2014-2020 Formula One 1.6l V6 turbo engine formula

[langwadt]

Apex, you are forgetting to add IMHO to most of your posts (IMHO).

Please answer a very simple question: What happens immediately after your driver shifts one gear up while running at 10500 RPM?

Most obviously you wanted to make a point about the RPM drop and it is a good point to make.
I have not (IMHO) ignored that fact while arguing that these new F1 engines will not be running at 15k RPM. So while we consider an average of 2000 RPM drop in engine revs during shift change, we may end up with an operating envelope of lets say 9500-11500 RPM, which is still not much more than 10500 RPM.

But surely we are way off from 15k.
IMHO of course.

but the 15K is sure nice to have so you don't end up like RB so many times; hitting the limiter and having trouble passing
with DRS

until we see some power curves it is hard to tell how high the power curve is "flat"

at higher RPM the friction is higher, but you only have same fuel so you need lower boost pressure, which means you can harvest more from the MGU-H, I think ...

[langwadt]

Then I don't get it please explain; one engine makes ~450nm@17000rpm, other ~600nm@11000rpm, to run the same speed they need different gear ratios, so in the end the torque at the wheels is about the same
How is that more at the pavement?

The remarks made by drivers concerned exit from corners, when they were accelerating. Usually at that point it is the net force trying to push the car forward that may or may not break the traction and it is torque that does it
Force -> Torque
This is why Pirelli had to make the tires harder, in order to resist those higher magnitude forces, even though the maximum power seemingly was to remain the same.

The 2014 spec has about 30% more torque than in 2013. So this is what the drivers (Massa among them) were describing.
IMHO of course.

Imagine you put a 17000/12000 reduction gear on the 450nm@17000rpm V8, output shaft would then be 637nm@11000rpm
that is same range as the 600nm@12000rpm, so the force pushing the car will be about the same

They might feel very different because they have that torque available more of the time, the MGU-K can add torque feed from the MGU-H, not only in burst like the KERS

IMHO

[wuzak]

Apex, you are forgetting to add IMHO to most of your posts (IMHO).

Please answer a very simple question: What happens immediately after your driver shifts one gear up while running at 10500 RPM?

Most obviously you wanted to make a point about the RPM drop and it is a good point to make.
I have not (IMHO) ignored that fact while arguing that these new F1 engines will not be running at 15k RPM. So while we consider an average of 2000 RPM drop in engine revs during shift change, we may end up with an operating envelope of lets say 9500-11500 RPM, which is still not much more than 10500 RPM.

But surely we are way off from 15k.
IMHO of course.

but the 15K is sure nice to have so you don't end up like RB so many times; hitting the limiter and having trouble passing
with DRS

until we see some power curves it is hard to tell how high the power curve is "flat"

at higher RPM the friction is higher, but you only have same fuel so you need lower boost pressure, which means you can harvest more from the MGU-H, I think ...

Correct.

The Cosworth curves shown in this, and other threads, show a peak power speed of 11,000rpm for the ICE by itself, and 12,000rpm in turbo-compound mode, using recovered energy from the MGU-H.

Also, the power in self sustaining mode is more at 15,000rpm than it is at 10,500rpm.

In ICE only mode, or ICE with full MGU-K, the power at 13,000rpm is roughly the same as 10,000rpm.

Maximum speed should be at around 11,000-12,000rpm. Power is about the same at those speeds with the full MGU-K allowance. In lower gears they will want to maximise power - which will mean between 11k and 13k rpm.

They will only use 8,000rpm to negotiate regions where they are traction limited.

[chip engineer]

...
Please realize that the figure quoted by Honda was attained at equivalence ratio of 1.02, which means just 2% richer from the stoichiometric ratio.
Under those circumstances the engine not only doesn't make it's "best" possible power, but at the same time is running very hot exhaust gas temperatures. In those days, the teams had no limit as to how many engines they use. Often they used as many in a weekend as is now allowed for a whole season.

Exhaust gas temperature is a highly important operational parameter in racing engines, which directly impacts engine life. In this case, when you are trying to run fuel efficient map, you will get high engine temps shortening it's life. Balancing those two will be this year's F1 science.

Back in the 80's, McLaren/Williams would install this fresh RA168 for Sunday and run it in the morning warm-up and the race itself, roughly less than 400 km total. Today we have 19 races with approx 700km per weekend and 5 engines, which works out to be around 2600 km engine life. That is 6-7 times longer life expectancy under comparable stress, which you may not have recognized.

It is hard to believe that with modern controls that running a stoichiometric ratio would be that difficult (although maybe that is part of the Renault problem). Certainly running richer reduces power when fuel is limited.

