2 stroke thread (with occasional F1 relevance!)

[manolis]

Hello Tommy Cookers.

You write:
“btw these (including the one from a Master's) contain at least 3 faults .....
they wrongly state that the inline 4 .....
inertia torque reaches the road .... and wrongly ....
one of its cylinder's inertia torque is shown as different to the other 3 .... and wrongly ....
are in denial over the crossplane 4's invention in Germany 50+ years ago (it then won 2 world championships)”



Here are the correct plots of the free inertia force, torque and moment of the Cross-Plane-Crankshaft Uneven-Firing Straight-Four of Yamaha-R1 (with a first order balance shaft) and of the Flat-Crankshaft Even-Firing Straight-Four (the typical car engine):



On the left axis you see what the plot is about; spot on the units per horizontal line.

The free inertia torque of the “Cross” is several times lower than what is presented in the plot:



of the https://mecaniblog.wordpress.com/2016/0 ... -approach/ article

In the plot:



of the same article the uneven firing of the “Cross” causes a significant asymmetry of the total (inertia and combustion) torque on its crankshaft: the expansion of the third cylinder continues after the ignition / start of expansion of the fourth cylinder, while after the end of the expansion of the first cylinder it is a period during which there is no combustion / expansion).
Is this the plot you mean by “one of its cylinder's inertia torque is shown as different to the other 3”?



You also write:
“regarding engine inertia torque pulses ..... attenuation by frequency separation isn't attenuation by damping .... isolation from either combustion or inertia pulses involves only tiny losses
some isolation is inherently and unavoidably present eg in the compliance of a transmission's shafts etc”


The idea behind the Yamaha-R1 project (as the Suzuki engineers – not the journalists - explain) is to compete / win the conventional arrangements by providing better feeling of the rear tire hooking to the driver.
Its uneven firing is a problem for the peak power.
Its inherent loads require stronger / heavier casing.
Its strong inertia-moment requires the addition of a 1st order balance shaft (more friction, more weight).
Despite all these, Yamaha was the permanent winner in moto-GP for several years. I.e. the idea of eliminating the inertia torque passing to the transmission / rear tire did work.

On the other hand:

The uneven firing is not accepted in car engines: the driver / passengers will feel as something is wrong with the engine.

If, using dampers, the second order inertia torque of the Even-Firing Flat-Crank Straight-Four is isolated, a part of the combustion torque is also isolated / consumed (the rest moto-GP teams / makers knew about dampers, isolation etc, etc, however they failed to win Yamaha’s R1).

And with the strong inertia torque of the Flat-Crank Straight-Four passing to the transmission, the friction in the transmission is increased, as well as the noise and the vibrations.



You also write:
“I have never said anything against the PatVRA as enabling what we might loosely call 'the poor man's crossplane'


As explained in the previous, the “crossplane” comes together with issues / side-effects.

The PatVRA does what the “crossplane” without having the issues / side-effects of the “crossplane”.

If you can have:

• the even-firing of the Flat Crank
  and
  the better-feeling (of the rear tire hooking) of the “crossplane”
  and
  the lower weight and cost of the Flat Crank,
  and
  the higher power of the Flat Crank,

why to chose the “crossplane”?

The one is a compromise, the other is a solution

So, what makes the PatVRA 'the poor man's crossplane' ?

Thanks
Manolis Pattakos

[63l8qrrfy6]

Manolis correlation does not imply causation. Shall we say you are falling for the causation fallacy?

You might want to check the motogp results in the last several years or the world superbike results where the yamaha crossplane inline 4 is being comprehensively beaten by a flatplane inline 4.

I am not implying one engine is better than than another, just highlighting that you are rushing to conclusions.

[gruntguru]

Manolis correlation does not imply causation. Shall we say you are falling for the causation fallacy?

You might want to check the motogp results in the last several years or the world superbike results where the yamaha crossplane inline 4 is being comprehensively beaten by a flatplane inline 4.

