No, my point is that physics in all aspects has to progress at the same rate because certain bits of the physics rely upon others to be acceptable. Therefore it doesn't make sense to make the turbo physics perfect before the tyre physics because the turbo physics relies on the tyre physics. I think that's fair enough, seing as tyre physics updates are on the way.

I'll buy that, no problem....

And my only point was that perhaps a problem exists, one that people don't talk about very often. That's all...

Quote from axus :

What you refer to as the bleeder valve - haven't found the proper name for it - could open more and more as revs build up until maximum boost is reached so that the car isn't as snappy to drive and you get a smooth toque curve. Is that good enough?

Bleeder valves (or bleed valves) are used to cheat the wastegate by bleeding out some of the air going into the wastegate actuator so that the wastegate opens on a bit higher boost levels.

On my engine, a VW 1.8T, the boost buildup is incredibly rapid at low pressures, and gradually ease into it's stable position. I believe this is a function of the ECU, as it controls the wastegate on this engine. This turbo (K3) can easily put out double the boost it's allowed by the ECU, and thus the buildup should be virtually instant. The fact that the K3 turbine is tiny does not affect the overall behaviour of the boost buildup, other than being faster overall. I believe if the ECU left the turbine and wastegate on their own, I'd see the accelerating buildup BBT was expecting.

Having said that, I somehow doupt Scawen has coded in ECU-controlled curves for the wastegate position like VW does. Thus I have no idea why the RAC does this, I only know that it can and does happen in real life.

Quote from Hyperactive :

The turbo in RA is barely noticeable for me. It feels almost NA...

Yeah, I get that feeling with nearly every turbo car in LFS. If anyone has played with Netkar, drive the Supra- it has some of the best turbo modeling I have ever seen in a sim. More importantly with the way boost builds and the way the behaviour of the car changes when the boost comes up. I've taken a ride in a street Supra (not anything like the GTR Supra in NetKar) that had an upgraded single turbo, boost maxed at 20 lbs at probably 2600 rpm and held there all the way to redline. Car was pretty fast too. I've also driven many 1.8T VW's (I used to work for VW) some stock, some chipped, and some with upgraded turbos, and the boost responce was nearly the same in all of them, although the upgraded turbo car was a bit more "laggy" down low, it still had nice mid torque but when the turbo came into full song there was that "kick in the back" feeling that hit much harder than the stock turbos did.

I would imagine turbo modeling in a sim would be a very difficult thing to do, especially on cars that are not really based on real-life counterparts with real-world data to compare them to. The exception of the RAC, and from what I have read the engine in that car is nearly identical to the Saab "Viggen" engine which had very quick turbo responce. So who knows.

Brendan

Quote from atledreier :

On my engine, a VW 1.8T, the boost buildup is incredibly rapid at low pressures, and gradually ease into it's stable position. I believe this is a function of the ECU, as it controls the wastegate on this engine. This turbo (K3) can easily put out double the boost it's allowed by the ECU, and thus the buildup should be virtually instant. The fact that the K3 turbine is tiny does not affect the overall behaviour of the boost buildup, other than being faster overall. I believe if the ECU left the turbine and wastegate on their own, I'd see the accelerating buildup BBT was expecting.

Having said that, I somehow doupt Scawen has coded in ECU-controlled curves for the wastegate position like VW does. Thus I have no idea why the RAC does this, I only know that it can and does happen in real life.

Interesting. When you say rapid building at low pressures, how low are you talking?

The phenomenon you've transcribed here is still different from the RAC, where the boost gauge needle pretty much moves in unison with the tach... and you need to shift pretty much the instant max pressure (which is quite low on that car) is acheived.. Which is what seems odd to me (talking about roll on from any speed or standing start)

With the K3 turbo being so small, I can get full ECU-allowed boost (.55 Bar) from 1400 rpm. If I go WOT at 1800rpm, I have full boost in about 2 seconds, but the boost build pretty fast in the beginning, and slower as I approach full boost. At higher rpm the boost hits full very fast, less than half a second, so it's hard to make out just how the curve is. I'd go film it, but it's dark out so it'd probably just be a blurred mess.

Edit: I've been looking at the turbo behaviour in LFS a little closer and made a few observations. It seems to me turbo spool time is a direct function of throttle position and rpm, and very linear. In real life the low exhaust pressure at low rpm would inhibit any rapid spool, which in turn inhibit a rapid increase in exhaust pressure, making the whole process pretty slow, which is the essence of turbo-lag in the first place.
The other cars seem more natural to me, but they have the power- peak really really high in the rpm band. But then they are race cars, and suppsed to have pointy torque curves.

Quote from Lord_Verminaard :

...boost maxed at 20 lbs at probably 2600 rpm and held there all the way to redline......I would imagine turbo modeling in a sim would be a very difficult thing to do, especially on cars that are not really based on real-life counterparts with real-world data to compare them to. The exception of the RAC, and from what I have read the engine in that car is nearly identical to the Saab "Viggen" engine which had very quick turbo responce. So who knows.

