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

laranjateam a écrit:

How do you calculate the speed at which the maximum power of a motor is achieved?

The answer to your question could easily become a complete book, Laranjateam, but I will try to keep it short and simple.
At each crankshaft revolution we need a certain amount of exhaust port area that has to be open during a certain amount of time, so all burnt gases can leave the cylinder.
After this, we need a certain amount of transfer port area that has to be open during a certain amount of time, so the fresh mixture can flow from the crankcase into the cylinder.
The port areas depend on the heights and widths of the ports.
The available time during which the ports are open, depends on their timing angles, divided by the engine rpm: when the crankshaft turns twice as fast, the available time is halved.
If the crankshaft turns too fast for the available port areas, the ports will close again before sufficient exhaust gas has been removed and substituted by fresh mixture; the engine runs out of breath.
=======================================
The exhaust gases must leave the cylinder before the transfer ports start to open; otherwise a part of the exhaust gas would flow into the transfer ducts. The exhaust port area above the transfer ports, multiplied by the port timing angle during which this area is open, is called the blowdown angle.area.
Blowdown angle.area is not simply the total blowdown area multiplied by the total blowdown angle.
It is the sum of a lot of small area steps, multiplied by the time during which each of these steps is open.
For example, let us assume that the exhaust port is a simple rectangle, 40 mm wide, that it opens 1 mm further for each degree of crankshaft rotation, and that the total blowdown angle from the point where the exhaust port is beginning to open till the point were the transfer ports are beginning to open, is 30°.
Then the first degree of exhaust opening will open an area of 40 mm x 1 mm = 40 mm², and this area will be open during the whole 30° of blowdown period. That first area thus has an angle.area of 30° x 40 mm² = 1200°mm².

When the crankshaft turns 1 degree further, an additional area of 40 mm² is opened. This second area will be open during 29°, so its angle.area is 29° x 40 mm² = 1160°mm².

Repeat this calculation for each crank degree until the end of the blowdown phase, and add all the angle.area values;
this will give you the total blowdown angle.area.

In reality this calculation is complicated by the fact that not every degree of crank angle gives the same port height difference, and even more complicated by the fact that the exhaust port is not a simple rectangle. But you will understand the principle of angle.area.
==================================================
The tranfer port area, multiplied by the port timing angle during which this area is open, is called the transfer angle.area.
As an example, let's assume a transfer port with a timing of 130°.
When you turn the crank 1º past Transfer Open, a certain area of port, say A1, will be exposed. I don't know or care what that area is, that's for you to measure.
That area of port A1 will be open for 130º. The angle.area for this area A1 is = 130 x A1.
Turn the crank another 1º and an additional area of port, A2, will be opened.
This additional area A2 will be open for 128º. The angle.area for this area is = 128 x A2.
The next degree of crank rotation will open Area A3 which will be open for 126º and its angle.area will be 126 x A3.
Keep going like this until BDC, then add up all the angle.areas you have calculated.
The total sum is the angle.area for that transfer port.
=====================================================
The exhaust port timing should be about 190°. Explaining the reason, called wave superposition, would make this simple story too complicated, but you can find the explanation elswhere on this forum and on  http://www.kiwibiker.co.nz/forums/showthread.php/86554-ESE-s-works-engine-tuner?p=1130172752#post1130172752

The transfer port timing should be about 130°. Any higher, and the blowdown angle, that starts when the exhaust ports begin to open, and ends when the transfer ports begin to open, would become too small.
Because the transfer timing is more or less fixed, we try to make the total transfer port width as big as possible:
almost the total cylinder circumference of a modern competition engine is occupied by transfer ports.
======================================================
When we have established the values for blowdown angle.area and transfer angle.area, we can divide these values by the engine rpm. That gives us the blowdown time.area and the transfer time.area.
And when we divide those time.areas by the volume of cylinder plus combustion chamber, we find the Specific Time.Areas, usually abbreviated to STA.
A modern competition two-stroke engine should have a blowdown STA of 8,72°mm² per cc per 1000 rpm, and a transfer STA of 66,16°mm² per cc per 1000 rpm.
You can find some more information about the subject here:
[Vous devez être inscrit et connecté pour voir ce lien]

Thank you Very much, Mr Frits.
I will read and reread to try to understand the information left in these lines.
All the best for you and your family.
Thank you.

First let me say thank-you to all that have contributed. I have made it to the end of this incredibly informative thread and have many questions. To start could Jan or Frits expand on the sparkplug modification? Does this yield a power gain throughout the RPM range? I assume that the plug must be lowered in the head an amount equal to what has been ground off. How close is too close to the piston crown?

As far as I remember the head was not modified.

Thank-you, Jan. Can you explain what the performance gain was?

