[GP125] All that you wanted to know on Aprilia RSA 125, and more, by Mr Jan Thiel and Mr Frits Overmars (PART 5)

Retard the ignition 28degrees on 13000 is a bit to much 18 to 15 Will do beter. Or try to make the exhaust longer a lot of heat is radiated into the piston head, which could be caused by a to short pipe.

A Beings Jan Thiel and Frits Overmars: Two escape that work well in motor 2 T um, it happens that the system generates more power produces a plateau in the curve, because? and pipelining the modificacióna which try to overcome that point?

Sr. Jan Thiel y Frits Overmar: De dos escapes que funcionan bien en un motor de 2 T, ocurre que el que genera más potencia produce una meseta en la curva de potencia, Por qué? y cuál sería la modificación a probar en ese escape?

alwoodman1 a écrit:

A Beings Jan Thiel and Frits Overmars: Two escape that work well in motor 2 T um, it happens that the system generates more power produces a plateau in the curve, because? and pipelining the modificacióna which try to overcome that point?
Sr. Jan Thiel y Frits Overmar: De dos escapes que funcionan bien en un motor de 2 T, ocurre que el que genera más potencia produce una meseta en la curva de potencia, Por qué? y cuál sería la modificación a probar en ese escape?

All two-stroke exhaust systems are designed for a certain rpm. At this rpm the reflection pulse from the exhaust pipe comes at exactly the right time to push fresh mixture back into the cylinder just before the exhaust port closes.
But at lower revs it takes longer before the ports close, and the transfer ports are still open when the return pulse from the exhaust pipe arrives back at the cylinder. Then the fresh mixture in the cylinder is pushed back into the crankcase.
This situation is worst at 2/3 x the rpm of maximum torque, and it is worse when the exhaust pulse is stronger. So the pipe that is most effective at the correct rpm, is the most disturbing at low rpm.

Thanks for the attention. I understand your explanation, but I must recalculate the exhaust system to prevent this happening to me?.


Muchas gracias por la atención. Comprendo su explicación, pero en que debo recalcular el sistema de escape para que esto no me suceda?

alwoodman1 a écrit:

Thanks for the attention. I understand your explanation, but I must recalculate the exhaust system to prevent this happening to me?

That depends on your wishes. If you want a two-stroke with a four-stroke character, you can build a less-effective pipe but then you will also have four-stroke power .

Gracias Sr Frits. Mi pregunta es cual es la sugerencia del calculo: modificar un sector determinado del sistema de escape, calcular con otra velocidad sonica which was used 525 m / s, the distribution of conical sectors, etc. Thanks

Thank you Mr. Fritz. My question is what is the suggestion of calculation: modify a particular sector of the exhaust system, another sonic speed calculate Which was used 525 m / s

alwoodman1 a écrit:

Thank you Mr. Fritz. My question is what is the suggestion of calculation: modify a particular sector of the exhaust system, another sonic speed calculate Which was used 525 m / s

Suggestions for a 'tame' two-stroke: do not make the exhaust port higher than 190°, keep the transfer ports low, about 120°, keep the header rather narrow and the end cone rather long.
A programmable variable ignition will also help low-down power.

There is an alternative pipe design that tames the powercurve down low. It uses two cones, one being perforated and is welded in position before the main cone.
[Vous devez être inscrit et connecté pour voir ce lien]

I havent built one yet but it may be helpful in some circumstances.

Gracias Frit y Howard Gifford.
Frits se trata de un motor 100cc (Sachs Televel, seguro lo recuerda) de 2 velocidades, mi curiosidad es porque usar una diferencia escape-transfer de 70º, pienso que podría quedar sin fuerza para mover.
Al Sr Howard, muy interesante, he usado el sistema de conos perforado n el tradicional sistema de kart, Si Ud dispusiera de mayor información sería muy bienvenida

Thanks Howard Frits and Gifford.
Frits is a 100cc engine (Sachs Televel, sure remember) 2 speed, my curiosity is because exhaust-transfer using a 70 ° difference, I think it could be powerless to move.
Mr Howard, very interesting, I used the perforated cone system n the traditional system kart, If you, providing greater information would be very welcome

alwoodman1 a écrit:

Frits is a 100cc engine (Sachs Televel, sure remember) 2 speed, my curiosity is because exhaust-transfer using a 70 ° difference, I think it could be powerless to move.

