[GP125] All that you wanted to know on Aprilia RSA 125, and more, by Mr Jan Thiel and Mr Frits Overmars (PART 4)

motoholic71 a écrit:

I think that any design that wont allow an effective squish head with a central spark plug will prevent high power because detonation will occur.
Probably the next evolution we would see on the tracks is fuel injection. There is a DIY kit available on the market but the electric fuel pump power draw will require a powerful generator and hence a weight penalty on the flywheel. Ideally it should have a mechanical pump.

This uniflow engine does indeed have a squish band although the plugs are in the side of the cylinder. This situation does not need to be so as there are ways of igniting the charge from the centre of the chamber. Out of my development budget though.
This engine may not be the ideal solution, but it does run, it is efficent and it can be done with carburetors, no need for EFI. Fill the cylinder at one end exhaust the other. The next engine has a fresh air buffer between exhaust and fuel air, cleaner fill still.
Cleaner cylinder fill, higher BMEP.

Frits Overmars a écrit:

motoholic71 a écrit:

Isn't there a government fun fund to apply to exploit this stuff?

The other day I visited an army museum and an airforce museum.
Man, the things we could have done with the money that got spent by the military....

while I'm not in favour of killing people the competition world owes A LOT to aviation developments and early aviation to the military. It's sad but war boosts technology.

uniflow a écrit:

motoholic71 a écrit:

I think that any design that wont allow an effective squish head with a central spark plug will prevent high power because detonation will occur.
Probably the next evolution we would see on the tracks is fuel injection. There is a DIY kit available on the market but the electric fuel pump power draw will require a powerful generator and hence a weight penalty on the flywheel. Ideally it should have a mechanical pump.

This uniflow engine does indeed have a squish band although the plugs are in the side of the cylinder. This situation does not need to be so as there are ways of igniting the charge from the centre of the chamber. Out of my development budget though.
This engine may not be the ideal solution, but it does run, it is efficent and it can be done with carburetors, no need for EFI. Fill the cylinder at one end exhaust the other. The next engine has a fresh air buffer between exhaust and fuel air, cleaner fill still.
Cleaner cylinder fill, higher BMEP.

Is there any publication of the details?

motoholic71 a écrit:

uniflow a écrit:

motoholic71 a écrit:

I think that any design that wont allow an effective squish head with a central spark plug will prevent high power because detonation will occur.
Probably the next evolution we would see on the tracks is fuel injection. There is a DIY kit available on the market but the electric fuel pump power draw will require a powerful generator and hence a weight penalty on the flywheel. Ideally it should have a mechanical pump.

This uniflow engine does indeed have a squish band although the plugs are in the side of the cylinder. This situation does not need to be so as there are ways of igniting the charge from the centre of the chamber. Out of my development budget though.
This engine may not be the ideal solution, but it does run, it is efficent and it can be done with carburetors, no need for EFI. Fill the cylinder at one end exhaust the other. The next engine has a fresh air buffer between exhaust and fuel air, cleaner fill still.
Cleaner cylinder fill, higher BMEP.

Is there any publication of the details?

No, although I do still have the engine and have pictures of it inside and out. This was developed on a shoe string budget, I even had to make my own CDI ignition. This is the middle one of three such engines built.
It is more complicated in that there is a drive needed to join the crankshafts but any application needs to have the speed reduced anyway, so you are already half way there.
I wanted a uniflow scavenged engine but no blower, just simple crank case pumping.
What I'm trying to say is it's all well and good speculating on what might or might not work, ideas need to be put into practice and investigated, have a go.

motoholic71 a écrit:

Frits Overmars a écrit:

motoholic71 a écrit:

Isn't there a government fun fund to apply to exploit this stuff?

The other day I visited an army museum and an airforce museum.
Man, the things we could have done with the money that got spent by the military....

while I'm not in favour of killing people the competition world owes A LOT to aviation developments and early aviation to the military. It's sad but war boosts technology.

