F1technical.net

Vortex37 wrote:A link for those interested in engine oil chemistry.

That's interesting... but it's a bit of mechanical engineering as well. Tribology.

Vary wrote:A first evaluation of engines power in test configuration Made By Eng. Benzing http://www.formula1benzing.eu/index.html (it's in italian, the paragraph were talk about power is the last one, "il calcolo delle potenze")

Good article. Mercedes 880 hp including MGUK. 3.3 bar MAP. Suggests they are possibly over 900 hp in "emergency mode".

Is he guesstimating the aero drag - I could not quite figure this out from the Google translate?

I wonder if you could infer drag to a first approximation from the rate of reduction in car acceleration on the approach to vmax, because the power curve is flat(ish)? I did not see this type of method mentioned, though.

Brian Coat wrote:Is he guesstimating the aero drag - I could not quite figure this out from the Google translate?

I wonder if you could infer drag to a first approximation from the rate of reduction in car acceleration on the approach to vmax, because the power curve is flat(ish)? I did not see this type of method mentioned, though.

He mentioned it in other articles, he basically calculate the aero drag from frontal area and tyre resistance, at max speed that resistance power is the "minimum" engine power

I don't he can infer anything with so many assumptions. Wheel spin, grip, other things are happening as the car goes down the straight.. rear wings etc. etc.

He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

Mass is known (perhaps)
Rolling resistance is known. (Proll)
acceleration (a) is known at various speeds on the straight
Drag is proportional to v^2
aero power (Paero) is proportional to v^3
Acceleration at any speed a = Pacc/(v x m) so Pacc = a.v.m where Pacc is remaining power available to accelerate the car
But Pacc is also = Ptot - Paero - Proll

So Ptot - Paero - Proll = a x v x m

By measuring acceleration at two different speeds you can produce two versions of this equation and since there are two unknowns (Ptot and Paero), we can solve for them. Better still, the acceleration can be continuously measured over the second half of the straight and by curve fitting at least one more unknown can be solved - possibly mass.

This is possible because mass has a different effect on acceleration than drag at different speeds.

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

Mass is known (perhaps)
Rolling resistance is known. (Proll)
acceleration (a) is known at various speeds on the straight
Drag is proportional to v^2
aero power (Paero) is proportional to v^3
Acceleration at any speed a = Pacc/(v x m) so Pacc = a.v.m where Pacc is remaining power available to accelerate the car
But Pacc is also = Ptot - Paero - Proll

So Ptot - Paero - Proll = a x v x m

By measuring acceleration at two different speeds you can produce two versions of this equation and since there are two unknowns (Ptot and Paero), we can solve for them. Better still, the acceleration can be continuously measured over the second half of the straight and by curve fitting at least one more unknown can be solved - possibly mass.

This is possible because mass has a different effect on acceleration than drag at different speeds.

Excuse my ignorance, with "Paero" you mean the power to overcome the drag? Why is it unknown since we have the aero drag (calculated by frontal area and drag coefficient)

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

Mass is known (perhaps)
Rolling resistance is known. (Proll)
acceleration (a) is known at various speeds on the straight
Drag is proportional to v^2
aero power (Paero) is proportional to v^3
Acceleration at any speed a = Pacc/(v x m) so Pacc = a.v.m where Pacc is remaining power available to accelerate the car
But Pacc is also = Ptot - Paero - Proll

So Ptot - Paero - Proll = a x v x m

By measuring acceleration at two different speeds you can produce two versions of this equation and since there are two unknowns (Ptot and Paero), we can solve for them. Better still, the acceleration can be continuously measured over the second half of the straight and by curve fitting at least one more unknown can be solved - possibly mass.

This is possible because mass has a different effect on acceleration than drag at different speeds.

If you corner exit is compromised because of lower traction, your top speed will be compromised as well. As i said it's too many variables. Also we don't know how the drivers are driving, or what they are driving to, or the fuel levels of the car, which would affect the acceleration.
We don't have a definitive idea of the power of the engines from last year, even with a full season of data, so i don't see how we are going to know this season with winter testing alone.

