The power loss with altitude thread.

[Jay]

There's a lot of half truths and incorrect physics going on in this thread But that aside...

What's the point in modelling altitude's affect on aero and power if the affect of temperature isn't taken into account?

Forget altitude, lets just talk about the same altitude but different temperatures. This will vary engine performance as well as Aero. Aero depends on density which in turn depends on temperature and pressure. A couple of months ago I was doing some wind tunnel testing and from the start of the day to the end of the day the density could vary around 3 to 4%, and that's within a room, not outdoors (its a closed loop wind tunnel but its an open test section).

Going from one track to another with wildly different temperatures and humidities will massively effect tires, which on will affect suspension, it massively affects engine performance, it can also easily affect aerodynamics by several percent (given that many of the aerodynamic devices on race cars will only affect aero a percent or two anyway). That's all without changing altitudes at all.

Lets also bring in cross winds which can totally screw things up

Indeed and good points but the thread was about power loss with altitude. Although if active weather is a factor in GT5 like the rumours suggest then this is very valid topic and would be nice to simulate, I am sure KY is well aware of it but if he will model it this time around is unknown.

I remember when asked (around GTHD days) what KY couldn't achieve on the PS3 and would like to on a future system he said he would like to model the air itself they way it flowed around the car but didn't go very deep into it. I assume he means actually simulate it which besides flow it includes pressures, temp etc etc. So it shows he has his mind on it.

 

[WolfRacer543]

Indeed and good points but the thread was about power loss with altitude. Although if active weather is a factor in GT5 like the rumours suggest then this is very valid topic and would be nice to simulate, I am sure KY is well aware of it but if he will model it this time around is unknown.

I remember when asked (around GTHD days) what KY couldn't achieve on the PS3 and would like to on a future system he said he would like to model the air itself they way it flowed around the car but didn't go very deep into it. I assume he means actually simulate it which besides flow it includes pressures, temp etc etc. So it shows he has his mind on it.

Yes true, but the reason I bought it up is because one of the reasons engine performance decreases with altitude is because of changes in temperature which bring about changes in density. Just pointing out how much these things can affect a car on the same track as the temperature changes, let alone going to a completely different track.

I think KY's dream of simulating air flow around a car will have to wait maybe 4 or 5 generations of consoles, maybe the PS8 or 9 will be able to do it Given that to solve the Navier-Stokes equations still has to be done iteratively and takes a mighty long time even on modern PCs.

 

[kayman]

*Warning thread may contain some physics*

Okay so this is thread about physics accuracy, now altitude affects a car in numerous way one of the most noticeable effects is the power output of your engine.

Perhaps you may wonder why does it matter?

Well while many tracks are barely higher than sea level, there are some in-game tracks which are sufficiently high enough to have a noticeable effect. At around 2000 metre's (6500feet) a naturally aspirated engine should experience a powerloss of close to 20%. So your 300bhp car should find itself with around 240bhp.

Clearly Gran Turismo would be more realistic if it took this into account, sadly GT5P doesn't appear to do this currently going by the quick-tune menu at Eiger and then Daytona in which the power remains the same.

If GT where to re-introduce the pike's peak hill climb the effect would be even more noticeable as the rally ranges from 9000-15000feet which would result in a 25-45% powerloss depending on the on what height you are at. It seems too much of an important detail to miss.

Anther reason why I think it is very important aside from accuracy of simulation, is it adds another factor to tuning, a forced induction engine will experience less of a powerloss than a naturally aspirated enigne, this could affect what car you choose for a certain race, how you tune and setup the car. For example you may choose to buy a turbo for your car instead of going down the naturally aspirated root if you planning on racing at a track like eiger.

Now for the physics, you've been warned.

A naturally aspirated looses approximately 3% power for every 1000feet, this is due to a decrease in air density at altitude and thus less oxygen available for combustion. For a forced induction engine its not so simple.

example

Let's say you have a 300bhp turbo charged engine at sea level. The turbo charger supplies 10psi of boost.

Using eiger as an example again, the powerless can be calculated. assuming sea level air pressure to be 14.7psi.

When car produces its peak power of 300bhp at sea level, the intake pressure is 14.7(air pressure) + 10psi provided by turbo charger = 24.7psi to get 300bhp.

at 6500 feet air pressure has decreased by 20% so air pressure is 11.8psi.

11.8psi (air pressure) + 10psi (assuming turbo can spin fast enough to maintain efficiency) = 21.8psi intake pressure.

21.8/24.7 = 89% power

=267bhp

this is an extra 27bhp advantage over a naturally aspirated engine. While this may not seem that huge, the higher the altitude the more power is saved by the turbo, this saving can be increased further by increasing the boost pressure of the turbo.

I think its an interesting factor which to consider when tuning a car, and of course it will add to the accuracy of the simulation so I am certainly hoping we will see the effects of altitude in GT5.

Now back on topic...

Wouldn't this also mean that turbo lag would be more evident the higher you are? As when the car is not on boost you would have the same loss as a NA engine.

 

[Raitziger]

I remeber Timo Salonen said in interview that in some rally in mountains there was very large power loss on hiss 600 HP Peugeot 205 Turbo 16. He even said in later stages car fel like it did not even have 300 HP and he did not bother to use helmet as it was so hot

Here is one clip of a one of our world champions:

 

[IsmokeGT]

Was his rally car a Carbureted engine or fuel injected?

Fuel injection compensates for less oxygen, the motor still loses power but not nearly as much as a carbuetor feed motor.

That video looks old, so it is hard to tell by the video, unless I missed a date somewhere.

 

[Stevisiov]

The date in the video is 1985.

