Whats the deal with torque curves?

[BD2003]

I'm trying to understand this a little better. What is the physical basis for there being peaks and dips in the amount of torque an IC engine produces? Wouldnt any friction be exactly proportional to the RPM, thereby canceling out?


And a related question - what exactly defines the peak?

Is it the peak energy transmitted to the wheels per unit of gasoline combusted? (ie the most efficient)

Or is it the absolute peak amount of torque that the engine can produce, not taking efficiency into account?

 

[Triumph]

No efficiency. Peak torque is just what it says, the point where the engine is putting out its most torque.

I'm not sure I understand why you think torque curves WOULD be perfectly flat?

 

[zerocool84]

Originally posted by: Triumph
No efficiency. Peak torque is just what it says, the point where the engine is putting out its most torque.

I'm not sure I understand why you think torque curves WOULD be perfectly flat?

He's probably seen many forced inducted vehicles that have come out in the past couple years with flat torque curves and wonders why every car doesn't have one.

My car stock has a flat torque curve from 2200-4400rpm because the boost is controlled by the ECU and that's how most get them from the engine controlling the boost.

 

[fbrdphreak]

It's all about the engine's design. Intake velocity & flow, valve timing, head flow & velocity, stroke, bore, and thus displacement. You could design pretty much any engine (I-4, V6, V8, V10, etc) to have a flat torque curve. But there are compromises to be made, including low vs high end power, fuel efficiency, emissions, and many, many more things.

 

[Bartman39]

One of the main things that is a direct factor in torque is "compression"... A prime example is a Diesel engine which fires off nothing but compression... Most diesel engines are atleast 18 to 1 compression or higher and add a turbo charger which basicly forces more in to make the compression effect even higher which means more torque...

An engine that is normaly aspirated (carb`d or EFI) is limited to its dynamic compression but through the careful selection of heads, valves and combustion chamber shape all combined with a well designed cam shaft a nice smooth torque curve can be developed...

Normaly race engines are designed for a very narrow torque band so that gearing, tire size, weight and even aero dynamics work together to make it from point A to point B at the quickest possible time...


In short... Compression and the camshaft combined have the greatest effects on torque... Throw in a turbo or supercharger and it all goes out the window and dont even mention Nitrous...

 

[natto fire]

An IC engine really just an air pump that changes pressure forces into twisting forces (torque) via it's crankshaft, or elliptical if you are talking about a rotary engine. How a crankshaft is designed can change this immensely. There are things to consider like rod to stroke ratio, and if taking power curves from a chassis dyno, drivetrain losses, etc.

The torque curve is because of an engines volumetric efficiency, or how well it can move air in an out of the engine. As noted in this thread, forcing the induction charge will dramatically change the torque curve, but it is not the end all. If there is too much exhaust backpressure then that will lower the V.E.

The peak again depends on which type of dyno power is being measured, but it is essentially when the engine is producing the most power because it simply cannot move more air. The reasons are pretty much unlimited as to what can cause it, but it simply cannot move any more air, which is what an IC engine relies on to make power.

An example of a motor that does produce torque natively is many electric motors. Not sure if you have seen a torque curve from those, but it is quite humorous.

 

[overst33r]

Originally posted by: Captain Howdy
An IC engine really just an air pump that changes pressure forces into twisting forces (torque) via it's crankshaft, or elliptical if you are talking about a rotary engine. How a crankshaft is designed can change this immensely. There are things to consider like rod to stroke ratio, and if taking power curves from a chassis dyno, drivetrain losses, etc.

The torque curve is because of an engines volumetric efficiency, or how well it can move air in an out of the engine. As noted in this thread, forcing the induction charge will dramatically change the torque curve, but it is not the end all. If there is too much exhaust backpressure then that will lower the V.E.

The peak again depends on which type of dyno power is being measured, but it is essentially when the engine is producing the most power because it simply cannot move more air. The reasons are pretty much unlimited as to what can cause it, but it simply cannot move any more air, which is what an IC engine relies on to make power.

An example of a motor that does produce torque natively is many electric motors. Not sure if you have seen a torque curve from those, but it is quite humorous.

