Ferrari Power Unit Hardware & Software

[izzy]

Izzy

again you are saying plastic deformation is the issue ....
what is the meaning of stiffness ?
the others are talking only about elastic deformation ?

steel or part-steel pistons exist for use at temperatures where the fatigue strength of al-alloy pistons is inadequate
you are saying this also applies to the cylinder ?
yield strength being the step beyond fatigue strength (and what we might call elastic strength)

lol plastic elastic baby, they are quite closely related and we hardly have any actual figures so it doesn't matter does it? and where does 'again' come from? The point is AM and how it opens up new ways of making an engine that can run hotter, under Mattia who is an engine engineering guru and knows that he can run steel hotter than he can run aluminium

[PlatinumZealot]

I dunno what you guys are talking about with plastic deformation... But i am talking about deformation due to clamping load and combustion and piston loads and engine as stressed memeber loads etc. These deformations make the cylinder less cylindrical and the bore less circular too and thus the piston rings will work less effectively to seal and also cylinder wear will be higher. The more true the cylinder is you can run less tension in the piston ring as well and reduce friction.

So in essence cylinder deformation must be minimised to increase reliability and efficiency.

If i were an engine manufacturer i would go for it even it means gaining a few grams here and there.

[gruntguru]

When thermal effects are considered though, Young's modulus decreases much faster with temperature for aluminium . . .

Yes but not in the normal temperature range for a cylinder. Having said that I am sure the F1 bore runs at a higher temperature than normal so presumably the temperature gradient from bore to coolant is also higher than normal. Whether this is achieved with insulating surface treatment, steel liners or some other means is an interesting topic.

if it's being AM'd, then it won't be a single 'wall thickness' will it, it'll be a super thin skin as the inside face of the cylinder, then some kind of lattice or foam type structure on the back to give it stiffness that'll be much lighter and they can flow coolant through, to give the right temperature gradient as you say even if it's steel, as thermal conductivity is generally linear with thickness

then the flame temperature will be about 2000 degrees or something? They want to run the combustion chamber hotter as that gives a bit more expansion of the charge and reduces heat rejection. Their limits are the oil, i guess, and fuel and avoiding pre-ignition, but aluminium's stiffness falls off a cliff with temperature:
https://www.researchgate.net/profile/Ol ... 219-T6.png

Yield strength is not the issue so your graph is not relevant. (The cylinders will always operate well below the yield limit.)

Stiffness is determined by Young's Modulus so a chart like this is what you need. (no cliff)

[izzy]

Yield strength is not the issue so your graph is not relevant. (The cylinders will always operate well below the yeild limit.)

Stiffness is determined by Young's Modulus so a chart like this is what you need. (no cliff)
https://www.engineeringtoolbox.com/docs ... ticity.png

what do you mean there's no cliff? The aluminium trace just stops at 400! And that's Fahrenheit which is only 200 Centigrade. It's not EVEN a cliff

yes okay Young's Modulus is a better measure as long as the load is in the elastic range, tho i don't think we can assume a new Ferrari cylinder definitely will be as the idea is to run it hotter and get more work out of it. But in any case we can see exactly why they might be going for AM'd steel can't we, as at that 200 degrees it's 3x as stiff and then keeps going

[63l8qrrfy6]

I agree there is little difference when loaded only with cylinder pressure and/or tension/compression along the cylinder axis. Loading of the cylinder is of course much more complex than that and all the other loading vectors (piston side-loads, localised thermal stress) will favour the aluminium cylinder due to the greater wall thickness. Further, the higher thermal conductivity results in lower temperature gradients and lower thermal stress.

Well if we say the thermal resistance of aluminium is 4 times less than steel and the steel is 3 times thinner then the temperatures can't be that different. Of course the liner loading is much more complex but hoop stress is the most representative case I can do with pen and paper

When thermal effects are considered though, Young's modulus decreases much faster with temperature for aluminium . . .

Yes but not in the normal temperature range for a cylinder. Having said that I am sure the F1 bore runs at a higher temperature than normal so presumably the temperature gradient from bore to coolant is also higher than normal. Whether this is achieved with insulating surface treatment, steel liners or some other means is an interesting topic.

Actually for high Cu (200 series) or high Si (300 series) typically used in such applications there is a pronounced drop in young's modulus after 200°C which is well within the operating range of a liner. The graph you are showing is for low alloy/ Mn alloy which is perhaps not representative.

