[GP125] All that you wanted to know on Aprilia RSA 125, and more, by Mr Jan Thiel and Mr Frits Overmars (PART 3) (Locked)

Vangelis,
If no one answers your questions, it is either that they are unable, unwilling, or too busy just now. Asking again will not change this, and it is frankly rather rude.

Your question about how to get more revs is a little bit silly: it depends on everything about your motor, so how can you expect to get a simple answer from someone who doesn't know your combustion chamber shape, port timings, transfer shapes, crankcase volume, temperature distribution, pipe profile and so on and so on?

You are right... I am sorry..

GrahamB a écrit:

Vangelis,
If no one answers your questions, it is either that they are unable, unwilling, or too busy just now. Asking again will not change this, and it is frankly rather rude.

Your question about how to get more revs is a little bit silly: it depends on everything about your motor, so how can you expect to get a simple answer from someone who doesn't know your combustion chamber shape, port timings, transfer shapes, crankcase volume, temperature distribution, pipe profile and so on and so on?

Frits Overmars a écrit:

We are currently testing a geometrical ratio of 9:1. I do not want to have anything to do with 'corrected' compression ratios  .

Hi Jan and Frits,
Thanks you for the dimensions of the RSA125 cylinder head you were kind enough to post – the bowl diameter is wide and shallow compared to the RS125 and TZ125 shape for unleaded fuel.
From your previous comments on the many tests you made, I image that you also tried a smaller/deeper bowl sizes – can you tell us what the advantages were with your final shape over the smaller/deeper bowl size. Was it design for a particular fuel and did you try the oxygenate unleaded in your tests?

Frits, you recently mention a compression ratio of 9:1 and a comment on cylinder filling being improved because of the additional volume. For this to work, you must have an interesting exhaust pipe design to give the increased boost pressure – can you tell us more about this concept?

GtG001 a écrit:

Thank you for the dimensions of the RSA125 cylinder head you were kind enough to post....  can you tell us what the advantages were with your final shape over the smaller/deeper bowl size. Was it design for a particular fuel and did you try the oxygenate unleaded in your tests?

All competition combustion chambers are developed for the mandatory type of fuel. And the final shape gave the best result - it is as simple as that. It would be impossible to sum up all the differences resulting from all the small and large variations in shape.

Citation :

Frits, you recently mention a compression ratio of 9:1 and a comment on cylinder filling being improved because of the additional volume. For this to work, you must have an interesting exhaust pipe design to give the increased boost pressure – can you tell us more about this concept?

There is nothing out of the ordinary in my concept. Every exhaust pipe will give increased boost pressure if it is fed with exhaust gas containing more energy. It is just a matter of optimizing the dimensions, like it has always been in pipe development.

vangelas a écrit:

Many kart drivers they try to get the carburettor as close as possible into the reed cage..
Where this help ( more power) ?

Where can I buy good quality Reeds and what is better to use Carbon or other material and in what thickness?

Let me have a try at these. I think Frits has mentionned that it is desirable to make the carbs as short as possible, because you don't want the intake tract to become an additional resonator at a frequency close to the main one: the pipe/cylinder/transfers. What you want instead is that the column of air up-stream of the reeds should react as quickly as possible to the opening of the reeds or disc valve: it's not like a 4-stroke where you use the momentum of that column to force in a bit more gas against the rising cylinder pressure before the valve closes. So moving the carb closer is another way of making the tract shorter.

For reed valves, I would think you are best to use whatever restricts the intake least, so whatever is more flexible and light. I put some carbon reeds in my TZ because I thought they were high-tech, but I think it might have been a mistake. Otoh, it's possible that using CF allows the material to be thinner and hence more flexible than fibreglass... I don't know.

Once upon a time, fibreglass reeds were called "digestible", because unlike the stainless ones of the time, they didn't destroy your motor if they broke. Don't know if CF reeds have the potential to do more damage than glass fibre...

Frits Overmars a écrit:

Haufen a écrit:

May I ask for details of the engine, such as displacement, bmep and piston speed range?

