2 stroke thread (with occasional F1 relevance!)

[gruntguru]

I agree that the poppet valves not affecting the detonation does not matter. But this means "Without exhaust valves ... hot spots ... HUCR increases a lot. " may not be correct.

That does not disallow other potential knock benefits from your concept though, including higher rpm itself.

My thoughts exactly Brian. The compact combustion chamber with its short flame path and high squish turbulence is the key feature for detonation control in Manolis design. At high revs the rapid combustion will produce a dramatic improvement over the Panigale design. (and not just in the area of detonation)

[gruntguru]

your plot of constant torque at all rpm seems odd for an NA engine without VVT and incompatible with the effect of your late inlet valve reducing VE as rpm is reduced

I think Manolis has plotted the maximum torque possible at all rpm to illustrate what is possible with optimum breathing at a chosen rpm. Pick a rev range, optimise cam and ports for that rpm and the lines will show what torque and power to expect across that range.

[manolis]

Hello Brian Coat.

In the question
“Where the knocking problem in the poppet valve engines comes from?”
my
“Does it matter?”
has to do with the fact that while every poppet valve design is around the knocking problem and how to avoid it, the guys behind the Bishop Rotary Valve project claim they never experienced knocking problems despite the fact they did try substantially high compression ratios.

After a century of development the design of the poppet valve engine is still focused on avoiding the knocking (and necessarily is limited to lower than optimum compression ratios), while the other design (Bishop rotary valve) is, from the beginning, “unable” for knocking.

Unless I am wrong, the 13:1 and 14:1 “geometrical” compression ratios used in some car engines are combined with late intake valve closing (Atkinson-Miller cycle), which means a substantially lower actual compression ratio

Thanks
Manolis Pattakos

[manolis]

Hello Tommy Cookers.

You write:
“the Bishop RoV seems to have offered Ilmor no better breathing than did their developed poppet valve head”


Quote from the http://home.people.net.au/~mrbdesign/PD ... echBRV.pdf

“In 1997 Bishop started working with Ilmor Engineering (later Mercedes-Ilmor) to develop their rotary valve technology for F1 engines.
The initial development was carried out on 300cc single cylinder bottom ends supplied by Ilmor at Bishop’s premises in Sydney Australia.
Bishop was responsible for the cylinder head design, development and demonstration of the required durability and performance.

By late 2000 back to back testing with the poppet valve single cylinder engine demonstrated a 10% power advantage and improved durability.

In 2002 the first V10 engines using this technology were built and tested exhaustively.

A completely new V10 engine was designed and manufactured in 2003.

Testing of these engines was prematurely terminated when the FIA announced changes to Article 5.1.5 of the engine regulations late in 2004 with the specific purpose of banning this rotary valve technology.

. . .

Whilst the rotary valve engine has demonstrated breathing capacity that is at least equivalent to the best F1 poppet valve engines, it has the huge advantage that it does this without the dramatic reduction in life that occurs with F1 poppet valve heads.

As inertia induced loads in the valve train are absent in the rotary valve, the forces that destroy the poppet valve heads are also absent.

Further the mechanical and gas loads seen by the valve are essentially independent of speed.

The only issue affecting durability that changes with engine speed is the peripheral speed of the sealing elements and the bearings. As the peripheral speed of the valve over the sealing elements is approximately 80% that of the maximum F1 piston ring velocity, this is of little concern.
In production poppet valve engines, engine life considerations require changes that greatly curtail their breathing capacity from the level achieved in F1. This is clearly not the case with the rotary valve and Bishop’s research suggests a production rotary valve has a breathing capacity up to 45% greater than that observed on current 4 valve production engines.

. . .

On the power front both the rotary and poppet valve engines were designed for the same maximum speed and both engines produced near identical power outputs despite far greater resources being devoted to the development of the poppet valve engine. From a F1 perspective the rotary valve had the potential to increase power faster as it could increase its breathing capacity by merely increasing the valve diameter and unlike the poppet valve it had no inertia issues preventing its operation at higher speeds.”

End of Quote


One can consider the claims of the Bishop team as questionable.
However the dates (bold text) and the facts (the banning, by FIA, of the rotary valves from F1) do support Bishop's claims.


The guys in Bishop claim they achieved a 10% power increase as compared to the original single cylinder Ilmor poppet valve engine.

And they focus, again and again, on the inertia loads in the cylinder head of the poppet valve engine.