I dont think so!
The militairy and governments killed more tech than they gave.
At least its out of sight
No pun intended

Watch the tube and look for tom bearden

I agree with Howard, the aim of all such system being to provide increased power, which in a 2 stroke engine, it is necessary to increase the mass of air or air/fuel in the cylinder at the time of exhaust port closure, this being referred to the trapped mass. Obviously this is not a simple as that, as it is also a function of the gas purity, meaning minimizing residue exhaust ga from the previous cycle. For a given cylinder gas pressure at the time of exhaust port closure, it can be seen that the lower the exhaust ports are, the greater the trapping volume, giving more opportunity for an increased trapped mass. However, this must be compromised by the blowdown and timing requirements of the exhaust system as Jan and Frits have pointed out on numerous occasions. They have indicated that the exhaust system can increase the air/fuel mass flow through the engine by 300%. So, perhaps using some form of atmospheric inlet (either via the head or side ports) system, eliminating the restrictions of drawing through the crankcase system, it might allow this figure to be increased.
Direct injection. Personally I can’t see any benefit in term of maximum power. I think that a homogeneous mixture, pre-cooled with the evaporation of the fuel, must offer the best chance for maximum trapped mass. The main benefits are minimizing unburnt charge loss out the exhaust, thereby reducing unburnt HC emissions and increased fuel economy. However, I think the mechanical system created by Marco Nuti of Piaggio is a neat system with unlimited rpm potential. It also offers a small supercharge effect. This was called the FAST system.
Back to Frits. When you tried the side entry, did you also have conventional crankcase scavenged ports in operation at the same time? I would think the if the C port as of a boost type, feeding through a hole in the piston, that there would be sufficient lubrication for the bearing.
A reed valve head. Consider a flat topped piston and a flat squish band in the head. Then consider an annular thin metal ring of the same size as the squish band. Cut this into 3 or 4 equal sectors. One end of each of these is secured to the head. Under each of these curved reed valves is a series of small holes. These holes connect to a separate inlet duct to atmosphere, giving 3 to 4 ducts. Many years ago I milled a flat face into a cyl head and fitted a small reed valve, sealing over a 5 mm hole. No leakage, no burning out. Lots of mechanical questions re reed stoppage, clearances and available area. Thinking more about this, I think this could work well in conjunction with the Overmars/Seeber/Foekema et al types of contra flow axial systems. This combination would provide starting, crankcase and piston lubrication and cooling.

uniflow a écrit:

No, although I do still have the engine and have pictures of it inside and out. This was developed on a shoe string budget, I even had to make my own CDI ignition. This is the middle one of three such engines built.
It is more complicated in that there is a drive needed to join the crankshafts but any application needs to have the speed reduced anyway, so you are already half way there.
I wanted a uniflow scavenged engine but no blower, just simple crank case pumping.
What I'm trying to say is it's all well and good speculating on what might or might not work, ideas need to be put into practice and investigated, have a go.

Off course. What I'm curious about is if you're using the exhaust side crankcase for scavenging aswell,if the cases communicate and if the cranks counter rotate

motoholic71 a écrit:

uniflow a écrit:

No, although I do still have the engine and have pictures of it inside and out. This was developed on a shoe string budget, I even had to make my own CDI ignition. This is the middle one of three such engines built.
It is more complicated in that there is a drive needed to join the crankshafts but any application needs to have the speed reduced anyway, so you are already half way there.
I wanted a uniflow scavenged engine but no blower, just simple crank case pumping.
What I'm trying to say is it's all well and good speculating on what might or might not work, ideas need to be put into practice and investigated, have a go.

