2 stroke thread (with occasional F1 relevance!)

[gruntguru]

Manolis,

I find it hard to believe that the potential of the valve springs has been reached.

Whatever force is needed, a suitable spring stiffness can provide it at the required displacement. I agree that the problems posed by the dynamic behaviour of the spring are by no means trivial but they are not unsolvable.

The main reason pneumatic valves are used in motorsport is because the work required to compress the working fluid adiabatically is less than the work required to compress a linear spring by the same amount. Simply put, it takes less power to turn the cams. Yes, it does side step potentially problematic spring dynamics, but that is a secondary effect.

If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration). Pneumatic springs permit significantly higher valve accelerations for several reasons. Just a few:
1. Spring dynamics (as mentioned)
2. Reduced reciprocating mass.
3. Pneumatic force can be increased with rpm. Fixed spring force can exceed lubricant film strength at low speeds if set too high.

[manolis]

Hello Gruntguru.

You write:
“If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration).”


In this plot ( from http://www.pattakon.com/pattakonPatRoVa.htm ) :



it is shown the way the PatRoVa rotary valve can give as square profile as desirable.


According the above plot, the (peak) valve area is the area of the “blue” window times two (because there are two such windows that cancel out the total force applied by the high gas pressure (during compression, combustion, expansion) on the rotary valve, and on the bearings of the rotary valve).

Keeping constant the area of the blue window (and its radial dimension),
and increasing the diameter of the rotary valve (the rotation axis of the rotary valve (at the cross, at top) moves away from the window),
the f1 angle decreases (and the f4 angle increases by the same amount to keep the inlet duration: 2*(f1+f4) constant).

As the diameter of the rotary valve increases, the duration the window remains completely open during the induction (which is: 2*(f4-f1) ) increases, while the duration of the two ramps (2*(2*f1)), which is when the blue window is “partially open”) decreases.

For instance, if the diameter of the rotary valve doubles (keeping the area of the blue window and its radial “height” unchanged), the f1 halves, which means the duration of the two ramps halves and the angle 2*(f4-f1) during which the blue window remains completely open increases by the same amount (simply talking: what is lost at the ramps is added to the “full open” duration).

Increasing more and more the diameter of the rotary valve, the “valve lift profile” gets more and more square.
Theoretically, it can be as square as desirable.


Increasing the external diameter of the PatRoVa rotary valve, and keeping unchanged the radial “height” of the window, the hub (or shaft) of the PatRoVa rotary valve gets even stiffer, while the exhaust port on the rotary valve moves away from the “high pressure sealing area” of the rotary valve (the area on the two opposed disks where the window gets red), which further helps in keeping the “sealing clearances” (between the rotary valve and the windows of the combustion chamber) too small:



Worth to note here:
The sealing of the PatRoVa is not based on the contact of some surfaces or sealing means (as happens, for instance, in the Cross-Bishop rotary valve, wherein the sealing is based on a set of immovable seals abutting onto the working cylindrical surface of the rotary valve).


According all the previous, the increase of the external diameter of the PatRoVa rotary valve (in order the valve lift profile to be more and more “square”) does not limit the rev limit of the engine. Because its bearings are rid of loads, and because its “sealing” is not based on the contact of surfaces (which means, it is a frictionless sealing).

The only limitation is the required space for the bigger diameter PatRoVa rotary valves.


By the way, in the moto-GP revs (r.p.m.), a substantially bigger clearance is allowed because the time for leakage is substantially reduced as compare to that of non-racing engines.

Achieving at, say, 200 rpm of the manual cranking, such a compression (the valve is dry):



the clearance required at, say, 100 (a hundred) times higher revs (20,000r.p.m), where a motoGP PatRoVa is to work, is by far bigger and way easier to be achieved and maintained.



QUOTE from http://www.pattakon.com/tempman/Bishop_ ... echBRV.pdf

Testing of these engines (the Bishop Rotary Valve F1engine) 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.

End of QUOTE.


Does anybody know whether the rotary valves are already banned in the motoGP engines?


Thanks
Manolis Pattakos

[henry]

Manolis. What are he advantages of the tapered valve?

An earlier poster, who seemed quite Cross, remarked on the awkward shape of the inlet tract involving two right angled turns into the cylinder. Does this tapered design improve that and so improve volumetric efficiency?

