Ferrari engine failures caused by pneumatic leaks

[xpensive]

To me the most likely explanation is they are running it so lean for economy purposes, last year they could use a lot more fuel to cool that engine, plus two cool charges per race should have helped some. I've seen Alonso looking for clean air in the middle of a straight too many times in these few races, especially Bahrain.

Ferrari secretly trying to optimize the 32 reciprocating pneumatic seals at 30 Hz and perhaps went a step too far for the longer run?

What's this thing about the frequency of the seals? i don't know much about pneumatic springing.

At 18 000 Rpm, the 32 valves are moving their reciprocating pneumatic seals back and forth 30 times per second.

[WhiteBlue]

The most simple explanation is found in the refueling ban. In 2009 Ferrari would have had no problem to recharge their pneumatic pressure at pit stops. They also might have been running the engines cooler. This year the pit stops are too short to charge the air tank with a bottle, so they run out of air. There may be more factors aggravating the situation. The most likely change is a material change to the seals to make them work better and longer at elevated temperatures.

Btw, this is already reported in the source in post #1 of this thread. viewtopic.php?p=164412#p164412

[alelanza]

To me the most likely explanation is they are running it so lean for economy purposes, last year they could use a lot more fuel to cool that engine, plus two cool charges per race should have helped some. I've seen Alonso looking for clean air in the middle of a straight too many times in these few races, especially Bahrain.

Ferrari secretly trying to optimize the 32 reciprocating pneumatic seals at 30 Hz and perhaps went a step too far for the longer run?

What's this thing about the frequency of the seals? i don't know much about pneumatic springing.

At 18 000 Rpm, the 32 valves are moving their reciprocating pneumatic seals back and forth 30 times per second.

hmmmm... my math gives me 150 hz @ 18k rpm for valve movement, what am I doing wrong?
And why do you say reciprocating seal? as i understand it, just like in a regular engine the valve seal does not move, only the valve and in this case the pneumatic spring actually move.

The most simple explanation is found in the refueling ban. In 2009 Ferrari would have had no problem to recharge their pneumatic pressure at pit stops. They also might have been running the engines cooler. This year the pit stops are too short to charge the air tank with a bottle, so they run out of air. There may be more factors aggravating the situation. The most likely change is a material change to the seals to make them work better and longer at elevated temperatures.

Btw, this is already reported in the source in post #1 of this thread. viewtopic.php?p=164412#p164412

Yeah saw that, but how did they carry out the refills last year? i never saw any evidence of that.

[xpensive]

@at aleanza
150 Hz, that would mean 9000 strokes per minute, which makes you entirely correct and me a mathematical moron.

Unless I have completely misunderstood these systems, there is a continous volume change in each pneumatic cylinder, but as they are all connected, the total volume for the 32 valves does not change over the cycle. Please enlighten me where and if I got things wrong?

[alelanza]

@at aleanza
150 Hz, that would mean 9000 strokes per minute, which makes you entirely correct and me a mathematical moron.

LOL, not at all, just a small mathematical hiccup

Unless I have completely misunderstood these systems, there is a continous volume change in each pneumatic cylinder, but as they are all connected, the total volume for the 32 valves does not change over the cycle. Please enlighten me where and if I got things wrong?

I think I get what you mean. Initially i thought you were referring to the valve seal, you know the one that goes on top of the valve guide, but i see you're referring to the sealing of pneumatic spring mechanism itself. And it wouldn't surprise to find out the valve seal is a continuous part of the pneumatic spring outer casing, so nothing wrong really

[xpensive]

@ aleanza;
It would seem reasonable that each individual pneumatic cylinder has some sort of axial compression in order to bring the valve back into sealing position, why there per definition has to be a reciprocating gas-seal somewhere? If that action is performed at 150 Hz for hours between rebuilds (how many races now?), no ordinary o-ring would do I'm afraid. The beauty of the whole system is the interconnection between each pneumatic cylinder trough a common cannister, guaranteeing a constant pressure and no hysteresis.

If you at the same time are trying to minimize the not so negligeable friction of the entire system, well there's a challenge.

[WhiteBlue]

The most simple explanation is found in the refueling ban. In 2009 Ferrari would have had no problem to recharge their pneumatic pressure at pit stops. They also might have been running the engines cooler. This year the pit stops are too short to charge the air tank with a bottle, so they run out of air. There may be more factors aggravating the situation. The most likely change is a material change to the seals to make them work better and longer at elevated temperatures.

