Boost and turbo lag not right

[nickg07]

Something that has annoyed me for a while in GT is the implementation of boost and in consequence turbo lag.

Now in GT, the amount of boost you receive when you press the throttle seems related to the throttle percentage but not ( and importantly ) mapped correctly to the rpms at which the accelerator is pressed.

Most cars only start to generate boost at a certain rpm, lower for diesels and modern cars. Higher for older more laggy motors.

For instance an MR2 Turbo will start to generate boost at say 2750-3000rpm with peak psi arriving at 4500rpm. In GT you can press the throttle half way and generate almost half its max psi boost at 1500-2000rpm in 4th gear, instantly as soon as you press the throttle half way and more by pressing the throttle further. This isn't really possible and neither is the instant movement of the boost meter with the thrittle pedal. It can take half a second after pressing the gas to see a real boost gauge start to move positively - actual lag - in GT it happens instantaneously.

Here is a skyline R34 MNP building a significant amount of boost with 1/2 throttle, in 4th gear at 900 rpm. No turbo is spooling on a 2000 car at less than 2000rpm.

Modern cars with better ECU and small turbos might be kicking in and starting to build boost at 1500-2000 rpm ranges and have a faster turbo response to throttle but the larger turbos on the older cars were not designed to operate in this way. This makes turbo cars in GT perform better against NA cars than they should be able to because the lag doesn't actually happen, you get instantaneous response from the turbo. And that's on top of receiving boost at RPMS that you shouldn't be able too, making your torque curve larger in effect.

 

[Flash_Sweden]

The game is not even close to simulate boost/turbolag in the right way.
If you are using a wheel with clutch, you can stall the engine. But before you start it up again, you can press down on the accelerator and see the boost increase.

 

[Griffith500]

There is merit in discussing whether the boost gauge actually represents anything the physics engine is using. Some believe it is just a graphic with its own superficial control. We have the data logger now, so I'm sure someone could come up with an experiment to test that.

Additionally, you can hear the boost building in the exhaust sound of the 97T, which is a nice demonstration of the expressivity of the method used. Why it would be in the sound and not the physics is not clear; it would be more understandable if it were in the physics and just not in the boost gauge.

 

[Tornado]

GT games generally do a pretty good job of modeling the power curve of a turbocharged engine (for stock vehicles, at least. GT2 was the last game where power tuning really changed the power curve); but even though the games have modelled the sounds of blowoff valves and turbo whine of a turbocharged engine from the first game they've never actually bothered to model the power delivery of a turbocharged engine. Hence why you can have a turbocharger that doubles the power output on an engine (mostly seen in the sub-100 HP cars, but there are a handful of more powerful vehicles with similar power gains), and the engine has a massively peaky torque number as a result; but you still have instantaneous throttle response no matter where you are in the rev range.


GT2, with its really loud turbo noises, is a particular offender. You could add something like 300 horsepower to a Skyline with a turbo alone, causing the torque peak to be at something like 5500 RPM, and when you drove it you heard the turbo s l o w l y build up boost; but the car still acted like it was on boost immediately.

 

[nickg07]

What the boost gauge does a good job of replicating is boost spikes, I.e if you bury the throttle it will over boost by a small percentage and then crawl back. And boost psi seems to reduce in some cars as you get to higher rpms showing the boost dropping off. This is really cool to watch.

It's just how and when the boost should build that they have seemingly done incorrectly if the boost gauge is a true reflection of how the physics work. Which I believe to be the case

 

[Master__Shake_]

Turbo lag is heavily based on natural power curve of the vehicle before the turbo is installed. Some cars have drastic changes to powerband, others have minimal. To expect all cars to have massive turbo lag simply because of a large turbo installed is expecting all motors to act the same. Small motors with large turbos will spool slower than large motors with large turbos. If you expect the boost gauge of the game to exert true boost numbers, you may be disappointed.

Anyone with any knowledge of internal combustion motors will tell you that, in order to determine horsepower output, you need a mathematical formula for RPM & torque. In that formula, torque & horsepower intersect at 5252 RPM, on all internal combustion motors. Very rarely do you see this on the power chart in this game, so how can you expect a simple gauge next to the speedo cluster to be accurate.