Of course you may. Displacement is 50 cc. I cannot answer the other questions at the moment, as it is a joint project.
Whereabouts in Germany do you live, Haufen? We will test the engine next sunday, 7 July, at the Oberlandring in Bernsgrün (http://simsongp.forumprofi.de/forum-anzeigen-termine-f7.html)

Thanks, Frits, I understand. I live in southwest Germany, but I am on holiday at the moment. Will you run it in Liedolsheim, this year?

GrahamB a écrit:

I think Frits has mentioned that it is desirable to make the carbs as short as possible, because you don't want the intake tract to become an additional resonator at a frequency close to the main one: the pipe/cylinder/transfers. What you want instead is that the column of air up-stream of the reeds should react as quickly as possible to the opening of the reeds or disc valve: it's not like a 4-stroke where you use the momentum of that column to force in a bit more gas against the rising cylinder pressure before the valve closes. So moving the carb closer is another way of making the tract shorter.

I try indeed to keep the inertia of the mixture column in the inlet tract as small as possible. But I do not worry about frequencies close to one another; except for piston port induction the inlet tract/crankcase/transfers/cylinder/pipe form one compound resonator with one root frequency.

Citation :

For reed valves, I would think you are best to use whatever restricts the intake least, so whatever is more flexible and light. I put some carbon reeds in my TZ because I thought they were high-tech, but I think it might have been a mistake. Otoh, it's possible that using CF allows the material to be thinner and hence more flexible than fibreglass... I don't know.
Once upon a time, fibreglass reeds were called "digestible", because unlike the stainless ones of the time, they didn't destroy your motor if they broke. Don't know if CF reeds have the potential to do more damage than glass fibre...

Carbonfibre reeds can be thinner than fibreglass reeds; steel reeds can be thinner still. Typical values are 0,2 mm for steel (or even less if the reeds have a wafer-structure like in the early Honda Elsinore), 0,32 for carbonfibre and 0,45 for glass.
It is not just a question of more flexibility, it is the ratio between mass and resilience that counts. Or, to rephrase it, the ratio between the reed's flexibility and root frequency.

Fibreglass reeds can pass through an engine without leaving any trace; steel will usually destroy the cylinder and the piston. I have next to no experience with CF reeds breaking up but I expect their damage-potential will be somewhere between glass and steel.

hi

frits in the picture from you, there is a geometric compression of 15.5:1.
is this out of date?

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

what is the gain of 9:1 vs. 15.5:1?

thank you manuel

Haufen a écrit:

Frits Overmars a écrit:

Whereabouts in Germany do you live, Haufen? We will test the engine next sunday, 7 July, at the Oberlandring in Bernsgrün (http://simsongp.forumprofi.de/forum-anzeigen-termine-f7.html)

I live in southwest Germany, but I am on holiday at the moment. Will you run it in Liedolsheim, this year?

If the engine does what I hope it will do, it will certainly appear in Liedolsheim but I will probably not be there. It will be run by Jan Schäffer of Langtuning.

Manuel Rainer a écrit:

frits in the picture from you, there is a geometric compression of 15.5:1.
is this out of date? what is the gain of 9:1 vs. 15.5:1?

The RSA-data in the picture are state of the art; certainly not out of date. But I hope to develop a different approach to generating two-stroke power. What the gain is? Time will tell.

Frits Overmars a écrit:

Haufen a écrit:

Frits Overmars a écrit:

Whereabouts in Germany do you live, Haufen? We will test the engine next sunday, 7 July, at the Oberlandring in Bernsgrün (http://simsongp.forumprofi.de/forum-anzeigen-termine-f7.html)

I live in southwest Germany, but I am on holiday at the moment. Will you run it in Liedolsheim, this year?

If the engine does what I hope it will do, it will certainly appear in Liedolsheim but I will probably not be there. It will be run by Jan Schäffer of Langtuning.

c'mon frits thats not fair to help Jan for Liedolsheim  he is already damn fast. you should rather help a clueless person which is slow - like me  

however, you mentioned finding the right curve with the ignitech. I don't want to go more than 18° BTDC on max revs. So for testing sth like 30° BTDC is it useful to set the rev limiter somewhere and then searching for the intersection with different angles?

koenich a écrit:

....you mentioned finding the right curve with the ignitech. I don't want to go more than 18° BTDC on max revs. So for testing sth like 30° BTDC is it useful to set the rev limiter somewhere and then searching for the intersection with different angles?