I am not implying one engine is better than than another, just highlighting that you are rushing to conclusions.

I don't think Moto GP is the target market for the Pat VRA. The benefits of an L4 with no inertia torque should not be understated.

There is a beautiful illustration of inertia torque and TV. Jack up the drive wheels of a 4cyl manual transmission vehicle. Idle the engine in the highest gear and try to accelerate.

[63l8qrrfy6]

I don't think Moto GP is the target market for the Pat VRA. The benefits of an L4 with no inertia torque should not be understated.

There is a beautiful illustration of inertia torque and TV. Jack up the drive wheels of a 4cyl manual transmission vehicle. Idle the engine in the highest gear and try to accelerate.

Haven't tried but I think the TV response will depend on whether or not the inertia torque will excite a driveline resonance hence it will be different for different engine/gearbox/driveline configurations.

If there is no resonance it should accelerate with minimum vibration.

As I've stated in the last post on the previous page there are cases where driveline resonances occur in the engine speed range. In some cases they can be traversed safely while in other cases they require damping or absorption mechanism. For example many inline 4s would use a dual mass flywheel to reduce response (only works at a fixed frequency) or CPAs (which work over the whole order like the PatVar) in the case of Merc 9g tronic gearboxes.

There are off the shelves solutions for flywheels incorporating both dual mass and CPA (Luk and Schaeffler).

I maintain that they are only needed in specific situations and where they are called for there are established solutions which work very well.

I think the PatVar could work just fine if it was to be balanced perhaps by mirroring the rocker to the other side of the flywheel or by attaching some form of bobweight 180deg apart. But I can't really see how it can be better than a CPA and I am not convinced by the effects on traction in a racing motorcycle application.

[manolis]

Hello MudFlap

You write:
“I think the PatVar could work just fine if it was to be balanced perhaps by mirroring the rocker to the other side of the flywheel or by attaching some form of bobweight 180deg apart.”



With one only "single-sided" rocker (as in the animation) the PatVRA system is fully balanced, i.e. you do not need to add a “mirroring” rocker etc to the other side of the flywheel (by the way: with a “mirror set” of parts at the other side of the flywheel, the system would turn to “super-static”, requiring extreme manufacturing accuracy and creating lots of friction).


Here is a slide from the PatVRA animation:



The left yellow pin is secured on the flywheel, the right yellow pin is secured to the (beige) crankshaft arm (it is secured to, or is a part of, the crankshaft), the cyan pin is secured to the, “free” to rotate, cyan (cyan-green) disk.

The red con-rod (connecting rod) connects the yellow pin of the flywheel with the cyan pin of the “free” disk.

The green con-rod connects the cyan pin of the “free” disk with the yellow pin of the crankshaft arm.

The mass of the red con-rod is “equivalent” (as regards the inertia forces it creates) with a pair of point masses m1 and m2, with the first located at the center of the left yellow pin and the latter located at the center of the cyan pin.

The mass of the green con-rod is “equivalent” (as regards the inertia forces it creates) with a pair of point masses m3 and m4, with the first located at the center of the cyan pin and the latter located at the center of the right yellow pin.

The flywheel can perfectly balance the m1 and the mass of the left yellow pin.
The “free” disk can perfectly balance the m2+m3 and the mass of the cyan pin.
The beige crank-arm (with its bob web located oppositely to the right yellow pin) can perfectly balance the m4 and the mass of the right yellow pin.

So, the “single sided” PatVRA can be perfectly balanced, adding no vibrations to the engine.



Worth to repeat here:

Besides the vibrations / noise reduction and the better feeling, it is also the power saved in the transmission line:

• If you pass 500mN of torque through a transmission line, with the 300mN being an idling / "oscillating" / inertia torque and the rest 200mN being the useful torque, the power loss (friction) is substantially higher than passing only the useful torque through the same transmission line.
  