An example of a high Si alloy:


An example of a high Cu alloy is on page 64 of this paper:
https://ntrs.nasa.gov/archive/nasa/casi ... 004069.pdf
.

. . . . while its coefficient of thermal expansion is roughly twice that of steel. At the usual operating temperatures there is just no way an aluminium bore could weigh as much as a steel one for the same stiffness.

Isn't that what I said?

You know what I meant

[gruntguru]

Yield strength is not the issue so your graph is not relevant. (The cylinders will always operate well below the yeild limit.)

Stiffness is determined by Young's Modulus so a chart like this is what you need. (no cliff)
https://www.engineeringtoolbox.com/docs ... ticity.png

what do you mean there's no cliff? The aluminium trace just stops at 400! And that's Fahrenheit which is only 200 Centigrade. It's not EVEN a cliff

"Only 200* C"?
I acknowledge we are talking about these engines running higher cylinder temperatures than normal but lets think about how hot.

The coolant side of the cylinder casting will be close to coolant temperature - maximum perhaps 150*C?

The combustion side of the cylinder casting carries an oil film - maximum perhaps 200*C? Even if this is higher it is likely due to some kind of thermal barrier with a strong temperature gradient meaning the bulk of the cylinder material is at less than 200*C.

[izzy]

"Only 200* C"?
I acknowledge we are talking about these engines running higher cylinder temperatures than normal but lets think about how hot.

The coolant side of the cylinder casting will be close to coolant temperature - maximum perhaps 150*C?

The combustion side of the cylinder casting carries an oil film - maximum perhaps 200*C? Even if this is higher it is likely due to some kind of thermal barrier with a strong temperature gradient meaning the bulk of the cylinder material is at less than 200*C.
https://i.imgur.com/jqCYaDx.jpg

oh great i was going to ask that exact question, about the wall temperatures. But as @Mudflap shows 200 is about the limit for aluminium alloy and anyway i reckon the ultimate limit for Ferrari will be the oil on one side, and the coolant on the other.

The boiling point of engine oil is about 300C, and they have Shell working on it, and if it's straight water for coolant then 6 bar of pressure gives about 160C , so that's a gradient of 140, which is completely doable in steel, which won't be a solid thickness anyway if it's AM'd, they'll basically be able to dial in the temperature drop they want and the stiffness too

[PlatinumZealot]

The oil temperature is not that hot. It is about 130 to 160 degrees coming from the engine. I think i got this from some telemetry screen in some spy photo or something.. But you can imagine the bulk temperature of the aluminum to be about that range.

For example.. At my power plant we had cylinder liner temperature sensors.. Granted these were large bore slow speed diesel engines.. But liner temperatures operating ranges were 130 to 165 degrees.. Cylinder pressures were about 130bar. Steel liners though. Two stroke. So it could very well be hotter than 4 strokes. Who knows. But 160 degrees is a safe guestimate I would say.

[gruntguru]

The oil temperature is not that hot. It is about 130 to 160 degrees coming from the engine. I think i got this from some telemetry screen in some spy photo or something..

The oil on the bore surface just prior to being scraped off would be the hottest oil anywhere in the engine - much hotter than the oil returning to the scavenge pumps.

[izzy]

The oil temperature is not that hot. It is about 130 to 160 degrees coming from the engine. I think i got this from some telemetry screen in some spy photo or something.. But you can imagine the bulk temperature of the aluminum to be about that range.

For example.. At my power plant we had cylinder liner temperature sensors.. Granted these were large bore slow speed diesel engines.. But liner temperatures operating ranges were 130 to 165 degrees.. Cylinder pressures were about 130bar. Steel liners though. Two stroke. So it could very well be hotter than 4 strokes. Who knows. But 160 degrees is a safe guestimate I would say.

oh, i wonder if they have cylinder liner temperature sensors in F1? But they have a whole department of Shell working away on it, and you couldn't have the sensor right on the surface could you, and as @Gruntguru says it'll be hotter on the surface of the cylinder than anywhere else, so my guess is they'll be working towards 300, as close as they can manage. I think there are a few dots, like with AM and the mystery coolant, that join up towards this range

[63l8qrrfy6]

oh great i was going to ask that exact question, about the wall temperatures. But as @Mudflap shows 200 is about the limit for aluminium alloy and anyway i reckon the ultimate limit for Ferrari will be the oil on one side, and the coolant on the other.

