Ferrari Power Unit Hardware & Software

[dans79]

Something I wanted to add, would be if I was trying to use EMI to manipulate the sensor readings, I wouldn't be trying for a steady state offset, I'd be trying for something like this .



The way to think about this, is that you inject the maximum amount of extra fuel in troughs, and you lead in and tail off as you get towards the peaks. Syncing frequencies becomes less problematic. If you are slightly out of sync, the anomalies will be wave like over time, and would probably be hard to discern from fuel pump pulses.

[izzy]

Something I wanted to add, would be if I was trying to use EMI to manipulate the sensor readings, I wouldn't be trying for a steady state offset, I'd be trying for something like this .

https://i.ibb.co/42Dj0qF/Untitled.png

The way to think about this, is that you inject the maximum amount of extra fuel in troughs, and you lead in and tail off as you get towards the peaks. Syncing frequencies becomes less problematic. If you are slightly out of sync, the anomalies will be wave like over time, and would probably be hard to discern from fuel pump pulses.

nice! and yes as you posted so much efficiency follows from this apparently simple flow rate rule, and so much complexity too. And once you know the sensor samples at 2kHz how could you not play with it, if you're an f1 engineer?

i bet they've all been playing with it, and that's how they've guessed what Ferrari have been up to. and is it 'cheating'? i mean a rate has to have a timebase doesn't it and it's not an hour or they could flow any amount for 10 minutes then make it up or down for 50 minutes. Plus every measurement ever has a tolerance

[ubuysa]

All that we know from Austin is this.....

1. The FIA issued a TD after Red Bull suggested some fuel flow trickery.

2. The Ferraris seemed to be down on their usual pace.

All else is conjecture and rumour. I would suggest we all wait and see what the Ferrari pace is like in Brazil and Abu Dhabi. We might all be surprised.

Sent using Tapatalk

[3jawchuck]

All that we know from Austin is this.....

1. The FIA issued a TD after Red Bull suggested some fuel flow trickery.

2. The Ferraris seemed to be down on their usual pace.

All else is conjecture and rumour. I would suggest we all wait and see what the Ferrari pace is like in Brazil and Abu Dhabi. We might all be surprised.

Sent using Tapatalk

What's the point of a technical forum if people don't discuss possible technical developments? None of us here know anything about what the teams are actually up to. All that can be done is speculate and solve the problem amongst ourselves.

[nzjrs]

What's the point of a technical forum if people don't discuss possible technical developments? None of us here know anything about what the teams are actually up to. All that can be done is speculate and solve the problem amongst ourselves.

The last few pages have been great reading. Reminds me of some of the good old times over in the Honda PU thread.

[dans79]

Where is the truth?
Maybe here:

Meanwhile, while by no any means conclusive, we have some data:
https://f1ingenerale.com/?attachment_id=97413
https://f1ingenerale.com/?attachment_id=97412
https://f1ingenerale.com/f1-power-unit- ... -qualcosa/

This analysis is very poor, as it is mostly trying to use the top speed delta as a way of justifying it's claims.

Given the limited data, integrating the delta curves would give a much better indicator, as to how much the Ferrari advantage has or has not changed.

With the given data, the best analysis would be to curve fit the speed traces, and then use the fits to calculate the distance it takes to accelerate from one velocity to another. Ideally you would do this at the lowest non-traction limited speed to minimize how different DF levels effects the data. For this data set say 175 to 250/275.

[izzy]

What's the point of a technical forum if people don't discuss possible technical developments? None of us here know anything about what the teams are actually up to. All that can be done is speculate and solve the problem amongst ourselves.

yes exactly, the fun is guessing, arranging the snippets we get, and the thing is just to keep an open mind and and know we don't actually know. And anyone who it turns out guesses right obviously gets to feel amazing in the end

[saviour stivala]

I have a question that hopefully one of the resident engine gurus could answer (GG, Mudflap, wuzak, dans, etc)

With all the furore surrounding the recent TD and the possible affects it had on the Ferrari PU and specifically on the measurement of fuel flow, what would likely be the maximum/optimum fuel flow in the ICE`s that Formula 1 is using at the moment?
Obviously as the engine is limited to 1.6 litres, the rpm is also limited and the amount of fuel allowed to be used per race is limited there must be a theoretical cut off point [efficiency accounted for]. Why do we have a need for the fuel flow sensor/meter. Couldn`t they just say, you have x amount of fuel for the race [with x litres left at the end for testing], use that how you see fit...

Or am I missing something blindingly obvious?

Cheers,

p.s. I don`t think this is how Ferrari have a better PU, I still think they have a better MGUH system than anyone else!

the 100kg/h is a measure to cap horsepower and probably to reduce spending outlets. Without it, teams would definitely run 15,000RPM.

There is no questioning the fact that this is a fuel limited engine formula, limited to a max power speed of 10500 RPM. And of course, without the 'maximum of 100kg/h fuel flow rule' they would definitely run to whatever maximum RPM was allowed. As also is a fact that without the added maximum RPM at which the maximum fuel flow is allowed, they would definitely run at a much slower maximum RPM.

[saviour stivala]

I have a question that hopefully one of the resident engine gurus could answer (GG, Mudflap, wuzak, dans, etc)

With all the furore surrounding the recent TD and the possible affects it had on the Ferrari PU and specifically on the measurement of fuel flow, what would likely be the maximum/optimum fuel flow in the ICE`s that Formula 1 is using at the moment?
Obviously as the engine is limited to 1.6 litres, the rpm is also limited and the amount of fuel allowed to be used per race is limited there must be a theoretical cut off point [efficiency accounted for]. Why do we have a need for the fuel flow sensor/meter. Couldn`t they just say, you have x amount of fuel for the race [with x litres left at the end for testing], use that how you see fit...

