Applying Torque vs Horsepower

[TexasRebel]

Multiply the torque x2 (which is what gears do) and you get exactly the same output at exactly the same speed in gear with the 10,000 rpm engine verses the 5000 rpm engine. So how do you get performance differences in a drag strip or LeMans (and why are they different?).

PS apparently, Top Fuel motors are limited to around 7900 rpm. Why limit them if low rpm and high torque is faster? Why don't high torque diesels win anything until they get major rules advantages? When they won LeMans, they were allowed more displacement, more boost, AND bigger restrictors. My grandma can win a race with that many advantages.

Between the two examples, one can do without a gearbox what the other needs a gearbox to accomplish. A gearbox is weight and friction. If you want the wheels to turn fast, you need RPM. If you want an increase in RPM you need torque.

Top fuel limits engine RPM for safety, not to hinder performance. Energy stored in a rotating assembly increases with the square of rotational velocity (1/2mv^2). Grenade an engine at 10,000 RPM and you have 4X the energy in parts flying as if it was turning 5,000 RPM. You'll find most regulations in any type of racing are for reasons of safety. Heck, you could win every open class drag race if you strap a Titan IV rocket booster on your car... unless surviving is a requirement to win.

The key to understanding horsepower and torque is understanding why a torque curve looks like it does. Bore diameter, fuel burn velocity, rod:stroke ratio, combustion gas expansion time, &c. all factor into it. It's why your idle speed changes if you adjust your timing. It's why you want your timing to change with RPM. It's why the torque curve increases and peaks instead of starting out as high as possible at 0 RPM like an electric motor.

Sponsored

 

[GT Pony]

The key to understanding horsepower and torque is understanding why a torque curve looks like it does. Bore diameter, fuel burn velocity, rod:stroke ratio, combustion gas expansion time, &c. all factor into it. It's why your idle speed changes if you adjust your timing. It's why you want your timing to change with RPM. It's why the torque curve increases and peaks instead of starting out as high as possible at 0 RPM like an electric motor.

The main reason the torque curve fall off in the upper RPM range on a NA motor is because of engine friction and loss of volumetric efficiency at high RPM - ie, the engine can't "breath" as good due to increasing air flow restriction that doesn't' fill the cylinders as efficiently. The cure for that is forced induction. When the torque curve peaks, that is the point where the internal friction is still relatively low and volumetric efficiency is the highest.

 

[Norm Peterson]

into competition tractor pulling. Getting a car moving is easy. When your goal is to move an increasing load as quickly as possible, low end torque gets you to high RPM horsepower... Cars are a fixed load with a huge power:weight ratio.

I think equating car acceleration to competition tractor pulling is biasing your thinking (sounds better than calling it being in a 'rut', no?). Given that we agree that getting a car moving is relatively easy, you don't need huge torque at the crankshaft to accomplish it.

A car's drag loading (aero + rolling) in the first couple of gears is essentially independent of vehicle position (IOW, distance from the 'start'), which is in contrast to loading in the tractor-pull case where loading is very much dependent on position. IOW, for a car there is no sharply increasing load for the car's engine to have to "get ahead of" in order to get some momentum going.

yes linear, and also infinite.

The limit as RPM approaches infinity of TQ*RPM/5252 (where TQ is constant) is infinity.

That's a pretty good illustration of the difference between mathematical theory and engineering practice - what the math says is true can conveniently overlook the fact that in "real life", RPM is very definitely limited to some finite value.

Gearing comes at a price... weight. Make an engine with a completely flat Horsepower curve and you have no reason for a gearbox. A completely flat horsepower curve has peak torque at zero RPM.

If you could even build an engine with a completely flat HP "curve" with enough HP to support a top speed above 100 MPH at any reasonable RPM, you'd have drivability difficulties at 5 MPH. Chances are that for drivability in general (and particularly in wet/worse weather and other poor traction situations) you'd have to do the kind of torque management that would end up giving you a much different actual crankshaft torque curve than the concave-up (1/RPM) shape corresponding to a perfectly flat HP line. Basically, you'd have to choke what the engine was capable of down to something you could actually use, separate from the throttle modulation you do with your right foot.

The closest current approximation to a flat HP engine might be to fit it with a CVT, which can be controlled in conjunction with the engine to put the engine at an rpm optimum for conditions. But they have their own limitations (especially at small drive-pulley diameters).


I hadn't given up/thrown in the towel/bailed from this discussion, just that for some unknown reason I'm not getting notices of reply to any of my subscribed discussions on the site.