This might be a misunderstanding but I did not mean 200°C is the limit for aluminium, I only wanted to show that there is a large drop in stiffness at such temperatures.

Aluminium cylinder heads and pistons do very well at 300°C in many racing engines!

[izzy]

oh great i was going to ask that exact question, about the wall temperatures. But as @Mudflap shows 200 is about the limit for aluminium alloy and anyway i reckon the ultimate limit for Ferrari will be the oil on one side, and the coolant on the other.

This might be a misunderstanding but I did not mean 200°C is the limit for aluminium, I only wanted to show that there is a large drop in stiffness at such temperatures.

Aluminium cylinder heads and pistons do very well at 300°C in many racing engines!

Oh okay sorry i took it as the same thing. There seems to be a lot of research going on, like with iron as an alloy. And they do use yield strength as the measure quite often, i suppose it's easy to measure. Anyway as an amateur it'd make me nervous using a material half way to being an actual liquid!! But let's hope we find out eventually what awesome things they're doing

[PlatinumZealot]

The oil temperature is not that hot. It is about 130 to 160 degrees coming from the engine. I think i got this from some telemetry screen in some spy photo or something.. But you can imagine the bulk temperature of the aluminum to be about that range.

For example.. At my power plant we had cylinder liner temperature sensors.. Granted these were large bore slow speed diesel engines.. But liner temperatures operating ranges were 130 to 165 degrees.. Cylinder pressures were about 130bar. Steel liners though. Two stroke. So it could very well be hotter than 4 strokes. Who knows. But 160 degrees is a safe guestimate I would say.

oh, i wonder if they have cylinder liner temperature sensors in F1? But they have a whole department of Shell working away on it, and you couldn't have the sensor right on the surface could you, and as @Gruntguru says it'll be hotter on the surface of the cylinder than anywhere else, so my guess is they'll be working towards 300, as close as they can manage. I think there are a few dots, like with AM and the mystery coolant, that join up towards this range

300C No way. The new f1 engines are highly efficient Lean burn.

Here is some information on the V10 cooling systems. Its hard to extrapolate this into the new engines but Imagine heat flow should be similar... ( slightly less heat but much smaller thermal mass? Same water pump flow?)

Water temperarures were increased up to130 C
To reduce radiator sizwe and unnecessary cooling. Jacket geometry and pump flows improved to increase cooling where it mattered. See graphs of temperature at different engine locations.

https://www.google.com/url?sa=t&source= ... HFmlnT1vWR

[izzy]

300C No way. The new f1 engines are highly efficient Lean burn.

Here is some information on the V10 cooling systems. Its hard to extrapolate this into the new engines but Imagine heat flow should be similar... ( slightly less heat but much smaller thermal mass? Same water pump flow?)

Water temperarures were increased up to130 C
To reduce radiator sizwe and unnecessary cooling. Jacket geometry and pump flows improved to increase cooling where it mattered. See graphs of temperature at different engine locations.

https://www.google.com/url?sa=t&source= ... HFmlnT1vWR

300 is the possible bore surface temperature from @mudflap, also the boiling point of oil, and then i was taking coolant temperature as pure water boiling point at 6 bar pressure being 160, so a gradient across the bore structure of 140. that's a great paper you linked, we can see lots of scope for using AM to enhance the cooling, with more smaller ways and perhaps no empty galleries any more but all foam or lattice type structures, much more compact

[PlatinumZealot]

300C No way. The new f1 engines are highly efficient Lean burn.

Here is some information on the V10 cooling systems. Its hard to extrapolate this into the new engines but Imagine heat flow should be similar... ( slightly less heat but much smaller thermal mass? Same water pump flow?)

Water temperarures were increased up to130 C
To reduce radiator sizwe and unnecessary cooling. Jacket geometry and pump flows improved to increase cooling where it mattered. See graphs of temperature at different engine locations.

https://www.google.com/url?sa=t&source= ... HFmlnT1vWR

300 is the possible bore surface temperature from @mudflap, also the boiling point of oil, and then i was taking coolant temperature as pure water boiling point at 6 bar pressure being 160, so a gradient across the bore structure of 140. that's a great paper you linked, we can see lots of scope for using AM to enhance the cooling, with more smaller ways and perhaps no empty galleries any more but all foam or lattice type structures, much more compact

Road car manufacturers have been using AM to make sand cores for casting. Likley in f1 too.

Not satisfied with that method?