Ferrari Power Unit Hardware & Software

[saviour stivala]

Of course it can, it's PR fluff written by someone maintaining a public facing website, not by a hundred engineers and ratified by the FIA. The exact same site in a different place says a different thing ffs

How can anybody imagine and go as far as to say that a team the caliber of Mercedes publishes its F1 car specification it will just be PR FLUFF written by someone maintaining a public facing website?. just because one of the numbers rubbishes your argumentation?.

[henry]

I don’t know if this has been posted before, but it’s a great read on the strategies that could be employed on the PU’s. It was written just prior to the 2014 season. Much of its contents are very much applicable today.

https://core.ac.uk/download/pdf/82194974.pdf

Thanks for this link.

This is the first paper I’ve seen that deals numerically with the issue of anti-lag. I can’t pretend to fully understand it but 2 things jump out

1. It refers to part of the MGU-K output under braking going to maintain the Turbine/compressor speed. I’ve not seen that mentioned before, although it seems obvious.

2. It gives a value for Turbine, Compressor,MGU-H rotational inertia at 5.62E-4 kgm2. Over in the Honda PU thread I used a value of 6.42 to get to an estimate of H shaft speed fluctuations for Honda’s extra harvest mode. I now feel more confident in those calculations.

[saviour stivala]

They don't eliminate back pressure, and they don't stop the turbine generating power from either flow or pulse tuning, they reduce back pressure.

When the waste gates are open and the turbocharger is being operated in electric supercharging mode the exhaust gases are bypassing the turbine scroll so there is no back pressure in the exhaust scroll, when the exhaust gases are bypassing the turbine scroll, how can the turbine generate power.

[saviour stivala]

So what’s going on with the 2015 Honda? It vents through the housing. No external pipes to the tailpipe. Did Honda not get the message?

according to the rules waste gate/s need have it's own dedicated exhaust pipe which means it is against the rules to vent/bypass waste gates exhaust gasses through the scroll housing.

Well that’s not how it was before 2016. Honda most definitely vented internally through the scroll housing. Renault, on the other hand vented from the top of the scroll through a pipe that joined the main pipe. (It had to do this because the regulations then required all exhaust gases to go through a right circular cylinder ar the rear of the car)

I’d show you a picture of the Renault but it’s a little fuzzy and even the clear Honda images seem open to (mis)interpretation.

The rules only constrain the tail portion of the wastegate pipes, the last 150mm. They say nothing at all about where they originate other than once they’ve passed the wastegate they must exit via the wastegate exhausts.


How are you getting on with that definition of “free load mode” you cited?

5.8.2: engine exhaust system must have only one single turbine tailpipe exite and one or two waste gates tailpipe exiteswhich must all be rearwardsand through which akk exhaust gases pass. All and only turbine exite exhaust gases must pass through the turbine tailpipe and all and only the waste gate exhaust gases must pass through the waste gate tailpipe (s).none of the tailpipes may be contained within any of the other tailpipes.

[saviour stivala]

There are plenty of ways, and no, it is completely, physically impossible for them to eliminate back pressure, if there's flow, there's a pressure gradient. Physics 101 for 6 year olds.
I don't know if you've noticed, but this is a technical site, so twisting rules, making rules up and blatent rubbish like that above isn't particularly relevent as contributions.

When exhaust gases are bypassed from scroll to waste gate tailpipe there is no exhaust gases in the scroll so there is no back pressure in the scroll.

[PhillipM]

When exhaust gases are bypassed from scroll to waste gate tailpipe there is no exhaust gases in the scroll

Yes, there is. The wasgates don't shut off the flow path through the turbine, that would generate even higher backpressure

[subcritical71]

1. It refers to part of the MGU-K output under braking going to maintain the Turbine/compressor speed. I’ve not seen that mentioned before, although it seems obvious.

2. It gives a value for Turbine, Compressor,MGU-H rotational inertia at 5.62E-4 kgm2. Over in the Honda PU thread I used a value of 6.42 to get to an estimate of H shaft speed fluctuations for Honda’s extra harvest mode. I now feel more confident in those calculations.

I would agree with you on point 1) seems obvious. Confirms for me that they are able to harvest more than 2MJ/lap and need someplace else to send it. And there has got to be fewer losses going from K > H vs K > ES > H.

point 2) - its always good when you can get validation, especially in this sport where everything is such a highly guarded secret. I was surprised how low the value was!

Is it known or suspected what the output of the H can produce given the limitations of the turbine size?

