PDA

View Full Version : TECH ARTICLE: what keeps torque from dropping off?



daniel
11-23-2006, 09:40 PM
the perfect turbo would spooll like stock and keep the torque flat to redline... *dreams*

daniel
11-25-2006, 09:26 PM
so, what keep the torque from dropping off? I'm guessing the ball bearings, right? Because of less rolling resistance? So if someone did a BB type 20G turbo it would work out better wouldn't it?

trinydex
11-26-2006, 07:29 AM
this is a serious lesson in turbocharger and internal cumbustion engine theory.

torque can be loosely associated as a product of exhaust gas velocity. horsepower can be loosely associated as a product of mass flow. an example to clarify is that big horsepower cars flow a lot of air and what we view as torquey cars generally make thier torque low when the volume flow has high velocity. now i say loosely associated because gearing has to do with both factors and also because of what i'll explain next.

high horsepower engines with enough compensative gearing make high amounts of torque, however we do not view these cars as torquey but peaky. take for example an f1 2.4 liter v8. it'd be hard to convince me that this motor doesn't make a ton of torque at 16k rpms... but that's the problem isn't it... 16k rpms makes your car sound like... well an f1 car. so snap back to the real world for a second.

a typically torquey motor is the lsX motor where X can be 1, 2, 6, 7 etc... basically the vette/fbody motor. this motor generally displaces more than 6 liters so it flows a ton of air. but generally it flows at lower cylinder pressures than a turbo car like the evo. now with that basic qualification out of the way it doesn't matter.... now i say this because if you flow 1 pound through a 2 inch^2 hole... the hole loosely dictates the speed of the gas. that said the lsx's optimum flow is TYPICALLY at the midrange which gives you that low end torque feel.

now why did i say optimum flow? well because the cross section of the pipes going in and pipes going out of the cylinder dictate the flow speed and hence the torque (loosely again). so when you achieve optimum flow you achieve some level of peak torque (this can be modified). so if you made all the ports REAL big in the lsx would it then become peaky? absolutely. this is what's known as shifting the powerband or torqueband around.

see in the end this is a limitation of engineering, you're asking to maximize two fundamentally opposing quantities. here's the punch line. small flow area increases velocity but limits maximum flow potential, large flow area decreases flow velocity but increases maximum flow potential. NOW this is relative... what i MEANT to say was small flow area increases LOW END gas velocity but limits HIGH END maximum flow (tha last two adjectives were redundant) and large flow area decreass LOW END flow velocity but increases HIGH END FLOW VELOCITY which produces a HIGH END TORQUE!!! a bigger hole is idealized to a bigger flow.

if you had a 2" hole and it produced 100 foot/second flow speed at 4000rpm then a 3" might do the same at 6000rpm.

this is why in drag racing it's always better to shift the rpm higher, because you're never going to lose torque unless you can't gear yourself into the powerband.

if you cut yourself short by getting low end torque, you're LOSING HIGH END TORQUE!!!

now to answer your question of what gets you to keep the torque from dropping off. well the answer to any limitation of engineering problem is likely variable geometry of some sort. you can have a variable geometry turbocharger, this allows you to flow fast at the low end and fast at the high end... this creates a nice torque band that is relatively flat the whole way through, never dropping off while your hp increases the whole time. the other way is variable valve timing, you can change the flow characteristics/speed of the engine by changing the valve timing (mivec).

ball bearings and various sizings of turbos would only change what they call "static" quantities. in order to get the best of both worlds you need dynamic quantities.

BHCevo
11-28-2006, 08:05 PM
I agree with everything trinyDex says. His stated correlations between exhaust gas velocity, mass flow, torque and horsepower more or less capture what happens in the real world and correctly highlight compromises needed between optimzing peak torque or optimizing peak horsepower.

That being said, I'll offer an equivalent but alternate explanation that better highlights the fundamental relationship between torque and horsepower.

First, I want you to think of your engine as something that makes torque. Forget about horsepower, think torque. Torque is force multiplied by a lever arm. In this case, force is the force supplied by the power stroke and lever arm is the geometry of the crank/rod/piston. In this case force is actually the time averaged force over all the power strokes of all the cylinders.

Okay, two things conecern us here
1) How do we increase torque? Well, most of the time we increase the power stroke force. The time averaged force goes up as
the total amount of fuel and oxygen we burn in all our cylinders increases. In general, we can inject arbitrary fuel (yeah EFI), so we are really limited by the amount of oxygen we can stuff into the cylinders. There are two ways we increase the total mass oxygen we burn
a) Increase combustion volume (displacement)
b) Increase combustion oxygen pressure (forced induction ie turbos or superchargers)
Or both hehe.

