32 replies to this topic

[manolis]

Engines balance and vibrations.

I saw a lot of replies sent for Renault’s V-10 vibrations problems.
In the discussion I find some mistakes. For instance, that a flat ten or a flat six cannot be balanced or even realized.
The flat six (like the one in Porsche 911) is even firing and well balanced. The quality of its balancing is absolutely equal to that of the straight six (which, even most people think it is perfectly balanced, actually has some strong third order torsional vibrations around crankshaft axis).
The ten cylinder flat (boxer), even firing, can be perfectly balanced: no inertia forces or moments or torques of any order. The only bad thing of the arrangement is the weak crankshaft with the ten crankpins, compared to the one of the normal V10 of 72 degrees with the five crankpins. And “no strong crankshaft” means “no racing use”.

It seems there is a lot of confusion about vibrations and balance, and it shouldn’t. It is just simple mathematics .
At the last page of the www.pattakon.com site there is the “BALANCE” program. It calculates EASILY the vibrations (or the unbalance) of any reciprocating engine arrangement and it also calculates EASILY the necessary balancing shafts.
With this program you can calculate even the balancing of the new Honda’s V5 SBK.

In return, we need your objection regarding the Variable Valve Actuation system presented in the site. Do you have some?

Thanks
Manolis

[Ben]

It's a Moto GP engine not a Superbike (I assume that's what you meant by SBK?)

Ben

[dosco]

Originally posted by manolis
Engines balance and vibrations.

The ten cylinder flat (boxer), even firing, can be perfectly balanced: no inertia forces or moments or torques of any order.

Are you saying the flat-ten doesn't vibrate because it's "perfectly" balanced?

[jpf]

well, not to re-open this can of worms, but a 180 degree v-10 wouldn't be, but a 10 cylinder boxer would be; it's a difference in the crankshaft...

[manolis]

Sorry, but here with flat and boxer we mean the same thing.

I see that by flat you mean the five crankpin crankshaft with 180 degrees Vee angle, so yes there are vibrations of first order. But again the proper selection of the firing order (or of crank pin arrangement along crankshaft) can actually cancel the second order inertia moment and leave pure first order. Then with a first order balancing shaft the engine becomes very similar in smoothness to the V-12.

But the same can happen in V-10, 72 degrees arrangement. A first order balancing shaft is what it needs to smooth it out.
The V-10 in 90 degrees has uneven firing. In VW they move for 18 degrees the crankpins to return to even firing, while the balancing stays similar to V-10, 72 degrees. And they use a first order balancing shaft.

The V-5 of Honda was mentioned as a strange arrangement with very succesfull balancing, difficult to be calculated. Or it isn't?

Thanks
Manolis

[perfectelise]

Originally posted by manolis
Sorry, but here with flat and boxer we mean the same thing. ...

flat and boxer are the same, and 180 degree V is different - depending on how con rods meet crank shaft - Right Is a boxer better than a 180 degree V, Porche think so, why?

[marion5drsn]

The statements of "manolis" must be taken in view of the fact that he or she is speaking with the fact that the transferring of Greek to English is difficult.

The use of ten separate conrod journals is not acceptable due to the fact you have already mentioned. Not only is it weak but also it is far too long to meet the requirement of the ‘Tub” and the positioning of the driver. And yes the six separate crank pins in a 6 cylinder 180 degree boxer flat engine is the same ‘fix” that other six cylinder engines have used to achieve equal firing patterns and equal spacing of the firing degrees at 120. Also as you state it does weaken the crankshaft in that it would require seven main bearings to strengthen the block and crankshaft. Which also would increase the engines overall length.
I am currently working on the program he speaks of and am finding that it is easy to use and merely a matter of finding the small input errors of mine to use it.
The drawing he has made are just beautiful, at least to me.
M.L. Anderson

[Kevin Thomas]

The ten cylinder flat (boxer), even firing, can be perfectly balanced: no inertia forces or moments or torques of any order. The only bad thing of the arrangement is the weak crankshaft with the ten crankpins, compared to the one of the normal V10 of 72 degrees with the five crankpins. And “no strong crankshaft” means “no racing use”.

