Ferrari Power Unit Hardware & Software

[gruntguru]

I think 400 bar peak cylinder pressure is a bit unrealistic. Maybe with moderate detonation.

400 is very high but it does agree with a statement from Mercedes last year (elephants on pistons) which suggested 400+ bar.

The v8 had 13-14 CR and that was a geometrical limitation. I don t see how you can get 16 without reducing valve lift considerably or designing a piston with massive recesses.

B/S ratio is currently ~1.5 compared to ~2.5 for the V8s.

Apart from that, engines making peak power at 11k rpm don't need as much valve area and lift per unit displacement as those making peak power at 17 or 18k rpm.

& especially with 'turbo-boost' - being pumped in, @ several times ambient-N/A's best volumetric levels..

Ahh that's the thing about VE - if you calculate it based on volumetric flow of air at manifold conditions, the VE of a particular engine at a particular rpm does not vary with boost (assuming exhaust back pressure remains equal to MAP). So the "turbo-boost" is merely increasing the density of the air - volumetric flow remains the same.

[gruntguru]

I thought the idea was to maximise the heat in the combustion chamber to create power in the expansion stroke or in expansion through the turbine.

Dissipating heat through the piston to the cylinder walls, oil or the air in the crankcase seems like something to be avoided if possible.

I'm guessing to maximise the average temperature in the cylinder you need to minimise the peak temperatures, the hot spots. Perhaps a "honeycomb" structure could be used to micromanage the heat flows in the piston reducing the hot spots, increasing the average temperature and doing so with limited weight penalty.

An enclosed honeycomb structure will work as an insulator, so its no good at dissapating heat. Unless it would be an open structure which oil was sprayed onto, but structually i can't see how that would work.

If it was an open structure it would't need to be 3D printed. I would assume it is closer to the 3D equivalent of an "I" beam than a "T" yet with some kind of "open" structure (mesh?) appearance from below.

I think Henry is suggesting the top surface spreads the heat horizontally to avoid "hot spots" while the insulated structure below makes the average temperature of the crown higher - as permitted by the use of steel.

[Holm86]

Would it be a benefit to have a heat barrier coating on the piston crown? Or would that just cause too high temperatures in the combustion chamber?

[gruntguru]

I think 400 bar peak cylinder pressure is a bit unrealistic. Maybe with moderate detonation.

400 is very high but it does agree with a statement from Mercedes last year (elephants on pistons) which suggested 400+ bar.

The v8 had 13-14 CR and that was a geometrical limitation. I don t see how you can get 16 without reducing valve lift considerably or designing a piston with massive recesses.

B/S ratio is currently ~1.5 compared to ~2.5 for the V8s.

Apart from that, engines making peak power at 11k rpm don't need as much valve area and lift per unit displacement as those making peak power at 17 or 18k rpm.

I remember looking into the elephants claim and I think it gives something like 350 bar +/- 100 depending on the species and gender of elephant.

So did I.
5 ton elephant = 400 bar.
viewtopic.php?f=4&t=23688&p=623622&hili ... ts#p623622

Shouldn't the smaller B/S drive higher lift to achieve equivalent vale curtain area thus making higher compression ratios even harder to get?

No. Smaller B/S ratio means smaller valves and less lift. Lift beyond 25% of valve diameter has no benefit except to extend the duration and increase the area under the rest of the lift curve. Of course that also means longer duration "off the seat" which is the opposite of what you need when peak-power-rpm is reduced by 35-40%.

Regardless, a 66% increase in stroke (from 1.5 to 2.5) will increase CR from 13:1 to over 20:1, all else being equal.

[MrPotatoHead]

The idea of the Steel piston vs Aluminum is to have a stronger piston with better heat rejection characteristics.
However the piston would be heavier if it was solid and this would cause more problems.

The concept of using "honeycomb" structures in 3D Printing is becoming more and more common. They will use "Topology Optimization" to reduce the weight whilst keeping the strength at an acceptable level. This is where much of the focus is right now on the software side of 3D Printing - especially being driven by the Aerospace industry.

[PlatinumZealot]

But isn't that the reason for running steel pistons? They can cope with higher temperatures.

I think the lure of using steel pistons in race cars is mainly to be a lighter piston than aluminum.

[godlameroso]

The idea of the Steel piston vs Aluminum is to have a stronger piston with better heat rejection characteristics.
However the piston would be heavier if it was solid and this would cause more problems.

The concept of using "honeycomb" structures in 3D Printing is becoming more and more common. They will use "Topology Optimization" to reduce the weight whilst keeping the strength at an acceptable level. This is where much of the focus is right now on the software side of 3D Printing - especially being driven by the Aerospace industry.

https://youtu.be/vPv7PwS50OE?t=6m3s

Funny you mention that.

[gruntguru]

But isn't that the reason for running steel pistons? They can cope with higher temperatures.

I think the lure of using steel pistons in race cars is mainly to be a lighter piston than aluminum.

The benefit is operating temperature. Anything you can do to lighten a steel piston (honeycomb structures etc), you can do to an aluminium piston. The aluminium piston will always be lighter provided the temperature is low enough.

[63l8qrrfy6]

No. Smaller B/S ratio means smaller valves and less lift. Lift beyond 25% of valve diameter has no benefit except to extend the duration and increase the area under the rest of the lift curve. Of course that also means longer duration "off the seat" which is the opposite of what you need when peak-power-rpm is reduced by 35-40%.

Regardless, a 66% increase in stroke (from 1.5 to 2.5) will increase CR from 13:1 to over 20:1, all else being equal.

I agree that at L=D/4 the curtain area equals the valve area, however with valves having included angles I am pretty sure that there is an advantage to increase lift past 25%D as the valve moves away from the bore and reduces shrouding effects.