2) The other thing that concerns us is ...why does the torque output vary with engine speed? Generally speaking.
Simple. Torque is the product of (oxygen mass being burnt) times lever arm. Your lever arm isn't changing so clearly the oxygen mass being burnt is changing. as a function of engine speed. Your displacement isn't changing, so your combustion oxygen pressure MUST be changing, regardless of the presence of forced induction.

Why does the combustion chamber oxygen partial pressure change as a function of engine speed? This is because in the real world the fluid mechanics of gas (oxygen) flow matter greatly for establishing the actual amount of oxygen that got into the cylinders during the brief time the intake valves were open. Look at your torque curve....it is actually the curve of the efficiency with which your cylinders fill with air, as a function of engine speed. This is also called volumetric efficiency.

Volumetric efficiency depends greatly on intake geometry, exhaust geometry, intake and exhuast PORT geometry, intake manifold geometry and exhuast manifold geometry. It also depends on cam shaft timing and valve timing (hence will change with MIVEC or different cam shafts or cam gear settings).

How volumetric efficiency varies with port geometry is what Triny was saying above. In his language, a "torquey engine" means an engine with the torque (volumetric efficiency) peak at low engine speeds. Because the torque peak occurs at low engine speeds, there is no point in a high redline, so you better have big displacement to give you torque at the crank since you cant depend on short gears to give you that torque boost at the wheels.

A "high horsepower engine" means an engine with a torque peak (volumetric efficiency peak) ocurring at high(er) engine speeds. Thus you can get away with smaller displacement, and a small value of the peak torque, because you can use short gears to boost your smaller crank torque. You can use shorter gears because your useful redline is higher. Why a high engine speed torque peak is called high horsepower i'll explain below.

I'll summarize the important point one more time: to increase HORSEPOWER you sacrifice your peak torque height (at midrange) in order to improve your redline torque. Your peak hasn't moved to redline, although it has moved up some, but your torque near redline is better.Â* Why does this increase horsepower? I will explain below.

Okay, torque torque torque. What about horsepower. Here ya go.

Horsepower is torque times engine speed. Thats it. Thats all you need to know. HP = Torque X RPMS.

From this simple fact follows many things:
1) Peak Horsepower occurs closer to redline than peak torque because as engine speed goes up horsepower goes up proportionally, independent of torque. Hence, in order to optimize horsepower usually you sacrifice your midrange torque peak in order to bolster your near redline torque. It usually turns out thatÂ* Peak Torque X 4000 RPMS < Redline Torque X 7000 RPMS. You can think about this.

3) A high horsepower engine, everything else being equal, has a higher redline. The max engine speed is high.

2) A high horsepower engine, everything else being equal, pulls hard because you can get away with SHORTER GEARING at HIGHER SPEEDS. Thats it. Thats why a high horsepower engine provides good acceleration. No other good reason.

3) A high horsepower engine with tall gears is slow(er).

5) Maximum velocity goes up with horsepower (not torque), everything else being equal. Of course, it might take you forever to get there because your torque (acceleration) sucks but...


By the way, I'm assuming you understand how your gear ratios act as a torque multiplier at the wheels.

My last point is: what makes a car pull hard is torque. Thats it. The only reason high horsepower makes your car pull hard is because it lets you stay in lower gear for a longer time (low gears provide more torque at the wheels). Thats it. EVERTHING ELSE BEING EQUAL.

BHCevo
11-28-2006, 08:22 PM
I realized I never really answered the OP question. Here ya go.

To keep torque from dropping off you need (infinitely) variable intake/exhaust valve timing/lift (go BMW!) and also variable intake/exhaust geometries (go Porsche!!). In both cases I believe the intake valve timing and intake physical geometries are more important than the exhaust sides. I might be wrong about the relative importance of the exhaust and intake sides. Hafta think about it.

In practice, if you want higher torque at redline just raise the boost (go C-16 fuel!!). You will still have a peaky torque curve but at least the whole curve will have increased by a constant factor and you will have better acceleration (torque) at redline. Of course, now your midrange torque might be a little
much...hehehe

Triny mentions variable geometry turbos (VATN turbos). These turbos have all the good characteristics of a small turbo (super fast spool, low end torque) and a large turbo (high end torque, flow huge CFM). Crazy, eh? Theoretically, a perfect VATN turbo needs no wastegate and no boost controller since you can control maximum boost by varying your turbo geometry in real time. Sounds too good to be true? Well, the newest carrera turbo utilizes twin VATN turbos. I believe truck turbos have utilized VATN for a while now. Of course, the benefits of VATN are only realized to the extent you can vary the turbo geometry. Big variationsÂ* mean big benefits, small variability means small benefits.