  With the loads on the gearwheels, on the power shafts, on the bearings etc of the transmission line being substantially lower, the transmission can be more lightweight and more reliable.

Thanks
Manolis Pattakos

[gruntguru]

Haven't tried but I think the TV response will depend on whether or not the inertia torque will excite a driveline resonance hence it will be different for different engine/gearbox/driveline configurations.

If there is no resonance it should accelerate with minimum vibration.

As I've stated in the last post on the previous page there are cases where driveline resonances occur in the engine speed range. In some cases they can be traversed safely while in other cases they require damping or absorption mechanism. For example many inline 4s would use a dual mass flywheel to reduce response (only works at a fixed frequency) or CPAs (which work over the whole order like the PatVar) in the case of Merc 9g tronic gearboxes.

There are off the shelves solutions for flywheels incorporating both dual mass and CPA (Luk and Schaeffler).

I maintain that they are only needed in specific situations and where they are called for there are established solutions which work very well.

I think the PatVar could work just fine if it was to be balanced perhaps by mirroring the rocker to the other side of the flywheel or by attaching some form of bobweight 180deg apart. But I can't really see how it can be better than a CPA and I am not convinced by the effects on traction in a racing motorcycle application.

Agree with most of that - although not the last paragraph.

I think "Inertia Torque" is a terrible misnomer for what is actually a cyclic angular variation of the output. The resulting variation in "torque" is a function of this variation but equally a function of engine speed, inertia of the load and elasticity of the connection to the load (drivetrain). Consequently the abundant dynamic solutions that have been used historically can only "fix" the problem at certain speeds. The problem is angular displacement and can be completely solved with a simple kinematic solution like the PatVAR.

[Tommy Cookers]

I think "Inertia Torque" is a terrible misnomer for what is actually a cyclic angular variation of the output. The resulting variation in "torque" is a function of this variation but equally a function of engine speed, inertia of the load and elasticity of the connection to the load (drivetrain). Consequently the abundant dynamic solutions that have been used historically can only "fix" the problem at certain speeds. The problem is angular displacement and can be completely solved with a simple kinematic solution like the PatVAR.

most or much of the cyclic angular variation of the output is caused by gas loads ie piston ICEs being serial explosion engines
hasn't Manolis already written here that this can't be solved by the PatVAR ?

btw the crossplane inline 4 was invented (by Fath/Kuhn) to be an inline 4 with neither primary nor secondary force imbalance

[manolis]

Hello Tommy Cookers.

You write:
btw the crossplane inline 4 was invented (by Fath/Kuhn) to be an inline 4 with neither primary nor secondary force imbalance


The Fath / Kuhn crossplane has its crankpins arranged at 0, 90, 180 and 270 degrees.
The free inertia force and the free inertia torque are as in the R1-Yamaha crossplane (wherein the crankpins are arranged at 0, 90, 270 and 180 degrees).
I.e. the inertia force and inertia torque plots on top of this page are for the Fath-Kuhn crossplane, too.

The difference is in their inertia moment:



The slightly lower inertia moment of the Fath-Kuhn crossplane comprises a first order component of 390Kp*m (3,900mN) and a second order component of 74Kp*m (740mN), while the inertia moment of the corssplane Yamaha R1 is 435Kp*m (4350mN) pure 1st order.

With a first order balance shaft the Yamaha R1 is perfectly balanced.
With a first order balance shaft, the Fath-Kuhn crossplane has a significant 2nd order unbalanced inertia moment.

I.e. if an external balance shaft is to be used, the Yamaha R1 crossplane is substantially better than the Fath-Kuhn crossplane.
Without external balance shaft, both crossplane arrangements have significant vibration issues.



You also write:
most or much of the cyclic angular variation of the output is caused by gas loads ie piston ICEs being serial explosion engines
hasn't Manolis already written here that this can't be solved by the PatVAR ?


No and yes.

What I wrote was about the inertia torque on the engine casing, not about the combustion torque pulses.