Off course. What I'm curious about is if you're using the exhaust side crankcase for scavenging aswell,if the cases communicate and if the cranks counter rotate

First engine the cranks are turning the same way, three gears. This engine has four gears so the cranks counter rotate.
Engine layout is, two split singles head to head but rotated 180 degrees. Gas flow is straight down the cylinder ( not like the split single ). All four crank cases pump ( four carbs and four reed blocks ), exhaust crank case runs double the fuel ( very rich to help piston cooling) and the inlet crank case runs just air, gases all mix in the cylinders at transfer. Two crank cases fill one cylinder while the other two crank cases fill the opposite cylinder from the opposite end, all at the same time. Pistons are different sizes, this creates a squish. Clear as mud?
No, crank cases do not comunicate directly, there is a labyrinth seal between them. Exhaust cranks leed the inlet by six degrees.

Ken I understand. Looking at the RSA transfers I think that it's necessary to have a huge amount of volume readily available to draw from to get an effective blowdown.
Correct me if I'm wrong:
On reed engines the exhaust neg. pulse only cracks open the reeds after the transfers are closed, what enters the cylinder is mainly the content of the transfer tunnels.
How about poppet valves instead of reeds, intake on top and exhaust on bottom.I know,we discussed it but know the flow here is inverted.

What about poppet valves on the head of a standard 2stk. engine, but no valve train.Simple light springs, valves open like reeds, like the very early 1 cam 4 strokes that had no control over the intake;
The neg pulse opens this valves.
They would be more adequate for a comb.chamer while safely providing more flow than a reed as withstands the high Bmep without losing section. Connect this directly to carbs, they would come on automatically when neg pulse is high enough? Sure you loose some ideal squish surface,but... I like it.

I would try a secondary explosion on the header, I think more energy should be created on the pipe without being at the expense of exhaust timing and spark delay.
Since if we move the energy to the pipe we're diverting it away from the comb.chamber/piston;less work done,still is worth it as we look at an ignition curve the spark is retarded towards the end to increase temp/resonance freq. but also to put more kinetic energy on the pipe, increasing supercharge?
If we look at catalyst 2 stroke pipes where a reeds chamber is connected to the headers to draw fresh air to convert monox into CO2.At the expense of power since less mix is pulled from the case.

If we could grant that there is enough mix left on the header to ignite it at the moment of highest blowback pressure after piston closed, the neg pulse would be higher upon exhaust blowdown.
Maybe connecting a comb.chamber or just a bulge to the header that would hold this extra charge and a plug, ignite it and dump it downstream.
A higher neg. pulse would be present on the header at blowdown time, more charge sucked out.
Then the blowback return pulse is also higher, more supercharge, this is the objective; put in more energy in the pipe; sort of a pulse jet in which the thrust/energy is used to supercharge the engine instead of moving a vehicle.

This is simple enough to build, but huge transfers should exist or a direct connection from case to chamber or from carb through reeds(danger) to assure an undiluted mix able to burn at blowback pressure inside this chamber.

Of course economy is completely ignored, as is pollution on this approach. And extra cooling would be needed,since were burning extra fuel. If this worked even on a temporary "power boost" switch.

I'm terribly sorry, I'm obnoxiously hogging the forum and ignoring the original subject feel free to let me know if I'm being rude, besides very boring

Ken Seeber a écrit:

Direct injection. Personally I can’t see any benefit in term of maximum power. I think that a homogeneous mixture, pre-cooled with the evaporation of the fuel, must offer the best chance for maximum trapped mass.  The main benefits are minimizing unburnt charge loss out the exhaust, thereby reducing unburnt HC emissions and increased fuel economy.

+1

Citation :

Back to Frits. When you tried the side entry, did you also have conventional crankcase scavenged ports in operation at the same time? I would think the if the C port as of a boost type, feeding through a hole in the piston, that there would be sufficient lubrication for the bearing.

The A-and B-ducts were blocked at the cylinder base. The side-mounted carbs were directly connected to the A- and B-ports. These ports also doubled as piston-controlled inlet ports. Only the C-duct & C-port remained original, connecting crankcase and cylinder for starting purposes. There was no hole in the piston to feed the C-duct but even if there had been, the big end bearing would not have taken advantage of it.

uniflow a écrit:

motoholic71 a écrit:

uniflow a écrit:

No, although I do still have the engine and have pictures of it inside and out. This was developed on a shoe string budget, I even had to make my own CDI ignition. This is the middle one of three such engines built.
It is more complicated in that there is a drive needed to join the crankshafts but any application needs to have the speed reduced anyway, so you are already half way there.
I wanted a uniflow scavenged engine but no blower, just simple crank case pumping.
What I'm trying to say is it's all well and good speculating on what might or might not work, ideas need to be put into practice and investigated, have a go.