I don't understand. Why would a flat 10 have more crankpins than a v-10?

[Wuzak]

Originally posted by Kevin Thomas


I don't understand. Why would a flat 10 have more crankpins than a v-10?

In a horizontally opposed, or boxer, engine a cylinder on the left hand bank reaches TDC at the same time as the corresponding cylinder on the right hand bank. And they both reach BDC at the same time. Hence the term "boxer".

This can't be achieved of a single crank pin - it requires 2 crank pins per cylinder pair 180° apart. So for a 10 cylinder boxer, 10 crank pins are required.

A conventional vee engine shares a crank pin for each pair of cylinders. So for a V10 there are only 5 crank pins.

This can be used in a 180° Vee engine, with cylinder pairs travelling in the same direction - as the LH bank piston heads towards TDC, the RH bank cylinder is going towards BDC.


The flat 12 engines used by Ferrari in the mid-late '70s were in fact 180° V12s, rather than "boxers". I am not sure if the BB (Berlinetta Boxer) and Testarossa engines were boxers or 180° V12s.

[manolis]

Originally posted by Wuzak


In a horizontally .... The flat 12 engines used by Ferrari in the mid-late '70s were in fact 180° V12s, rather than "boxers". I am not sure if the BB (Berlinetta Boxer) and Testarossa engines were boxers or 180° V12s.

The flat 12 cylinder with six crankpins crankshaft (0, 120, 240, 240, 120 and 0 degrees) is absolutely balanced (neither inertia forces of any order, nor inertia torques, nor inertia moments).
That is, with a relatively strong and short crankshaft it achieves full balance and even firing.

So it becomes bad selection the use of the "boxer", as you call it, arrangement in 12 cylinder engines : you gain nohing in smoothness while you loose a lot in crankshaft strength and weight(as it has twelve crankpins with each pair at 180 degrees angular distance). You loose also a lot in overall length of the resulting engine.

So all 12 cylinder in 180 degrees Vee were made as flat (and not as "boxer").

Thanks
Manolis

[jpf]

Exactly, because a straight-six is inherently balanced, and if I'm not mistaken a V-12 of any angle is too. So with a 12 cylinder, you don't gain anything by adding the extra crank pins to make it a boxer. But with 10 or 8 or 6 cylinders, then the trade off is there: better balance for the longer, more complicated, more fragile crank.

In any case I don't think there's any doubt the compromise isn't worth it for F1, as marion5drsn pointed out.

[manolis]

Originally posted by jpf
Exactly, because a straight-six is inherently balanced, and if I'm not mistaken a V-12 of any angle is too. So with a 12 cylinder, you don't gain anything by adding the extra crank pins to make it a boxer. But with 10 or 8 or 6 cylinders, then the trade off is there: better balance for the longer, more complicated, more fragile crank.

In any case I don't think there's any doubt the compromise isn't worth it for F1, as marion5drsn pointed out.

No, the straight six is not inherently balanced.
Yes, the straight six has no inertia forces and it has no inertia moments of any order. But its inertia torque around crankshaft is strong and of third order. Its amplitude is more than half compared to the inertia torque of the straight four having same pistons, connecting rods and stroke (which is of second order).
The inertia torque could be regarded as a messure of the variability, during a cycle, of the kinetic energy strored into the basic mechanism (flywheel, crankshaft, connecting rods and pistons) when the crankshaft and flywheel rotate at constant speed.

On the other hand, the inertia torque of any EVEN firing twelve cylinder is completely balanced.

Thanks
Manolis

[jpf]

OK, I knew I should have just kept my mouth shut. I did not know about third order inertia torque, my balance knowledge is very basic. Thanks for humoring me.

[JwS]

As far as the VVA design, I saw it earlier on another thread, I think it is ingenious, nice and simple, the only thing thing that would concern me is the angle/load that the pushrod operates at, looks a little scary.
JwS

[manolis]

Hello, is anybody home?

Let's warm the discussion, after all I still need objections about the new VVA.