However, yes the PatVAR cannot balance out the combustion torque pulses, which is also the case for all single-crankshaft engines that power the drive wheel(s) though a gearbox.
Even the best Wankel-rotary and the best V-12 and V-8 engines cannot cancel out the combustion torque pulses: the transmission line transfers these pulses to the drive wheels and the casing of the engine “tries” to turn the opposite direction (action – reaction: the crankshaft / con-rods / pistons, supported on the cylinder walls and on the casing, push the primary shaft of the gearbox to rotate at one direction, and push the casing of the engine to rotate at the opposite direction).

So, this is not a weakness of the PatVAR, but a characteristic of all single-crankshaft engines.


About the “Cyclic angular variation of the output caused by gas loads”

Here comes one more advantage of the PatVRA as compared to the crossplane (of Yamaha, or of Fath-Kuhn): in the PatVRA the power pulses are equal, equally spaced (equally distributed), while in the crossplane the power pulses are uneven (un-equal), unevenly spaced.
In the PatVRA the maximum of the combustion torque is lower than in the crossplane, and the maximum time-distance between successive combustion pulses is shorter than in the crossplane, giving a better / smoother operation and decreasing the maximum loads in the transmission line.


Worth to note here:

With two counter-rotating crankshafts having zero phase-difference, say as in the OPRE Tilting of the Portable Flyer:



and symmetrical load (the two counter-rotating propellers of the Portable FLyer) the basis of the engine can get rid of combustion torque loads (i.e. from combustion vibrations).

For instance:

A misfiring (or the abrupt opening /closing of the throttle valve) of the conventional engine of an ultra-light can destabilize it, causing even its turn upside-down; a misfiring (or the abrupt opening/closing of the throttle valve) of the OPRE Tilting with the two counter-rotating propellers (powering the same ultra-light) is anything but dangerous: the pilot may not even notice it.

Thanks
Manolis Pattakos

[Tommy Cookers]

.....
1 Without external balance shaft, both crossplane arrangements have significant vibration issues.
.....
However, yes the PatVAR cannot balance out the combustion torque pulses ....

2 A misfiring (or the abrupt opening /closing of the throttle valve) of the conventional engine of an ultra-light can destabilize it, causing even its turn upside-down; a misfiring (or the abrupt opening/closing of the throttle valve) of the OPRE Tilting with the two counter-rotating propellers (powering the same ultra-light) is anything but dangerous: the pilot may not even notice it.

well .....
1. neither engine has significant vibration issues .....
their moment imbalance (ie equivalent forces at crankcase) is far less than conventional engine's forces at crankcase - always
always except maybe in steam locomotives (typically there's 2 cylinders eg 400mm bore but spaced at 2m centres)
(NOTE TO SELF - steam locos are crossplane so have no need for Pat VRA)
2. the most numerous WW1 aircraft had Monosoupape type (unthrottled) engines - having only full power or zero power
their power control for eg landing or formation flying done by blipping the ignition on/off - even today

engine cutting etc produces mainly yaw effects not roll effects
uncontained yaw effects can produce further roll effects tipping planes over ie into a spin

[gruntguru]

I think "Inertia Torque" is a terrible misnomer for what is actually a cyclic angular variation of the output. The resulting variation in "torque" is a function of this variation but equally a function of engine speed, inertia of the load and elasticity of the connection to the load (drivetrain). Consequently the abundant dynamic solutions that have been used historically can only "fix" the problem at certain speeds. The problem is angular displacement and can be completely solved with a simple kinematic solution like the PatVAR.

most or much of the cyclic angular variation of the output is caused by gas loads ie piston ICEs being serial explosion engines
hasn't Manolis already written here that this can't be solved by the PatVAR ?

Obviously my post refers only to the bit referred to as "Inertia Torque".

There is a big difference. Gas loads on pistons generate a torque in direct proportion to gas pressure and crank position. "Inertia torque" results in a torque which is a function of many variables as I said above.