In any case, when I calculate the thermal efficiency of the Honda RA168E from the paper you linked, I get 32.2%. That is 272 g/kW-hr (at 1.02 fuel equivalence ratio, 12000rpm) and 9817 Kcal/kg (84% toluene), as given in the paper. Efficiency improves to 32.5% at stoichiometric ratio using data from the paper.

So let's calculate hp with 100 kg/hour with the same efficiency using fuel optimized for energy/mass (assume 45 MJ/kg) rather than the 84% toluene Honda used (41.1 MJ/kg optimized for energy/volume):
32.2% gives 540 hp
32.5% gives 545 hp

Since the Honda RA168E fuel was just 90 motor octane (101.8 research octane), I would expect considerable improvement in efficiency just from higher compression ratio with unlimited octane 2014 fuel and direct injection. Also reduced friction at 10500 rpm vs.12000 rpm for the Honda RA168E should also help power. Some here claim 46 MJ/kg or more fuel energy as well. Maybe a total improvement of 6 to 10 %?

So nearly 600 hp from the piston engine alone, and 750 hp with the 160 from the MGU-K seems reasonable to me.

[321apex]

but the 15K is sure nice to have so you don't end up like RB so many times; hitting the limiter and having trouble passing
with DRS

until we see some power curves it is hard to tell how high the power curve is "flat"

at higher RPM the friction is higher, but you only have same fuel so you need lower boost pressure, which means you can harvest more from the MGU-H, I think ...

As you probably know, cam timing determines the characteristics of engine's power envelope in terms of RPM. Unless you have switchable cam timing a'la Honda V-TEC, you have a relatively fixed "power envelope". I don't know if such solution is/will be practiced in F1, but the 2014 do not forbid it.
What that means is that the power peak and torque become "cast in stone" as an event in rev range. By adding or subtracting boost you may move these peaks up or down in value but not much in RPM.

So if you optimize cam timing of and engine to have power peak at say 11000 RPM and you wanted to make it still be able to run to 15k, it may not be able to get there due to cam timing. IMHO the power loss at 15k would be so substantial that this engine would not be "pulling" so in practical sense it will be useless to try to make it run there.

I would like to touch upon one more thing that has not been talked about much.
Namely, since this engine is a 90 deg odd firing V6, it will have uneven firing combustion cycles, which isn't good from the perspective of "tuning" inlet or exhaust tracts. Moreover, this engine will be prone to vibrations, specifically 1st order and 2 order rotating couples.

I have not read any forbidding remarks in the rules, which may prevent the use of a balancing shafts, so such a solution may be applied here...., adding friction IMHO . This vibratory phenomena is proportional with RPM, so the higher you rev the engine the greater it's magnitude.

I had a feeling at some point that Renault's early problems may have been related to this vibration and quite possibly their engine was shaking the ancillaries to pieces. If you recall, Renault was producing a 90 deg V6 in road cars, w/o balance shaft. Perhaps that did something similar and it wasn't shaking on the dyno

[Holm86]

but the 15K is sure nice to have so you don't end up like RB so many times; hitting the limiter and having trouble passing
with DRS

until we see some power curves it is hard to tell how high the power curve is "flat"

at higher RPM the friction is higher, but you only have same fuel so you need lower boost pressure, which means you can harvest more from the MGU-H, I think ...

As you probably know, cam timing determines the characteristics of engine's power envelope in terms of RPM. Unless you have switchable cam timing a'la Honda V-TEC, you have a relatively fixed "power envelope". I don't know if such solution is/will be practiced in F1, but the 2014 do not forbid it.
What that means is that the power peak and torque become "cast in stone" as an event in rev range. By adding or subtracting boost you may move these peaks up or down in value but not much in RPM.

It is true that the cams do determine much of the engines power curve. But much more so in an NA engine. In a boosted engine you can compensate alot with the boost.

And any sort of variable valve mechanism is prohibited by the regulations.

[321apex]

Correct.

The Cosworth curves shown in this, and other threads, show a peak power speed of 11,000rpm for the ICE by itself, and 12,000rpm in turbo-compound mode, using recovered energy from the MGU-H.

Also, the power in self sustaining mode is more at 15,000rpm than it is at 10,500rpm.

In ICE only mode, or ICE with full MGU-K, the power at 13,000rpm is roughly the same as 10,000rpm.

Maximum speed should be at around 11,000-12,000rpm. Power is about the same at those speeds with the full MGU-K allowance. In lower gears they will want to maximise power - which will mean between 11k and 13k rpm.

They will only use 8,000rpm to negotiate regions where they are traction limited.