When I wrote above that the exhaust timing should be no more than 190° , I did not know that you are dealing with a two-speed engine. For such an engine 190° may be too much. A safe starting point would be 170°.
How much higher you can go, will largely depend on the grabbing rpm of the centrifugal clutch.

Frits Overmars a écrit:

alwoodman1 a écrit:

Thank you Mr. Fritz. My question is what is the suggestion of calculation: modify a particular sector of the exhaust system, another sonic speed calculate Which was used 525 m / s

Suggestions for a 'tame' two-stroke: do not make the exhaust port higher than 190°, keep the transfer ports low, about 120°, keep the header rather narrow and the end cone rather long.

currently I am using 180/130° on my to 8.500/9000 rpm limited Honda Sky MX moped. I figured since the revs ain't that high, there is more time between exh en transfer opening . So basicly more blowdown time because of the lower revs and figured I could use that to make the transfers a bit higher to gain more area.

would it better to stick to 120 ? It's not that I want a 'tame' engine, I definitly want a very fast one but it also needs to be quick of the line and the clutch must last 1,5h races (and preferably a whole day of free practises to) with uncountable starts.

I would leave it as it is Jan.

I am very interested to know what would be the most suitable grade of aluminium to use for a cylinder head insert for two stroke race engines. I appreciate that the environment is hostile and I assume the material requires strength to resist combustion pressure; resilience to withstand the shock of detonation; thermal properties to resist melting and/by transferring heat - and not be super expensive. I understand that 7000 series is very strong but loses that strength at elevated temperatures compared to other Aluminium grades. When considering the (past ?) practice of installing (anti) detonation rings, would 2000 series which has copper added be desirable? or perhaps a grade with more silicon. Jan, Frits & all, your conclusions or opinions would appreciated on the optimum material.

JanBros a écrit:

currently I am using 180/130° on my to 8.500/9000 rpm limited Honda Sky MX moped. I figured since the revs ain't that high, there is more time between exh en transfer opening . So basicly more blowdown time because of the lower revs and figured I could use that to make the transfers a bit higher to gain more area.

would it better to stick to 120 ? It's not that I want a 'tame' engine, I definitly want a very fast one  but it also needs to be quick of the line and the clutch must last 1,5h races (and preferably a whole day of free practises to) with uncountable starts.

I always read such questions with only mentioning degrees, but to do it correctly in relation to the rpm than portsurface of transfers, exhaust and blowdown cannot be ignored!

LucF a écrit:

I always read such questions with only mentioning degrees, but to do it correctly in relation to the rpm than portsurface of transfers, exhaust and blowdown cannot be ignored!

I can totaly understand your pount if view, when tuning engines for the dutch, european or world championships.
I'm talking about dressed moped racing which is supposed to be more fun than competition. And as I'm no good at MX, I need a faster moped than the others to at least try to be competitive

in other words, it's all very basic and after crawling, I'm just trying to walk a bit faster in 2-stroke tuning, not to run very hard

[Vous devez être inscrit et connecté pour voir ce lien]

JanBros a écrit:

LucF a écrit:

I always read such questions with only mentioning degrees, but to do it correctly in relation to the rpm than portsurface of transfers, exhaust and blowdown cannot be ignored!

I can totaly understand your pount if view, when tuning engines for the dutch, european or world championships.
I'm talking about dressed moped racing which is supposed to be more fun than competition. And as I'm no good at MX, I need a faster moped than the others to at least try to be competitive

in other words, it's all very basic and after crawling, I'm just trying to walk a bit faster in 2-stroke tuning, not to run very hard

[Vous devez être inscrit et connecté pour voir ce lien]

Jan, if you ignore it, than you take the risk that the result of changing anything can go in the opposite way you wanted.
It doesn't matter at what level of tuning your engine is.