 

[FLASH48]

Was his rally car a Carbureted engine or fuel injected?

Fuel injection compensates for less oxygen, the motor still loses power but not nearly as much as a carbuetor feed motor.

That video looks old, so it is hard to tell by the video, unless I missed a date somewhere.

Going by the date, i'm pretty sure there is a carburetor on the car!

 

[Raitziger]

Well that was only timo salonen incar clip i found. I am not sure if that car has the same spec as one that had major power loss in the interview.

Here is peugeot 405 T16 driven by Ari Vatanen in classic film festival winner clip. I believe this car had fuel injection.

In this next clip Ari makes co-driver **** in his pants :

 

[Marf]

Going by the date, i'm pretty sure there is a carburetor on the car!

The Peugeot T16 engine is electronically fuel injected with a Magnetti Marelli ECU running the show.

 

[sumbrownkid]

^In fuel injectors defense, the technology only really took off in the 90s, especially the late 90s and maybe early 2000s. Direct injection might fight power loss more effectively than carbs or regular fuel injectors.

 

[Patrocles]

good idea! But as others have stated I, too, would like to see other items fixed first.

 

[niky]

Fuel Injection, by itself, won't fight power loss with altitude. Your car needs a MAP sensor to enable the ECU to raise the boost in response to thinner air. Without one, it will not be able cope with differences in air pressure.

O2 sensors also help change fuel trims in response to changing conditions... but I don't think they'd work fast enough to have an effect on a fast hill climb like Pikes' Peak... you'd simply just program the car to run pig rich at the bottom and just right at the top.

 

[Marf]

Fuel Injection, by itself, won't fight power loss with altitude. Your car needs a MAP sensor to enable the ECU to raise the boost in response to thinner air. Without one, it will not be able cope with differences in air pressure.

Most advanced ECUs(MOTEC, Solaris, Pectel) which are used in motorsport have two map sensors, one which measures barometric pressure, the other which measures the vacuum/boost condition the engine is experiencing. The delta of these two values is then used to apply modifiers to the ignition/fuelling/boost control maps stored in the ECU in an attempt to minimise power loss at altitude.

Of course the ultimate limiter here is the turbocharger, you can only increase boost so much to compensate for lower atmospheric pressure(and hence overall airflow) until the turbo is overspeeding and operating outside of its efficiency range.

O2 sensors also help change fuel trims in response to changing conditions... but I don't think they'd work fast enough to have an effect on a fast hill climb like Pikes' Peak... you'd simply just program the car to run pig rich at the bottom and just right at the top.

A decent wideband oxygen sensor can work fast enough to enable the ECU to make alterations based on the ratio it reads under low/medium loads, however running a boosted engine under high load in closed loop is still somewhat dangerous as yes, it may not respond fast enough to changeable conditions, hence open loop(i.e. using programmed values) is best.

The ideal for an ECU would be a system that was fast enough run both closed loop ignition and fuelling, where AF targets rather than injector opening times are programmed into the ECU which the ECU would then try and match based on sensory inputs. I believe Saab already have the technology available for closed loop ignition, the system uses pressure sensitive rings installed between the head and spark plug which measure the combustion pressure of each cylinder and can increase/decrease ignition advance to acheive MBT(minimum best timing) and also detect preignition/knock.

 

[Punknoodle]

Most advanced ECUs(MOTEC, Solaris, Pectel) which are used in motorsport have two map sensors, one which measures barometric pressure, the other which measures the vacuum/boost condition the engine is experiencing. The delta of these two values is then used to apply modifiers to the ignition/fuelling/boost control maps stored in the ECU in an attempt to minimise power loss at altitude.

Of course the ultimate limiter here is the turbocharger, you can only increase boost so much to compensate for lower atmospheric pressure(and hence overall airflow) until the turbo is overspeeding and operating outside of its efficiency range.



A decent wideband oxygen sensor can work fast enough to enable the ECU to make alterations based on the ratio it reads under low/medium loads, however running a boosted engine under high load in closed loop is still somewhat dangerous as yes, it may not respond fast enough to changeable conditions, hence open loop(i.e. using programmed values) is best.

The ideal for an ECU would be a system that was fast enough run both closed loop ignition and fuelling, where AF targets rather than injector opening times are programmed into the ECU which the ECU would then try and match based on sensory inputs. I believe Saab already have the technology available for closed loop ignition, the system uses pressure sensitive rings installed between the head and spark plug which measure the combustion pressure of each cylinder and can increase/decrease ignition advance to acheive MBT(minimum best timing) and also detect preignition/knock.

This man knows what he is talking about

 

[Marf]

Thanks

I should also add that my post was talking from the point of view of a turbocharged engine. There is very little that can be done for a naturally aspirated engine to minimise power losses at higher altitudes as they have no ancillary methods of being fed more air. All that can be done is to use the same barometric delta values to ensure the AFR and ignition timing are optimal as the amount of air decreases with altitude to ensure the engine is not running rich and wasting what power it can produce with the lower amount of oxygen available to it.

Also full closed loop is far easier to implement on a naturally aspirated engine as it will not experience anywhere near the same load and temperature differentials as a turbocharged engine does.

 

[niky]

My car actually runs closed loop all-the-time... O2 disconnected because my cams have so much overlap that the ECU keeps trying to enrichen the mixture way too much.

When I was talking about O2, I was talking about stock narrowband ones on street-cars, which are definitely not sensitive enough for those purposes.

Excellent post! We need more techy people like you on-board. Welcome to GTP!

 

[Marf]

Thanks

I've got some experience of the more basic "consumer" level ECU systems as I had one on my old car which I'd given the Gran Turismo treatment.