I will take with a grain of salt the "self sustaining" estimates being higher at 15k IMHO

This area is a great unknown and no measure of simulation is going to validate it without hardware testing. Chemical fuel is the primary energy source. By running at close to 10500RPM you minimize the energy waste.

I would like to see it to believe it that by wasting more of it at 15k you reclaim more of it thru MGU-H.

[321apex]

It is true that the cams do determin much of the engines power curve. But much more so in an NA engine. In a boosted engine you can compensate alot with the boost.

Not true!
Boost affects VE. It does not move power/torque events along rev range.
If at all - very unsubstantially.

[Abarth]

It is true that the cams do determin much of the engines power curve. But much more so in an NA engine. In a boosted engine you can compensate alot with the boost.

Not true!
Boost affects VE. It does not move power/torque events along rev range.
If at all - very unsubstantially.

What????? Variation in cam timing and/or valve lift profile do exactly what boost variation does, it allows more or less combustion air at a given operating point.

If you can control boost via software (which is standard these days) you can shape torque (and therefore power) curves very arbitrarily, of course within the constraints of engine robustness, allowed fuel flow or other driveability demands (i.e. response, especially at low revs, etc)

[wuzak]

So if you optimize cam timing of and engine to have power peak at say 11000 RPM and you wanted to make it still be able to run to 15k, it may not be able to get there due to cam timing. IMHO the power loss at 15k would be so substantial that this engine would not be "pulling" so in practical sense it will be useless to try to make it run there.

I doubt that cam timing would prevent the engine from going from 11,000rpm to 15,000rpm.

The power may fall off too much and make it impractical to use 15k, but the cam timing wouldn't prevent it. Maybe if it were optimised for 5k, but not when it is set up for 10-11k.

I would like to touch upon one more thing that has not been talked about much.
Namely, since this engine is a 90 deg odd firing V6, it will have uneven firing combustion cycles, which isn't good from the perspective of "tuning" inlet or exhaust tracts. Moreover, this engine will be prone to vibrations, specifically 1st order and 2 order rotating couples.

The inlet and exhaust systems will be tuned as 2 sets of three. The turbo should be set up such that the exhaust from the two banks do not mix - ie a twin scroll design. This has certainly been obvious from the pictures Renault have released and the video animation by Ferrari.

The V8 used a flat plane crank - which is also not ideal for vibrations. In the past they have run 90° V-10s (before bank angles were defined in the V8 era), 80° and 90° V6s.

The RA168E was an 80° V6, btw.

I have not read any forbidding remarks in the rules, which may prevent the use of a balancing shafts, so such a solution may be applied here...., adding friction IMHO . This vibratory phenomena is proportional with RPM, so the higher you rev the engine the greater it's magnitude.

Nothing in the rules to prevent them, but doubtful they will be used.

The vibrations will be worst at certain rpm, and have bad periods in multiples of that. From memory, and it is a long time since I looked at this, the first order will have the greatest magnitude. The vibrations are not directly proprtional to rpm.

I had a feeling at some point that Renault's early problems may have been related to this vibration and quite possibly their engine was shaking the ancillaries to pieces. If you recall, Renault was producing a 90 deg V6 in road cars, w/o balance shaft. Perhaps that did something similar and it wasn't shaking on the dyno

It may be due to vibrations, but we cannot know for sure. Rumours are that there is an issue with the crankshaft, which may be a torsional vibration issue, but we really won't know unless they make a statement to the effect.

[wuzak]

I will take with a grain of salt the "self sustaining" estimates being higher at 15k IMHO

This area is a great unknown and no measure of simulation is going to validate it without hardware testing. Chemical fuel is the primary energy source. By running at close to 10500RPM you minimize the energy waste.

I would like to see it to believe it that by wasting more of it at 15k you reclaim more of it thru MGU-H.

It is not that unknown.

Turbines have been around for many centuries.

Steam turbines have been around for about 125 years, maybe more. Gas turbines have been around for 80 years. Turbochargers for almost 100 years. They have been studied extensively.

The engine designer can, if he wants a quick estimate, go to a turbo manufacturer and get a bunch of maps of the compressor and the turbine. He can then match the characteristics of these to what he wants to achieve with the engine. These maps will tell him the power the turbine will produce for a give mass flow and rpm, and the amount of boost for a given mass flow required and rpm, and the power to achieve that.

Experienced engine designers like Cosworth can predict quite well what the engine will do under different consitions, how much boost it will need, what the mass flow is, etc.

No, it is not a huge mystery to them.

The time and money would be in optimising them. Which I doubt Cosworth have done, as yet. Just a first order approximation.