VATN turbos eliminate the peakiness associated with the turbo sizing. However, you still have the volumetric efficiency variances associated with static valve timing and static intake/exhaust geometries. so you'll needs VVT on top of VATN to further flatten the torque curve.

daniel
12-20-2006, 02:39 PM
cool, i'm a tech topic. thanks for the info guys.

biggiesacks
05-17-2007, 09:07 AM
Lets not forget about our friend N20 (Nitrous oxide) Which is another great way to add oxygen. You can control torque pretty well with the spray and ontop of that you can boost it for a little extra something. I remember reading something about using pure oxygen but i believe they found it to be to unstable.

gt40
05-17-2007, 09:12 AM
I realized I never really answered the OP question. Here ya go.

To keep torque from dropping off you need (infinitely) variable intake/exhaust valve timing/lift (go BMW!) and also variable intake/exhaust geometries (go Porsche!!). In both cases I believe the intake valve timing and intake physical geometries are more important than the exhaust sides. I might be wrong about the relative importance of the exhaust and intake sides. Hafta think about it.

In practice, if you want higher torque at redline just raise the boost (go C-16 fuel!!). You will still have a peaky torque curve but at least the whole curve will have increased by a constant factor and you will have better acceleration (torque) at redline. Of course, now your midrange torque might be a little
much...hehehe

Triny mentions variable geometry turbos (VATN turbos). These turbos have all the good characteristics of a small turbo (super fast spool, low end torque) and a large turbo (high end torque, flow huge CFM). Crazy, eh? Theoretically, a perfect VATN turbo needs no wastegate and no boost controller since you can control maximum boost by varying your turbo geometry in real time. Sounds too good to be true? Well, the newest carrera turbo utilizes twin VATN turbos. I believe truck turbos have utilized VATN for a while now. Of course, the benefits of VATN are only realized to the extent you can vary the turbo geometry. Big variationsÂ* mean big benefits, small variability means small benefits.

VATN turbos eliminate the peakiness associated with the turbo sizing. However, you still have the volumetric efficiency variances associated with static valve timing and static intake/exhaust geometries. so you'll needs VVT on top of VATN to further flatten the torque curve.


The theory is better than the reality:Â* The go fast porsche crowd puts the older NON variable turbos back on for more power.Â* You may spool faster by constricting the intake of the turbo but you are also flowing less air. More Air + fuel at the right ratio burning makes the power. You lose top end with the variable turbos and ultimately make less power because they are not as efficent up top as an ordinary garrett gt. There is a significant penalty you are paying in power levels with variable turbos. VATN is just not ready for prime time yet...

LiquidLife
08-02-2007, 11:44 AM
Why can't we just use a regular electric compressor tank instead of a turbo...Â* =o)

trinydex
08-02-2007, 12:13 PM
how does that keep torque from dropping off? you still have the problem of an engine's maximum flow, a limited range of on the fly adjust ability with cams. it is unfortunately a fundamental problem of engineering. if you put a super charger on a car you still have the problem of idealizing the top end vs the bottom end.

LiquidLife
08-02-2007, 12:20 PM
well my point, however dumb it may be, is that it would be anywhere from 50 to 200 psi give or take either way, hahaha. Then no one would care about drop off, if it were possible to have that much boost running anyway.

trinydex
08-02-2007, 12:29 PM
well that's what i'm saying is you still have to worry about drop off. if you want to flow 800 hp at the top end then you'll have something like a 4" exhaust. what is that gonna do to your low end :[

LiquidLife
08-02-2007, 12:31 PM
So we need a synthetic variable exhaust pipe. Let me jump into my time machine real quick, I'll be back.

LiquidLife
08-02-2007, 12:46 PM
I can't draw, oh well....

http://i193.photobucket.com/albums/z127/LiquidLife949/exhaust.jpg

LiquidLife
08-02-2007, 01:02 PM
A variable intake and exhaust control valve system would be nice for maximum power at all ranges.

trinydex
08-02-2007, 02:42 PM
it's what's known as an iris. it'd be very expensive and hella transformers to implement.

LiquidLife
08-02-2007, 02:47 PM
it's what's known as an iris. it'd be very expensive and hella transformers to implement.


Yeah, I forgot the name for it. It would work well though don't you think?

trinydex
08-02-2007, 03:35 PM
it would work very well if you could get it to work