Mmmm Tesla Roadster http://www.teslamotors.com/ima...motor_torque_curve.gif

 

[BD2003]

Originally posted by: Triumph
I'm not sure I understand why you think torque curves WOULD be perfectly flat?

Because intuitively I'm thinking that it should be all dependent upon the energy you put into the system via gasoline. Any inefficiency would be directly proportional to the RPM, which without considering anything else I dont understand (which is clearly why I'm asking), would theoretically lead to a flat curve.

I'm not really asking "why arent they flatter?" so much as "what makes them curved in the first place?"

In other words, are the peaks due to the ability for the engine to combust more gasoline (due to an increased airflow at a certain RPM perhaps), or due to losing less of the energy from the already burnt gasoline? (friction in the drivetrain)

 

[BD2003]

Originally posted by: Captain Howdy
An IC engine really just an air pump that changes pressure forces into twisting forces (torque) via it's crankshaft, or elliptical if you are talking about a rotary engine. How a crankshaft is designed can change this immensely. There are things to consider like rod to stroke ratio, and if taking power curves from a chassis dyno, drivetrain losses, etc.

The torque curve is because of an engines volumetric efficiency, or how well it can move air in an out of the engine. As noted in this thread, forcing the induction charge will dramatically change the torque curve, but it is not the end all. If there is too much exhaust backpressure then that will lower the V.E.

The peak again depends on which type of dyno power is being measured, but it is essentially when the engine is producing the most power because it simply cannot move more air. The reasons are pretty much unlimited as to what can cause it, but it simply cannot move any more air, which is what an IC engine relies on to make power.

An example of a motor that does produce torque natively is many electric motors. Not sure if you have seen a torque curve from those, but it is quite humorous.

Ah, got it.

So, it has absolutely squat to do with efficiency?

This is a sample dyno for the same model car I have.

http://www.dragtimes.com/image...Toyota-Celica-Dyno.jpg

The bump from 3.5-4.5k is very noticible, the car just feels more alive in that range, and you can definitely feel the drop off until 6k.

So while most of my asking is just pure curiosity how it works, I also wonder whether I'm burning proportionally more or less gas in that range as opposed to keeping it under 3k.

 

[natto fire]

Originally posted by: BD2003
Ah, got it.

So, it has absolutely squat to do with efficiency?

This is a sample dyno for the same model car I have.

http://www.dragtimes.com/image...Toyota-Celica-Dyno.jpg

The bump from 3.5-4.5k is very noticible, the car just feels more alive in that range, and you can definitely feel the drop off until 6k.

So while most of my asking is just pure curiosity how it works, I also wonder whether I'm burning proportionally more or less gas in that range as opposed to keeping it under 3k.

It has quite a bit to do with volumetric efficiency. Mechanical efficiency does have a bit to do with it, but not nearly as much as the volumetric efficiency.

One way to think of it (prepare for a crappy analogy) is if you were flicking a pinwheel. Think of each of your finger flicks as a powerstroke of a cylinder. There is only a certain amount of speed you could flick the pinwheel, because of the air resistance (major) of the fins and friction (minor) at the hub. Now let's pretend that the ability to overcome the air resistance is your volumetric efficiency. The torque curve would be the amount of torque exerted on the hub during these finger flicks. Try to imagine the finger flicks happening at extremely higher rates than you can do. This is essentially how a reciprocating piston engine creates torque.

To answer your second question, you will be burning the most fuel at your peak power output, but it is slightly more efficient because you are minimizing pumping losses (which are caused by all surfaces in the intake tract, but especially the throttle butterfly.

 

[Zenmervolt]

Originally posted by: Captain Howdy
To answer your second question, you will be burning the most fuel at your peak power output, but it is slightly more efficient because you are minimizing pumping losses (which are caused by all surfaces in the intake tract, but especially the throttle butterfly.

Actually, you're burning the most fuel at redline and wide open throttle, which is often not the peak for either horsepower or torque. Higher RPM always equals more gasoline used assuming that the throttle position is held constant (that said, RPM is the "800 pound gorilla" in the fuel use equations and RPM has a much greater effect on fuel consumption than does throttle opening).