Hi Jan, Frits! my name is Jonathan Lambrecht, I’m from Argentina (the land of Vaifro Meo, not Messi nor Francisco…) I’m 26 and I’m trying to learn what you teach since the beginning of this thread, I think I’m doing it well enough lol (gave up from high school, but not from 2 strokes…).

This is my first post since I didn’t felt in the need of directly ask a question until now:
I’m currently working on a programmable ignition and my circle of "2 stroke professional tuner-misled-informed friends” know that, and the question that I got the most is something like "I know that you can set whatever angle of advance/retard at whatever rpm you want, but...what advance curve should I use for my xxxxxxxx model xxxCC bike? Does it has a Thor/Zeuz-like powerful Taser spark? ". So I shift my focus from the ignition module itself and the “glass ball” of ignition angle too, to a most almost-scientific aproach and I decide to build a "machine" (kind-of) for measure the "best" ignition curve for any given motor (motor? engine? idk, you get the idea doe...).
What I end up with is a 1 point variable ignition via potentiometer and a prony brake for measure torque and power at any given speed, since I’m no mathematician or engine specialist (the reality is that I’m a hobbyist, I work for money, and then burn my future in this kind of "power seeking" "developments") what I do is the following; turn on the engine, give it a load at full throttle till the rpm goes to the desire amount to measure, then I play around with the potentiometer varying the advance angle and watching my load on the Prony brake AND my rpm gauge till I get the most torque at the rpm I desire and save that data (sheet of paper and a pencil, top tech…), the same thing over and over again on (in?) the whole rpm range. Then I use all that data to program the advance curve that I got on the CDI module, so my question is: am I on the right path to figure advance curve? Or my method is a "better than nothing" non scientific method? (With which I would be proud of myself anyway, better than nothing is a lot for my level of “hobbyism” lol)

In other hand I would like to mention 2 things: first, sometimes, the "best" advance angle is beyond point of detonation and what I do is to use up to 12,5% methanol and that seems to solve the problem without creating a new one (remember, hobby), I think that this is reasonable since I got no money to redesign half engine... And 2nd: the thing about spark power and turbulence I figured out by myself, I was trying to get a lot of spark power (800v 3.3uf on the CDI) but I got a lot of problems and 0 gain, so I gave up and ended up with 400v 1.5uf (decent) and then I read frits confirming what I had imagine previously, so the next step is to work on the cylinder head I guess...I hope my rubbish English level doesn’t get in the way...

PD: I’m not the only Argentinean among us, I recognize some names here in this thread, AND I know that Diego Silva, from Uruguay, would read this eventually too!

PD2: you could, should, must write a book, it is fair for us mortals to have a bible, a 2014/15/16 one, not my 1973 books (which I love and I’m very grateful to Graham, and the other one that I never remember his name…tuners handbook, you know…) I would love to pay for it knowing that jan would get money for his Nissan GTR and I’m going to get to touch the heaven of two strokes with my own hands…

jona2t a écrit:

Hi Jan, Frits! my name is Jonathan Lambrecht, I’m from Argentina...
...I turn on the engine, give it a load at full throttle till the rpm goes to the desire amount to measure, then I play around with the potentiometer varying the advance angle and watching my load on the Prony brake AND my rpm gauge till I get the most torque at the rpm I desire and save that data (sheet of paper and a pencil, top tech…), the same thing over and over again on (in?) the whole rpm range. Then I use all that data to program the advance curve that I got on the CDI module, so my question is: am I on the right path to figure advance curve? Or my method is a "better than nothing" non scientific method? (With which I would be proud of myself anyway).

Welcome to this forum, Jonathan. I noticed that your name sounds Belgian rather then Argentinian;
any European ancestors in your family?
The good news is that you can be proud of yourself. The method you describe is exactly the same as the method Jan and I used when experimenting with variable ignition. (the bad news is that it didn't work).