Or am I missing something blindingly obvious?

Cheers,

p.s. I don`t think this is how Ferrari have a better PU, I still think they have a better MGUH system than anyone else!

The maximum/optimum fuel flow in the ICE that the formula is using at the moment is that allowed by the maximum fuel flow rules (100kg/h at 10500 RPM).

[saviour stivala]

That's surely a more interesting conjecture than the one on the invisible fuel cache or the one on the nearby electric cable that can alter the flow-meter function without causing noticeably anomalous readings.

In order to anti-synchronize the flow illegal peaks against the measurements, they would need to detect the ultrasound impulses and modulate the pump accordingly, using a sensor for the sensor (so to speak) which should pass unnoticed through scrutineering together with the relative electronics and software, and the whole secret system should work almost perfectly on a sub-millisecond time-scale. Even if the time intervals between measurements were always exactly equal (no drift, no thermal effects, etc) and that therefore the system would have to be fine-tuned only occasionally during the race, it is imo quite unlikely to be feasible, especially considering that the movement of mechanical parts is involved.

I think there's a more practical way of doing it (obviously I am arguing purely for the sake of it and because I find it interesting - I am in no way implying that any team does it).

One could simply "calibrate" the pumps by running the engine at a known operating point (let's say 95 kg/h) and then just slightly tune the phase angle. Adjusting the phase angle in one direction would cause an increase in FFM measurement (even though the operating point and hence real fuel flow is unchanged) while going the other way would cause the reading to decrease. Such calibration could probably be built into a closed loop control system that ensures the frequency and phase never drift too far.

This way all that is really required is that the real fuel flow rate is very well mapped for the entire range of engine operating conditions which isn't such an unlikely proposition.

There's also multiple ways of creating the 2kHz pulses - abrupt diameter changes, PRV flutter, etc but I proposed the electrical gear pump since it offers the simplest way of controlling the phase angle.

Yes, stabilizing the frequency and phase of the peaks per se is fairly unproblematic, at least in principle, but how would you propose (hypothetically, of course) to control the _relative_ phase of flow peaks vs FFM measurements? If your proposal involved a U/S detector, we would be back into the first reply of mine.

The part of the fuel system called “all in the fuel tank”. Low pressure electrical lift pumps normally 4 in quantity, lifts fuel and transfer said fuel into a fuel pot at low pressure. The fuel pressured in the fuel pot goes through the fuel flow sensor/meter on its way to the engine. So no fuel pump is pumping fuel directly through the fuel flow sensor/meter.

[saviour stivala]

Well, one thing people seem to forget, although the FIA standard fuel flow sensor is the relevent one for the fuel flow regarding the rules, it's not the only way they can determine the fuel flow.
Does anyone remember the Australian GP of 2014 where Daniel Ricciardo was disqualfied after the race because he breached the flow limit? RBR argued against it by referencing their own fuel flow sensor within the engine. Basically you can calculate the fuel flow via the injector duty cycles. That's all visible in the telemetry. So even if Ferrari would manipulate the signal from the standard fuel flow sensor, the additional fuel would show up elsewhere in the telemetry.

All fuel supplied to the injectors must be supplied to the FIA data logger.

[Tzk]

i bet they've all been playing with it, and that's how they've guessed what Ferrari have been up to. and is it 'cheating'? i mean a rate has to have a timebase doesn't it and it's not an hour or they could flow any amount for 10 minutes then make it up or down for 50 minutes. Plus every measurement ever has a tolerance

It is cheating. The rules state that you can't use a procedure/system/device to raise the actual fuel flow above the stated limit of 100kg/h. If you use a system to jam the sensors signal, then you clearly violate that rule.

Note that while the flow uses an hour as base you can't assume that you can use 90kg in 1min and the rest of the 100kg/h in the remaining 59mins. You could even convert the number to grams per second (it's ~28g/s). It doesn't work this way and any engineer you ask should agree on this.

[saviour stivala]

Further to “the part of the fuel system called all in the fuel tank”, The other part of the fuel system “called all on the engine”, The high pressure mechanical pump normally driven by the intake camshaft which gets all its fuel through the fuel flow sensor/meter, delivers fuel flow fundamentally proportional to engine RPM. The fuel consumed by the engine is also approximately proportional to RPM at full throttle. But at closed throttle, the engine uses no fuel. To match fuel supplied to the fuel required, the high pressure mechanical pump has a variable displacement mechanism activated by a sophisticated pressure regulating device, this high pressure pump pressures fuel up to a maximum of 500bar into the fuel rail. Homologated sensors directly measures the pressure and temperature of the fuel supplied to the fuel injectors. All of these signals must be supplied to the FIA data logger.

[izzy]

It is cheating. The rules state that you can't use a procedure/system/device to raise the actual fuel flow above the stated limit of 100kg/h. If you use a system to jam the sensors signal, then you clearly violate that rule.

Note that while the flow uses an hour as base you can't assume that you can use 90kg in 1min and the rest of the 100kg/h in the remaining 59mins. You could even convert the number to grams per second (it's ~28g/s). It doesn't work this way and any engineer you ask should agree on this.

yes i see what you mean but it could be a legendary grey area, a bit. they only have 105kg so over 1.5 hours their rate has to be less than 100kg/hr, on average. A rate is always an average, it's a mass over a period of time which is not infinitely small. However small you slice there's a period and the regs don't say what it is. if it's 1/2000s then they needed to say so.

it does depend what they've done, if anything, and it's probably illegal, but there might be enough of a defence to save them from it being a flat cheat. I don't know. it does seem to be rather clever

[ncx]

Further to “the part of the fuel system called all in the fuel tank”,

Do you think there is no way to modulate significantly the fuel flow into the FFM by means of devices that are inconspicuous enough to pass scrutiny?