Off course. What I'm curious about is if you're using the exhaust side crankcase for scavenging aswell,if the cases communicate and if the cranks counter rotate

First engine the cranks are turning the same way, three gears. This engine has four gears so the cranks counter rotate.
Engine layout is, two split singles head to head but rotated 180 degrees. Gas flow is straight down the cylinder ( not like the split single ). All four crank cases pump ( four carbs and four reed blocks ), exhaust crank case runs double the fuel ( very rich to help piston cooling) and the inlet crank case runs just air, gases all mix in the cylinders at transfer. Two crank cases fill one cylinder while the other two crank cases fill the opposite cylinder from the opposite end, all at the same time. Pistons are different sizes, this creates a squish. Clear as mud?
No, crank cases do not comunicate directly, there is a labyrinth seal between them. Exhaust cranks leed the inlet by six degrees.

Thank you, I understand.it is a clever design, a bit complex but not too much. You should pursuit it.

motoholic71 a écrit:

Looking at the RSA transfers I think that it's necessary to have a huge amount of volume readily available to draw from to get an effective blowdown.

Blowdown is the phase where burnt gases exit the cylinder. The blowdown angle.area (= the sum of each step of open exhaust area times the number of crankshaft degrees that it is open) should be such that the cylinder pressure has dropped below the transfer duct pressure when the transfer ports open. So blowdown and transfer are (or should be) completely separate phenomenae.

Citation :

On reed engines the exhaust neg. pulse only cracks open the reeds after the transfers are closed

On any engine the exhaust suction can only reach the reeds while the transfers are open.

Citation :

How about poppet valves instead of reeds.

Poppet valves have a lot more inertia than reed valves so they are way more reluctant to open.

Citation :

I would try a secondary explosion on the header, I think more energy should be created on the pipe without being at the expense of exhaust timing and spark delay.

Spent gas in the header hardly contains any oxygen. If, like it was proposed yesterday, you use a blower to enter additional air into the header, then why not do the logical thing and blow into the crankcase instead?
Besides, in many forms of motorsport a blower is not allowed, or only at the expense of half the original cylinder capacity.

Citation :

If we could grant that there is enough mix left on the header to ignite it...

How would you prevent that part of the mixture from taking part in the in-cylinder combustion?

Citation :

I'm terribly sorry, I'm obnoxiously hogging the forum and ignoring the original subject  feel free to let me know if I'm being rude, besides very boring

I'm glad to see new ideas here, even if it may seem that I'm only knocking them down. I hope you'll come up with answers to all my objections .

Come to think of it, there have been patents where in a twin-cylinder engine the positive exhaust pulse from one cylinder is directly used to charge the other cylinder.
You may want to look for Curtil-patents, and the pictures below may already give a clue.

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

Frits Overmars a écrit:

motoholic71 a écrit:

Looking at the RSA transfers I think that it's necessary to have a huge amount of volume readily available to draw from to get an effective blowdown.

Blowdown is the phase where burnt gases exit the cylinder. The blowdown angle.area (= the sum of each step of open exhaust area times the number of crankshaft degrees that it is open) should be such that the cylinder pressure has dropped below the transfer duct pressure when the transfer ports open. So blowdown and transfer are (or should be) completely separate phenomenae.

Citation :

On reed engines the exhaust neg. pulse only cracks open the reeds after the transfers are closed

On any engine the exhaust suction can only reach the reeds while the transfers are open.

Citation :

How about poppet valves instead of reeds.

Poppet valves have a lot more inertia than reed valves so they are way more reluctant to open.