It looks to me that it does but try as I may I cannot make stereoscopic imaging work. Perhaps I need to return to childhood.

[J.A.W.]

Hello Gruntguru.

You write:
“If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration).”

Does anybody know whether the rotary valves are already banned in the motoGP engines?


Thanks
Manolis Pattakos

Hi Manolis, I think gg was referring to the "dynamics" - limitations of camshaft driven poppet valves..

& at 1st glance, it appears that Moto GP regs, while explicitly deeming 4cyl 4Ts, do not actually ban rotary valves..

Moto GP regs here: http://www.motorsport-total.com/motorra ... t-2016.pdf

Note: forum member: "uniflow" - has empirical experience of variable timed rotary valve use in 2Ts..

[63l8qrrfy6]

Manolis,

I find it hard to believe that the potential of the valve springs has been reached.

Whatever force is needed, a suitable spring stiffness can provide it at the required displacement. I agree that the problems posed by the dynamic behaviour of the spring are by no means trivial but they are not unsolvable.

The main reason pneumatic valves are used in motorsport is because the work required to compress the working fluid adiabatically is less than the work required to compress a linear spring by the same amount. Simply put, it takes less power to turn the cams. Yes, it does side step potentially problematic spring dynamics, but that is a secondary effect.

If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration). Pneumatic springs permit significantly higher valve accelerations for several reasons. Just a few:
1. Spring dynamics (as mentioned)
2. Reduced reciprocating mass.
3. Pneumatic force can be increased with rpm. Fixed spring force can exceed lubricant film strength at low speeds if set too high.

The square profile is not possible because it can t be made into a lobe regardless of wether the valve is pneumatic or sprung..

If you were to calculate the pneumatic valve seal friction force I think it would be very close to cancelling out any reciprocating mass advantage.

Varying the pneumatic pressure is a strong advantage. You can even dump the pressure out while the valve is opening provided you have a compressor rather than a limited air supply.

[gruntguru]

Manolis,

I find it hard to believe that the potential of the valve springs has been reached.

Whatever force is needed, a suitable spring stiffness can provide it at the required displacement. I agree that the problems posed by the dynamic behaviour of the spring are by no means trivial but they are not unsolvable.

The main reason pneumatic valves are used in motorsport is because the work required to compress the working fluid adiabatically is less than the work required to compress a linear spring by the same amount. Simply put, it takes less power to turn the cams. Yes, it does side step potentially problematic spring dynamics, but that is a secondary effect.

If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration). Pneumatic springs permit significantly higher valve accelerations for several reasons. Just a few:
1. Spring dynamics (as mentioned)
2. Reduced reciprocating mass.
3. Pneumatic force can be increased with rpm. Fixed spring force can exceed lubricant film strength at low speeds if set too high.

The square profile is not possible because it can t be made into a lobe regardless of whether the valve is pneumatic or sprung..

It turns out that "infinite acceleration" is not possible either. Half-square profile and infinite acceleration is theoretically possible with roller followers. It won't work of course!
.

If you were to calculate the pneumatic valve seal friction force I think it would be very close to cancelling out any reciprocating mass advantage.

The seal friction force is probably less than the friction deliberately introduced to high performance valve springs for damping purposes.

If friction was a concern, pneumatic springs could use a bellows or diaphragm.

[63l8qrrfy6]

If dynamics were not an issue, valve opening events would have a square profile (infinite acceleration). Pneumatic springs permit significantly higher valve accelerations for several reasons. Just a few:
1. Spring dynamics (as mentioned)
2. Reduced reciprocating mass.
3. Pneumatic force can be increased with rpm. Fixed spring force can exceed lubricant film strength at low speeds if set too high.

The square profile is not possible because it can t be made into a lobe regardless of whether the valve is pneumatic or sprung..

It turns out that "infinite acceleration" is not possible either. Half-square profile and infinite acceleration is theoretically possible with roller followers. It won't work of course!
.

If you were to calculate the pneumatic valve seal friction force I think it would be very close to cancelling out any reciprocating mass advantage.

The seal friction force is probably less than the friction deliberately introduced to high performance valve springs for damping purposes.

If friction was a concern, pneumatic springs could use a bellows or diaphragm.