[gruntguru]

40hp gain is something huge in terms of engine development, especially in this stage of the development cycle! Is it possible though that this is not only engine related and Ferrari found a new clever way to stall aerodynamic parts to gain this speed increase?

Exactly what I was thinking, maybe we are looking at the wrong area when we should be looking at the bargeboards.... Just a guess and ot from the engine thread

The gain is not aerodynamic (drag reduction) in nature. A reduction in drag coefficient would look like an power gain at all speeds and proportional to the cube of velocity. ie there would not be sudden change at 150 k/hr.

[gruntguru]

With the waste gates situated before the turbine scroll any/if any, exhaust exhaust gasses enters the scroll the gasses energy inside the scrol is as good as dead.

Wright made 14000 aircraft engines that gained 6 to 18% exhaust power recovery without reduction of crankshaft power
6% more power at low power sea level cruise and 18% more power at takeoff
this (slightly) changed the world

it is blowdown recovery
there is energy turbine-recoverable from the exhaust pulses in an exhaust stream at ambient pressure

Trying to compare the 18 cylinder double row cyclone exhaust turbo compounding to justify an argument that “any/if any” exhaust gasses finds their way into the turbine scroll from a fully open waste gate is not “as good as dead” energy wise, is worse than trying to compare apples to oranges. Because in the cyclone turbo exhaust compounding each of the three power recovery turbines was driven by the direct exhaust gasses from three front row and three rear row cylinders directly and not from the “as good as dead” energy wise exhaust gasses from a fully open waste gate.

The point you are missing is that it is possible to harness significant energy from the exhaust blowdown pulse (the pulse of high pressure high velocity exhaust gas that exits the exhaust valve while the piston is still descending on the power stroke) while maintaining a much lower pressure in the cylinder during the exhaust stroke. This has been discussed here many times and the consensus is that it is possible to do this during electric supercharger mode and the gains would be substantial.

If you think of the exhaust header tubes pointing at the turbine inlet you can imagine that some of the high velocity, high pressure pule energy will still make its way to the turbine even with the wastegate open and educing the average pressure in the exhaust to atmospheric.

[gruntguru]

according to the rules waste gate/s need have it's own dedicated exhaust pipe which means it is against the rules to vent/bypass waste gates exhaust gasses through the scroll housing.

Erm, if you close the wastegate thats exactly what happens!!! All the gas goes through the turbine. Is it now illegal to close the wastegate?

[henry]

1. It refers to part of the MGU-K output under braking going to maintain the Turbine/compressor speed. I’ve not seen that mentioned before, although it seems obvious.

2. It gives a value for Turbine, Compressor,MGU-H rotational inertia at 5.62E-4 kgm2. Over in the Honda PU thread I used a value of 6.42 to get to an estimate of H shaft speed fluctuations for Honda’s extra harvest mode. I now feel more confident in those calculations.

I would agree with you on point 1) seems obvious. Confirms for me that they are able to harvest more than 2MJ/lap and need someplace else to send it. And there has got to be fewer losses going from K > H vs K > ES > H.

point 2) - its always good when you can get validation, especially in this sport where everything is such a highly guarded secret. I was surprised how low the value was!

Is it known or suspected what the output of the H can produce given the limitations of the turbine size?

The only published information I have seen on H output comes from a couple of sources.

The Limebeer paper on optimal control (2013) uses a figure of 40kW. It also suggests the H power to drive the Compressor is 60kW. Like other values in the paper I think these are conservative. E.g. it also uses an ICE power of 440kW.

race info that Magnetti Marelli used to publish. I think they stopped at the end of 2015. From memory I think they used to give a rough energy balance round a lap. From that it could be deduced that the H generated 45 to 50kW. At the time I thought MM only supplied Ferrari, but it might have been Renault as well.

MM also included some K energy from braking numbers. Taken in conjunction with race card info Brembo published around the same time this suggested a recovery of around 50% of the potential (taken as total stop time x 120kW)

FWIW I use 60kW as H output when I do my own energy balance calculations. Nice easy number and a reasonable extrapolation from MM’s numbers.

It has been mooted that the H output might be somewhat higher. If it is we might quickly get to the situation where the only way they can deploy 4MJ to the K on a race lap is to use some electric supercharger.

[turbof1]

So let's get some facts, yes facts, straight:

I qouted the AMG PETRONS MOTORSPORT PU106C specification which says 4MJ per lap (33.33 seconds full power).