Citation :

I would try a secondary explosion on the header, I think more energy should be created on the pipe without being at the expense of exhaust timing and spark delay.

Spent gas in the header hardly contains any oxygen. If, like it was proposed yesterday, you use a blower to enter additional air into the header, then why not do the logical thing and blow into the crankcase instead?
Besides, in many forms of motorsport a blower is not allowed, or only at the expense of half the original cylinder capacity.

Citation :

If we could grant that there is enough mix left on the header to ignite it...

How would you prevent that part of the mixture from taking part in the in-cylinder combustion?

Citation :

I'm terribly sorry, I'm obnoxiously hogging the forum and ignoring the original subject  feel free to let me know if I'm being rude, besides very boring

I'm glad to see new ideas here, even if it may seem that I'm only knocking them down. I hope you'll come up with answers to all my objections .

Blowdown was a"typo",a mistake. I also lac plot in my writing, please excuse me, I need more practice.

I was under the impression that at high rpm the scavenge negative pulse wasn't fast enough to open the valves during actual transfer, I thought it would reach them later on the piston way up.

Poppet valves heavy, I know, even worse without command, it would work for slow speed/big bore engines? but if the alternative is reeds in a combustion chamber pressures, possible knock events. would it hold?

A blower to counter lack of O2 in the header. But I'd like a system without mechanical external chargers, for several reasons including compliance to rules. I don't believe injecting fuel in a hot exhaust was prohibited in motorsports (because it's crazy?)

How to keep combustion separated. There is a phenomenae on 2 stroke that was hard for me to understand, how the hot exhaust gases don't ignite transfer charge on engines with bad TA blowdown, safe the od backfire.

If the pipe is designed to, on normal run, develop a low back pressure, or overall pressure in order to leave fresh mix "wasted " in the header (which 2 strokes are famous for) but without changing resonance, an ex. chamber with a "too wide" bleeder tip, so the volume of scavenged mix exceeds that of the trapped.
then the ignition must be timed to match blowdown, the bulk of the 2nd combustion must reinforce and combine with the blowdown and by the time transfer opens there's only vaccum there. The extra volume would now be "right" for the pipe and achieve a greater back pressure, supercharging the cylinder.

I don't think backfire will happen often if a stable run is achieved.

The challenge would be to ignite the mix in the header, if the back pressure is high enough at the correct time to be in phase with blowdown, and if we timely inject a little more fuel and lots of Volts then maybe a significant combustion will happen.
If it runs it will run hot and header colling will be necessary or autoignition is possible.Maybe wrap the injector feed around it to facilitate ign...If I had like a 4 trace scope and pressure transducers, I could see what's happening in an eventual prototype,but...I don't.

Is a pleasure to counter your objections as it helps me to "keep it real", I'm no expert, if anything I learn more and avoid mistakes, with luck and common sense.

Thank you for those links, I'll check them out as soon as I can, sounds interesting.

frits back on page 9 you posted a link for the rsa parts diagram. is the rsa cylinder available for purchase ? if so how much is it ? there is no prices given on the aprillia site

I don't know Brokedown. I never occupied myself with commercial matters.
And you don't need to ask Jan either; he is allergic to this type of questions.

Frits as a curiosity that patent reminded me of this, I don't quite understand the idea behind it but it was the cute little KR250
[Vous devez être inscrit et connecté pour voir ce lien]

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

@ motoholic:
I wonder how you are going to calculate/make the expansion pipe work, when all the time you are going to (or at least try to ...) create new waves by igniting injected fuel in the pipe. How are you going to make those waves work together with the original wave from the cylinder ? how are you going to tell your "created wave" it can only go down the pipe and can not "wave" into the direction of the cylinder ?

The quote-buttons have somehow disappeared which makes it rather troublesome to answer the various question.
Marc, Help!