You're more likely to estimate boost output by the powerband shown on a given car by the under/over-the-peak curve. The Diablo GT/GT2 have heavy turbo lag, which is demonstrated by incredible rate of power increase after 4k RPM. If you try placing this car in a high gear, it will not go faster than ~7MPH from a dead stop. Other cars, like certain WRX STIs will have medium peak power and slow arching power loss before and up to 2k RPM after peak, even with a high-range turbo. This gives them a torquey off-the-line response from the motor itself, not just the power from the turbo, then slowly loses power after peak. Because of these natural power curves, certain cars prefer medium turbo over high turbo.

I believe the lack of turbo lag is to do with gearing & car specifications rather than a game fault, but the boost gauge itself is most likely showing much higher than it actually is.

 

[Griffith500]

...

Anyone with any knowledge of internal combustion motors will tell you that, in order to determine horsepower output, you need a mathematical formula for RPM & torque. In that formula, torque & horsepower intersect at 5252 RPM, on all internal combustion motors. Very rarely do you see this on the power chart in this game, so how can you expect a simple gauge next to the speedo cluster to be accurate.

...

That probably just means the graphs aren't drawn using lbf.ft and bhp, or they use different zeros.

Also, you can't accurately guess turbo lag from looking at the torque curve. Lag is how long it takes the turbo-charger to spool up to its operating speed for a given (constant) engine speed. What you're describing is more accurately known as "boost threshold"; supercharged cars can do the same, and mega-cammed NA engines, also.

The GT40 has horrendous "lag".

 

[Master__Shake_]

That probably just means the graphs aren't drawn using lbf.ft and bhp, or they use different zeros.

Also, you can't accurately guess turbo lag from looking at the torque curve. Lag is how long it takes the turbo-charger to spool up to its operating speed for a given (constant) engine speed. What you're describing is more accurately known as "boost threshold"; supercharged cars can do the same, and mega-cammed NA engines, also.

The GT40 has horrendous "lag".

The problem with their power chart is that its not graphed correctly and therefore shows torque nearly as high as horsepower when, in many cases, its several hundred lower. I happen to have desktop dyno on my laptop, and could accurately chart torque & power output of the cars on this game if I have the specs of the intakes, cams, carbs/throttle bodies, and forced induction measures as well as the cubic inch of the block & compression ratios. And yes, dyno charts do accurately show turbo lag from slight throttle increases because power output shows when boost kicks in and how much it kicks in with fuel/air ratio.

When comparing GT6 to real life, turbo lag is from off-idle to part or full throttle, but never really off-idle on this game unless you're several gears too high for the speed you're travelling. If you have a stock transmission car in 6th gear with stock power, you will have very similar pull numbers to real life. Once you start throwing on parts & turbo kits, the numbers radically change and therefore need higher ratios to simulate the same thing. Shorter gears eliminate both turbo lag & shift lag, which would be the same as real life but they don't have custom ratio transmissions you can run on the street. Once the turbo loses too much exhaust pressure, it will slow and lose boost, which just so happens during large gaps in gears during a shift. If you're circuit racing and the RPMs never drop below 7k, there will be no noticeable turbo lag because it never falls below the threshold.

That being said, I only have a DS3 and cannot compare with wheel users who have a clutch. For DS3, the RPMs are always fixed for what gear you're in, and never off-idle. The wheel drivers may experience significantly more turbo lag than others, or may not.

 

[Griffith500]

The problem with their power chart is that its not graphed correctly and therefore shows torque nearly as high as horsepower when, in many cases, its several hundred lower. I happen to have desktop dyno on my laptop, and could accurately chart torque & power output of the cars on this game if I have the specs of the intakes, cams, carbs/throttle bodies, and forced induction measures as well as the cubic inch of the block & compression ratios. And yes, dyno charts do accurately show turbo lag from slight throttle increases because power output shows when boost kicks in and how much it kicks in with fuel/air ratio.

When comparing GT6 to real life, turbo lag is from off-idle to part or full throttle, but never really off-idle on this game unless you're several gears too high for the speed you're travelling. If you have a stock transmission car in 6th gear with stock power, you will have very similar pull numbers to real life. Once you start throwing on parts & turbo kits, the numbers radically change and therefore need higher ratios to simulate the same thing. Shorter gears eliminate both turbo lag & shift lag, which would be the same as real life but they don't have custom ratio transmissions you can run on the street. Once the turbo loses too much exhaust pressure, it will slow and lose boost, which just so happens during large gaps in gears during a shift. If you're circuit racing and the RPMs never drop below 7k, there will be no noticeable turbo lag because it never falls below the threshold.