Norm

 

[Norm Peterson]

My understanding of Top Fuel is that there is so much power that it is unnecessary to provide a transmission at all (do the torque times axle gear divided by tire radius math and compare that to whatever the current weight requirement is). Axle gearing and RPM are specified as vehicle spec measures to keep trap speeds in check. Shortening their runs to 1000' was a track spec measure intended for the same reason.

Actually, I don't think full engine power is allowed to reach a TF's tires until well down the strip where aero loading and exhaust thrust have added their downforce contributions. Clutch slip & tuning is apparently a pretty big deal.


Norm

 

[TexasRebel]

The main reason the torque curve fall off in the upper RPM range on a NA motor is because of engine friction and loss of volumetric efficiency at high RPM - ie, the engine can't "breath" as good due to increasing air flow restriction that doesn't' fill the cylinders as efficiently. The cure for that is forced induction. When the torque curve peaks, that is the point where the internal friction is still relatively low and volumetric efficiency is the highest.

Not really. In lower RPM the engine fights itself a little until it gets into the designed band. In the designed band, pressure from the expanding gas is acting on the pistons efficiently, the rods are at optimum angles to transfer the force provided by the pistons into torque on the crankshaft. As RPM increases, fuel continues to burn and expand at the same velocity and the rotating assembly is outrunning it. Think about a bicycle. You don't run out of RPM because of friction. You have to upshift because your leg muscles can't keep up.

I think equating car acceleration to competition tractor pulling is biasing your thinking (sounds better than calling it being in a 'rut', no?). Given that we agree that getting a car moving is relatively easy, you don't need huge torque at the crankshaft to accomplish it.

But the more torque you have, the faster it happens... remember F=ma?

A car's drag loading (aero + rolling) in the first couple of gears is essentially independent of vehicle position (IOW, distance from the 'start'), which is in contrast to loading in the tractor-pull case where loading is very much dependent on position. IOW, for a car there is no sharply increasing load for the car's engine to have to "get ahead of" in order to get some momentum going.

But the theory is the same. You want to transfer energy into the load as quickly as possible. With a car, you have a relatively constant load.

That's a pretty good illustration of the difference between mathematical theory and engineering practice - what the math says is true can conveniently overlook the fact that in "real life", RPM is very definitely limited to some finite value.

yep, and if valves could close faster and combustion could happen instantly, that limit would only happen when friction in the rotating mass and peak torque found equilibrium... As it happens, torque falls off for much different reasons.

If you could even build an engine with a completely flat HP "curve" with enough HP to support a top speed above 100 MPH at any reasonable RPM, you'd have drivability difficulties at 5 MPH. Chances are that for drivability in general (and particularly in wet/worse weather and other poor traction situations) you'd have to do the kind of torque management that would end up giving you a much different actual crankshaft torque curve than the concave-up (1/RPM) shape corresponding to a perfectly flat HP line. Basically, you'd have to choke what the engine was capable of down to something you could actually use, separate from the throttle modulation you do with your right foot.

That already happens. Clutch modulation is a huge input into getting going on slick/icy roads. The torque at the wheels does not directly correspond to available engine torque. Even in an automatic, the torque converter modulates output.

 

[TexasRebel]

My understanding of Top Fuel is that there is so much power that it is unnecessary to provide a transmission at all (do the torque times axle gear divided by tire radius math and compare that to whatever the current weight requirement is). Axle gearing and RPM are specified as vehicle spec measures to keep trap speeds in check. Shortening their runs to 1000' was a track spec measure intended for the same reason.

Actually, I don't think full engine power is allowed to reach a TF's tires until well down the strip where aero loading and exhaust thrust have added their downforce contributions. Clutch slip & tuning is apparently a pretty big deal.


Norm

Yep. Here is a blog post that explains them pretty well.
https://www.highpowermedia.com/blog/3669/top-fuel-clutch-construction

These cars want a pretty high RPM torque peak in the engine. Application of torque to the ground is modulated by the clutch and nothing else. The optimal torque application to the ground is constant... just below the slip line. Maximum torque is maximum force where the tire meets the pavement; is maximum acceleration for a constant weight.

 

[GT Pony]

Not really. In lower RPM the engine fights itself a little until it gets into the designed band. In the designed band, pressure from the expanding gas is acting on the pistons efficiently, the rods are at optimum angles to transfer the force provided by the pistons into torque on the crankshaft. As RPM increases, fuel continues to burn and expand at the same velocity and the rotating assembly is outrunning it. Think about a bicycle. You don't run out of RPM because of friction. You have to upshift because your leg muscles can't keep up.

Yes, there are some other minor factors involved, but the main reason the torque curve is shaped the way it is (and hence the HP curve) on a NA engine is mainly due to the volumetric efficiency of the engine.

