21 replies to this topic

[hydra]

I've been wondering about this for quite a while now... What are the effects, both qualitative and quantitative, of piston speed on engine life (as measured in terms of piston/ring/bore wear and NOT catastrophic failures) ? What is it that seems to be preventing top flight engines from going much above 5000fpm? Surely its not just max piston accelerations (I don't want to go into those) as you could go with a long-stroke lowish-rpm motor. Is it gas dynamics or wear/longevity considerations?

Secondly, what is it that F1 engine makers are doing to make their engines last this much longer without lowering their rev limits much? Makes me wonder...

[McGuire]

I don't think there is a 5000 fpm piston speed limit, any more than there is a 20,000 rpm limit. The troubles lie not in the piston's speed but stopping and reversing its direction twice per crank rotation.

[McGuire]

Below are two F1 engines. Both run at 19000 rpm and have 41mm strokes, so their mean piston speeds are identical at just over 5000 fpm. However, the first uses a rod/stroke ratio in the normal range for current production engines, 1.62:1. The second employs the very long connecting rod (120mm) now in vogue in F1, for a rod/stroke ratio of nearly 3:1. As the two traces show, while the mean speeds are identical the accelerations differ remarkably.

Due to the extreme bore/stroke ratios now used in F1, former Cosworth chief Geoff Goddard likes to describe the piston and rod as essentially a big dinner plate with with a wire hook scotch-taped to the underside. Naturally, twice per crank rotation the dinner plate must stop and change direction, and that's when the piston is most prone to pull off its hook. Of course these extreme accelerative loads apply equally not only to the piston and pin but to the rings and ring lands as well, and even to the oil lubricating the piston. So as Goddard says, it's not so much a piston speed limit as a piston acceleration limit, which he and others have stated is currently in the neighborhood of 10,000 g.

[shaun979]

I have it on good authority that rod ratio in F1 engines would not be so high if their block heights could be shorter without interfering with port,runner,airbox, dimensions/spacing. For at least one major F1 engine supplier, rod ratio is not even a top 10 consideration. Rod length is something they design to, not around.

Going wide angle would allow shorter block heights and shorter rods, but their are problems that accompany such a change.

The extra acceleration of short rods undoubtedly affects all reciprocating compoenents above the crank, but increased crank loads are offset by the lower mass of the shorter rod. Then again that might be offset once again by the need to beef up the pin and piston.

[hydra]

McGuire,
You're missing my point. I clearly stated that I wanted to completely disregard the effect of piston acceleration and the resulting forces on the rotating assembly and focus purely on the tribological/wear aspect of such a high piston speed... i.e. how much worse - from a longevity point of view - is a 5000fpm piston speed than a 4000fpm piston speed? Around where does the speed/longevity balance become unfavorable for a street engine?

[McGuire]

Quote

Originally posted by shaun979
I have it on good authority that rod ratio in F1 engines would not be so high if their block heights could be shorter without interfering with port,runner,airbox, dimensions/spacing.

Sorry, but I am not following you at all. For an engine of dimensions X tall and Y wide, how can a longer connecting rod/taller deck height save space and allow more room for port runners, airbox etc?

[McGuire]

Quote

Originally posted by hydra
McGuire,
You're missing my point. I clearly stated that I wanted to completely disregard the effect of piston acceleration and the resulting forces on the rotating assembly and focus purely on the tribological/wear aspect of such a high piston speed... i.e. how much worse - from a longevity point of view - is a 5000fpm piston speed than a 4000fpm piston speed?

Acceleration is the real issue in piston and cylinder wall wear, not mean piston speed.

[shaun979]

Quote

Originally posted by McGuire


Sorry, but I am not following you at all. For an engine of dimensions X tall and Y wide, how can a longer connecting rod/taller deck height save space and allow more room for port runners, airbox etc?

Not a problem. I'm not sure what you mean by X tall and Y wide though.. the overall dimensions vary with design. To answer the question.. Imagine the block height at the extreme end of shortness. There is no space in the valley for proper runner approach angle, runner spacing, and they draw too closely from the airbox - they're all bunched together. One could change runner angle or make them curve to space them out but that changes flow characteristics and F1 engine designers have found that the best compromise is to lengthen the decks and run longer rods to make space.