Right, unless it's a drag motor I don't understand this idea of not reaching max boost unless I'm over 4K, wouldn't seem like a drivable street car
but again I am no expert which is why I am asking all these questions...

Quote from atledreier :

With the K3 turbo being so small, I can get full ECU-allowed boost (.55 Bar) from 1400 rpm. If I go WOT at 1800rpm, I have full boost in about 2 seconds, but the boost build pretty fast in the beginning, and slower as I approach full boost. At higher rpm the boost hits full very fast, less than half a second, so it's hard to make out just how the curve is. I'd go film it, but it's dark out so it'd probably just be a blurred mess.

Edit: I've been looking at the turbo behaviour in LFS a little closer and made a few observations. It seems to me turbo spool time is a function of throttle position and rpm, and very linear.

Cool now the discussion is actually going somewhere....

When it's light out sometime I'd love to see your film!

And good point about the linearity of what happens in LFS, that's exactly what I was getting at in my last post.

Quote from atledreier :

In real life the low exhaust pressure at low rpm would inhibit any rapid spool, which in turn inhibit a rapid increase in exhaust pressure, making the whole process pretty slow, which is the essence of turbo-lag in the first place.
The other cars seem more natural to me, but they have the power- peak really really high in the rpm band. But then they are race cars, and suppsed to have pointy torque curves.

Turbochargers are driven more via heat and the pressure differentials created through the restriction on the turbine (exhaust) side. Exhaust pressure in terms of kinetic energy is a tiny fraction of what spools turbochargers up. In essence they run on heat... And when you open the throttle all the way even at low RPM there is magnitudes more heat dumped into the exhaust manifold, which is why they function more the opposite of what you stated and are referred to as "positive feedback devices". As soon as pressure starts to build, serious amounts of heat are generated along with massive cylinder pressures, accounting for the copious amounts of torque generated. More heat causes a faster spool time not a slower one...

The response on the GTR cars I can't comment on, they might just be accurate since they are tiny motors making so much power...I don't know, but the road cars should have much more boost available more evenly through the RPM band.

Snaps to Ball Bearing for generating such an informative thread, and to all those replies.

I can throw away the rest of the internet now and learn everything I need to know right here
I guess we need to keep in mind that LFS is a work in progress......so any deficiencies, whether real or percieved, could simply be a result of unfinished or un-tweaked modelling.
And besides, the RAC is the offspring of Satan, I would rather drive a UF1000 in a 6 hr enduro than spend 10 min in a RAC

Quote from Sapient :

Snaps to Ball Bearing for generating such an informative thread, and to all those replies.

I can throw away the rest of the internet now and learn everything I need to know right here
I guess we need to keep in mind that LFS is a work in progress......so any deficiencies, whether real or percieved, could simply be a result of unfinished or un-tweaked modelling.
And besides, the RAC is the offspring of Satan, I would rather drive a UF1000 in a 6 hr enduro than spend 10 min in a RAC

Thanks!

I think this topic is cool but considering the lack input by many of the local physics gurus (bob? tristan? TODD? skiingman? and others...) perhaps it's only me that finds it fascinating.... Oh well

I'd love to come and chat, but as I don't regularly drive turbo cars, especially ones with a boost gauge, I can't say how it's meant to behave on a general basis, but it sounds as though you do.

I'll keep an eye on this thread, but unless you seriously balls it up, I doubt my limited knowledge will help.

All I can input is that I tried making the entire smart range with Mechanic (well, with what little we can adjust), and once I had all the engine parameters right, the peak torque and power were spot on for the base model. The rest of the range simply have more boost, and just by altering the boost number to what the cars use in real life, I didn't have to tweak anything else to get power or torque to match up.

So what I'm saying is, while I don't know anything about spool times, the amount of extra torque the turbo creates at a certain boost pressure, seems spot on.

I drive a turbo car, and I have a boost gauge. Problem is, the turbo is running about idle, as the engine is tax-restricted to hell and back. So my turbine is almost constantly in it's linear range, and not very useful for this discussion. If someone could send me a new ECU, I'd be happy to show you how it behaves...

Wow it's been a while since I've been on here... enoyed reading the thread. In my experience, a turbo should spool up rather quickly once you're into it's working rev-range. Obviously that is dependant upon the turbo used, and how the ECU is mapped. It's a bit weak, but I have a vid that I used recently when diagnosing where I was losing boost:

http://www.rossburton.com/castlecoombe/Boostleak.MOV

you can only really see the boosguage at the beginning (hard to hold the camera steady and drive at the same time!) but what you'll notice, is that on idle it sits about -0.5, and as soon as I get above about 2200 rpm the boost climbs steadliy. The fact it holds at about .8bar for a few secs is not normal, that was the problem i was trying to fix, but you'll then see it suddenly climb up to 1.5bar (max). This then starts bleeding off at about 5000 rpm to 1bar @ approx. 7000rpm, but I don't see that too often as that'd be breaking the speedlimit even in 2nd

If anyone is interested, I'm sure I could get some footage of boost characteristics, but I think the point is seen that once the car is up in the revs, the boost should climb quite rapidly to max boost. In fact, the turbo would continue to build boost until it died were it not for the wastegate frequency valve that electronically limits boost (if it's not ecu controlled, then the tightness of the spring in the wastegate will limit boost).

cheers
Ross

This is a scooby, may be of use in the descussion

http://media.putfile.com/Scooby60

Troy

Quote from Bob Smith :

So what I'm saying is, while I don't know anything about spool times, the amount of extra torque the turbo creates at a certain boost pressure, seems spot on.