You are probably thinking of interfered concentric valve springs they used in the past. That was back when spring dynamics were not properly understood - it was just a band aid to make engines last for a race. Nowadays for example the 1000rr engine which has the same bore as an f1 engine revs to 14000 with fairly simple single springs.

Dead coils also produce a wee bit of friction but nothing significant as far as I am aware.

There is yet another disadvantage of the pneumatic spring - it is very hard to make the valve rotate.

If I have time this weekend I'll have a go at writing a little piece of code to calculate dynamic spring response just to have a look at what can be achieved with non uniform pitch, diameter and even cross section.

[J.A.W.]

Ok, now, back to 2T stuff..
Here's a Canadian 2T maker; http://www.compactradialengines.com/mz301.html
..which along with conventional engines such as an inline triple, also makes cruciform 'radial' 2T mills..

[J.A.W.]

An interesting Rotax-based disc-valve twin extended to an inline 3, using the 2nd disc for 2 cylinders..

[Tommy Cookers]

[quote=manolis post]Hello J.A.W. You write:
“Hi Manolis, I think you may be a bit too quick in dismissing the useful gains to be had from 2T exhaust energy,
even the simple fixed dimension expansion chamber offers a 'boost' of ~0.3 bar over ambient atmosphere..”

The reasonable choice for a 2-stroke is the compression ignition because, by definition, a Diesel engine runs well from extra lean (idling), to lean, to slightly lean (at peak torque, peak power). ........
In comparison, a spark ignition 2-stroke has to operate at “strange conditions” wherein the residual gas causes unstable combustion and varies periodically the conditions the fuel is burnt at (this was what the Honda Radical Combustion was trying to fix).
One more advantage of an efficient direct injection Diesel is the low temperature and pressure of the exhaust gas. With more energy of the fuel going to the crankshaft as mechanical energy, the thermal loss on the exhaust gas reduces.
The sooner the combustion completes, the better. We talk for progressive and not impact combustion......
With a longer piston dwell around the combustion Dead Center (say, TDC), more fuel is burnt at higher expansion ratios.
a molecule of fuel burnt at 15 degrees after the TDC (wherein the piston has covered only 2% of its stroke downwards) “sees” an expansion ratio of 107/(2+7)=12:1, while a molecule burnt at 30 degrees after the TDC (the piston has covered 8% of its stroke downwards) “sees” an expansion ratio of only 7:1.
The more efficient the burning of the fuel, the less energy is left to the exhaust gas and the quieter the exhaust.
This is what the OPRE and PatOP prototype engines do.

I think there are no 2-stroke Diesels using expansion box for tuned exhaust.
The gain of using a well exhausted (literally speaking) gas to “boost”, by an expansion chamber, a high efficiency di Diesel is questionable.
Without any exhaust system (the PatRon 2-stroke di Diesel of the animations), the noise from the relatively quiet exhaust is covered by the noise from the rotating propeller.ThanksManolis Pattakos[/quote]


'strange conditions' ie combustion variability at partial torque an inevitable result of tuned length exhaust systems (without variable port timing anyway)
maybe why aviation regulatory authorities are sceptical of 2 strokes in aviation
variability causing subharmonic vibrations potentially affecting airframes and variability causing power instability eg in instrument flying
ie some of our apparent progress is illusory
btw the tuned exhaust system will give benefits of .3x ambient ie .3 bar at sea level and less at altitude
btw don't our F1 cars have exhaust system benefits of ?x 4 bar from tuned exhaust action upstream of the turbine ?

Compact Radial includes ungeared 3 and 4 cyl 2 stroke radials of low rpm and specific output (I feel vindicated)
does the 3 have crankcase compression-type induction ??
(the 4 is presumably a doubled HO twin so no problem ?)
people should remember that 90% of light aircraft engines are ungeared

NOTE TO SELF - the PZL-3S 7 cyl 1265 cu in was ungeared, was replaced for agplanes by the geared -3SR version 1976 on, now both disappeared ?
but ASz 62 is still going - 1820 in 'Cyclone' geared w 2 speed supercharging c1150 hp t/o 96 octane ?eg re-engining Otters



regarding the importance or otherwise of instantaneous valve opening and closing ......
VVT (and/or other VG) is surely rather more important with the ever-larger relative valve area that we seem to be seeking
as the lowered gas velocities give much reduced dynamic pressure and so tend to reduce gasflow wherever valve timing is non-ideal

[J.A.W.]