That being said, I only have a DS3 and cannot compare with wheel users who have a clutch. For DS3, the RPMs are always fixed for what gear you're in, and never off-idle. The wheel drivers may experience significantly more turbo lag than others, or may not.

I think you're using "lag" when you don't actually mean anything to do with a time delay. By your definition, a 2-stroke lags like hell. Turbo lag is a dynamic phenomenon; a torque curve is static, steady-state (certainly in terms of throttle position, which is the important control regarding lag).

Read this. I think you're thinking of "boost threshold".


From your description, the power / torque graphs just use different zeros / scales for the two curves; it's not that difficult to adjust to, just the same as different units, literally.

Now, back to the lag!

 

[DolHaus]

I've noticed some spool up time on some stock cars but never any lag or throttle response delays.
All the custom turbos feel sort of like a variable vane unit, the power delivery is very smooth and not peaky as you would expect with a High-Rpm setup.

 

[Tornado]

All the custom turbos feel sort of like a variable vane unit, the power delivery is very smooth and not peaky as you would expect with a High-Rpm setup.

Basically, the issue with that is that all the turbocharging upgrade does in GT5 and GT6 is apply a power multiplier on a specific "third" of the rev range (as opposed to the NA tune modifications, which apply the power multiplier on the entire thing +100 rpm). It doesn't really change the characteristics of the car at all outside of that rev range it acts on, like would happen if you replaced an OEM turbo with a larger one (which would usually make it so the engine had to rev higher to get the turbo to act); nor does it model how you might need to lower compression in real life so the engine would produce less power out of the RPM the turbo activates on.

It basically just makes the the entire RPM range more powerful, but a specific part is made substantially more powerful than the rest; so when combined with the absence of turbo lag it's sorta more like they modeled a really efficient supercharger than a turbo. I think I'll whip a graph together to show the differences.

 

[Master__Shake_]

I think you're using "lag" when you don't actually mean anything to do with a time delay. By your definition, a 2-stroke lags like hell. Turbo lag is a dynamic phenomenon; a torque curve is static, steady-state (certainly in terms of throttle position, which is the important control regarding lag).

Read this. I think you're thinking of "boost threshold".


From your description, the power / torque graphs just use different zeros / scales for the two curves; it's not that difficult to adjust to, just the same as different units, literally.

Now, back to the lag!

If you're going by the vague definition of hesitation of throttle response, then not this game nor any current racing games have accurate turbo lag simulation. Throttle hesitation happens for a number of reasons and isn't exclusive to turbocharged applications. I was spending my time, in vain, explaining why turbo lag happens and how it affects the powerband, but I see you're not interested in how PD actually implemented it into the game.

The reason why they couldn't implement real hesitation to throttle is that people would complain non-stop about how the game is broken and doesn't respond to controls, have to hand-replot throttle response for each car they painstakingly made lag for, then implement them while fixing current bugs and adding new content. I don't blame them for excluding it, it would've been time wasted. At the same time, it makes low and most mid-range turbos obsolete for purposes of avoiding turbo lag; this game is becoming more and more for the casual, arcade-y gamer, so no surprise there.

 

[Tornado]

If you're going by the vague definition of hesitation of throttle response

That "vague definition" is literally the definition. The time it takes for the turbo to provide full boost pressure on WOT application while in the RPM range that the turbo operates.

I was spending my time, in vain, explaining why turbo lag happens and how it affects the powerband, but I see you're not interested in how PD actually implemented it into the game

Because they didn't implement it into the game, and you're not describing turbo lag, best represented by this bit here:

The Diablo GT/GT2 have heavy turbo lag, which is demonstrated by incredible rate of power increase after 4k RPM.

Where you described a normally aspirated car as suffering from it.

 

[nickg07]

What it comes down to is that responsiveness to throttle I agree. For instance NA car there's usually and intake pipe-> throttle body. With FI there's an intake pipe to turbo compressor and more pipe, intercooler, pipe and then throttle body. The compressor will only increase in velocity when more exhaust gas is thrown out of the engine and through the turbo. The larger the turbo the longer the spool time. And the slower the intake compressor responds to feeding the exhaust gasses the slower the whole process is.