I would concur with that, the output at any given boost seems to be accurate, indicating that the "cyber air / fuel" ratio behaves right since power output / fuel consumption are related in LFS...

This part of the boost modelling seems spot on I agree

Quote from Ross Burton :

Wow it's been a while since I've been on here... enoyed reading the thread. In my experience, a turbo should spool up rather quickly once you're into it's working rev-range. Obviously that is dependant upon the turbo used, and how the ECU is mapped. It's a bit weak, but I have a vid that I used recently when diagnosing where I was losing boost:

http://www.rossburton.com/castlecoombe/Boostleak.MOV

you can only really see the boosguage at the beginning (hard to hold the camera steady and drive at the same time!) but what you'll notice, is that on idle it sits about -0.5, and as soon as I get above about 2200 rpm the boost climbs steadliy. The fact it holds at about .8bar for a few secs is not normal, that was the problem i was trying to fix, but you'll then see it suddenly climb up to 1.5bar (max). This then starts bleeding off at about 5000 rpm to 1bar @ approx. 7000rpm, but I don't see that too often as that'd be breaking the speedlimit even in 2nd

If anyone is interested, I'm sure I could get some footage of boost characteristics, but I think the point is seen that once the car is up in the revs, the boost should climb quite rapidly to max boost. In fact, the turbo would continue to build boost until it died were it not for the wastegate frequency valve that electronically limits boost (if it's not ecu controlled, then the tightness of the spring in the wastegate will limit boost).

cheers
Ross

More footage is always better! Thanks for the input and the movie. Another thought following through with your observations: The way the turbochargers behave now in LFS, for any given car, boost would eventually cease to build which also illustrates my point very well. Watching the boost gauge needle ease into position @ max PSI implies that it's limited by "it's own restrictions" (poor wording, gotta be quick @ work I should be coding hehe ) rather than violently trying to build more pressure and suddenly being limited by a wastegate. One could argue that the retardation of the pressure increase near max boost is due to the impeller running out of it's efficiency range, but then boost should start to drop again right away, but it doesn't; it holds for a couple thousand more RPM... That's the other thing we don't see is boost dropping when the engine exceeds peak power output (approach or exceed redline in most cars in LFS). I'm reasonably certain this should happen as well.....

Gotta run back soon got more to say

Note in these vids we can't see people's feet but they still tell a tale to back up my ramblings.

Here is a quick link.... Super high horsepower car so this could compare to how the GTR may act, although this car has more power.... EVO 8, 2L turbocharged motor... putting out 826hp WOW. (No, not Rice Tristan, it's cool really)
http://media.ams-evo8.com/sevo826awhp.wmv

Here's another one.
http://www.oroinvestments.com/ ... gtir/andy_gtir_speedo.avi

Note on this link, most of the way down the page he describes the cars boost characteristics.
http://www.powerlabs.org/jdmsubaru.htm

Here's a race prepped Supra, boost gauge visible in the lower right of his cluster.
http://www.gscdownloads.com/steve/007_2.wmv

If you roll your FX0, and keep your foot on the throttle out of spite and bitterness for having rolled the car, (guess you could test this without the spite and bitterness, it would be healthier), you will notice also that full boost is never acheived, which seems odds since there is no engine damage to impede such shinanigans

*cough*

coughity cough cough

Quote from axus :

It may be a shortcoming what we are trying to say is that the results that LFS produces are believable and explainable and since the RA is the only real life reference point we have, someone who has driven one must say what the turbo lag is like in it. The rest of the cars are non-existent in real life so you can easily use these explainations to say that what LFS does is believable. Therefore, it is not the most important thing on the list.

I know that the engine in a RA is a Saab B204L - the same as in for example a 9-3.
http://video.google.com/videop ... 7241632962&q=saab+9-3

Thats a vid of a quite tuned version, it has alot of spool up time too, but when it's there, it's there.

Being a saab-fanatic, i can tell you, that the RA's base-engine behavior is way of, it needs to build boost quicker, way quicker, but maybe the RA-developers have designed the enginemanagement to disallow fast turboboost, to prevent spinning.

EXCELLENT, a post!

Thanks for the info, that video is insane!

Interesting theory about about the RA design limiting the pressure build rate... Most cars that do this accomplish it via limiting the AMOUNT of boost available, usually in first gear only. So if the RA behaved that way in first gear only, it might be beleiveable but as it is it's silly.