Two-stroke engines do not need rev limiters. But using the rev limiter to avoid the danger of too much ignition advance in an rpm zone with good cylinder filling is an excellent idea.
I was planning to attach a file called Zündkurven.PDF that explains the why and how of programmable ignitions, but I cannot find any option to do so....

I have the Zündkurven PDF  if u watch the name you should know me from another forum

I don't want to use the rev-limiter in general, just for a dyno'ing with a fixed pre-ignition of 30° for example. This way it should also be possible to find the intersections in the low-rpm/high-pre-ignition area. Just shutting the throttle at a certain rev-number seems a bit too dangerous...

BTW - seriously, if u ever come by Karlsruhe you are very welcome for a coffee and a nice dinner.

Frits, any thoughts on ignition pick-ups? 4-strokes typically use ~15° spaced pulses from the crank to get frequent updates of crank speed. The Honda and Yamaha production racers & MX bikes made do with 1 pulse/rev until the very end, although I think the Nakano/Jacque YZR250 used a multi-tooth pick-up on the balance shaft.

There must be a bit of timing inaccuracy when accelerating in lower gears, trying to guess when 18°btdc (for eg) will next occur. What did the RSA use?

GrahamB a écrit:

Frits, any thoughts on ignition pick-ups? 4-strokes typically use ~15° spaced pulses from the crank to get frequent updates of crank speed.....There must be a bit of timing inaccuracy when accelerating in lower gears, trying to guess when 18°btdc (for eg) will next occur. What did the RSA use?

Multiple pulses per revolution should give a better resolution (pun intended) but in practice one pulse per rev works fine. The RSA125 has a sensor below the crankcase, looking at a single notch in the circumference of the left crank wheel.
[Vous devez être inscrit et connecté pour voir cette image]

[quote="GrahamB"]Frits, any thoughts on ignition pick-ups? 4-strokes typically use ~15° spaced pulses from the crank to get frequent updates of crank speed. quote]
 
Hello Graham,
 
The reason 4-strokes use frequent pulses is to synchronize injection&ignition with the engine actual cycle, in order to recognize the power stroke. An acceleration threshold is used to sort both strokes. With this strategy, ECU does not require any camshaft position sensor anymore, ignition is performed only one rev out of 2, and injection can be phased in the appropriate crank angle window.
 
Arnaud

Frits Overmars a écrit:

Haufen a écrit:

Frits Overmars a écrit:

'A high secondary compression ratio improves power in a four-stroke;
in a two-stroke with an efficient exhaust system it is the other way around'.

I am wondering where would you draw the line between a (too)high, and a suitable compression ratio in a two-stroke application? 11? 13? 15? Or in corrected values: about 6 / 7,5 / 9?

We are currently testing a geometrical ratio of 9:1. I do not want to have anything to do with 'corrected' compression ratios  .

I am still confused to this day knowing that RSA run 15:1 geometrical and Honda JHA Kit has in it's paper that it comes with two 12 and 12,4 cc heads (for flat top piston) giving really low compression ratios

aolivry a écrit:

The reason 4-strokes use frequent pulses is to synchronize injection&ignition with the engine actual cycle, in order to recognize the power stroke. An acceleration threshold is used to sort both strokes. With this strategy, ECU does not require any camshaft position sensor anymore, ignition is performed only one rev out of 2, and injection can be phased in the appropriate crank angle window.
 
Arnaud

Thanks Arnaud, but many 4-strokes also use a cam sensor (eg GSXR) or a manifold vacuum sensor (SV650) in addition to a 24-2 tooth wheel. If you disconnect the vacuum sensor on an SV it changes to injecting and sparking every revolution, it cannot recognise the power stroke from the crank data alone.
If you think about it, engine speed only needs to increase by 3% per rev and the ignition timing will be 10° retarded. Yamaha made a fuss about this by adding "acceleration compensation" to the 95 model TZ...