With a given cylinder capacity, the bore increases to make room for bigger valves providing freer breathing and more power at higher revs.

The piston gets bigger and heavier. But its stroke is shorter limiting the increase of the resulting inertia loads at the higher revs.

Together with the piston, the intake and exhaust poppet valves get bigger and heavier. But in this case the stroke the bigger valves perform (i.e. the “valve lift”) is not shorter but longer than the stroke of the smaller valves. The longer stroke a valve performs combined with the increased mass of the valve and with the increased revs of the engine give substantially heavier inertia loads into the cylinder head.

For instance, the increase of the Panigale 1299 bore from the stock 116mm to 128mm (as described in a previous post) allows the increase of the intake valve diameter from 46.8mm to 51.5mm (46.8*128/116) (i.e. by some 10%) and the increase of the intake valve lift from 16mm to 17.6mm (i.e. by some 10%, too).
Similarly, the exhaust valve diameter increases from 38.2mm to 42,1mm and the exhaust valve lift increases from 14.3 to 15.8mm.
To keep the same per cylinder capacity, the piston stroke of the PatRoVa Panigale 1299 drops from the stock 60.8mm to 50mm, i.e. it drops by more than 17%.

According the previous calculations, see what happens in case Ducati decides to “destroke” their Ducati Panigale 1299 for the sake of more specific power:

At 14,000rpm (wherein the rev limit of the PatRoVa Panigale 1299 “was set” to give the same mean piston speed with original Panigale 1299 at 11.500 wherein its rev limiter is) the acceleration the piston undergoes is increased by 22% (revs ratio square times strokes ratio, i.e. ( 14,000 / 11,500 )^2 * ( 50.0 / 60.8 ) )

At the same 14,000rpm the acceleration the bigger poppet valves undergo is increased by 63% (revs ratio square times valve lifts ratio, i.e. ( 14,000 / 11,500 )^2 * ( 17.6 / 16.0 ) ).

This is about three times more than the increase of the piston acceleration.

If I were Ducati and wanted to seriously "destroke" the engine, I would worry three times more for the Desmodromic mechanism than for the piston.


In simple words for the non technically oriented:
The acceleration increases with revs square if everything else remain unchanged.
In the case of the piston the sorter piston stroke decreases the acceleration the piston undergoes.
In the case of the valve train, the longer valve lifts increase the acceleration the valve train parts undergo.


Compare the case with the Bishop rotary valve wherein:

“From a F1 perspective the rotary valve had the potential to increase power faster as it could increase its breathing capacity by merely increasing the valve diameter and unlike the poppet valve it had no inertia issues preventing its operation at higher speeds”.

Compare the case with the PatRoVa rotary valve.
Either at 11,500 rpm, or at 14,000rpm, or at 24,000rpm of the crankshaft it runs smoothly and nicely without vibrations, without inertia loads. .
The total inertia force and the total combustion force on the bearings of the PatRoVa rotary valve at, say, 24,000rpm of the crankshaft is zero, no matter how much it weighs..
In comparison, calculate the increase on the inertia and combustion loads on the bearings of the camshafts of the Desmodromic poppet-valve cylinder-head of the short stroke Ducati (128mm bore, 50mm stroke) at 14,000rpm (the combustion load is significant for the initial opening of the exhaust valves).


Do I miss something?

Is the dominant limiting factor the acceleration of the piston or the accelerations related with the valve train?



You write:
“your 3cc '18' OS engine example supports my original argument (surprisingly well considering that it's 200x smaller)
whilst it has 3.37x the piston acceleration of the Ducati, this is due to its b:s ratio of only about 1.04 if debored to this b:s ratio the Ducati piston is about 30% of its present mass”

The weight of the piston of the Panigale 1299 is 480gr, only 2% more than the piston of the Panigale 1199 (116mm vs 112mm diameter).

The bore to stroke ratio of the OS.18 engine is 1.066, not 1.04

Question: where the “30%” come from?


Thanks
Manolis Pattakos

[manolis]

Hello Gruntguru.

Quote from my last post:

“At 20,000rpm there is time shortage for the filling of the cylinder.
The piston moves too fast.
The air / air-fuel entering from the intake ports is not enough to fill the vacuum in the cylinder.
When the piston passes from the BDC, the vacuum is still significant.
At a crankshaft angle, say at 75 degrees after the BDC, the pressure inside the cylinder equals to the ambient pressure.
But the revs are extreme and the inertia of the entering gas stream causes the supercharging of the cylinder till the 90 degrees after the BDC when the intake closes and the fresh charge gets trapped into the cylinder (here plays the streamlining of the intake track).”