T-C, as you must realize,
..the current F1 machines are boosted by a complex system far removed from a simple fixed geometry pipe..
..in cost also..
A properly designed adaptive 2T expansion pipe might well maintain its advantage over ambient pressure..
..as altitude is gained, much as a turbo also does...

The simple fact is that any/all 2T exhaust is - per se - expelled energetically, compared to the laboured flow
endured by the 4T - pumping out through its exhaust port, & usually obtunded by a poppet valve head..

Kindly take those interminable 4T poppet valve travails to an apropos thread...

[Tommy Cookers]

the turbocharger simply adjusts itself to altitude (by rotating faster) - as Rateau said
yes the 2 stroke 'tuned exhaust' with increasing outlet restriction for altitude would seem able to maintain notable effect
(I guess that even at sea level outlet dimensioning is set to produce mean exhaust pressure above ambient, but no-one will say)
but it's hardly a recipe for cycle-cycle consistency at partial powers of scavenge and thereby combustion ?
and a 'tuned' induction system is less effective at altitude
in F1 a 2 runner tuned exhaust 'seeing' the pre-turbine exhaust as a 4 bar 'atmosphere' is simple and not implausible ?
I suggested simultaneous firing 2 stroke layouts for reduced crankcase and exhaust system quantity but it's part simplification

it's physics (ie a gas can't be in 2 places at once) that gives the illusion that valves and ports should always be bigger and/or faster opening
ie even if the exhaust port or valve was infinitely large and opened instantaneously we would need to open it at the time we do now
and 99% of the time our engines would be worse

[J.A.W.]

T-C, such matters as air density drop with altitude/rated height of aero-engines is a functional parameter..
..given due consideration in all aircraft operations..

& per the thread topic..
..one of the notable R-R Crecy 2T design features - a very potent contribution of exhaust thrust to flight power...

R-R certainly felt that for high-speed flight by piston engines - running hard at high power settings..
..the contribution of exhaust thrust more than compensated for parasitic mechanical blower losses..
..& negated the expense/complexity of turbos - for their purposes...

[manolis]

Hello J.A.W.

You write:
“Ok, now, back to 2T stuff..
Here's a Canadian 2T maker; http://www.compactradialengines.com/mz301.html
..which along with conventional engines such as an inline triple, also makes cruciform 'radial' 2T mills..”


The radial-three (SC430) seems an interesting engine design:





but it also seems a weak engine, too: 80mN/lit specific torque from a 2-stroke!.

They write :
”Each cylinder has its own separate crankcase chamber, interdivided by the crankshaft's sealing discs.”

Are the crankpins in a line?

Thanks
Manolis Pattakos

[manolis]

Hello all.

A square valve lift profile is meaningless.

However a “more square” valve lift profile may define the winner of a race (motoGP etc)

QUOTE from http://www.hotrod.com/articles/ctrp-080 ... selection/ (do take a look):



End of QUOTE


According the above plot, at 80% of the maximum valve lift the flow is already maximized.

What an even higher valve lift, or a "more square" valve lift (or both) can offer, is the earlier maximization of the air flow and the later drop of the air flow (which means: increased area under the gas-flow curve).


Here are some of the available valve lift profiles:



of the pattakon VVA-roller Honda Civic prototype engine:



https://www.youtube.com/watch?v=-zzW8YkReLU

The substantially bigger valve lift (compare the top blue curves with Honda's original red curves) is better basically because it maintains the maximum air-flow for longer.



However, with a “more square” valve lift profile, the valve train “feels” like operating at substantially higher revs (because the valve has to accelerate / decelerate in shorter time intervals as compare to the case with “less square” valve lift profile).
And the inertia loads (this is the enemy) increase with the revs square.


No doubt, with pneumatic valve springs the designer has fewer limitations than with the wire valve springs.

The real question is how much better a motoGP engine performs with pneumatic valve springs (or how far the limitations go).


With the Cross-Bishop rotary valves, the problems related with the high inertia loads in the valve train at extreme revs is eliminated.
The valve lift profile can get even more “square” than with pneumatic valve springs, in expense of a bigger diameter rotary valve that gives higher speed of the relative sealing means (friction, wear etc)).

In comparison, with the PatRoVa rotary valve, the “more square” valve lift profile is for free.

Thanks
Manolis Pattakos