I feel in the media like on top gear with the evo fq400 they tried to show lag as being a function of how long it takes you in a high gear to get the turbo to spin up from low rpm. So sit in 4th at 20mph and put your foot down and complain about low acceleration until you hit the rpm where the turbo spools.

PD need to remodel it imo. They need to understand that an engine has a power curve and that the turbo sits on top of that curve. For instance at 3000rpm a 2 litre non turbo could be 100bhp at WOT. with a turbo on the same car at the same 3000rpm at wot it could be anywhere from 100-150bhp depending on whether the turbo boost pressure. You don't always have 150bhp atw as soon as you press the gas. If youd been at WOT since 2000rpm you might be at full boost. If you were down or up shifting ime off the gas and then floored it you wouldn't have 150bhp right there and then at the same rpm.

 

[Master__Shake_]

That "vague definition" is literally the definition. The time it takes for the turbo to provide full boost pressure on WOT application while in the RPM range that the turbo operates.

Because they didn't implement it into the game, and you're not describing turbo lag, best represented by this bit here:

Where you described a normally aspirated car as suffering from it.

It is "vague" because throttle hesitation occurs for many other reasons than turbo lag. Cars with single plane intakes & carburetors have input lag because of lack of low-end grunt, building most power high in the band. Same goes with large plenums on fuel-injected cars, although fuel injection suffers less from it. Other reasons can include lean or rich fuel/air mixture, bad throttle linkage, flawed ECU management, leaking injectors/carb jets, etc. Many cars in this game, especially cars over 20 years old, will have one reason or another to have hesitation on off-idle acceleration. The simple fact is that this is a game, and PD doesn't have the time or possibly the ability to implement all real-life physics to a driving simulator; which is not a mechanical simulator.

I should have been specific when using the GT/GT2 as an example: They have incredible lack of low-end power when equipped with high range turbo. I shouldn't have assumed that everyone uses modified cars, because many circuit guys prefer driving stock equipped cars and not the heavily upgraded ones I'm used to in the drag community. That was my mistake.

 

[332i]

The torque curve between different turbochargers is pretty obvious. You can also see the boost gauge taper off at higher RPMs with various torque curves. It's definitely not a freewheeling infographic. Maybe on some cars, but for the most part...seems somewhat correct. Some cars also put out more boost and that can be noted on the gauge as well. Just because it's not a dramatic level of lag, does not mean it is incorrect. A lot of the stock, older premium cars have wicked boost lag. With the ECUs and turbo kits, a lot of the lag gets eliminated...which is what we want...

If you're using a dog box, you're not going to experience much, if any boost lag...which is how it should be. On the other hand, using a stock trans makes for lag, which is also how it should be. The less time between shifts, the less the turbo spools down. I don't think it is as flawed as you think. Not across the board, at least.

All the turbo kits provide the most torque where they say they do and that is pretty evident when using pedals.

 

[Griffith500]

Again, turbo lag has nothing expressly to do with the torque curve. You might be able to make guesses, but they won't be reliable in all cases.

An example: if you're motoring the engine off-throttle at 5000 rpm, the moment you open the throttle fully, there will be no boost, even if the car makes peak torque (i.e. peak boost) there. It takes time for exhaust gas to spin up the turbine, and hence accelerate the compressor, which pushes more air into the engine, which produces more exhaust gas, which accelerates the turbine further still, and so on, creating a feedback cycle of increasing turbo-compressor shaft speed, and hence boost, until it is checked somehow (e.g. wastegate, friction, gas leak through hole in engine block, surge).

That is turbo lag; it's a dynamic consideration that a torque curve cannot tell you about unless the torque curve is taken from a dynamic test (i.e. a fast ramp). But it's not perfect even then; you'd have to be a skilled water brake dyno tester to get that info reliably. Easier would be to do it on the road, possibly, where it matters.
People say of the F40 that you can jab the throttle mid-corner, which overwhelms the rears even off-boost, so you feather the throttle slightly, but the boost still builds anyway (mechanical boost control). So,in effect, you've reduced your throttle input, but you get more power anyway because the engine is taking time to build up to the steady-state balance for the control inputs supplied. How does the torque curve account for that behaviour?


