. radiusing edges should always be a rule to follow, except in the exhaust/transfer ports where i always stay on the sharp side (but keeping an eye on the piston ring's life). I believe this makes sharper pressure waves travelling in the ducts... if i am right.

thinking about your simmetric design... how is low/mid range operation?
i would couple it with a CVT, which is very nice because keeps the engine on almost fixed rpm, but some torque is needed because the clutch engages at *max* 9000 rpm then the engine has got to rev until its working rpms (which can be very high with this design i suppose) on an almost fixed ratio with the rear wheel.

well if the engine runs at very high rpms than it could be possibile to use a shorter gear ratio by the way, partially o totally solving the problem (if any)

also it might work better with a slightly oversquare design? (bore>stroke)
with a larger bore should also be easier to make all the transfer ducts act simmetrically (thinking about a small 70cc, typically 47,6 x 39,3)

Tuned two-strokes without exhaust power valves always have the same type of power band: at 2/3 of the rpm of maximum torque there is a deep gap in the torque curve because the positive exhaust pulse arrives back at the cylinder at exactly the wrong moment.
With my symmetrical cylinder design the maximum torque rpm can be about 25% higher than for a conventional cylinder, but the same goes for the torque dip, so the engine character is conventional, only with higher rpm figures.

An oversquare engine is not a good idea for a two-stroke; I will try to explain this with an exaggerated example.
Let us assume an engine with a bore and stroke of 40 mm.
Now we make the bore twice as big: 80 mm. That means that all ports can be twice as wide. But if we want to keep the cubic capacity the same, the stroke must be 10 mm instead of 40 mm. And that means that all ports will be ¼ as high.
So all port areas are halved, as are the rpms of maximum torque and maximum power. If we assume that the quality of the cylinder filling remains the same, the oversquare engine develops only half as much power as the square engine...

What do you thing ,Frits, about injectors in the transferts ports?
Do they cool the charge ?
What fuel pressure is used generally ? I know there is a increase when is above 8 bars but it is forbidden in most of champs but set up to 30 to 50 bars in dragsters (better gasification)

Fuel pressure is about 3 to 5 bar for indirect injection and about 20 times more for direct injection.
Injectors in the transfer ports do cool the charge - but not very effectively, because there is very little time available for mixing and evaporation of the fuel. If you want to cool the charge, injection in the intake duct is much more effective because the fuel is much longer on its way to the combustion chamber, and meanwhile the charge is intensely stirred by the crankshaft, especially by the conrod. And the big-end bearing needs that fuel.

This is one of the two big problems with direct fuel injection in a racing two-stroke: an Aprilia RSA125 at full throttle on the test bench consumes 1 drop of oil every 30 crankshaft revolutions. That may be enough for lubrication but it certainly is not enough to cool the bearing. Without fuel passing through the crankcase the engine will not live very long...

The second big problem is creating a homogeneous air/fuel-mixture. With direct injection, even when you start injecting at BDC, there is only one half crankshaft revolution available for mixing and evaporation. In the RSA at 13000 rpm that would be 0,002 sec...

With indirect injection or carburation the situation is much, much better. Let us follow a droplet of fuel on its journey through the engine. The journey begins when the droplet exits the carburettor or injector. Four or five crankshaft revolutions later that same droplet appears in the combustion chamber (this is a mean value; some droplets travel quicker, some are slower).
And before you ask: "How do you trace a single droplet?" You label it - with radioactivity. Do not try this at home... 8)

Tha main interest of direct injection is stratified charge, but I don't know if it works well or not on two stroke.

You can find wall guided strategy or spray guided ones.
Wall guided can be used to cool down piston crown as well...

If you want power, stratified charge is not the best option. The potential power delivery of an engine is limited by the amount of air it can breathe, and every molecule of oxygen in the cylinder should be used. That means: a rather rich mixture and absolutely no weak spots anywhere in the combustion chamber.

OK Frits, it seems to be logical!
But does that mean amking power with direct injection is almost impossible?
It should have some advantages?

If I right remember i heard that Ktm use injection system on their 125cc, but it was indirect and both with carburator, but i never seen it.

Camus14 a écrit:

OK Frits, it seems to be logical! But does that mean amking power with direct injection is almost impossible? It should have some advantages?

Direct injection certainly has big advantages. If we want our beloved two-strokes to survive, direct injection may be the only way. But the combination with high rpms still presents a lot of problems that need to be solved.

pierre a écrit:

If I right remember i heard that Ktm use injection system on their 125cc, but it was indirect and both with carburator, but i never seen it.

That was a solution that Harald Bartol used on his GP-racers instead of a power jet. There was a normal carburettor and an auxiliary jet that fed into the crankcase to make the mixture richer at certain circumstances.
Rotax is also deeply involved in injection systems for two-strokes, as are Evinrude and a well-known Yapanese company. No, not Honda ....

Good news!

But about Rotax, they will not have a chance to use this technology on the famous aprilia rs 125, as I have heard that they are going to use four stroke on it.
Maybe on the new HM125 bike (that use Rotax engines now)...

The Japaneese is Yamaha, isn't it? Anyway, Yamaha is the only one in the big four that still sell two stroke dirtbikes and it works well!! So I bet it's them

Camus14 a écrit:

The Japaneese is Yamaha, isn't it? Anyway, Yamaha is the only one in the big four that still sell two stroke dirtbikes and it works well!! So I bet it's them

You win

There are some new DI snowmobiles that have grease filled bearings combined with oil injection, that operate up to quite high rpm(about 10,000) for the bore size. The packaging would never be compatable with a racing motocycle (the cylinder head is very tall). They have been on the market for at least 2 yrs the botom ends have proved very reliable, despite a lot of initial skepticism in the market.
OSSA have released a trials bike that has replaced the carburetor with an injector with great success.
Frits anymore info you wish to share on you flow bench as I wish to build one. As I'm unsure if 'blowing' from under the transfers or 'sucking' out the exhaust duct is the correct method. Also I note on the Aprilia cylinder the unusual shape to the exhaust duct outlet(pipe step) , is that to minimise exhaust gas and fresh gas mixing or to keep flow velocity up (or both).


Il y a quelques nouveaux DI motoneiges qui ont rempli graisse roulements combinée à l'huile l'injection, qui opèrent jusqu'à très haute rpm(environ 10,000) pour l'alésage taille. L'emballage n'aurait jamais être compatible avec une racing moto (la tête de cylindre est très tall). Ils ont été sur le marché pour au moins 2 ans la classe extrémités se sont avérées très fiables, malgré une beaucoup de scepticisme initial sur le marché. OSSA ont publié un procès-bike qui a remplacé le carburateur avec un injecteur avec un grand succès.

Frits est qu'il y plus informations que vous souhaitez partager sur votre flux pulsé banc comme j'aimerais construire un. Comme je suis pas certain si «souffle» dans le cadre du transfert ou de «sucer» de l'échappement gaine est la bonne méthode. J'ai également note sur la Aprilia cylindre la forme inhabituelle de l'échappement gaine débouché(pipe étape) , est que de minimiser gaz d'échappement et fraîche mélange de gaz ou de les garder écoulement (ou les deux).

@ Institute of TwoStrokes
You can find a exemple of this type of flow bench in SAE paper 2006-32-0061 [Vous devez être inscrit et connecté pour voir ce lien]