Does anybody know why nobody uses the staight five with crankpins at 0, 72, 288, 144 and 216 degrees?
This five cylinder arrangement has, for sure, lower length than straight six, permitting its mounting transverselly even in small front wheel drive cars.
With just one counterbalancing shaft of first order (rotating in opposite direction than crankshaft)this five cylinder has significantly less inertia vibrations than the in line six.
And its tortional loads along crankshaft are light compared to the rest straight five and to the straight six.
So?

Thanks
Manolis

[VAR1016]

Originally posted by manolis
Hello, is anybody home?

Let's warm the discussion, after all I still need objections about the new VVA.

Does anybody know why nobody uses the staight five with crankpins at 0, 72, 288, 144 and 216 degrees?
This five cylinder arrangement has, for sure, lower length than straight six, permitting its mounting transverselly even in small front wheel drive cars.
With just one counterbalancing shaft of first order (rotating in opposite direction than crankshaft)this five cylinder has significantly less inertia vibrations than the in line six.
And its tortional loads along crankshaft are light compared to the rest straight five and to the straight six.
So?

Thanks
Manolis

I cannot comment on the five, but as you are obviously very knowledgable, can you explain how Ferrari makes a flat-plane crankshaft V-8 work properly in a production car ?(I believe that all the Ferrari V-8s since the 308 have had 180 degree crankshafts).

The VVA looks jolly clever I must say - have you managed to test the engine for horsepower?

PdeRL

[AndreasL]

Originally posted by manolis
Hello, is anybody home?

No, I'm at work...

Originally posted by manolis
Does anybody know why nobody uses the staight five with crankpins at 0, 72, 288, 144 and 216 degrees?

?

Volvo, Audi, Fiat, Hoda and Mercedes does/did.

Or do they have another order of the pins? I'm getting unsecure about the orders, and don't have the references to look it up...

/Andreas

[pessimist]

This is for the question regarding flat plane v8 engines. Actually this site has explanations on engine smoothness on most configurations found in production cars.

http://autozine.kyul...ine/smooth4.htm

[JwS]

The Flat crank V8 gives up a little on vibration but allows superior exhaust scavengeing (sp?) The engines are relatively small displacement, so I guess vibration is ok (I have driven one, didn't notice any probems). I think they fire like two separated 4cyl engines and have headers without need of any crossover, basically two 4cyl headers.
JwS

[VAR1016]

Originally posted by pessimist
This is for the question regarding flat plane v8 engines. Actually this site has explanations on engine smoothness on most configurations found in production cars.

http://autozine.kyul...ine/smooth4.htm

Thanks for that very good link. Marion Anderson insists that the flat-plane V-8 is incorrect - I suppose it is worth noting that the Cosworth DFV was noted for vibrating!

PdeRL

[Wuzak]

Originally posted by VAR1016


Thanks for that very good link. Marion Anderson insists that the flat-plane V-8 is incorrect - I suppose it is worth noting that the Cosworth DFV was noted for vibrating!

PdeRL

I might be wrong about this, but I thought that the DFV was a 75° Vee, which would have made the vibration problem worse.....

TVR's production V8 is a 75° Vee with a flat plane crank......................

[desmo]

The DFV was 90°.

[manolis]

For VAR1016


The flat-plane crankshaft of Ferrari gives strong second order inertia forces.
It also gives a small inertia moment of second order, due to the offset of the two banks of cylinder, but it is relatively small.
The inertia force has amplitude 1.4 times higher than the inertia force produced by the one bank of cylinders (which actually is a normal straight four). It is not at all negligible. The light pistons and connecting rods improve somewhat the situation, but there are significant vibrations.
The inertia force has direction not on the Vee bisecting plane (or “vertical” plane), but on the “horizontal” plane. So the engine body tries to move not up and down, but horizontally. I suppose they put some Kg of soft rubber in the mountings of the engine.