About 0,5HP

polcat88 a écrit:

First let me say thanks to all that have contributed. I have made it to the end of this incredibly informative thread and have many questions. To start could Jan or Frits expand on the sparkplug modification? Does this yield a power gain throughout the RPM range? I assume that the plug must be lowered in the head an amount equal to what has been ground off. How close is too close to the piston crown?

Hi Steve, welcome here. I just read your presentation [Vous devez être inscrit et connecté pour voir ce lien]
How about some pictures of your three-cylinder engine? (yeah, I know: posting pictures here is not the easiest thing to do).

If you compare the modified plug with the standard plug, you can see that the ground-off part was not threaded. It was the part that originally intruded into the combustion chamber. So the modified plug need not be lowered and the distance from the electrodes to the piston will not change.
How close is too close to the piston crown? I try to locate the spark about halfway between the combustion chamber ceiling and the piston crown; definitely not any lower.
The power gain with the modified sparkplug is concentrated on the higher rpm zone. It also raises the maximum revs by about 500 rpm.
[Vous devez être inscrit et connecté pour voir cette image]

Hi guys, I finally found a way of making large files available to the forum. Now I have an opportunity to pester you guys with files that I couldn't show until now. You should be able to download them from the Jumpshare website.
For starters here is a collection of tips and concepts: [Vous devez être inscrit et connecté pour voir ce lien]

Wanna see what HCCI really is? Here is a video, not an animation but the real deal, filmed inside firing engines, showing HCCI, spark ignition and diesel combustion modes. The huge difference in burn speed between the various modes should be an eye opener: [Vous devez être inscrit et connecté pour voir ce lien]

And while I'm at it, you'll find my FOS Freeware collection here: [Vous devez être inscrit et connecté pour voir ce lien]
I wrote these programs in the course of over four decades for my own personal use and I never bothered to make them user-friendly, so you may have to do some experimenting. The lot runs on DOS 6.2 or Windows 98 (I recommend Win98SE) so either dig up an old computer or use an emulator. On my Win7-laptop I use VMWare Player. I advise against DOSBox 0.74 because that is essentially a stripped-down version of DOS 5; it will run some of the programs, but not all of them.

By the way, everything that I make available on open forums such as Pit-Lane and KiwiBiker, may be freely distributed. Happy Easter .
[Vous devez être inscrit et connecté pour voir cette image]

Hi Frits and thank-you for your reply.  All snowmobile engines that I have seen have the bottom of the non-threaded part of the plug even with the combustion dome, so I guess we have been doing it wrong for a long time. I will have to machine off an amount equal to the ground off length, from the top of the plug threads on my combustion inserts. Can you explain the increased RPM and do you mean peak power or peak revs? I will attempt to get some pictures of my three cylinder engine posted tonight.

When experimenting with these modifications you must make sure that no part of the plug thread enters the combustion chamber. Otherwise carbon can form a deposit in the exposed part of the thread, which would ruin the thread in the cylinderhead when you unscrew that plug.

Some pics for Frits of my Promod 1047cc engine

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

That looks like superbike power at half the size and half the weight. Yummy!

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


Frits, can you suggest the appropriate angles for the roof of the exhaust duct? The floor of the duct? Ratio of length of the exhaust duct to bore?

polcat88 a écrit:

Frits, can you suggest the appropriate angles for the roof of the exhaust duct? The floor of the duct? Ratio of length of the exhaust duct to bore?

The roof angle of the exhaust duct should be about the same as the roof angle of the A-transfers: around 25°.
The reason: between transfer closure and exhaust closure the exhaust port itself becomes a transfer port, feeding washed-through mixture back into the cylinder. The exhaust duct floor angle is less critical.
The length of the exhaust duct should be such that the duct can hold about 75% of the cylinder volume. That way the amount of washed-through mixture that will get shoved back, does not come into contact with the hot exhaust header;
it can pause in the cooled part of the exhaust tract, provided the duct is really cooled, which is important for power and for detonation-prevention.

Thanks, Frits. Still more questions. Should the auxiliary duct roof be at a 25 degree angle as well? In most of the pictures I have seen, the auxiliary tunnels end near centreline height at the flange which leads me to believe they are at more of an angle than the main port. How are the auxiliary tunnels sized and at what inward angle? Are the tunnels basically a semicircle the height of the auxiliary port?  Angled inward to carry through to the flange face? I assume the port floor should have at least a degree or two of divergence from the roof for better flow.

Like the main exhaust roof angle, the aux.duct roof angle should also be of 20° to 25° for optimum flow. But because the aux.duct floor is so close to its roof, the aux.duct floor angle is much more important than the main exhaust duct floor angle. However, the aux.duct floor angle cannot be much more than the usual 25° of the A-transfer roof that's right below it.
I often see aux.duct floor angles close to 0°, which is definitely not optimal. If the floor angle is 25°, try a 20° roof angle.
You can find data about the aux.tunnels' outward angle in the 'FOS tips & concepts' that I offered last week.
The inward angle is not quite so critical. The tunnels basically have a rectangular cross flow area.