Which is larger? Under many operating conditions "inertia torque" can be significantly larger than torque variation due to gas loads.

[manolis]

Hello Tommy Cookers.

You write:
“well .....
1. neither engine has significant vibration issues .....”



We talk for the 4-cylinder in-line crossplane without balance shaft (either the Yamaha R1 or the Fath-Kuhn

The following plot:



Is the comparison of the unbalanced inertia moment of the crossplane Yamaha R1 and of the twin made after removing the two outer (side) cylinders of the R1 engine (i.e. a twin having crankpins at 0 and 180 degrees).

The vibrations of the twin with a balance shaft and the vibrations of the twin without a balance shaft are like comparing the “day with the night”:

• Quote from page 222 of this discussion:
  
  This engine suffered typical parallel twin type vibration, i.e. quite bad. I tried to fix it with changing the crankshaft balance factor but all that did was to relocate the bad vibration elsewhere within the operating rpm range.
  
  I knew this might happen, and in fact it only took 15 minutes of run time to start cracking the exhausts.
  
  I pushed RON back into the corner of the workshop and that was that for about a year.
  
  I had been considering using a counterrotating balance shaft but I didn’t want the extra weight.
  
  What to do however?, as without this shaft the vibration was a show stopper.
  
  One Sunday afternoon (several actually) I built a balance shaft and bolted it up externally to the engine as a test - with counter rotating bob weights flying around in mid-air. The difference was like night and day! I could not believe how smooth this new addition made the engine feel.

By adding two more cylinders at the ends of the twin (to make the Yamaha R1 engine) the unbalanced inertia moment increases more than three times.

So, the crossplane 4-cylinder does have significant vibration issues and does need a 1st order balance shaft.

Thanks
Manolis Pattakos

[Tommy Cookers]

.... By adding two more cylinders at the ends of the twin (to make the Yamaha R1 engine) the unbalanced inertia moment increases more than three times ......

the R1 only needs to run at half the rpm of the twin to make the same power as the twin
doing that the R1 will make less unbalanced inertia moment than the twin

[manolis]

Hello Tommy Cookers.

You write:
“2. Suzuki RG500 (+other 2 stroke sq4s/near-sq4s ?) balances combustion torque pulses (simultaneous firing on both cranks)”



Here is the Suzuki RG500 engine:



The two crankshafts rotate at the same direction and transmit their power to an intermediate shaft rotating at opposite direction.

The arrangement can’t balance the combustion torque pulses. On the contrary, the combustion torque pulses, due to the simultaneous combustion in two cylinders, double in size.


If the two crankshafts of an engine counter-rotate, and each one drives its own load (with the two loads being equal and counter-rotating), then the combustion pulses can be balanced.
Say like:



wherein a 500cc OPRE Diesel engine drives two symmetrical counter-rotating propellers (the two crankshafts counter-rotate at zero phase difference).

The basis of the engine, which stands free on the floor, is not only perfectly rid of inertia forces, moments and torques, but it is also perfectly rid of combustion torque pulses.



You also write:
”the R1 only needs to run at half the rpm of the twin to make the same power as the twin
doing that the R1 will make less unbalanced inertia moment than the twin”


With the same pistons and piston stroke, the R1 will run at the same rpm as the twin, to make two times the power of the twin.

The limitation of the rpm of an engine (and especially a sport motorcycle engine) has to do with the breathing efficiency of the engine and with the reliability of the engine (if the mean piston speed goes over a limit, the reliability is hurting). It has nothing to do with the vibrations reduction.


***********

EDIT:

When two motorcycles, the one powered by the Twin and the other powered by the 4-cylinder crossplane, go side by side on the same road, then the big one can operate at lower revs.

When the rider of the crossplane will use high revs, or will go fast, or will race, the vibrations will be too strong and the riding will be uncomfortable.
It is supposed that a bigger / more expensive engine provides more power having better NVH at all conditions.