Now that I think a bit more about it, a larger diameter valve (eg for larger B/S) will stick out further for the same lift and same included angle, so maybe a CR of 16 is achievable.

Regarding CR, would it not be more efficient to keep it low in order to reduce work done by piston during compression and instead increase PR and change inlet timing to shift more work to the compressor?

[godlameroso]

You have to think about why the regulations limit CR to 18:1, you have to assume that the manufacturers are already somewhere in the ballpark of 14-16:1

By the same token you have to wonder why they chose 1,000 volts as the ERS limit, when they're operating at less than half that.

[63l8qrrfy6]

I'm guessing to maximise the average temperature in the cylinder you need to minimise the peak temperatures, the hot spots. Perhaps a "honeycomb" structure could be used to micromanage the heat flows in the piston reducing the hot spots, increasing the average temperature and doing so with limited weight penalty.

An enclosed honeycomb structure will work as an insulator, so its no good at dissapating heat. Unless it would be an open structure which oil was sprayed onto, but structually i can't see how that would work.

If it was an open structure it would't need to be 3D printed. I would assume it is closer to the 3D equivalent of an "I" beam than a "T" yet with some kind of "open" structure (mesh?) appearance from below.

I think Henry is suggesting the top surface spreads the heat horizontally to avoid "hot spots" while the insulated structure below makes the average temperature of the crown higher - as permitted by the use of steel.

This is what really baffles me about the 3D printing rumors.

It implies that the piston is either an enclosed honeycomb/ fancy oil gallery or has some sort of massive undercuts that just can't be machined after forging. The problem with these is that the 'dry' honeycomb is poor at conducting heat, while any complicated geometry that involves oil circulation is very inefficient.

A few years ago I looked at gallery cooled steel pistons and the fill ratio of the galleries was as low as 30% with squirt jets pointing straight at the inlets. Not only that, but oil circulation was very poor too, as at high engines speeds the oil was sticking to the walls and was not being displaced by fresh 'cool' oil.

I can hardly imagine anything better than an 'open' structure that is saturated by a dozen cooling nozzles.

Maybe 3D printing is just used for convenience, surely it must be a lot faster than having to wait on forging tooling for every single iteration ? Several suppliers offer powdered sintered tool steels with proof strength in excess of 2GPA and excellent cleanliness, can't see why similar quality can't be achieved with SLS/SLM.

[gruntguru]

Regarding CR, would it not be more efficient to keep it low in order to reduce work done by piston during compression and instead increase PR and change inlet timing to shift more work to the compressor?

Not sure of the typical isentropic efficiency for the in-cylinder compression process but I suspect it is better than the 80% we usually assume for the turbocompressor.

Of course the real benefit of higher CR is actually the higher expansion ratio that accompanies it. In-cylinder expansion is definitely preferable to turbine recovery.

[J.A.W.]

To be fair, there are a few 'fictional' process functions being assumed here.. such as..
..the effects of comp' ratio - dynamic/running versus static, & density/heat of charge air, on flow metrics..

[stevesingo]

An enclosed honeycomb structure will work as an insulator, so its no good at dissapating heat. Unless it would be an open structure which oil was sprayed onto, but structually i can't see how that would work.

If it was an open structure it would't need to be 3D printed. I would assume it is closer to the 3D equivalent of an "I" beam than a "T" yet with some kind of "open" structure (mesh?) appearance from below.

I think Henry is suggesting the top surface spreads the heat horizontally to avoid "hot spots" while the insulated structure below makes the average temperature of the crown higher - as permitted by the use of steel.

This is what really baffles me about the 3D printing rumors.

It implies that the piston is either an enclosed honeycomb/ fancy oil gallery or has some sort of massive undercuts that just can't be machined after forging. The problem with these is that the 'dry' honeycomb is poor at conducting heat, while any complicated geometry that involves oil circulation is very inefficient.

A few years ago I looked at gallery cooled steel pistons and the fill ratio of the galleries was as low as 30% with squirt jets pointing straight at the inlets. Not only that, but oil circulation was very poor too, as at high engines speeds the oil was sticking to the walls and was not being displaced by fresh 'cool' oil.

I can hardly imagine anything better than an 'open' structure that is saturated by a dozen cooling nozzles.

Maybe 3D printing is just used for convenience, surely it must be a lot faster than having to wait on forging tooling for every single iteration ? Several suppliers offer powdered sintered tool steels with proof strength in excess of 2GPA and excellent cleanliness, can't see why similar quality can't be achieved with SLS/SLM.

Would it not be correct to state that for best thermal efficiency in the combustion chamber, one would want to minimise heat losses to the surfaces the flame front comes in to contact with. In such a case, a cooler piston would be a greater absorber of CC heat than a hot piston.

The reason one would cool an aluminium piston is to maintain a bulk material temperature which is appropriate to the properties of the material and the loads applied.

A steel piston could be equally strong at a higher temperature as an aluminium piston at a lower temperature.

For sure, the steel piston would have more mass, but these engines only rev to 13500rpm as opposed to 20000rpm which has been possible in the past, therefore the engine designers are more than capable of managing much higher conrod loads than we currently see.

[PlatinumZealot]

But isn't that the reason for running steel pistons? They can cope with higher temperatures.

I think the lure of using steel pistons in race cars is mainly to be a lighter piston than aluminum.

The benefit is operating temperature. Anything you can do to lighten a steel piston (honeycomb structures etc), you can do to an aluminium piston. The aluminium piston will always be lighter provided the temperature is low enough.

I was going to post that but i thought about it and i came to the conclusion that if it were only high surface temperature you can get that with coatings on the aluminum. But the high temp strength and resilience of the steel is the real advantage at the end of the day (able to machine to smaller features).