Btw, this is already reported in the source in post #1 of this thread. viewtopic.php?p=164412#p164412

Yeah saw that, but how did they carry out the refills last year? i never saw any evidence of that.

They had twenty people on the pit stops last year, so you may not have seen the guy with the bottle. It would probably have looked like a small fire extinguisher and people will probably not have focussed on him. Do some work on your digital records of races and a search on overhead pictures of pit stops and you will find him.

Edit: Found a pic with the bottle in plain view.



You even see that they have no time to recharge while the tyres are changed. That is done later while the refueling continues.

[xpensive]

Thanks WB, perhaps this supports my theory in a way, when it does not seem to be a problem for the other Ferrari-engined cars,
an over-optimized pneumatic-spring sealing system, constanly leaking?

[riff_raff]

Pneumatic valve spring systems are basically a reciprocating piston and thus have a compression ratio. The initial pressure is determined by the feed pressure, and the spring chamber has a check valve controlling its flow. The piston in the spring system has a circumferential seal. These seals must be capable of operating under very high accelerations and must have very low friction. If the sliding frictions are too high the seal material can easily fail due to overheating. The surface finish, fits and clearances of these pneumatic valve spring parts are are very critical. Even a small seal leak would affect the rate of the valve spring. The likely result being valve float and seat bounce.

[autogyro]

Pneumatic valve spring systems are basically a reciprocating piston and thus have a compression ratio. The initial pressure is determined by the feed pressure, and the spring chamber has a check valve controlling its flow. The piston in the spring system has a circumferential seal. These seals must be capable of operating under very high accelerations and must have very low friction. If the sliding frictions are too high the seal material can easily fail due to overheating. The surface finish, fits and clearances of these pneumatic valve spring parts are are very critical. Even a small seal leak would affect the rate of the valve spring. The likely result being valve float and seat bounce.

So presumably the seals are designed to leak or there would be no need to repressure the system at stops. This would be to assist with valve cooling using Nitrogen. If this were not the case, it would be possible to design a 'balloon' pneumatic (Nitrogen) spring for each valve that could be fixed to the valve stem at one end and subject to no sliding or leaking. Or perhaps this has yet to be tried? I still do not understand why the Ferrari problem has not appeared in other teams.

[xpensive]

The reason why you want a common valve pneumatic "spring" system, it's not realy a spring as you have a constant pressure, is the hysteresis when compressing and de-compressing. If you had each individual chambers, you would quickly heat up the gas to unacceptable levels.

[autogyro]

The reason why you want a common valve pneumatic "spring" system, it's not realy a spring as you have a constant pressure, is the hysteresis when compressing and de-compressing. If you had each individual chambers, you would quickly heat up the gas to unacceptable levels.

It would be possible to design a balloon valve spring with no sliding seals and also connect it to the other valves as a common system.
It would also be possible to fit a leak off valve to give a flow of gas and also control it far more efficiently for cooling.

[gonek]

The reason why you want a common valve pneumatic "spring" system, it's not realy a spring as you have a constant pressure, is the hysteresis when compressing and de-compressing. If you had each individual chambers, you would quickly heat up the gas to unacceptable levels.

It would be possible to design a balloon valve spring with no sliding seals and also connect it to the other valves as a common system.
It would also be possible to fit a leak off valve to give a flow of gas and also control it far more efficiently for cooling.

There are several solution regarding your idea. One you can see here: http://www.freepatentsonline.com/5586529.pdf You can try to find another patents. I have seen the patent with pneumatic spring which looks exactly like a half of balloon but I can't find it now.

Regarding pneumatic sealing. Ferrari has used in their V10 engines static sealing. It can be seen in Peter Wright's book. Maybe they still use the same arrangement in V8 and maybe this is the reason of the gas leakage problem?

[autogyro]

Thanks gonek.
I have had sight of a number of such patents before.
Do you know if there is an 'off the shelf' balloon valve spring product, that could be used as a replacement on a fairly conventional engine?
I have been trying to gain an extra 500rpm for a race engine I was partly responsible for, the design of which has won a number of championships.
Because of heavy valves and springs, we have come up against an rpm limit.
Valve bounce at the end of some straits that limits our gearing choices.

On the Ferrari issue, I still cannot see how they have the problem and other users of the engine do not.

[xpensive]

Thanks gonek.
On the Ferrari issue, I still cannot see how they have the problem and other users of the engine do not.

Where is it written they have the same engine?