Looking again at the Ducati valve lift profiles:



the “75 degrees” seems too much (there is neither adequate time nor adequate valve area for the supercharging of the cylinder by exploiting the inertia of the entering gas stream).

“45 degrees” instead of “75 degrees” seems more realistic.

By the way, at 45 degrees after the BDC the piston has already covered some 10% of its stroke, while at 75 degrees after the BDC the piston has already covered some 30% of its stroke.

Thanks
Manolis Pattakos

[gruntguru]

Manolis.
Agreed. If the cylinder pressure reaches atmospheric at 75* ABDC and no further flow occurs, the VE is only 70%.

If flow ceases at 90* ABDC (approx 42% of stroke), 100% VE requires a cylinder pressure of about 2 atmospheres.

(Surprised myself with that result! Maybe its wrong?)

[Tommy Cookers]

Whilst the rotary valve engine has demonstrated breathing capacity that is at least equivalent to the best F1 poppet valve engines ........
........For instance, the increase of the Panigale 1299 bore from the stock 116mm to 128mm allows the increase of the intake valve diameter from 46.8mm to 51.5mm (46.8*128/116) (i.e. by some 10%) and the increase of the intake valve lift from 16mm to 17.6mm (i.e. by some 10%)...exhaust valve diameter increases from 38.2mm to 42,1mm and the exhaust valve lift increases from 14.3 to 15.8mm.
To keep the same per cylinder capacity, the piston stroke of the PatRoVa Panigale 1299 drops from the stock 60.8mm to 50mm, i.e. it drops by more than 17%.
.....see what happens in case Ducati decides to “destroke” their Ducati Panigale 1299 for the sake of more specific power:
At 14,000rpm (wherein the rev limit of the PatRoVa Panigale 1299 “was set” to give the same mean piston speed with original Panigale 1299 at 11.500 wherein its rev limiter is) the acceleration the piston undergoes is increased by 22% (revs ratio square times strokes ratio, i.e. ( 14,000 / 11,500 )^2 * ( 50.0 / 60.8 ) )
At the same 14,000rpm the acceleration the bigger poppet valves undergo is increased by 63% ........

Do I miss something?
Is the dominant limiting factor the acceleration of the piston or the accelerations related with the valve train?

You write:
“your 3cc '18' OS engine example supports my original argument (surprisingly well considering that it's 200x smaller)
whilst it has 3.37x the piston acceleration of the Ducati, this is due to its b:s ratio of only about 1.04 if debored to this b:s ratio the Ducati piston is about 30% of its present mass”
The weight of the piston of the Panigale 1299 is 480gr, only 2% more than the piston of the Panigale 1199 (116mm vs 112mm diameter).
The bore to stroke ratio of the OS.18 engine is 1.066, not 1.04
Question: where the “30%” come from?

the 30% is the estimated mass of the notional piston of a bore of 1.066 x 60.8 mm, as a % of the mass of the 116 mm piston

more importantly .......
what you are missing is that the available poppet valve 'curtain' (periphery x lift) area increases greatly with increase in b:s ratio
ie potentially larger valves lifting further (as you assume above)
this would be necessary to maintain curtain area in proportion to massflow if the rpm was increased in proportion to the b:s ratio (as you suggest)
imo designing for higher stress/accelerations is unrealistic - a 50 mm stroke V twin crankshaft,rod and 128 mm piston - what fatigue life at 14000 rpm ?

assuming as I do that rpm should be proportional to something like the square root of the b:s ratio, ie broadly designing to the original stress/accel
we could eg keep the curtain area massflow-proportional without any increase in lift
or keep the curtain area massflow-proportional with some small increase in lift and some small increase in diameter
or improve the curtain area/massflow proportion but with less increase in lift and less increase in diameter than you suggest

as imo the rpm etc in later NA F1 eg the higher (engine freeze) b:s ratio that Ilmor used to equal their (earlier) Bishop-valved engine
also a higher b:s ratio benefited the later poppet-valved radial aircraft engines (but would be a disbenefit to sleeve-valved ones)

no-one said it was easy to get the 13.2:1 CR used by these F1 engines
Gilles? Simon said the current F1 engines would have a conflict between valve envelope and compression volume (much lower b:s of course)
thermodynamics would support the view that there was negligible benefit from any higher a CR (in NA F1)

[manolis]

Hello Tommy Cookers

You write:
“the 30% is the estimated mass of the notional piston of a bore of 1.066 x 60.8 mm, as a % of the mass of the 116 mm piston “

What I was asking is: why 30% and not 10% or 50%?