@motoholic71: I was not familiar with your Kawa-publication and my japanese is a bit rusty.
I suppose the above scheme works on the lines of Kawasakis KIPS-system where an additional volume is coupled to the exhaust system at certain revs, lowering its resonance frequency, kind of like the Honda TRAC-system. And it looks like Kawa uses one pipe as an additional volume for the other one, and vice versa. But I am just guessing here.

@JanBros: there is an exhaust pulse that begins at exhaust opening and returns just before exhaust closure, but after that there is a residual pulse that departs from the cylinder at exhaust closure, runs up and down the pipe, and returns to the cylinder just when the exhaust starts to open for the next blowdown event.
As long as the blowdown pressure is double the residual pulse pressure or higher, the residual pressure does not hinder the blowdown outflow and it enhances the amplitude of the new primary pulse.
The same might happen with the pulse that motoholic wants to create by igniting mixture in the pipe.

Frits : you can copy the text before, and use the "quote" on the responding page, like this :

Frits a écrit:

The same might happen with the pulse that motoholic wants to create by igniting mixture in the pipe.

I agree it might happen, but how big are the chances ?

as the wave from the "afterburner" will always start later (and probably won't be as fast), won't there always be a clash between the returning positive wave from the blowdown pulse with the negative wave from the "afterburner" , resulting in a weakened presure wave ?

I "re-red"(?) my previous posts and found them all but very confusing, Frits got it completely.
I expect that the 2nd explosion has much smaller pressure than the blow down, not enough to hinder the flow and also that this pressure differential ratio  remains constant since this energy will boost the next cylinder charge, resulting in a stronger blow down.

The ignition should ideally happen at a moment of the highest pressure possible and amongst still gasses, right in front of the yet closed piston, perhaps inside a water jacketed exhaust port cylinder, with the plug and injector threaded bungs welded to the header as close as possible to the cylinder.
I don't like having to use a DI hardware to atomize the fuel into a fine mist.
But without a richer mix chances of ignition are probably none. With the feed line coiled around the header fuel gets very hot, more prone to ign. Who knows? talk is cheap and everytime soothing is different.

This was not the original idea of the 1st post, I veered away from it, 1st was a combustion chamber with reeds and injector, connected to the header by a small length of tube in a "y" config. facing the cylinder to push back the charge into the cylinder, but then realised that probably further down the pipe, the pressure may never fall below ATM, so scavenging would not happen.

Wow. All very interesting, but I just uprated my old Bultaco with three Ti bolts.

Was hoping the expensive weight savings would remain !

I'm prepping a Honda Sky for moped racing, but after only +/-6 runs down the street, the (original 0.5mm aluminium) head gasket is already blown/burned (guess I did my job well).

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

head and cylinder are bolted to the cases with only 4 long M6 bolts. so I'm wodering if they aren't strong enough for the increased power, and that the bolts stretch too much ?
If so, there is not enough meat in the cases to put in M8's so I was thinking of turning down the threads of 10.8 alen M8's to M6, drill out the holes in the head and cylinder to 8mm and bolt them like that. those M8's shouldn't stretch so much ? Or is it not a good idea to use plain 10.8 steel bolts ?

and if it's not the bolts stretching, copper gaskets are stronger than aluminium ? does the thickness matter ? I've got 0.8mm and 1mm thick copper.
I'd prefer to ust O-ring the head, but I don't want to reshape the combustion chamber (because of losing the thickness of the head gasket, the sparkplug isn't straight in the center so it's difficult to get the head correct on my lath). Or should I use a copper gasket AND O-ring the head and cylinder ? using a thicker gasket, I only need to mill the head to the correct height.

Can't you use M7-bolts? They are much stronger than M6-bolts. And use more bolts at the head. Even M5-bolts in-between the original cylinder studs will make a lot of difference. And try to fit an O-ring instead of a flat head gasket. You can get very thin O-rings and make the groove in the cylinder if the head has not enough meat.

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

Frits Overmars a écrit:

The quote-buttons have somehow disappeared which makes it rather troublesome to answer the various question.
Marc, Help!