Ferrari Power Unit Hardware & Software

[KAIZEN]

Small question, rumours went Ferrari had a sudden increase of ?40?hp some races ago, which I think started this discussion. Though it is a valid discussion how the energy system could potentially be used, it does not explain the sudden 40hp increase does it? I mean the mgu-k just isn't allowed to deliver more than 120kW, I assume all teams use this power, so the secret must be somewhere else?

The FERRARI horsepower gains are not electric horsepower gains but are produced by the ICE (combustion).

Very well possible. They definitely made huge steps with the electrical horsepower deployment across a longer span of time, but the increase qualifying horsepower can't just be explained away with an increase in electrical deployment. I personally think both mercedes and ferrari (those 2 at the very least) are already at the point they can deploy a constant 120kw electrical power across a single lap. Just make sure you have 4MJ on board when starting your hot lap, and spend it fully plus the additional harvested energy. You should easily be able to have enough energy for 120kw the whole lap round when on the throttle.

One possible explanation is Ferrari is having more than enough surplus of electrical energy to constantly keep the turbo spooled up electrically and can just open up the wastegate constantly, reducing back pressure and increasing ICE horsepower.

An other explanation could be a breakthrough regarding fuel.

Or a mixture of several options. The last few weeks I have seen truly intense and very eye opening discussions about what is actually possible with electrical energy harvesting and deployment. Although a piece of Ferrari's puzzle might manifest in the electrical side due an engineering breakthrough being able to suddenly to harvest a big chunk more energy, conceptually everything we discussed will probably have been known by the manufacturers even before 2014. If they have the technology to do it, they will do it.

So either:
-They made an engineering/technological leap that unlocked a concept we discussed here (but again should be known to Ferrari for a very long time already).
-Or they made the improvement somewhere else, and we are looking at something like fuel.

Exhaust sound by electric turbo.
https://youtu.be/T7qxqcuodio
I agree with your consideration.

[hollus]

Always better to deploy your energy at the lowest speed possible, because you deploy energy and kinetic energy varies with the square of the speed, so speed changes with the square root of the energy change.

Say you are going at 150 km/h and you deploy energy instantly to increase kinetic energy by 10%. This increases your speed by 4.88%, that is, by 7.3 km/h.
If you instantly deploy the same energy which going at 300 km/h where your kinetic energy is 4 times as large, now this only increases your kinetic energy by 2.5%. And this only increases your speed by 1.24% or 3.73 km/h.
The same energy that takes you from 150km/h to 157.3 km/h only takes you from 300km/h to 303.7 km/h.
The first one is a change of speed of 4.87%, meaning that it reduced the time to cover a fixed piece of track by 4.87%.
The second one changes your speed by 1.24% meaning that it reduced the time to cover a fixed piece of track by 1.24%.

So: you deploy at the lowest possible speed because:

1) The speed change is magnified at lower kinetic energy
2) The time change resulting from (1) is further magnified by changing speed at lower speeds.
3) (No numbers here) as mentioned above, higher drag wastes your energy deployed as higher speeds as well.

I have made a thought experiment with instant deployment, others have provided simulation numbers for more realistic scenarios above.


<Now in Mod mode: this discussion of when to deploy is not completely off topic here, IMO, es we are still discussing the implications of a certain speed trace where Ferrari's magic started at 250 km/h, which appears illogical. But it it continues for too long, it distracts too much form the focus of this thread, it will be split.
In fact, I am flagging this very post so that the other mods can decide if it needs to be split here and now, because a detailed discussion of the exact moment to deploy appears to be borderline off-topic to me.>

[saviour stivala]

I already expressed my opinion on here that: There is very little if any differences between the top two teams capabilities as regards the two possible maximum power unit outputs extractions. That these two power unit max power output modes both involve the combined power of the electrical engine and that of the ICE, but that they (the electric engine and ICE) both differ in their percentage contribution into the two different maximum power output modes. I also believe that one cannot deploy more electrical power than one can harvest. I also believe that both the top two teams are now capable of harvesting past the 60% mark of ERS needs by the MGU-H. Considering the above top two teams capabilities as regards the extraction of maximum power output from their power units in both possible max power output modes and the very little differences if any between them, it leaves no doubt in my mind that the FERRARI power advantage during the race is solely due to their new found capability of deploying during the race the extra ICE combustion power of which both the top two teams have and can deploy during qualifying. In my opinion for FERRARI to be able to do this, they must have found the extra reliability needed to do this and also must have also found the fuel consumption improvement to be able to do this. For Mercedes not being in a position to do the same during a race, when they can deploy the same power output in qualifying, it only means they are yet lacking those two important things that FERRARI have found and can now use during the race.