As for the relation of the low revs with the vibrations, the Norton Commando with the elastic mounts of its Twin engine (crankpins at 0, 360 degrees) vibrated strongly at the lower revs:

• Quote from https://thevintagent.com/2019/03/28/the ... o-miracle/
  
  Back at his drawing board, Bob Trigg got to work. Rubber specialists Metalastic were too busy to help, but did tell him that the rubber had to be a high enough rate to absorb the vibration – if the rate was too low, the vibes would simply destroy the mounts. In quick time, an initial prototype was put together and tested on the factory’s internal roads. The new system did cut vibration, but only over 6500rpm. Norton’s Engineering Director Stefan Bauer, who had little motorcycle experience but was a lateral thinker, told them to slice the mounts in half. Trigg thought that would reduce their life, but did as he was told – now they cut vibration over 4000rpm. Again, Dr Bauer said, slice them in half.
  
  “We took the bike up the road,” said Bob Trigg, “and it was like being in an aeroplane, bumping along at low revs and then at 2300rpm it would smooth out. And that was great, because you could still feel a big twin under you at very low revs, but it took all the hassle out of vibration when riding.” At that point, the Norton Commando was born.

A crossplane 4-cylinder without a balance shaft does not fit with modern vehicles.
The addition of a balance shaft makes it perfect as regards its inertia vibrations; the pain is worth the gain.

***********

Thanks
Manolis Pattakos

[manolis]

Hello Tommy Cookers.

More about the inertia and combustion torque pulses.


The Suzuki RG500 (photo in the last post) has two crankshafts rotating at the same direction.


Here are two famous engines, each utilizing a pair of counter-rotating crankshafts.

The Brough Superior:



and the Ariel Square:




Due to their symmetry and to their counter-rotating crankshafts, the inertia torque on the engine casing is zero.

However the transmission line is loaded by a heavy inertia torque (as heavy as in the typical straight four car engines).
Why?
Because all four pistons stop together and then (after 90 crank degrees) all four piston move at high speed.
I.e. the angular speed of the crankshafts cannot help maximizing (and minimizing) two times per crank rotation.

The same for the combustion torque pulses: they pass to the transmission and to the drive wheel, pushing the engine to rotate the other way.

Thanks
Manolis Pattakos

[63l8qrrfy6]

I think "Inertia Torque" is a terrible misnomer for what is actually a cyclic angular variation of the output. The resulting variation in "torque" is a function of this variation but equally a function of engine speed, inertia of the load and elasticity of the connection to the load (drivetrain). Consequently the abundant dynamic solutions that have been used historically can only "fix" the problem at certain speeds. The problem is angular displacement and can be completely solved with a simple kinematic solution like the PatVAR.

What I call inertia torque is simply the torque produced by the piston acceleration. It is part of the excitation torque on the drivetrain. I think what you call inertia torque is the dynamic torque at different points in the driveline and represents the response to the excitation. As you say this quantity is much more difficult to calculate since the inertia, stiffness and damping of the system need to be considered. However if the driveline was perfectly rigid then the two would be equal.

I think the fact that PatVRA is a kinematic solution is actually part of the problem. That means that the torque amplitude and frequency it produces are monotonic functions of engine speed and the phase angle is fixed.
Going back to TC's point about dealing with the firing excitations which in an even firing inline 4 produce the same 2nd order excitation as what I call the inertia torque but their phase angle is different and changeable with engine speed and load.

So for example if the inline 4 engine is boosted it would produce a high 2nd order firing excitation at a low engine speed. The PatVRA would only produce a weak "cancelling" torque with a phase angle which would be different than the ideal 180 degrees. By contrast a CPA tuned to the 2nd order has free rollers which will respond with the correct phase angle to whatever the perturbation is because they are not kinematically constrained.

Sure the PatVRA will work in an inline 4 where the inertia torque has a similar amplitude to the firing torque amplitude such a high speed N/A engine but does such an application really benefit that much from such a device ?