Am I making sense? Can anyone explain this in English better? As a last resort I will draw it in a paint program and post the picture but I warn you that my drawing skills are terrible.

[desmo]

Gotcha Sean I asked someone to run some F1 simulations in WAVE on VEs with differing rod ratios and the present ratios don't appear to be chosen with that in as a priority. Rod ratios also significantly affect port velocities and dwell time near TDC, both of which are probably a consideration for F1 too.

[Engineguy]

At McGuire's two example rod lengths, I dont see any packaging problems for the ports, airbox, etc. with the common 90° Vee angle.

Actually, I don't think the rod can be less than about 3.2" due to piston pin boss to crank counterweight issues, based on the current F1 crank I've seen.

W.A.G. from pics>> Toyota 89mm (3.5")..... BMW 114mm (4.5")

[McGuire]

Quote

Originally posted by shaun979


Not a problem. I'm not sure what you mean by X tall and Y wide though.. the overall dimensions vary with design.

What I mean: note Engineguy's two drawings. It would be very difficult to force the longer rod/taller deck height into the smaller engine package without imposing serious compromises on the rest of the hardware, especially in the cylinder heads.

[McGuire]

Quote

Originally posted by McGuire


Acceleration is the real issue in piston and cylinder wall wear, not mean piston speed.

What I mean there: the forces fluttering the rings and vibrating the ring lands, side-loading the pistons, applying tangential loads on the crankpin etc are the result of accelerations, not the piston's mean speed.

Basically, piston speed is good. At midstroke there is enough speed and pressure to maintain hydrodynamic lubrication between the piston and cylinder wall. The trouble is the piston must come to a dead stop and change direction twce per crank rotation. (And oil has inertia too.) So the lubrication breaks down into thin-film as the piston slows down, then into very spotty boundary lubrication through TDC and BDC. Not surprsingly, that is where we find the majority of wear as well.

[shaun979]

Quote

Originally posted by Engineguy
At McGuire's two example rod lengths, I dont see any packaging problems for the ports, airbox, etc. with the common 90° Vee angle.

Actually, I don't think the rod can be less than about 3.2" due to piston pin boss to crank counterweight issues, based on the current F1 crank I've seen.

W.A.G. from pics>> Toyota 89mm (3.5")..... BMW 114mm (4.5")

Thanks for the picture engineguy.

McGuire's examples at 41mm stroke would mean a 2.61" rod vs a 4.84" rod. Packaging may not be a problem, but I am not sure about overall flow dynamics not being adversely affected at the short lengths. The picture that you posted shows the obvious changes in runner spacing, aixbox dimensions (assuming similar final volumes). I believe this would definitely change flow dynamics but since I cannot run the exact calculations or testing myself, I have no choice but to rely on my source who is connected.

[shaun979]

Quote

Originally posted by McGuire


What I mean: note Engineguy's two drawings. It would be very difficult to force the longer rod/taller deck height into the smaller engine package without imposing serious compromises on the rest of the hardware, especially in the cylinder heads.

ah I see now what you mean. I never said overall engine dimensions had to be fixed. Cylinder head efficiency can be maintained if they are not changed (or perhaps improved if they make them taller!). If they are not changed, the engine, like you say, would be larger. I believe the reason they accept this is because the unit makes more power in this slightly larger (X, Y dimension) form, and the overall car with that engine is determined to be the best obtainable result.

[shaun979]

Quote

Originally posted by McGuire
vibrating the ring lands

McGuire, is there any more you can tell us about vibrating the ring lands? That certainly is interesting!

BTW, I do agree with you that accelerations are the limit. I do also believe that piston speed ties loosely to overall loads faced since big engines that have the longer strokes have bigger components with more mass, and the larger a component (assuming similar materials), the harder it is to maintain integrity at a given acceleration. Beefing things up makes loads on the other components grow rapidly. I believe the mentioned loose relation is the reason mean piston speeds usually end up around the 25m/s mark for medium life engines.