You also write:
“assuming as I do that rpm should be proportional to something like the square root of the b:s ratio, ie broadly designing to the original stress/accel
we could eg keep the curtain area massflow-proportional without any increase in lift
or keep the curtain area massflow-proportional with some small increase in lift and some small increase in diameter
or improve the curtain area/massflow proportion but with less increase in lift and less increase in diameter than you suggest”


Take it the “opposite way”.

Suppose we want to destroke the Panigale 1299 Ducati engine in order to shift the red line from the current 11,500rpm to 14,000rpm.

Unless I am wrong, according your way/approach, the stroke should reduce from 60.8mm to 41mm and the bore should increase to 141mm. This way, the piston acceleration remains unchanged.

With 141mm bore, i.e. 21.5% larger than the stock, the intake valves can increase from the stock 46.8mm to 57mm.

In order to maintain the same accelerations in the valve train (i.e. as in the stock engine) the lift of the intake valve should reduce from the stock 16mm to 10.8mm.

I hear you asking: “And why to maintain the stock accelerations in the valve train?”

Because this is what you propose for the piston: to maintain the original piston acceleration unchanged. The same principle should apply to the valve train, too.

Think of it: a 57mm intake valve with only 10.8mm lift!


There is more:

While the “curtain” (i.e. the periphery of the valve times the valve lift) of the stock intake valve is 46.8*pi*16=23.5cm2, the “curtain” of the bigger diameter valve (bigger by 22%) is 57*pi*10.7=19.3cm2, i.e. only 82% of the stock!

If you decide to maintain the same “curtain” (say, the same flow capacity) with the stock, you need 13mm lift for the big valve, which means 20% higher accelerations in the valve train (the engine still cannot breath freer).

If you decide to maintain the stock lift, the accelerations in the valve train increase by 48% as compared to the stock.


According the previous, the “destroking” of a conventional poppet valve engine for the sake of more power at higher revs is a difficult problem to really solve.

In comparison, with the PatRoVa rotary valve on the cylinder head the engine gets rid of the poppet valves and of their inertia and other problems.

If you decide to keep the piston acceleration unchanged, all you have to do is to divide the stroke by the revs ratio square and multiply the piston diameter by the revs ratio. It results in extremely over-square design.
The PatRoVa rotary valve is OK with it.

In you decide to keep the mean piston speed unchanged (in expense of a relatively small increase in piston acceleration), all you have to do is to divide the stroke by the revs ratio and to multiply piston diameter by the square root of the revs ratio.
The PatRoVa rotary valve is OK with it, too.


But there is more:

Besides the inertia loads there are also the combustion loads on the piston.

With 141mm piston diameter instead of the stock 116mm, the piston bears 48% heavier combustion loads (bore ratio square).

The travel of the flame from the spark plug to the ends of the combustion chamber is by more than 20% longer (141mm bore vs 116mm bore), while the available time is by more than 20% shorter (14,000rpm instead of 11,500rpm), which means need for even longer spark advance and inevitably worse efficiency and hotter engine.

I.e. by reducing the piston stroke as required, you achieve to keep unchanged the piston acceleration in expense of a huge increase of the force acting on the piston crown due to the gas pressure into the cylinder.

With the other approach (the one that maintains the mean piston speed) the piston acceleration does increase but only by 22%, while the force on the piston crown (due to the gas pressure into the cylinder) increases, too, but only by 22%.


Do I miss something?

Thanks
Manolis Pattakos

[manolis]

Hello Gruntguru.

You write:
“If flow ceases at 90* ABDC (approx 42% of stroke), 100% VE requires a cylinder pressure of about 2 atmospheres.
(Surprised myself with that result! Maybe its wrong?)”


The dead volume has to be taken into account, too.

While it seems extraordinary, it also seems the inertia of the inlet gas stream does achieve it, otherwise they would not exist naturally aspirated engines with Volumetric Efficiencies substantially higher than 100%.