This is factually wrong. 4MJ does indeed provide 33.33 seconds of 120kw on the driveshaft, but it's not limited to 4MJ. For instance, energy harvested by the mgu-h can immediately be deployed through the mgu-k on driveshaft as well. Whoever wrote that on Mercedes' website definitely was only schooled in pr. Also, what Mercedes writes on their website is highly irrelevant: they are not F1's authority and the only reliable source of regulations is the pdf document provided on the fia's website. Mercedes can write on their website the power unit is a fusion reactor for all they care.

There is also a whole bunch of misunderstandings going on about the mgu-k limitations. I think it speaks for itself when 4-5 people are basically all telling where these limitations don't apply, there is more to it then just a 4MJ limit. My advice is to read the flow chart very well and read what is stated on it. Apply the limits on the part it is stated on and on absolutely nothing else. These limits are highly specific!

There are plenty of ways, and no, it is completely, physically impossible for them to eliminate back pressure, if there's flow, there's a pressure gradient.

Unless we find a way to break down physics laws, than this is true. This is a physics basic principle even of "action and reaction".

[xtremeclock]

Mark Hughes: https://www.motorsportmagazine.com/repo ... rix-report

Ferrari has been doing great work on its power unit, giving it a clear advantage. There are suggestions that its twin battery arrangement might have allowed a recent breakthrough in allowing it to simultaneously deploy and harvest. So, for example, one battery could be deploying to the MGUk while the MGUh is harvesting for the other battery.

The recent gains, whatever they are, have been spectacular and apply just as much to the spec 1 engine still in Räikkönen’s car as to the spec 2 units in everyone else’s.

[saviour stivala]

The 2014 F1 turbo hybrid power unit formula has totally changed the design, way of use and control of the turbocharger as used in F1. The previous use and control was solely of boosting by exhaust gases and control by waste gate and pop-up valve. The present use is boost by exhaust gases as well as by electrical means (MGU-H), while control is by means of electrical MGU-H. Because there is always the risk of something going wrong with the MGU-H controlling the turbo boosting as well as slowing the turbo down, plus the possibility of the engine going from max power speed to minimum power speed in milliseconds with the sudden change in the intake consumption which means one is better off with some turbo RPM very quickly, more quickly than the turbo controller (MGU-H) can wind it down. And also because controlling of max impellor speed with MGU-H is risky, like applying a brake to the turbine instead of letting it free wheel, which is the more efficient and reliable, both a waste gate as well as a pop-up valve are employed as a safety measure. With the added benefit of being able by means of electrically spinning of the turbo instead of by means of the exhaust gases, of making use of the fully open waste gates bypassing exhaust gases from the turbine scroll. For a free load mode of operation.

[saviour stivala]

1. It refers to part of the MGU-K output under braking going to maintain the Turbine/compressor speed. I’ve not seen that mentioned before, although it seems obvious.

2. It gives a value for Turbine, Compressor,MGU-H rotational inertia at 5.62E-4 kgm2. Over in the Honda PU thread I used a value of 6.42 to get to an estimate of H shaft speed fluctuations for Honda’s extra harvest mode. I now feel more confident in those calculations.

I would agree with you on point 1) seems obvious. Confirms for me that they are able to harvest more than 2MJ/lap and need someplace else to send it. And there has got to be fewer losses going from K > H vs K > ES > H.

point 2) - its always good when you can get validation, especially in this sport where everything is such a highly guarded secret. I was surprised how low the value was!

Is it known or suspected what the output of the H can produce given the limitations of the turbine size?

The only published information I have seen on H output comes from a couple of sources.

The Limebeer paper on optimal control (2013) uses a figure of 40kW. It also suggests the H power to drive the Compressor is 60kW. Like other values in the paper I think these are conservative. E.g. it also uses an ICE power of 440kW.

race info that Magnetti Marelli used to publish. I think they stopped at the end of 2015. From memory I think they used to give a rough energy balance round a lap. From that it could be deduced that the H generated 45 to 50kW. At the time I thought MM only supplied Ferrari, but it might have been Renault as well.

MM also included some K energy from braking numbers. Taken in conjunction with race card info Brembo published around the same time this suggested a recovery of around 50% of the potential (taken as total stop time x 120kW)

FWIW I use 60kW as H output when I do my own energy balance calculations. Nice easy number and a reasonable extrapolation from MM’s numbers.

It has been mooted that the H output might be somewhat higher. If it is we might quickly get to the situation where the only way they can deploy 4MJ to the K on a race lap is to use some electric supercharger.

"the use of 4MJ on a race lap" whe in electric supercharging mode both "K" and "H" are sharing "ES" power, which means they are both being used as motor mode.