This arrangement has some advantages for use in sport cars.
The independent exhaust system of each bank (which works like an independent straight four) is the best tool for improving the aspiration of the engine, compared to the case of normal V-8 crankshaft.
It is also the weight and inertia of the crankshaft.
The Ferrari arrangement could work without any counterweights secured on the crankshaft. They normally use some counterweights just to reduce the inertia loads on the main bearings of the crankshaft, but they are relatively light.
On the contrary the conventional V-8 with the crankpins at 0, 90, 270 and 180 degrees absolutely needs heavy counterweights secured to its crankshaft, otherwise it suffers from extreme inertia moments of first order. The heavy crankshaft of normal V-8 is not good enough for Ferrari sport cars, which above all need instant response form the engine and as less as possible resistance when the car changes direction. And it is not the easier thing to change the direction of a heavy crankshaft rotating at 7500 rpm.

You need nothing more than balance.exe program in www.pattakon.com to calculate the previous.

For the VVA we do not have torque or power plots.
The mixture is not the correct one neither the spark. They are just roughly adjusted.
And the construction accuracy is not good.
So it would be unfair to take plots from this first prototype.





For AndreasL and anybody else

No, they use other arrangement of crankshaft. Their crankpins arrangement is 0, 144, 216, 288 and 72 degrees. And even using a first order counterbalancing shaft, they still suffer from significant second order inertia moment.

By the way, the first order inertia moment of these engines has an amplitude three times shorter than the amplitude of their second order inertia moment.
Then why automakers balance the first order inertia moment and not the second order one? It is not easy.

Thanks
Manolis

[desmo]

Manolis: Their crankpins arrangement is 0, 144, 216, 288 and 72 degrees.

That was also the prevailing crank geometry in F1 V-10s in the days of 72° engines. I am not sure if that same crank geometry for a 90° V-10 would be optimal. Any thoughts?

[VAR1016]

Many thanks Manolis; I'm afraid thinking in 3-D has never been my strong point, but you have explained the key points which is what I wanted -

I suppose another advantage would be that the simpler and more compact (180 degree)crankshaft would reduce "windage" in the crankcase as well has having less surface area to which oil could cling, further reducing inertial effects?

PdeRL

[JwS]

Manolis:
I gave some further thought to your VVA, I think the weakness is in the pushrod arrangement, not only in the mechanical sense, loading on the rod itself and possible lube/wear issues, but it also adds to the height of the engine, which is generally undesireable. Most modern cars want to tuck the engine in as tight as possible.
Obviously longer pushrods would help the side load issues, why not design it around a ohv pushrod engine and put the VVA mechanism on the side of the block?
JwS

[marion5drsn]

This shows a true boxer engine doing its thing in rotation. Cylinders adjacent to one another and not separated by other cylinders. In german. M.L. Anderson

http://www.boxermoto...es_technik.html

[manolis]

For JwS

The new VVA adds to the engine height something between 30 to 40 mm.
This is about 3 to 4 times less than the height added to the engine by the BMW’s valvetronic, the unique system in market that offers infinite lifts (but only for intake valves and only for relatively low revs). And yet, nobody blames valvetronic for its significantly increased height, as the benefits it offers are so many that the additional height is not even mentioned.
In most cases there is plenty of space at top of engine for locating the new VVA. In many new cars the long induction pipes pass over the cylinder head, adding more height than the VVA. And installing the VVA, these induction pipes normally are substituted by a set of short bell mouths (without throttle valves).
But again, if the height increase is so critical, there is the option of locating the system lower in the head and act on the valves by classic rocker arms. It seems a solution, but the best way is to act on the valves as directly as possible.
The system can be combined to side camshaft and keep the height unchanged. But is this a step ahead?

If there was problem in the loading of the small push rods or of the valve actuator, their dimensions could be significantly increased without changing the basic design. For instance the present pushrod stem diameter of 4 mm could change to 8 mm. Also the 6 mm diameter of its spherical ends could change to 12 mm. The valve actuator also could be the complete bucket lifter of conventional system (there is a relevant animation). But there was no indication from theory, nor from practice that the mechanism is overloaded. And as the new VVA will work most of its life at low to medium lifts (where the surface loading of the cam lobes is from negligible to low), the mechanism could be more reliable than conventional.


