Pressure pulses and turbo engines

[Pisobiker]

As sayd in another thread, seen there is some interest over this topic, lets start a specific thread:

Here my contribution.
Would point out this topic is actually very complex, but main general functioning could be explained even without getting too theorethical. Will try my best. Comments and add-ons of course very welcome.



FIRST PART: GENERAL INFOS


Think of the inlet pipe as a straight, even diameter tubing.
Lets say starting conditions are inside charge still => velocity=0, pressure=external pressure.
Intake valve is closed but piston is starting to travel to BDC decreasing cylinder presure.
When valve opens difference between internal (of cylinder) and duct pressure start moving charge inside the cylinder and this phenomena generates a pressure (actually a "depression") wave that travels the intake duct all the way back till the exit.
Now this pressure pulse travels at about speed of sound (actual value depends on pressure and temp of charge).
Which doesn't mean actual inlet flow is mooving at the speed of sound, but only the "signal" of motion. This signal of course carryes with him energy and substance: its "specific" energy is much greater than the "specific" energy of intake flow, but of course much smaller than total energy of intake flow.
This pressure pulse will be stronger the smallest (diameter) the duct because "suction" from cylinder will have a bigger influence thus inducing a much greater kinetic energy.

What happens to pressure (depression) waves while they find obstacles like open or closed ends?

A) OPEN END: wave will keep it's energy and reverse it's direction and sign. This means the outcoming "depression" pulse will travel back as a incoming "pressure" pulse through the duct towards the intake valve;

B) CLOSED END: wave will "bounce back" keeping both its energy and its sign (pressure will remain pressure; depression will remain depression).

What does this mean in practical terms?

1) compared to constant undisturbe flow, pressure wave incoming towards the valve carries more specific energy (thus charge).

2) If engine's speed at that moment, and valve timing would be right, valve will open EXACTLY when pressure pulse is reaching the valve itself thus allowing a higher mass of charge to flow inside the cylinder compared to the constant undisturbed flow;

BUT

3) If engine's speed and valve timing will have opened the valve lets say immediatly AFTER pressure wave reaching the valve, then it would have bounced back as a pressure wave thus actually REDUCING intake charge entering the cylinder compared to undisturbed flow.


WHAT ABOUT DUCT DIAMETER?

Well as sayd the smallest the diameter the strongest the pulse, so that from a theoretical standpoint you want the smallest ducts possible.
But unfortunatly this goes against flow so that section is always the best compromise with strong wave pulse and good overall flow (expecially on higher rpm).



This already explains beauty and misfortune of pressure waves:

since theyr speed is about FIXED total time for them to travel back and forth depends only on duct lenght.
To use them on the widest possible rpm ranges then, you'll need variable lenght intake ducts.
Optimum diameter increses with rpm, so you would also need a "VARIABLE SECTION" duct (almost impossible to do mechanically.... but might be done something aerodynamically ).

The two latter statements explain why on normal production engines those phenomenon are kept well away: not the good effects nor the bad ones, just a good compromise.



END OF FIRST PART. (time to go to sleep)


Marco

[madmk4]

My small contriution i copied and paste frpm the other thread that i'd wrote before this one.

Might be a bit less confusing, and a hell of alot more condensed
Long runners = torque is a very sweeping statement indeed.


For anyone who knows, in very, very veeeery short simple terms the idea of pulse tuning Is, when the air pulses backwards out of the inlet system, it'll refract back, if you can time this with the opening of the inlet valve you're onto a winner.

All engines do this, its just a trade off, and about harnessing the power where you want it, u8sing difference lengths and size trumpets. Sadly, you rarely get something for nothing. Might gain 10bhp beetween 5000-6000rpm but loose it somewhere else.

Ever see guys runnign silly long trumpets on their throttle bodies/carbs..Now you know why (probally the reason anyway)

People who say they have 'favourite' lengths for any perticular engine are generally talkin BS.

Cam timing, port diameter, valve sieze, port shape etc etc etc, anything that alters the TIME it will take the air to get from the inlet valve and back out, will alter what length will be to achive what you want.

This also works in the exhaust, but it pulls the gasses out, not pushes em back in, that'd be plain silly

[Pisobiker]

SECOND PART: EXHAUST SIDE

Similar considerations can be made on the exhaust, while taking into account pulse speed is in this case much greater due to the fact actual pressur pulse speed is the sum of "it's speed" plus speed of the fluid in which it travels (that for the exhaust is a lot higher than the intake).

there is a very important difference in what exhaust waves are supposed to do compared to intake ones:

Intake ones are generated when valve opens and are supposed to come back when valve opens again;

Exhaust ones are of course still generated when exhaust valve opens but are supposed to come back before it closes.
More into detail, what pressure pulse in the exhaust is actually supposed to do is to "come back" on the exhaust valve during the overlap period to generate a strong "suction" effect that will estract exhaust gas eventually entered the intake , help intake charge to start its motion in the cylinder (adding it's energy to the one coming from the intake pulse), while providing best possible scavenging.


END OF SECOND PART (sometimes during the day I need to "simulate" I'm working...

)


P.S. I'm trying to be as practical and concise as I can. but if you think I'm just wasting my and your time just tell, please.