How do flat spots figure into this backward definition of lag? E.g. the engine pulls strongly up to, say, 5500rpm, then it bogs until 6500 rpm whereupon it takes off again on account of making peak torque there. (I encountered an engine just like this recently.) Is that lag, or is it just a flat spot? What about electric motors - do they "lag" at high speed instead? Or how about opening the throttle too fast on a carburettor, "skipping" the progression jets and bogging it that way? Is that lag, or is it bogging? People use different words for a reason.


It seems that when people say "lag", they usually just mean "power band".

"Because the car lags behind my expectations of acceleration when I floor it in a high gear at low speed..."

 

[Tornado]

It is "vague" because throttle hesitation occurs for many other reasons than turbo lag

It's not vague in a discussion talking about turbo lag and whether or not it is modeled in the game when people keep referring to boost threshold when talking about it instead, like right here:

I should have been specific when using the GT/GT2 as an example: They have incredible lack of low-end power when equipped with high range turbo

Where you are again talking about a car having a really high boost threshold; on a car that has very little low end power to begin with before you add the turbo that only acts on the high RPM.


A carburetor that isn't running right, or a lazy throttle tip in like a 1970s Cadillac, or a plugged air filter or whatever is going to have very different feel from a car with a laggy turbo even if both can be described as throttle hesitation; particularly since something like a Thunderbird Turbo Coupe is always going to have the obvious delayed off/on sensation when you get on the gas but throttle response on an engine with a carb might improve as the engine warms up.

The torque curve between different turbochargers is pretty obvious. You can also see the boost gauge taper off at higher RPMs with various torque curves. It's definitely not a freewheeling infographic.

The boost gauge does correspond to what the turbo in the car should be doing (assuming the turbo was an unrealistically efficient one) in terms of where it acts on the RPM range and where it bleeds off. It just doesn't correspond to what the turbo in the car is actually doing, because for all intents and purposes there is no turbo modeled. Just a power multiplier. Even Forza 4, while doing a much better job showing boost buildup (visually through the boost guage slowly building, and aurally) and having a twin turbo vs. a single turbo upgrade system, mostly just makes it so the twin turbo system acts on an earlier RPM than actually reducing lag.

 

[332i]

Again, turbo lag has nothing expressly to do with the torque curve. You might be able to make guesses, but they won't be reliable in all cases.

An example: if you're motoring the engine off-throttle at 5000 rpm, the moment you open the throttle fully, there will be no boost, even if the car makes peak torque (i.e. peak boost) there. It takes time for exhaust gas to spin up the turbine, and hence accelerate the compressor, which pushes more air into the engine, which produces more exhaust gas, which accelerates the turbine further still, and so on, creating a feedback cycle of increasing turbo-compressor shaft speed, and hence boost, until it is checked somehow (e.g. wastegate, friction, gas leak through hole in engine block, surge).

That is turbo lag; it's a dynamic consideration that a torque curve cannot tell you about unless the torque curve is taken from a dynamic test (i.e. a fast ramp). But it's not perfect even then; you'd have to be a skilled water brake dyno tester to get that info reliably. Easier would be to do it on the road, possibly, where it matters.
People say of the F40 that you can jab the throttle mid-corner, which overwhelms the rears even off-boost, so you feather the throttle slightly, but the boost still builds anyway (mechanical boost control). So,in effect, you've reduced your throttle input, but you get more power anyway because the engine is taking time to build up to the steady-state balance for the control inputs supplied. How does the torque curve account for that behaviour?


How do flat spots figure into this backward definition of lag? E.g. the engine pulls strongly up to, say, 5500rpm, then it bogs until 6500 rpm whereupon it takes off again on account of making peak torque there. (I encountered an engine just like this recently.) Is that lag, or is it just a flat spot? What about electric motors - do they "lag" at high speed instead? Or how about opening the throttle too fast on a carburettor, "skipping" the progression jets and bogging it that way? Is that lag, or is it bogging? People use different words for a reason.


It seems that when people say "lag", they usually just mean "power band".

"Because the car lags behind my expectations of acceleration when I floor it in a high gear at low speed..."