In 1987 I was sitting next to a Garelli engine on the test bench while Jan was operating the dyno controls.
I could vary the ignition timing all I wanted, but it did not have any influence at all on the engine power!
We could not understand this at the time but a couple of days ago I discussed this strange phenomenon with Jan once more and we both came to the conclusion that we had probably encountered Homogeneous Charge Compression Ignition.
Nowadays HCCI is hot, but it was totally unheard of when we first stumbled onto it.
Full throttle HCCI will only occur if there are very high temperatures or pressures in the combustion chamber, so maybe you haven't encountered it yet.

The method I would use nowadays is to set the ignition timing to a fix number of crankshaft degrees and perform a dynamic power measurement. Then I repeat the procedure with a different fixed timing, and so on. For example, I do the first measurement with 14° advance and the second one with 10° advance. Where the two power curves intersect, you may conclude that 12° advance was probably the optimum, which gives you one rpm/degree point on the ignition graph that you are trying to establish. Etc.

jona2t a écrit:

And 2nd: the thing about spark power and turbulence I figured out by myself, I was trying to get a lot of spark power (800v 3.3uf on the CDI) but I got a lot of problems and 0 gain, so I gave up and ended up with 400v 1.5uf (decent)

Hi Jonathan, Thats what we also find out, but it is a too fast conclusion that this is always the case. There were also tests that a low spark energie gives a better result, so exactly the opposite way.

Both conclusions are related to the speed of the combution in which many more issues who are involved. To get a better view on this
subject read this.
Ignition-and-combution-in-a-2-stroke [Vous devez être inscrit et connecté pour voir ce lien]

If necessary I can copy this to this forum too, but only without pictures.

Frits Overmars a écrit:

jona2t a écrit:

Hi Jan, Frits! my name is Jonathan Lambrecht, I’m from Argentina...
...I turn on the engine, give it a load at full throttle till the rpm goes to the desire amount to measure, then I play around with the potentiometer varying the advance angle and watching my load on the Prony brake AND my rpm gauge till I get the most torque at the rpm I desire and save that data (sheet of paper and a pencil, top tech…), the same thing over and over again on (in?) the whole rpm range. Then I use all that data to program the advance curve that I got on the CDI module, so my question is: am I on the right path to figure advance curve? Or my method is a "better than nothing" non scientific method? (With which I would be proud of myself anyway).

Welcome to this forum, Jonathan. I noticed that your name sounds Belgian rather then Argentinian;
any European ancestors in your family?
The good news is that you can be proud of yourself. The method you describe is exactly the same as the method Jan and I used when experimenting with variable ignition. (the bad news is that it didn't work).

In 1987 I was sitting next to a Garelli engine on the test bench while Jan was operating the dyno controls.
I could vary the ignition timing all I wanted, but it did not have any influence at all on the engine power!
We could not understand this at the time but a couple of days ago I discussed this strange phenomenon with Jan once more and we both came to the conclusion that we had probably encountered Homogeneous Charge Compression Ignition.
Nowadays HCCI is hot, but it was totally unheard of when we first stumbled onto it.
Full throttle HCCI will only occur if there are very high temperatures or pressures in the combustion chamber, so maybe you haven't encountered it yet.

The method I would use nowadays is to set the ignition timing to a fix number of crankshaft degrees and perform a dynamic power measurement. Then I repeat the procedure with a different fixed timing, and so on. For example, I do the first measurement with 14° advance and the second one with 10° advance. Where the two power curves intersect, you may conclude that 12° advance was probably the optimum, which gives you one rpm/degree point on the ignition graph that you are trying to establish. Etc.

Nice ear Frits, it doesn’t only sound Belgian, it is Belgian :), about the European ancestors I can tell you that i know no "Argentinean" without a European last name, 0, the only real Argentinean root people lives in the north of this vast country and they are a little % of the population.

Well, good news and bad ones always come together ha-ha. About HCCI I never heard of it till now, with a quick reading i can tell that this is not going to be a problem for me anytime soon from now, because i use "Zanella" cylinders, those things are designed before the wheel was invented/discovered, so homogeneous charge sounds like sci-fi movie lol (I’m working on cylinder making, i would shoot you some questions about that eventually) and pressures are low because of the 9:1 RC used with 3rd world 70 MON gas. I hope that someday i will encounter this HCCI problem, like Grant Cardone says "new big a$$ problems are a indicator that you're going in the right direction"

I understood your method, but I don’t understand if you solve the HCCI thing or the HCCI thing is not a problem? (i need more reading on this, I’m on it right now) I would have to build a inertial dyno eventually i guess...