[Juzh]

I personally think both mercedes and ferrari (those 2 at the very least) are already at the point they can deploy a constant 120kw electrical power across a single lap. Just make sure you have 4MJ on board when starting your hot lap, and spend it fully plus the additional harvested energy. You should easily be able to have enough energy for 120kw the whole lap round when on the throttle.

Unlikely. There are still obvious drops in acceleration present at the end of straights, even on the ferrari and mercedes.

Pole lap in germany had a solid 50 seconds of full throttle, further suggesting a constant 120kw is a bit too much to handle for the ers.

[Juzh]

I'm not aware of a standard location for the microphones. My guess is that Ferrari has theirs mounted near the gearbox, other teams having theirs mounted near the radiators. Engine exhaust, turbocharger, and gearbox noise dominate the sounds we hear. Why would the electric motors, their controllers, and/or the ES be audible over these high-dB sources?

Since 2018 all cars are equipped with 2 onboard microphones, one in the regular old school position and one near the exhaust. Those 2 audio feeds are then broadcast sepparately to the left and right channel. They did this in order to improve the onboard sound and to allow easier differentiation between cars/engines.

ERS winding noises are very obvious on the ferrari, much more so than on other cars, i guess they positioned the second mic not near exhaust, but rather somewhere else in the car.

Winding in what sense? How would the electrical components be heard over gears, exhaust, turbcharger? Regardless of mic position.

Lost this post somehwere..

MGUs can easily be heard as a high pitched wind. Check any ferrari onboard from this year, it's very obvious.
It's actually similar to the 919 hybrid evo when the "boost" kicks in, just a bit less pronounced on the F1

[Gr1ff]

This! 100%!
The Electric horsepower is limited.
It acts through the MGUK and only develops in duration of usage.
The ICE is where Ferrari is doing the damage.

How do you explain the sudden increase with no change to kimi's ice then?

The engine could always do it. It's more reliable now so they can do it more often. Squeeze out more power using engine tuning, lubricants, fuel. The ICE is the big differentiator in qualifying.

I also do believe Ferrari has a "trick" chassis and "trick" aero that works the tyre better at certain tracks.

I understand your assumption. But it's not always been there so they aren't using it more often... they made a huge leap to find 40hp at a narrow window (lower - mid speed) with the exact same ICE from the start of the year. That's not performance from Fuel and lubricants. I could be wrong but I wouldn't say it's 100% ICE.

[Muniix]

All energy used/harvested comes from the fuel and how the ICE combusts it. The improvements might be on the turbine side along with fuel/combustion?

There is only so much ES; they are constantly trying to optimize its use at each corner of a given track. More ES usage requires more ES store which then requires either burning more fuel or taking an ICE derate. That trade off is probably shifted during the race depending on track position, strategy, etc.

You can achieve greater engine pressure from the same fuel, it's called pressure gain, or more specifically the 'Low-T QOOH amplification' injecting additional fuel after ignition of first pre-mixed fuel, as the radicals are starting to rapidly increase, as temperature rises into the Negative Temperature Coefficient range where additional heat slows down the reaction rate depending on fuel obviously that's between 800K-1400K, there's two pathways the reaction could take, it's the change in molar fractions of intermediate species, but cooling with additional fuel trigers the tripple radicial production from each O radical. This turns up the reaction rate, increasing heat release and pressure.

Giving extra pressure for piston, exhaust gas turbine for boost and harvesting to go to MGU-k while everyone else is effectively 'wasting' fuel. It's Mahle's and Ferrari's far greater resources going into basic TJI research finally paying off as they were destined to do.

[Muniix]

....

Is it better to do 120kW for 3 seconds, or 30kW for 12 seconds(?).

IMO it’s better to do the former, 120kW for 3 seconds, so long as the 3 seconds you choose are at the the lowest possible speed. Three 1 second bursts of 120 kW at the start of straights are better than 12 seconds of 30kW along a straight. This is because at higher speeds most power is used to push the air out of the way.

Of course in the context of Ferrari and their supposed power advantage this would mean they would deploy the extra power at the lowest speed at which the competition runs outfoxed power.

There have been conflicting reports about the threshold speed at which Ferrari gain their advantage. Some say 150kph, which would be as soon as they are not traction limited. This would suggest extra ICE power since everyone would be using the MGU-K at 120kW then. Some say it’s at 250kph which might be either more sustained use of either extra ICE power or MGU-K power, or both. Either way I think they will deploy all of any extra power they have at as low a speed as possible.

Grip limits play a major role on when to deploy, in lower gears all teams can exceed grip limits, only when speed increases requiring higher gears, can additional crankshaft power Ferrari have be deployed.