At peak HP, the engine is most efficient at turning fuel into work, and at peak torque, the engine is most efficient at turning fuel into force. I think. I may be wrong.

Pumping losses are minimised at all points on a torque/hp curve because engines are dyno-ed at wide-open throttle. If your engine's dyno chart shows 150 hp at 3,000 RPM and you're cruising down the highway at 3,000 RPM with light throttle, the engine is not making 150 hp. The throttle butterfly is choking off the engine and you're only actually making around 25-30 hp if you're moving at freeway speeds.

ZV

 

[Nyati13]

Originally posted by: BD2003
Because intuitively I'm thinking that it should be all dependent upon the energy you put into the system via gasoline. Any inefficiency would be directly proportional to the RPM, which without considering anything else I dont understand (which is clearly why I'm asking), would theoretically lead to a flat curve.

I'm not really asking "why arent they flatter?" so much as "what makes them curved in the first place?"

In other words, are the peaks due to the ability for the engine to combust more gasoline (due to an increased airflow at a certain RPM perhaps), or due to losing less of the energy from the already burnt gasoline? (friction in the drivetrain)

Getting gas into the engine is the easy part. Getting the air (without air the gas won't burn) into the engine is the hard part. And, once you've burned that gas in air, getting the exhaust out of the engine is another hard part (but not as hard as getting the air in).

The peaks are due to the fact that air doesn't always behave the same way. At different velocities (RPM) and different pressures it does different things. It's amazing that most engines do have a relatively smooth and flat torque curve, given how hard it is to get the air to do what you want it to.

 

[BD2003]

Originally posted by: Zenmervolt

Originally posted by: Captain Howdy
To answer your second question, you will be burning the most fuel at your peak power output, but it is slightly more efficient because you are minimizing pumping losses (which are caused by all surfaces in the intake tract, but especially the throttle butterfly.

Actually, you're burning the most fuel at redline and wide open throttle, which is often not the peak for either horsepower or torque. Higher RPM always equals more gasoline used assuming that the throttle position is held constant (that said, RPM is the "800 pound gorilla" in the fuel use equations and RPM has a much greater effect on fuel consumption than does throttle opening).

A higher RPM would clearly have more internal friction, but is that the only main factor as to why higher RPM = higher consumption?

The extra friction would certainly mean while cruising at a fixed speed, lower RPM = less consumption. (assuming theres no lugging)

But what about acceleration? Hypothetically, would it necessarily take more fuel to accelerate from 30-40mph at a low RPM vs. a medium RPM, especially assuming that the engine can produce more torque at medium RPM?

At peak HP, the engine is most efficient at turning fuel into work, and at peak torque, the engine is most efficient at turning fuel into force. I think. I may be wrong.

Its been years since I took HS physics. What exactly is the difference between work and force? IIRC, work is a measure of the force thats actually translated into kinetic energy?

Pumping losses are minimised at all points on a torque/hp curve because engines are dyno-ed at wide-open throttle. If your engine's dyno chart shows 150 hp at 3,000 RPM and you're cruising down the highway at 3,000 RPM with light throttle, the engine is not making 150 hp. The throttle butterfly is choking off the engine and you're only actually making around 25-30 hp if you're moving at freeway speeds.

ZV

I'm of the understanding that in just about every car, the fuel mix is slightly richened at WOT for performance issues.

Clearly you're not making 150hp at 300rpm with the throttle at 1/4th, but would the actual configuration of the curve change (as compared to the WOT curve), rather than just translating down to a lower power range?

 

[Zenmervolt]

Originally posted by: BD2003
A higher RPM would clearly have more internal friction, but is that the only main factor as to why higher RPM = higher consumption?

The extra friction would certainly mean while cruising at a fixed speed, lower RPM = less consumption. (assuming theres no lugging)

Higher RPM will always mean more fuel consumption per unit of time. It's a simple matter of there being more power pulses per unit of time. WOT means more are per pulse, which means more fuel per pulse, but that doesn't make nearly as much difference (on a non-forced-induction engine) as does pure RPM.