When you put your foot down to the floor, the torque curve is dependent upon turbo lag. At the same time, the turbo spool is dependent on where you are in the motors NA torque curve. So, tell me again how they not correlated. It may not make peak torque at max boost for said package and settings, but that is due to combining various turbos with various platforms. I didn't read much of your post, but by the looks of it, you're over complicating things and interneting too hard. Turbo lag has everything to do with the torque curve. Trim and IDs...you're under the impression that these do not affect or create a torque curve in conjunction with engine timing and displacement? For real? That's what your first sentence says.

A turbo dramatically shifts torque curves. It has everything to do with torque curves. If you have a decent curve to begin with and pair it to the right turbine, you're going to have less lag. Example - BMW N54. Max boost at 1500rpm, nearly zero lag upon getting back on the throttle. The smoother the curve on the motor with a turbo...with the CORRECT turbine, the less lag you will produce. So, yes. Lag has everything to do with torque curves.

Not only is boost dependent on it, but it also modifies it.

 

[332i]

It's not vague in a discussion talking about turbo lag and whether or not it is modeled in the game when people keep referring to boost threshold when talking about it instead, like right here:

Where you are again talking about a car having a really high boost threshold; on a car that has very little low end power to begin with before you add the turbo that only acts on the high RPM.


A carburetor that isn't running right, or a lazy throttle tip in like a 1970s Cadillac, or a plugged air filter or whatever is going to have very different feel from a car with a laggy turbo even if both can be described as throttle hesitation; particularly since something like a Thunderbird Turbo Coupe is always going to have the obvious off/on sensation when you get on the gas but throttle response on an engine with a carb might improve as the engine warms up.





The boost gauge does correspond to what the turbo in the car should be doing (assuming the turbo was an unrealistically efficient one) in terms of where it acts on the RPM range and where it bleeds off. It just doesn't correspond to what the turbo in the car is actually doing, because for all intents and purposes there is no turbo modeled. Just a power multiplier. Even Forza 4, while doing a much better job showing boost buildup (visually through the boost guage slowly building, and aurally) and having a twin turbo vs. a single turbo upgrade system, mostly just makes it so the twin turbo system acts on an earlier RPM than actually reducing lag.

Hm. Interesting. This make a lot of sense - the turbines not being modelled. That's the answer for sure. All good points, sir.

 

[DolHaus]

Lets just make sure we're on the same page when using terms (not aimed at anyone, just trying to avoid anyone getting hurt feelings)

Turbocharger lag

Turbocharger lag ("turbo lag") is the time required to change power output in response to a throttle change, noticed as a hesitation or slowed throttle response when accelerating as compared to a naturally aspirated engine. This is due to the time needed for the exhaust system and turbocharger to generate the required boost. Inertia, friction, and compressor load are the primary contributors to turbocharger lag. Superchargers do not suffer this problem, because the turbine is eliminated due to the compressor being directly powered by the engine.

Turbocharger applications can be categorized into those that require changes in output power (such as automotive) and those that do not (such as marine, aircraft, commercial automotive, industrial, engine-generators, and locomotives). While important to varying degrees, turbocharger lag is most problematic in applications that require rapid changes in power output. Engine designs reduce lag in a number of ways:

• Lowering the rotational inertia of the turbocharger by using lower radius parts and ceramic and other lighter materials
• Changing the turbine's aspect ratio
• Increasing upper-deck air pressure (compressor discharge) and improving wastegate response
• Reducing bearing frictional losses, e.g., using a foil bearing rather than a conventional oil bearing
• Using variable-nozzle or twin-scroll turbochargers
• Decreasing the volume of the upper-deck piping
• Using multiple turbochargers sequentially or in parallel
• Using an antilag system
• Using a turbocharger spool valve to increase exhaust gas flow speed to the (twin-scroll) turbine

Boost threshold

The boost threshold of a turbocharger system is the lower bound of the region within which the compressor operates. Below a certain rate of flow, a compressor produces insignificant boost. This limits boost at a particular RPM, regardless of exhaust gas pressure. Newer turbocharger and engine developments have steadily reduced boost thresholds.