[Pisobiker]

THIRD PART: HOW DO A NA ENGINE USE PRESSURE PULSES.

Pressure pulses that have accomplished theyr job do not die. Dissipating forces of course lower progressively theyr energy, but they keep on travelling back and forth in the duct therefore adding (or reducing) energy of newer generated pulses. This could build up very strong pulses.
A NA engine is very sensible of pressure pulses because of it's realatively small intake absolute pressures which make even small changes turn into quite big inprovements (Formula 1 engines can go as high as 1.3 of volumetric efficiency or even higher on resonant engine rpm). They also don't have any restriction in ducting dimensioning and layout so that they can just be made specifically for maximizing pressure pulses phenomenon.

As sayd, since phisical dimensions of ducts actually fix rpm at which it gets max efficiency, while, as a drawback, lowering efficiency in any other rpm, sometimes very complex exhaust (but expecially intake) layouts are built. On the intake, variable lenght bellmouths are used for racing purposes while on production cars variable lenght is achived while opening or closing (throug throttle bodyes) different sections of ducts thus reducing or increasing theyr overall lenght.
On the exhaust, 4 to 1, 4 to 2 to 1 layouts, compensating bypasses and variable section devices (Yamaha's ex-up is an example), help generating puses at different duct lenghts to multiply rpm ranges where pressure pulses give theyr contribution.

(Just as a tech tip, but found it very tricky when first read many years ago, Honda, on it's GP 500 two stroke engine, used CPU controlled exhaust water injection. Amount of water determined cooling off of exhaust gasses thus reduction of theyr velocity and thus velocity of pressure pulse, therefore having a much wider rpm range of resonance !!! :eek2: )

END OF THIRD PART

[Pisobiker]

FOURTH PART: WHAT ABOUT FORCED INDUCTION ENGINES?

Unfortunatly forced induction engines have devices that on the intake side or on both intake and exhaust sides disrupt flow or anyway greatly reduce theyr effect.
Superchargers only affect intake so that any contribution of the exhaust side is still available, while turbocharged engines have turbine stopping almost any incoming pressure pulse. Actually a small part of pulsating is still available from outlet of turbine, but energy is greatly reduced both because of converting kinetic energy of gas into pressure thus reducing pulse effect (constant flow doesn't generate pulses) and converting part of both kinetic and pressure energy into mechanical energy (delivered to cold side).
From a theoretical point of view it could still be possible to use pressure pulses in the exhaust ducts before turbine, but this would require far too long runners than needed for max performances of turbine and will also greatly amplify lag issues.

On the intake side, however some more engeneering is possible.
Of course both cold side impeller and intercooler turn to be very tough obstacles for pressure pulses. Left part is only runners between intake valves and manifold. If manifold has enaugh volume to allow reflection of pulses, tuning can be made here.

The most could argue (and they might be right) why taking all this hassle for 0.1-0.2 bar pressure increase over a small rpm range, when putting in a monster turbo can blow ten times as much over a wider range of rpm.
Also, top flow available is given by turbocharger cold side. If you already run at it's max flowing capabilityes, there is no gain on having higher VE: you'll just draw in the same charge (thus same power) while having less boost delivered.

What is the deal then with forced induction engines?
Well, on most of the purposes probably not much. But what about high speed racing (for example oval track), maybe where max boost allowed on manifold is setted by competition rules? Pressure waves could increase actual VE of engine having 5-10% more intake charge @ same manifold boost pressure.
Another possibility would be using pulses to maximize actual turbocharger cold side performances, given it is at it's top boost but not at top flowing capabilityes.

Gains of course could be considered negligeble by the most. But there actually seems to me there is some space left for discussion.
I personally would probably never think of messing up with this stuff on my engine because there would be loads of other things to do before. But if you want to explore any possibility of improving performances this might give the very last few percent to the lot....


END OF FOURTH PART

[Pisobiker]

FIFTH PART: LETS GET PRACTICAL

Do you want numbers?
Well there you go:

our 2.226 L engines, 81mm bore, 86.5mm stroke will resonate on the intake this way (with good approx, then must go for tryal and error on a dyno):

@ 3000 rpm: runner lenght=98 cm; runner diameter Ø 35.1 mm
@ 4000 rpm: runner lenght=73 cm; runner diameter Ø 40.6 mm
@ 5000 rpm: runner lenght=59 cm; runner diameter Ø 45.4 mm
@ 6000 rpm: runner lenght=49 cm; runner diameter Ø 49.7 mm
@ 7000 rpm: runner lenght=42 cm; runner diameter Ø 53.7 mm
@ 8000 rpm: runner lenght=37 cm; runner diameter Ø 57.4 mm

runner lenght is meant from valve seat, including bellmouth;
runner section is meant as average of smallest cylindrical part of it, not the bellmouth increasing section. On short runner layouts the whole runner could constitute the bellmouth so that it's real diameter is the smallest section immediatly outer of the splitting going to the valves.

Of course the much shorter runners of our engine setups mean that no consideration was given to wave phenomenon, while keeping runners and manifold the shortest possible for fast build up of boost.
If there isn't any other good reason (which I actually don't happen to know) Imsa manifolds, showing long runners, demonstrate some nice effort on taking advantage of resonant waves.


END