[Muniix]

IMO it’s better to do the former, 120kW for 3 seconds, so long as the 3 seconds you choose are at the the lowest possible speed. Three 1 second bursts of 120 kW at the start of straights are better than 12 seconds of 30kW along a straight. This is because at higher speeds most power is used to push the air out of the way.

Can you prove it with, ahem, calculations?
Say, you have 20 secs of max power, which will be faster 3secs of 120kw or 12 secs of 30kw?
You say the answer is simple...

I did a little calculation. Very basic.

I assumed:
Car mass 840kg (that’s 730 basic + 50kg fuel + 60kg for rotational inertia, wheels etc)
ICE power 550kW, 670 with MGU-K fully deployed (910hp)
Car max speed 360kph (at which speed all the ICE power is consumed by the drag)
Drag proportional to road speed cubed
Traction limit speed 150kph

I used your profiles:
Run 1 670kW for 3 seconds 550kW for 17
Run 2 580kW for 12 seconds 550kW for 8

After 20 seconds
Run 1 speed 357kph Distance 1763m
Run 2 speed 357.8 kph Distance 1755m

So the short burst of 120kW is 8m further down the road.

8 metres further down the road with less speed to reduce before the rapidly approaching corner, Run 2 going to need a huge straight to take advantage of that 0.8 kph velocity advantage.

[dren]

Injecting more fuel after the initial ignition would require more fuel, right? Unless the initial charge is lessened. From what I read, this is more for soot reduction. It must be hard to control since it's pretty simple in theory.

[subcritical71]

Grip limits play a major role on when to deploy, in lower gears all teams can exceed grip limits, only when speed increases requiring higher gears, can additional crankshaft power Ferrari have be deployed.

What does everyone think, could a form of traction control be implemented by way of the MGU-K? If so are the teams using it?

Call it 'K Over Acceleration Protection'. My thinking. At low speeds you will always be traction limited. Generally above 150kph you would no longer be traction limited. Lets say between 45 and 150 you have a throttle map that delivers ICE power which is less than that needed to break traction (but not more than 120kW less). Since you know gear ratio, speed, and general grip characteristics (from practice), the MGU-K could then be used as a variable output that just prevents wheelspin. The K could be controlled with very good precision to detect any over acceleration and then therefore reduce output.

I would think it could be used in the opposite role also, as a form of rear ABS during braking.

I'm thinking it will depend if section 5.5, Power Unit Torque Demand, takes into account MGU-K torque to the wheels.

[Tommy Cookers]

there's 3 unavoidable mechanisms that do this 'not traction control' or the equivalent braking thing
it's not traction control unless there's a wheel rpm anomaly sensor etc etc - which there isn't

these 3 (overlapping) mechanisms are
2d maps (ie steady state relationship) of PU torque vs accelerator demand (this varied according to rpm - the 2nd dimension)
inherent limit on dynamic response of K machine due to laws of physics
CU-K setup limit on dynamic response of K machine by nested direct or indirect current control - necessary for system stability

maybe a 4th mechanism is engineered/designed collapse of torque under dynamic conditions - depending eg on K machine type

they cannot not have this behaviour - but it's extent is limited by the limited size (in torque terms) of the K machine

[Nonserviam85]

Grip limits play a major role on when to deploy, in lower gears all teams can exceed grip limits, only when speed increases requiring higher gears, can additional crankshaft power Ferrari have be deployed.

What does everyone think, could a form of traction control be implemented by way of the MGU-K? If so are the teams using it?

Call it 'K Over Acceleration Protection'. My thinking. At low speeds you will always be traction limited. Generally above 150kph you would no longer be traction limited. Lets say between 45 and 150 you have a throttle map that delivers ICE power which is less than that needed to break traction (but not more than 120kW less). Since you know gear ratio, speed, and general grip characteristics (from practice), the MGU-K could then be used as a variable output that just prevents wheelspin. The K could be controlled with very good precision to detect any over acceleration and then therefore reduce output.

I would think it could be used in the opposite role also, as a form of rear ABS during braking.

I'm thinking it will depend if section 5.5, Power Unit Torque Demand, takes into account MGU-K torque to the wheels.

It is definitely possible if the regulations allow a slip measurement sensor in the wheels. It is common practice in 4Q drives in Rolling Stock (trains, trams etc)

[Tommy Cookers]

isn't that exactly what the regulations don't allow ? (in declaring the features that constitute traction control)

btw and afaik this is a 2 quadrant system

[ian_s]

I was wondering what RPM the MGU-K actually does, as for some reason I thought it was restricted to 200nm torque, but I cant find that in the technical regs.

edit:
doing some quick calcs 160hp is doable from under 6000 ice rpm at 1:1 ratio