[hydra]

McGuire,
As usual, you make an excellent point... To tell you the truth I was thinking of the effects of P*V on wear, where P is a function of side force (which in turn is a function of L/R and max cylinder pressure) and contact area, and V being piston speed...

[McGuire]

Quote

Originally posted by shaun979
McGuire, is there any more you can tell us about vibrating the ring lands? That certainly is interesting!

Not often regarded but nonetheless true...if the rings can flutter they can vibrate the ring lands. One of the things they can do to prolong engine life is to back off on the ring package to a more conservative combination... when Cosworth converted its competitive CART engine into a long-life spec package, it went from a trendy two-ring setup to a traditional three-piece kit, which helped take the rebuild cycle out to something like 1800 miles I believe. I have to presume similar things are happening in F1, producing marginal decrease in performance vs. significant increase in reliability and service life.

[shaun979]

Thanks McGuire. I see what you mean, but just in case anyone potentially misinterprets what was mentioned about the Cosworth CART engine longevity, the factor that by far made the most difference in extending life was the reduction of operating engine speed by 25% - from 16,000 RPM down to 12,000 RPM.

[McGuire]

Quote

Originally posted by shaun979
Thanks McGuire. I see what you mean, but just in case anyone potentially misinterprets what was mentioned about the Cosworth CART engine longevity, the factor that by far made the most difference in extending life was the reduction of operating engine speed by 25% - from 16,000 RPM down to 12,000 RPM.

Quite right. I would never suggest the extended service cycle was due solely to the more conservative ring package. Another interesting thing about the CCWS Cosworth spec engine...despite a reduction in power from around 810 bhp (the public figure for the spec package is +/- 750 bhp) the cars actually became slightly quicker on most of the road and street courses they run, due to increased torque and driveability. Of course at the bigger courses where the champ cars can really stretch their legs -- Road America and Cleveland, namely -- the hp loss was clearly evident.

[Dynojet]

Mcguire

It is a expectable effect a reprocating part that does have high mean velocity, to have high accelerations due de the reversions!!

That's why you can talk about engine wear thinking in terms of piston mean speed. And both stress from de inertial loads as the frictional wear is important.

If you think that speed does not affects the wear, thnks abou you sanding a wood piece. The more quickly you move the sandpaper, the quicker you 'll have you job done.

[McGuire]

Quote

Originally posted by Dynojet
Mcguire

It is a expectable effect a reprocating part that does have high mean velocity, to have high accelerations due de the reversions!!

That's why you can talk about engine wear thinking in terms of piston mean speed. And both stress from de inertial loads as the frictional wear is important.

If you think that speed does not affects the wear, thnks abou you sanding a wood piece. The more quickly you move the sandpaper, the quicker you 'll have you job done.

Wouldn't dispute anything you say...the sandpaper analogy is especially good. Just pointing out there is more to it in this case.

For a component rotating at a uniform speed, say a crankshaft at the main bearings, is there any critical point or speed barrier in the range from 5,000 to 20,000 rpm with respect to wear, or stress & fatigue? Not really...we can manipulate the bearing diameters/speeds, apply harder surfaces to the journals and bearings, suppply more oil pressure and volume etc.

Same for the piston...the piston gets optimal lubrication at mid-stroke, when its speed is the greatest. Through TDC the piston is at minimal speed (in fact it stops) but is nearly impossible to lubricate properly at that point. To me the limitation of the mean piston speed figure as an indicator is the piston must come to a dead stop and change direction hundreds of times per second. In that context the piston's average speed doesn't tell us very much.

[clSD139]

"Trinity Sirio .23" in "Radio Control Car Action" magazine (11/04):

"Experimenting with a variety of tuned pipes would certainly increase the peak horsepower figure and peak torque up in the rpm range. But that will probably result in loss of performance just for the sake of a big horsepower number."

Bore x Stroke 17 x 17mm.
Power: 2HP @ 28.000 rpm.
Max rpm: 40.000 !!!!

At max.rpm. the engine will also be worn out faster.

The piston has 3 grooves, for lubricating. Other hi-rev engines have also lesser piston rings.