For VAR1016
I think so, too. The less obstacles in air motion into the crankcase the easier the “windage”.
For the oil droplet hitting the rotating counterweights of the crankshaft what I think is that they act as a brake and energy absorbers.


For Desmo

(a) The V-10 at 90 degrees is not far in balancing than the V-10 at 72 degrees. The problem is that having just five crankpins the engine is not even firing.
(b) The solution is to split the crankpins to have 18 degrees offset (which is a small angle). The balancing is close to V-10 at 90 degrees and the engine is even firing. The drawback is the relatively lower strength of the crankshaft.
© There is also the solution of VW V-10 at 90 degrees, with crankpin offset of 18 degrees to achieve even firing, and with the main crankpins at 0, 144, 216, 288 and 72 degrees. This arrangement needs one first order balance shaft to work. With this balance shaft it is very close to smoothness to V-12.
(d) But also with a V-10 at 72 degrees having crankpins at 0, 144, 216, 288 and 72 degrees, again with a first order balance shaft, you get even better smoothness than VW and also stronger crankshaft.

To get the plots of inertia loads in case (a) just:
Open the program balance.exe and just press on the keyboard (enter is the enter key):
M enter enter 10 enter 100 enter 144 enter 216 enter 288 enter 72 enter 20 enter 90 enter –90 enter enter (the inertia force plot is on screen) enter (the inertia torque plot is on screen) enter (the inertia moment plot is on screen) fcccfcccc (to get the Fourier analysis and resynthesis of inertia moments) and finally ESC to leave program.
EXPANATION
M: for Multicylinder arrangements
10: for ten cylinder
100: for 100 mm distance from cylinder axis to next cylinder axis on same bank.
144: the angle of second crankpin
216: the angle of third crankpin
288: the angle of fourth crankpin
72: the angle of the fifth crankpin
20: for 20 mm offset along crankshaft axis of the two banks of cylinders
90: for the 90 degrees Vee angle
-90: so when the first piston of first bank is at TDC, the first piston of second bank to be at –90 degrees (and share, in this, the same crankpin)

To get the plots of inertia loads in case (b) just:
Open the program balance.exe and just press on the keyboard:
M enter enter 10 enter 100 enter 144 enter 216 enter 288 enter 72 enter 20 enter 90 enter –72 enter enter (the inertia force plot is on screen) enter (the inertia torque plot is on screen) enter (the inertia moment plot is on screen) fcccfcccc (to get the Fourier analysis and resynthesis of inertia moments) and finally ESC to leave program.

To get the plots of inertia loads in case © just:
Open the program balance.exe and just press :
M enter enter 10 enter 100 enter 72 enter 288 enter 144 enter 216 enter 20 enter 90 enter –72 enter enter (the inertia force plot is on screen) enter (the inertia torque plot is on screen) enter (the inertia moment plot is on screen) fcccfcccc (to get the Fourier analysis and resynthesis of inertia moments) and finally ESC to leave program.

To get the plots of inertia loads in case (d) just:
Open program balance.exe and just press :
M enter enter 10 enter 100 enter 72 enter 288 enter 144 enter 216 enter 20 enter 72 enter –72 enter enter (the inertia force plot is on screen) enter (the inertia torque plot is on screen) enter (the inertia moment plot is on screen) fcccfcccc (to get the Fourier analysis and resynthesis of inertia moments) and finally ESC to leave program.

You could also enter counterweights on crankshaft and reverse rotating balance shaft.
For more, there is the balance.htm file on last page of site giving many examples.

Finally, I do not know what is “optimal” in designing a V-10 for formula1. Who knows? Do they really need the even firing? Can they accept split crankpins? Can they accept a balance shaft?

Thank you
Manolis

[Wuzak]

Dr Mario Thiessen: "We decided in 2000, having studied bank angles from 72 degrees to 100-plus degrees, that 90 degrees was the optimum. This was seen as the optimum if you look at chassis performance on one side and engine performance - gas exchange and vibration issues - on the other side and that view hasn't changed."