Electrical boosting ("E-boosting") is a new technology under development. It uses an electric motor to bring the turbocharger up to operating speed quicker than possible using available exhaust gases.[24] An alternative to e-boosting is to completely separate the turbine and compressor into a turbine-generator and electric-compressor as in the hybrid turbocharger. This makes compressor speed independent of turbine speed. In 1981, a similar system that used a hydraulic drive system and overspeed clutch arrangement accelerated the turbocharger of the MV Canadian Pioneer (Doxford 76J4CR engine).[citation needed]

Turbochargers start producing boost only when a certain amount of kinetic energy is present in the exhaust gasses. Without adequate exhaust gas flow to spin the turbine blades, the turbocharger cannot produce the necessary force needed to compress the air going into the engine. The boost threshold is determined by the engine displacement, engine rpm, throttle opening, and the size of the turbocharger. The operating speed (rpm) at which there is enough exhaust gas momentum to compress the air going into the engine is called the "boost threshold rpm". Reducing the "boost threshold rpm" can improve throttle response.

 

[Ridox2JZGTE]

I'll just pretend that all turbo kit upgrades in GT6 is a perfectly setup turbo + intercooler - best possible setup - boost + ECU tuning ( don't know why GT6 do not allow intercooler upgrades like in older GT )
Until GT allows custom turbo + intercooler kit ( single, double, triple, parallel, sequential, wheel trim, turbine housing A/R and size, internal & external wastegate + size, intercooler size + air to air/water to air, tube&fin, bar & plate, delta fin/hybrid fin - affects efficiency, power+boost level etc, BOV with different sounds, compression ratio + boost adjustment, anti lag, and detailed ECU tuning - A/F ratio tuning : rich/lean, timing, boost level on various rpm range, overboost scramble switch, etc ), there won't ever be accurate turbocharging phyiscs, also engine overheating feature and ambient temp + altitude affecting turbo + intercooler performance.

I really want the overboost function like in Tokyo Extreme Racer - gotta love 30s 2 bar full boost for the long final straight side by side, blow the engine at the finish line

 

[Master__Shake_]

Where you are again talking about a car having a really high boost threshold; on a car that has very little low end power to begin with before you add the turbo that only acts on the high RPM.

The discussion wasn't originally aimed at you, and I have nothing to prove to you, but Griffith said that powerband has nothing to do with turbo lag. The Lamborghini was one example of a high RPM car that takes a long time to build power because of both high threshold and horrendous low-end power. A motor with low or mid RPM power, possibly a diesel, will make high exhaust pressure at low RPM and have very little turbo lag - even with a large turbo setup and the complete opposite side of the spectrum from the Lambo.

To detour slightly, I would like to add that the supercharging in this game is pathetically underwhelming. They've turned an alternative forced induction method into a simple bolt-on like intake or exhaust swapping. Why not provide multiple pulley setups, much like ranges for turbos, to incur higher torque at low RPMs and more horsepower throughout the powerband.

 

[Griffith500]

The discussion wasn't originally aimed at you, and I have nothing to prove to you, but Griffith said that powerband has nothing to do with turbo lag. The Lamborghini was one example of a high RPM car that takes a long time to build power because of both high threshold and horrendous low-end power. A motor with low or mid RPM power, possibly a diesel, will make high exhaust pressure at low RPM and have very little turbo lag - even with a large turbo setup and the complete opposite side of the spectrum from the Lambo.

To detour slightly, I would like to add that the supercharging in this game is pathetically underwhelming. They've turned an alternative forced induction method into a simple bolt-on like intake or exhaust swapping. Why not provide multiple pulley setups, much like ranges for turbos, to incur higher torque at low RPMs and more horsepower throughout the powerband.

No, I said the converse, actually. There's a difference.

The Lamborghini is NA; how can it have turbo lag?


As for the simplicity of the supercharging, the situation with the turbos is worse; partly because of the absence of real, dynamic lag, which crank-driven superchargers don't suffer from (unless you count belt slippage).

This thread is about turbo lag.

When you put your foot down to the floor, the torque curve is dependent upon turbo lag. At the same time, the turbo spool is dependent on where you are in the motors NA torque curve. So, tell me again how they not correlated. It may not make peak torque at max boost for said package and settings, but that is due to combining various turbos with various platforms. I didn't read much of your post, but by the looks of it, you're over complicating things and interneting too hard. Turbo lag has everything to do with the torque curve. Trim and IDs...you're under the impression that these do not affect or create a torque curve in conjunction with engine timing and displacement? For real? That's what your first sentence says.