I would like to step down a little bit, and get some things that i assume but I don’t take for granted out of the way, like a stage burning kind of thing:
1st why did you used the ignition trigger on the crank? (It’s a space thing?)
2nd it was a VR sensor or hall effect? (Almost sure that is VR),
3rd what type of battery (size) did you used?
4th did you considered eddy currents on the primary coils at some point? (I bet not, but I’m trying to get a very thin sheet for making a less eddy currents laminated core for coils)
5th neodymium magnets on the flywheel?
6th your primary coils only charge the battery right? so the CDI is a DC one with a inverter? Did you use flyback inverter?

And finally, the class that I’m trying to get in to with my ignition setup is a 125 cc, no-gears, no-clutch kart class, engine is not mine I don’t participate in any aspect of it, only ignition, the engine is winning races already, 33hp@13000 (17,XXhp average) with a PVL fixed analog ignition, to get you in perspective on what I’m dealing with here, i truly believe that I’m not going to be able to get anywhere at this instance but I’m learning...

LucF a écrit:

jona2t a écrit:

And 2nd: the thing about spark power and turbulence I figured out by myself, I was trying to get a lot of spark power (800v 3.3uf on the CDI) but I got a lot of problems and 0 gain, so I gave up and ended up with 400v 1.5uf (decent)

Hi Jonathan, Thats what we also find out, but it is a too fast conclusion that this is always the case. There were also tests that a low spark energie gives a better result, so exactly the opposite way.

Both conclusions are related to the speed of the combution in which many more issues who are involved. To get a better view on this
subject read this.
Ignition-and-combution-in-a-2-stroke [Vous devez être inscrit et connecté pour voir ce lien]

If necessary I can copy this to this forum too, but only without pictures.

Hi Luc, how are you? 1st off, thank you very much for stepping up and throw this priceless info on the (virtual) table, in 1 day I’m jumping like 5 years of headaches LOL, what i can see on your post is that a variable power ignition is a good TOOL, to find different aspects of the engine, like exhaust length and squish area like you mentioned, please give me some days to get all this info in my head and really understand all of it, I’m not that good with engines nor ignitions, I tend to spend like a week thinking in things till I get them, and I don’t feel comfortable asking further more questions if I don’t get the base of things, soon I will get to you again Luc.

I think I getting something with Luc's wonderful FB post, past peak torque HP increase because rotating mass, then I always retard ignition, but Luc is weakening the spark, so the exhaust can still be useful AND maximum pressure remains near optimal mechanical point, when in my case maximum pressure builds later and affect hp like crazy, I’m going to read it again like 20 times, but it made a lot of sense to my already!

Jonathan, did you measure the voltage across your larger capacitor? With an oscilloscope for example?
Maybe the rest of your CDI just can't charge such a big capacitor at high revs, as 800V 3,3uF holds a lot more energy than 400V 1,5 uF...

Another question I have is why do you want to reduce eddy currents in your primary coil? You can buy normal ferrite coils that will certainly do.

Senne s a écrit:

Jonathan, did you measure the voltage across your larger capacitor? With an oscilloscope for example?
Maybe the rest of your CDI just can't charge such a big capacitor at high revs, as 800V 3,3uF holds a lot more energy than 400V 1,5 uF...

Another question I have is why do you want to reduce eddy currents in your primary coil? You can buy normal ferrite coils that will certainly do.

Hey Senne, how are you? that were the problem, not even close to charge that cap at high speeds, I didn’t try hard enough to charge it because I wasn’t getting anything out of it (like Luc article explains) 800v 3.3uf would be around 1 joule if I’m not wrong, that’s a lot.

For what I understand (which is little to nothing) ferrite cores reduces magnetic force at the coil, laminate is better if you want a high energy spark. In other hand YOU can buy whatever you want, I have to make, because communist laws here in Argentina :) (I’m not mad at this law, they force you to make things)