From Racetech, April/May 2003, "Engine Tech Special Focus: BMW P83 Formula One V10"



Clearly BMW feels that any vibration issues that the 90° layout induces can be lived with, with a payback in other areas, such as chassis dynamics. I would imagine that to be similar to Ferrari, etc.

[VAR1016]

Originally posted by Wuzak
Dr Mario Thiessen: "We decided in 2000, having studied bank angles from 72 degrees to 100-plus degrees, that 90 degrees was the optimum. This was seen as the optimum if you look at chassis performance on one side and engine performance - gas exchange and vibration issues - on the other side and that view hasn't changed."

From Racetech, April/May 2003, "Engine Tech Special Focus: BMW P83 Formula One V10"



Clearly BMW feels that any vibration issues that the 90° layout induces can be lived with, with a payback in other areas, such as chassis dynamics. I would imagine that to be similar to Ferrari, etc.

Perhaps the extremely short strokes and lightweight components used, have a significant part in this decision?

PdeRL

[manolis]

Originally posted by Wuzak
Dr Mario Thiessen: "We decided in 2000 .... Clearly BMW feels that any vibration issues that the 90° layout induces can be lived with, with a payback in other areas, such as chassis dynamics. I would imagine that to be similar to Ferrari, etc.

So, it seems for the moment that they did made the best selection, as this arrangement is the winner. By the way, is this V-10 in 90 degrees uneven firing, or it has double crankpins in 18 degrees offset?

The time will show if this is true.
I suppose that if a good team decides to use a different V-10 arrangement and to exploit to the limit all its benefits (compared to the present winner), we could talk for a new champion.

After all if the criteria were so obvious and the evaluation so simple, every team had to use this arrangement from the beginning. But only the "try and error" method for evaluating the alternatives - always in combination to existing rules - gave this present winner.


And a question for everybody

Is it possible a single cylinder engine (without balancing shafts) to be, as regards inertia vibrations, equivalent to Wankel rotary engine and better than all conventional engines, no matter how many cylinders and how many balancing shafts they use?

Actually this single cylinder exists and is better balanced than all V-8, V-10 and V-12 arrangements. It can operate for hours standing free on the floor, from idling to the revs limit (24 m/sec mean piston speed), without any tend to move or "jump". Some photos of the prototype are in the "Theory and Images" in Pulling Piston Engine page of "pattakon" site.

Comments? Objections?

Thanks
Manolis

[Wuzak]

Originally posted by manolis

So, it seems for the moment that they did made the best selection, as this arrangement is the winner. By the way, is this V-10 in 90 degrees uneven firing, or it has double crankpins in 18 degrees offset?

I doubt that any team is using offset crankpins for their V10s.

[manolis]

If they do not use offset crankpins, then the best arrangement of V-10 in 90 has five crankpins at 0, 72, 288, 144 and 216 degrees and counterweights only on crankshaft.
The balancing of this arrangement is similar to normal V-8 in 90 degrees, but also the crankshaft counterweights are necessarily heavy.
The firing intervals are 54 degrees, then 90 degrees, then 54 degrees, then 90 degrees and so on.

To get the plots of inertia loads for this arrangement:

Open the program balance.exe and just press on the keyboard:
M enter enter 10 enter 100 enter 72 enter 288 enter 144 enter 216 enter 20 enter 90 enter –90 enter enter (the inertia force plot is on screen) enter (the inertia torque plot is on screen) enter (the inertia moment plot is on screen) then press SPACE BAR (to enter counterweights) and make the array seem like
1____-200_____180____1____1_____0_____0
2____-100____-108____1____ 1_____0_____0
3_____0______108____1_____1_____0_____0
4____100_____-36____1_____1_____0_____0
5____200_____36_____1_____1_____0_____0
and then press END key on keyboard (to exit from counterweights), then enter, enter etc. The engine now has proper counterweight secured to the crankshaft and the final balancing of the engine is excellent. Finally press ESC key to quit balance program.

Thanks
Manolis