A turbo dramatically shifts torque curves. It has everything to do with torque curves. If you have a decent curve to begin with and pair it to the right turbine, you're going to have less lag. Example - BMW N54. Max boost at 1500rpm, nearly zero lag upon getting back on the throttle. The smoother the curve on the motor with a turbo...with the CORRECT turbine, the less lag you will produce. So, yes. Lag has everything to do with torque curves.

Not only is boost dependent on it, but it also modifies it.

So a small cast iron turbine running in plain bearings, bolted on a low-speed torquey big lump of an engine will always have less lag than a roller-bearing exotic metal assembly on a peaky, small displacement engine running at high speeds, when both engines are at peak torque rpm? No, the torque curve is less than half of the lag equation, when most of it is the throttle position history, and half of the rest is the mechanical design of the rotating parts and their housings.

Obviously turbine "sizing" is important in determining the torque curve, at steady state, but you can "size" a turbo-compressor from the thermodynamics alone, without even considering whether that turbine will spin up in a reasonable time. And that time taken in "spinning up" is lag, regardless of engine speed and torque.

By the way, if you're not going to read my posts, don't bother replying.

 

[Ridox2JZGTE]

Fitting Greddy T88 SPL 34D single on a Lamborghini Gallardo, run close to 2 Bar, set AFR to very rich, then do rolling drag from 60kmh on 2nd gear - will be hilarious

 

[Master__Shake_]

No, I said the converse, actually. There's a difference.

Again, turbo lag has nothing expressly to do with the torque curve.

..Ok.

I already discussed the reason why I used the Diablo as an example. Maybe the Escudo would be a better example?

 

[nickg07]

Also the rate that boost builds and even it's total is linked to the gear you're in. For example a 3sgte in an mr2 turbo might not reach the peak 0.9bar boost in first gear (that's with tsv disconnected to remove 1st gear boost limit) because its not flowing enough and actually you can build 0.9 in the other gears due to higher engine load.

What GT doesn't replicate is that boost would say take longer to maximise at 40mph in 2nd compared to 3rd gear. In GT6 they both hit boost peak as soon as you press full power. As boost bis generated depending on load and rpm.

 

[Griffith500]

..Ok.

I already discussed the reason why I used the Diablo as an example. Maybe the Escudo would be a better example?

I know how torque creates acceleration.

What we're discussing in this thread is turbo lag. Engineers call it the transient response:

The knowledge of steady performances is not sufficient to estimate transient performances, especially for supercharged engines, because they depend on dynamic phenomena which are difficult to take into account.

turbo-moteurs.cnam.fr/publications/pdf/ASME99.PDF

Turbocharger lag is the most notable feature of diesel engine transient operation that drastically differentiates the torque pattern from the respective steady-steady [sic] conditions.

link.springer.com/chapter/10.1007%2F978-1-84882-375-4_6

Engineers believe that the steady-state torque curve is not a great indicator of turbo lag. Perhaps turbo lag doesn't mean what you believe it should mean.


Turbo lag is a transient / dynamic concern, aside from the "power band". GT does not model this, so far as has been ascertained. Other games have attempted it, but nothing (I've seen) is "perfect".
I have not attempted to model it myself, so I don't know the precise difficulties, but I'm sure any model developed for turbo engines would help with the dynamic modeling of NA / mechanically supercharged engines also.

 

[Exorcet]

The reason why they couldn't implement real hesitation to throttle is that people would complain non-stop about how the game is broken and doesn't respond to controls,

Too bad for them, they're wrong. Also easy enough to tell them to their face they're wrong with a license test or something.

have to hand-replot throttle response for each car they painstakingly made lag for,

What? They could just use any old model. Something like If aspiration is turbo, then lag = .03 s. Not a great model, but it wouldn't be by hand. I'd argue that almost anything is better than now* where lag doesn't exist.

* I don't own GT6, so I'm going off GT5 and before

then implement them while fixing current bugs and adding new content.

This is what they're going to be doing anyway.

I don't blame them for excluding it, it would've been time wasted.

How is it time wasted to model something rather important for one entire class of engine that is easily felt from the driver's seat? It's a simulator, it is a complete joke that the game turned 15 before lag was added.

 

[nickg07]

They don't need to map every car to such a granular level. But they must be able to ball park say, 80s turbos might have more lag than a 2014 turbo. And an 90s turbo somewhere in between. If there's 2 turbos that operate sequentially will have less lag. Parrallel turbo will have more lag but amazing boost when it does hit.