GrahamB a écrit:

aolivry a écrit:

 
Arnaud

Thanks Arnaud, but many 4-strokes also use a cam sensor (eg GSXR) or a manifold vacuum sensor (SV650) in addition to a 24-2 tooth wheel.

I was actually talking of systems I use at work on single and twin cylinders. On a 4-inline, crank speed variations are probably too small to distinguish power stroke from gas transfer stroke, and I guess that's what makes the can sensor mandatory. I do not know the SV650 system, so I cannot figure out the strategy behind the use of manifold pressure sensor to recognize power stroke.

We're off topic... but the pressure drops when the intake valve opens...

hi

Jan or Frits why whas the optimal power in the RSA set on 13000 rpm?

Would you get more power iff the engine has the peak on higher rpm?

For what reasons has not been made?

To high mechanical stress for piston and bearings?

Or is the cylinder filling over 13,000 not so perfectly?

Or are there other reasons?

Thanks Manuel

Like you suppose, the cylinder filling is deteriorating after 13000 rpm, Manuel.
In fact it is optimal at 12500 rpm where torque is at its maximum. The reason is that the time.areas (= the product of open port areas * the amount of time that these areas are open) becomes too small at high revs.
You can find a more thorough explanation as well as numerical and graphical values elsewhere in this topic.
Mechanical stress for piston and bearings has never been a problem.

Hello all

I stumbled upon this epic thread. I have read every single message and have learnt so much. Very cool that we can have this kind of experts answering to our questions. I would probably have a ton of questions, but they would probably be too simple and not worth answering but I think I could ask these:

"Mechanical stress for piston and bearings has never been a problem."

I would want to ask about the connecting rod and the big end bearing. Were they somehow specially treated or any special materials used? If compared to a lower output 125cc, say 250hp/liter, engine, did the RSA:s parts need to be designed much stronger and heavier?  

Would it be possible to achieve a high MEP at a relatively low(at least compared to the RSA) piston speed? Like 12-13 bar MEP at a speed of 11 m/s?

Thank you a lot

Hi Frits. Is there a formula you use to calculate the target maximum rpm of a 2 stroke engine where it makes its best power? This would be helpful to design the pipe to this rpm.
It sounds like a simple question but undoubtedly the answer will be long.

palezu a écrit:

I stumbled upon this epic thread. I have read every single message and have learnt so much. Very cool that we can have this kind of experts answering to our questions. I would probably have a ton of questions, but they would probably be too simple and not worth answering but I think I could ask these: "Mechanical stress for piston and bearings has never been a problem."

I would want to ask about the connecting rod and the big end bearing. Were they somehow specially treated or any special materials used? If compared to a lower output 125cc, say 250hp/liter, engine, did the RSA:s parts need to be designed much stronger and heavier?  

Would it be possible to achieve a high MEP at a relatively low(at least compared to the RSA) piston speed? Like 12-13 bar MEP at a speed of 11 m/s?

Hyvää huomenta, Palezu, thank you for your kind words.
Mechanical stresses are not so much a consequence of combustion pressure, but mainly of piston speed. One way of reducing those stresses is to keep the reciprocating masses low, so making parts heavier would be contraproductive. Nevertheless all parts should be as strong as possible, so it is important to select suitable materials. The RSA connecting rod for example is produced out of a french type of steel. Jan may be able to tell you more about it.

Achieving a high Mean Effective Pressure (and thus a high torque) at a relatively low piston speed would not only be possible; it would be even easier.
For optimal exhaust pulse utilization the exhaust timing of a two-stroke should be about 190°, but at the 13000 rpm of the RSA such a relatively low exhaust would result in insufficient blowdown time.area. Therefore the RSA has an exhaust timing of 202°.
At a lower piston speed this compromise would be unnecessary and moreover the 190° exhaust timing would result in a longer power stroke, so BMEP and torque could be even higher than it is now.
But power is torque * rpm so if you want power, you need to search for the compromise where torque * rpm is maximal.