Originally posted by: BD2003
But what about acceleration? Hypothetically, would it necessarily take more fuel to accelerate from 30-40mph at a low RPM vs. a medium RPM, especially assuming that the engine can produce more torque at medium RPM?

Actually, if there is no full-throttle enrichment, the most fuel-efficient way to accelerate is to apply full throttle and upshift as soon as it can be done without lugging the engine in the next gear. (So, something like driving a Civic, flooring it, and shifting at 2,500 RPM on level ground.) Remember though that engine efficiency and overall car efficiency are not the same things. At torque peak, the engine is most efficient at creating force from fuel. However, that doesn't account for gearing and drag and such that all come into play for the overall car. At highway speeds, the engine is actually very inefficient in terms of the power being produced per unit volume of fuel consumed. The car overall, however, is quite efficient in terms of distance traveled per unit volume of fuel consumed.

Originally posted by: BD2003
Its been years since I took HS physics. What exactly is the difference between work and force? IIRC, work is a measure of the force thats actually translated into kinetic energy?

Work is force multiplied by distance. Power is work done over time. I'm trying to come up with an analogy, but am failing, I will try to come back to this later.

Originally posted by: BD2003
I'm of the understanding that in just about every car, the fuel mix is slightly richened at WOT for performance issues.

Yes.

Originally posted by: BD2003
Clearly you're not making 150hp at 300rpm with the throttle at 1/4th, but would the actual configuration of the curve change (as compared to the WOT curve), rather than just translating down to a lower power range?

That I cannot say for sure, but I think that it would since the dynamics of the intake tract are different with the throttle butterfly partially closed.

ZV

 

[BD2003]

Actually, if there is no full-throttle enrichment, the most fuel-efficient way to accelerate is to apply full throttle and upshift as soon as it can be done without lugging the engine in the next gear. (So, something like driving a Civic, flooring it, and shifting at 2,500 RPM on level ground.)

I've heard exactly the same thing a number of times, and I actually went and tried it myself a few times. I have one of those scangauges that can output fuel consumption directly in terms of gallons per hour, as well as efficiency in mpg. I've been told that due to the enrichment at WOT, to use 3/4 throttle.

Yet what I've found is practically entirely the opposite. Fuel consumption was pretty much directly proportional to throttle position. At 2000rpm or so, the car is not lugging (barely - 1600ish rpm seems the limit), but there is very little difference in acceleration regardless of throttle position - it pulls nearly as hard at 1/8th throttle as it does at 3/4. Yet the fuel consumption skyrockets when I push it down, the car doesnt accelerate any faster, and mpg plummets.

I've also tried it on the freeway, in top gear at low rpm, to see how throttle position effects acceleration at 2k-3kish rpm. Same deal - consumption goes up, output stays virtually identical, and efficiency plummets at wider throttle, completely contrary to the notion that its most efficient at high throttle/low rpm.

And its a completely mechanical throttle, a cable linking the pedal and the throttle body, not any of the newfangled computerized drive by wire stuff.

Theres only a few realistic explanations for this:

1) My car (2000 Celica Gt-s 6spd) is designed differently than most, or its completely out of tune - which seems unlikely to me - its far from exotic, and I keep it in good shape.

2) The gauge is off - which also seems unlikely, because its been accurate within 3% to the pump over several months, even including tanks where I drove lots of short trips with the engine not getting much use while fully warm - mpg is expectedly much lower.

3) I'm not properly implementing this technique - which also seems unlikely, because its not exactly rocket science to push down on the pedal more and shift earlier.

4) The way of thinking is based on a theory that works on paper, but not in practice. One major thing that leads me to favor this option (or equally #1, but this holds for every car I've ever driven) - while using cruise control, which has always provided the best efficiency for me, acceleration takes place at what appears to be as little throttle as possible (as measured by free play in the pedal with CC engaged). I would imagine that if it truely was more efficient to burn fuel at wider throttle, the CC would be tuned to do so.

What do you think?