Frits Overmars a écrit:

If you compare the modified plug with the standard plug, you can see that the ground-off part was not threaded. It was the part that originally intruded into the combustion chamber. So the modified plug need not be lowered and the distance from the electrodes to the piston will not change.
How close is too close to the piston crown? I try to locate the spark about halfway between the combustion chamber ceiling and the piston crown; definitely not any lower.
The power gain with the modified sparkplug is concentrated on the higher rpm zone. It also raises the maximum revs by about 500 rpm.
[Vous devez être inscrit et connecté pour voir cette image]

Thanks, again Frits. I think I am almost out of questions!

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

Is this your suggestion for the auxiliary exit angle even at 100% bore?

How do we decide on the size of the auxiliary tunnel semicircle in the exhaust duct? Is the semicircle area equal to the area of the auxiliary port at the bore?

polcat88 a écrit:

Thanks, again Frits. I think I am almost out of questions!
Is this your suggestion for the auxiliary exit angle even at 100% bore?
How do we decide on the size of the auxiliary tunnel semicircle in the exhaust duct? Is the semicircle area equal to the area of the auxiliary port at the bore?

Yep, at 100% bore the position angle is 90°, so the aux. exit angle becomes 70°. But you'll definitely need to plug the gudgeon pin bores in the piston, to avoid leakage from the aux. exhaust ports to the transfer ports or vice versa.
As I wrote three days ago, the aux.tunnels basically have a rectangular cross flow area, so unless I misunderstand your question, there are no semicircles. The cross flow area of the tunnels should remain more or less equal to the cross flow area at the ports, expanding no more than 3°; more expansion would not help the flow and increase the volume of the complete exhaust duct, which we must try to keep to a minimum.

I think Google's translator is making this question seem more difficult than it is. I understand the auxiliary ports are rectangular in shape to begin with. They end up being semicircles on each side of the exhaust duct at the exhaust flange face. What is the percentage of the original auxiliary port area at the bore compared to the area of the semicircular exit at the flange?

Did the pin plugs take the place of the wristpin circlip? Or did they fit inside the the circlip?

polcat88 a écrit:

I understand the auxiliary ports are rectangular in shape to begin with. They end up being semicircles on each side of the exhaust duct at the exhaust flange face. What is the percentage of the original auxiliary port area at the bore compared to the area of the semicircular exit at the flange?

That is an impossible question to answer, Steve. The auxiliary ducts connect tangentially to the main exhaust duct and those semicircles disappear into nothing along the length of the main duct.
Try to look at it another way: there are three exhaust ducts that have to join into one duct with little or no change in collective cross section and with as little volume as possible.

Citation :

Did the pin plugs take the place of the wristpin circlip?

Mine did; some other versions did not.

Sorry Frits, I was under the impression that the auxiliary tunnels extended all the way to the flange face like the RSA cylinder. Is there no gain in carrying the tunnels all the way through to the face of the flange? Would you have the exhaust duct finish in a circular opening of the correct area, or the exhaust duct shape transiton to circular over the length of the exhaust flange?

polcat88 a écrit:

Sorry Frits, I was under the impression that the auxiliary tunnels extended all the way to the flange face like the RSA cylinder. Is there no gain in carrying the tunnels all the way through to the face of the flange? Would you have the exhaust duct finish in a circular opening of the correct area, or the exhaust duct shape transiton to circular over the length of the exhaust flange?

The aux.tunnels start joining the main duct directly after their bridges which are only about 12 mm long. Then they gradually become part of the main duct. At the cylinder/flange face they are almost done; only a small portion is still visible there.
That is probably what you called semicircles. You would see other shapes if you were to cut the exhaust duct+flange at another position. There's a definite optimal duct shape and for the flow it doesn't matter were the cylinder duct ends and the flange starts (I'm not talking about the duct cooling now; that is a different matter).
[Vous devez être inscrit et connecté pour voir cette image]

I can see your next question coming, so before you ask: when you look into the fitted exhaust flange towards the piston, like in the picture below, you can also see two semicircles, where the duct face does not match the bigger shape of the flange.
These top and bottom steps may seem bad for the flow, especially for the backflow, but grinding these steps away has proven bad for power. However, I consider them superseded. Grinding them increased the duct volume, which I think was the main reason for the power loss. Adapting the flange shape to those steps decreases the volume and improves the flow.
[Vous devez être inscrit et connecté pour voir cette image]

If polcat88 is ok to give me the microphone, I have a question to the Doctor Twostroke. :

Why two stroke have a so small engine braking compare to the four stroke ?

(I know the first answer: the goal of an engine is to accelerate, not to brake, and two stroke is the best engine. I'm looking for the mechanical reason... )

I'm only Student Twostroke, but low effective compression (without the pipe doing it's job) and low inertia.