[Tommy Cookers]

..... since this engine is a 90 deg odd firing V6, it will have uneven firing combustion cycles, which isn't good from the perspective of "tuning" inlet or exhaust tracts. Moreover, this engine will be prone to vibrations, specifically 1st order and 2 order rotating couples.
....... balancing shafts, so such a solution may be applied here...., adding friction IMHO . This vibratory phenomena is proportional with RPM, so the higher you rev the engine the greater it's magnitude.
...... Renault was producing a 90 deg V6 in road cars, w/o balance shaft. Perhaps that did something similar and it wasn't shaking on the dyno

all F1 V6s (and F2) have been odd firing, this since the 1950s (except as below)
because any true V6 (ie one with a 3 throw crank) will have odd firing
unless it has a 120 deg V angle, otherwise (eg 60 deg V) even firing V6s needs 6 throws, and are too bulky and heavy for F1 etc
eg the classic 65 deg Ferrari was more odd firing than the 90 deg of 2014 or the turbo Renaults or the 80 deg that Honda used

there seems no induction issue
there is an exhaust issue only now in 2014, due to the single turbo
exhaust pulse spacing is of course even within each bank's manifold/header
but the spacing would become uneven when the 2 bank's systems merge, unless the 2 bank's pipe lengths are suitably different
this was clearly done by Renault in those early years when they used a single turbo
to my eyes the pipes are so treated in 2014
the even spacing is primarily important for the turbo
true the pipe length cannot give ideal equal spacing at all rpm etc

there is surely no vibration factor that hasn't been lived with for years
and far less vibration than with the recent V8s

someone has in the past posted a link (to old RCE or another product of Mr Bamsey) that definitely says ......
Honda never raced at 1.02 equivalence ratio, for response reasons they always raced at 1.08
though on paper the 1.02 engine was better for power under the 150 litre rules

[321apex]

What????? Variation in cam timing and/or valve lift profile do exactly what boost variation does, it allows more or less combustion air at a given operating point.

If you can control boost via software (which is standard these days) you can shape torque (and therefore power) curves very arbitrarily, of course within the constraints of engine robustness, allowed fuel flow or other driveability demands (i.e. response, especially at low revs, etc)

In essence what you are saying in practical terms is not true. Please read carefully what I wrote - we are speaking of "location" along the RPM axis where power or torque peaks occur.

Holding inlet and exhaust geometry as a constant, the cam timing has the overwhelming impact in determining these RPM areas. Boost only move it up or down in "general vicinity" of those areas. The higher you dial in the boost, there is even a tendency to move the locations of those peaks LOWER. As an example in low boost mode of a given engine you may see a power peak at 10000 rpm, and at the maximum possible "safe" boost limit you may make tons more power but the peak may slide down to 9700 RPM.
Do you follow?

IMHO of course.

[321apex]

all F1 V6s (and F2) have been odd firing, this since the 1950s (except as below)
because any true V6 (ie one with a 3 throw crank) will have odd firing
unless it has a 120 deg V angle, otherwise (eg 60 deg V) even firing V6s needs 6 throws, and are too bulky and heavy for F1 etc
eg the classic 65 deg Ferrari was more odd firing than the 90 deg of 2014 or the turbo Renaults or the 80 deg that Honda used

there is surely no vibration factor that hasn't been lived with for years
and far less vibration than with the recent V8s

I didn't add this earlier, while you didn't recognize the fact that 2014 F1 engines have more delicate hardware attached to them susceptible to vibration.

In mechanical engineering, when you study "shafting" you learn of a phenomena called "critical speed" of a shaft. Well, the MGU-H to my mind contains insidevery much the sort of "simply supported rotating shaft". Long in length to maximize power potential, bearing at each end and small diameter of the rotor. The critical speed formula for such a shaft takes into account "g" as in earths gravity. In this case we have a much more significant accelerative effect in the form of engine vibration.

In the link you may read more about this phenomena
http://www.google.pl/url?sa=t&rct=j&q=& ... 0604,d.ZGU

The point is, that outside of FEA which everyone uses at this stage to predict vibratory regimes, you never know for sure until you mount the engine in the car and run it. The vibration in the past may have been tolerable, while today with this new hardware attached to the engine, the same vibration may be far more risky and damaging. Let's agree that this isn't insurmountable problem and given time this issue will get solved or at least contained.

[Tommy Cookers]

my guess is if the mgu-h is safe from the whirling phenomenon at 125000 rpm a bit of 12000 rpm engine vibration won't hurt it
speaking as one who was paid to design stuff before FEA even existed
one who has 'always' found the direct drive mgu-h rather implausible