It seems strange for a gas, yet after a velocity, say 300mph (RAM air), it starts behaving as a liquid.

140% volumetric efficiency for NA engines seems unrealistic: the specific torque should be near 150m*Nt / l ( 125 bhp per litter at 6,000rpm ).

Thanks
Manolis Pattakos

[manolis]

Hello all.

Talking for 2-strokes,
yesterday it was received from the UK Intellectual Property Office the final confirmation for the granting of a patent for the PatATi engine presented at http://www.pattakon.com/pattakonPatAT.htm





The Opposed Piston PatATi in the above video has only one combustion per crank rotation.
Listen to the sound.

The Cross Radial:



has four combustions per crank rotation.
Imagine the sound.

Thanks
Manolis Pattakos

[Brian Coat]

Hello Brian Coat.

In the question
“Where the knocking problem in the poppet valve engines comes from?”
my
“Does it matter?”
has to do with the fact that while every poppet valve design is around the knocking problem and how to avoid it, the guys behind the Bishop Rotary Valve project claim they never experienced knocking problems despite the fact they did try substantially high compression ratios.

After a century of development the design of the poppet valve engine is still focused on avoiding the knocking (and necessarily is limited to lower than optimum compression ratios), while the other design (Bishop rotary valve) is, from the beginning, “unable” for knocking.


Thanks
Manolis Pattakos

Rotary valve engines will not be "unable to knock".

The claim is an anecdote partly based an engine running at an rpm where conventional engines don't knock either.

It is not impossible to that the Rotary valve engine has superior knock performance but
1) it it is unlikely to be due to reduced local component temperatures because this is not normally a dominant factor in the onset of knock
2) it is unproven? : some real engine-lab-quality data would be nice but has anyone published any?

The attached may provide interesting background reading about knock.

http://digbib.ubka.uni-karlsruhe.de/vol ... nts/799528

[gruntguru]

The reduced knocking tendency of the BRV engines is due mostly to higher burn rates resulting from increased tumble.

[manolis]

Hello Brian Coat.

As you claim, there is no engine “unable to knock”.

What the guys behind the Bishop rotary valve claim, is that for the range of the compression ratios used and tested, they did not observe knocking.

For a spark ignition engine the friction loss and the loads on the kinematic mechanism and the need for more robust (and heavier) parts increase rapidly for compression ratios above, say, 15:1.

According Bishop:
“The F1 single cylinder engine ran compression ratios as high as 17:1 using standard F1 fuels before settling on 15.3:1 as optimum.”

The guys in Bishop had no reason to experiment with a, say, 20:1 compression ratio, for they knew the problems / issues with such high compression ratios.


In comparison the knocking in the modern conventional poppet valve engines is the rule for way lower compression ratios.

With knock sensors and feedback control, the modern poppet valve engine is continuously tuned by its ECU to run just “before” knocking.

A dozen of years ago I remember driving a new VW Golf 1.4 of a relative. Every time the gas pedal was pressed, the knocking was more than noticeable for a couple of seconds. The owner returned it to the dealer for inspection and the service said it was OK; it was the way it was built to run.


As for the cause of the knocking in the conventional poppet valve engines, the hot spots in the combustion chamber seem to play a significant (if not the dominant) role, even if the knocking does not start there.
The motion of the gas plays also a significant (if not the dominant) role.

Here is the "single" tumble in the Bishop Rotary Valve:



According Bishop:

"This oblique flow through the window is responsible for one of the rotary valves most useful attributes: its strong in-cylinder tumble flow.
The tumble ratio on engines with near square bore/stroke ratios is typically twice that reported for similar 4 valve engines.
Unlike the poppet valve this high tumble flow is generated without any loss of volumetric efficiency (VE) and is responsible for very fast burn rates observed."

The PatRoVa Rotary Valve proceeds one more step.
With its symmetrical design it creates a "double" symmetric tumble in the cylinder:



Thanks
Manolis Pattakos

[Brian Coat]

The reduced knocking tendency of the BRV engines is due mostly to higher burn rates resulting from increased tumble.

Any test data?

[gruntguru]

The reduced knocking tendency of the BRV engines is due mostly to higher burn rates resulting from increased tumble.

Any test data?

There is a paper out there, a PhD thesis, try University of Newcastle (Australia). CFD of flow and combustion in the BRV cylinder.

Here it is. https://opus.lib.uts.edu.au/handle/2100/248