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

That is the way to do it, but he doesn't have the data to do it. Not in this test in Barcelona anyways. To do it right yo need to know that the car pushed through (with ICE, with KERS, how much?), ideally the gears, its mass, ideally a whole acceleration curve, the wind and you also need to assume that aero values change with v2, something that we don't know as allegedly they all have flexible aero and selective stalling.
With onboard speeds from a race, over several laps... they I'd believe those numbers much better, but in the Barcelona test?

That said... an approximation with lots of assumptions is better than no guess at all!

An interesting article on turbine design

https://www.google.co.in/url?sa=t&rct=j ... 1401,d.c2E

hollus wrote:

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

That is the way to do it, but he doesn't have the data to do it. Not in this test in Barcelona anyways. To do it right yo need to know that the car pushed through (with ICE, with KERS, how much?), ideally the gears, its mass, ideally a whole acceleration curve, the wind and you also need to assume that aero values change with v2, something that we don't know as allegedly they all have flexible aero and selective stalling.
With onboard speeds from a race, over several laps... they I'd believe those numbers much better, but in the Barcelona test?

That said... an approximation with lots of assumptions is better than no guess at all!

Yeah, he said that during tests with so many unknown variables it has not a lot sense, but he did this calculations also during last year (and previous years) and will continue to do them with more data from races

Vary wrote:

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

Mass is known (perhaps)
Rolling resistance is known. (Proll)
acceleration (a) is known at various speeds on the straight
Drag is proportional to v^2
aero power (Paero) is proportional to v^3
Acceleration at any speed a = Pacc/(v x m) so Pacc = a.v.m where Pacc is remaining power available to accelerate the car
But Pacc is also = Ptot - Paero - Proll

So Ptot - Paero - Proll = a x v x m

By measuring acceleration at two different speeds you can produce two versions of this equation and since there are two unknowns (Ptot and Paero), we can solve for them. Better still, the acceleration can be continuously measured over the second half of the straight and by curve fitting at least one more unknown can be solved - possibly mass.

This is possible because mass has a different effect on acceleration than drag at different speeds.

Excuse my ignorance, with "Paero" you mean the power to overcome the drag? Why is it unknown since we have the aero drag (calculated by frontal area and drag coefficient)

Drag coefficient can vary enormously with aero setup, so is unknown.

ringo wrote:

gruntguru wrote:He will be looking at the acceleration at higher speeds on the straight, not where the car is grip-limited.

Mass is known (perhaps)
Rolling resistance is known. (Proll)
acceleration (a) is known at various speeds on the straight
Drag is proportional to v^2
aero power (Paero) is proportional to v^3
Acceleration at any speed a = Pacc/(v x m) so Pacc = a.v.m where Pacc is remaining power available to accelerate the car
But Pacc is also = Ptot - Paero - Proll

So Ptot - Paero - Proll = a x v x m

By measuring acceleration at two different speeds you can produce two versions of this equation and since there are two unknowns (Ptot and Paero), we can solve for them. Better still, the acceleration can be continuously measured over the second half of the straight and by curve fitting at least one more unknown can be solved - possibly mass.

This is possible because mass has a different effect on acceleration than drag at different speeds.

If you corner exit is compromised because of lower traction, your top speed will be compromised as well. As i said it's too many variables.

It is not necessary to know top speed, you don't see it in the equation above. It is easy to calculate top speed once the simultaneous equations are solved.

It's like you guys are using a spring balance to find the number of carbon atoms in a coconut. Commendable effort but no cigar. Maybe you can use your energies for some other aspects of the engine please.

Once again Ringo, you don't have to be interested in everything on the forum or even every topic discussed in this thread. There are a lot of members who delight in speculating on what the power outputs actually are. I am not one of them, but I do admire the technical challenge of trying to distill some realistic numbers from very limited information.