Why do you think the torque curve changes when you increase the volumetric efficiency by installing better breathing intake, exhaust and heads/valves? When you do that, you increase the volumetric efficiency a lot, and even in the higher RPM range that give a flatter and longer torque curve. Go do some googling on volumetric efficiency and you'll see what I'm talking about.

 

[TexasRebel]

When you open up the restriction on the intake and exhaust you generally lose peak torque, but shift the peak upward in RPM.

Other than that, your post is in complete agreement with mine, so why the hostility?:shrug:

 

[Norm Peterson]

But the theory is the same. You want to transfer energy into the load as quickly as possible.

And this is accomplished by operating around the power peak to the maximum extent possible/practical. A little less engine torque times a bigger gearing multiplier still beats a little more engine torque times a smaller gearing multiplier.

With a car, you have a relatively constant load.

Meaning that there isn't the same urgency in developing early momentum at very low speeds, because in a car the rate of load increase is in no danger of outrunning the torque curve and grinding everything to a halt.

But the more torque you have, the faster it happens... remember F=ma?

To a point. Once you've got enough low-end torque to spin the tires without having to shock them loose with harsh clutch engagement, still more torque at low to maybe midrange revs pays no dividends. Crutching that with tall gearing just makes it trickier to get just a little tire slip without it turning into a full-blown smoke show.

Not to mention being harder on the clutch in normal driving. I have a little real-world experience here, 500 miles from home having to replace a busted axle gearset (dropped C-clip) on a car whose transmission had a 2.85 low gear and 26" tall tires . . . and ending up with 2.56's because that's all the shop could get hold of. 10.5 mph/1000 rpm is more like where 2nd gear belongs - do-able but not really advisable for always starting up from a full stop.


Norm

 

[Norm Peterson]

yep, and if valves could close faster and combustion could happen instantly, that limit would only happen when friction in the rotating mass and peak torque found equilibrium... As it happens, torque falls off for much different reasons.

I was thinking more in terms of mechanical limitations. F1 was at one time operating somewhere beyond 19,000 rpm with near-unlimited budgets, so I'm not seeing an ICE being developed with an inherent limit much higher than that. Certainly not at the average enthusiast level.

Other than valve spring limitations, you've got matters of con-rod strength, rod and bolt stretch vs piston to head clearance vs squish considerations, vibration modes/adequacy of balancing, lubrication, ring stability, etc., to contend with as potential limiting factors. And in a practical engine you aren't going to want to be refreshing it every few hundred (or fewer) miles. I've built up a couple of engines with file-fit rings and attention to bearing clearances, just so it's clear I'm not talking solely from a theoretical point of view.


Norm

 

[TexasRebel]

I was thinking more in terms of mechanical limitations. F1 was at one time operating somewhere beyond 19,000 rpm with near-unlimited budgets, so I'm not seeing an ICE being developed with an inherent limit much higher than that. Certainly not at the average enthusiast level.

Other than valve spring limitations, you've got matters of con-rod strength, rod and bolt stretch vs piston to head clearance vs squish considerations, vibration modes/adequacy of balancing, lubrication, ring stability, etc., to contend with as potential limiting factors. And in a practical engine you aren't going to want to be refreshing it every few hundred (or fewer) miles. I've built up a couple of engines with file-fit rings and attention to bearing clearances, just so it's clear I'm not talking solely from a theoretical point of view.


Norm

here's where we get to differentiate reciprocating piston ICEs to rotary ICEs and turbine ICEs. A Wankel Rotary engine can hit 25,000 RPM or better, while gas turbines can get into the 100,000s of RPM. Each of these examples use low torque applied at a very high rate and are geared down tremendously at the cost of weight.

Those ultra high rev F1 engines use a ridiculously short stroke and have minimal reciprocating mass (pistons, con-rods) as well as rotating mass (con-rods, crankshaft). Short strokes also keep the reciprocating velocity low so the energy that goes into reversing the direction of anything reciprocating is minimized. As for those other factors, it's all in the design. Rods, springs, bolts, lubrication... The only thing you can't really change is the fuel burn/expansion rate.

I've built and been involved in parts refurbish for engines anywhere from United hit-miss engines to Model Ts, to air-cooled Wisconsins, to counter-rotating marine V8s, with many race applications in between both road and strip. Theory is useless without application, and observation of application refines theories.

 

[Grintch]

So 182hp, 109 ft-lb vs 176hp, 127 ft-lb vs 179hp, 190 ft-lb. All geared to reach the same top speed. Which wins, which loses, and by how much? Any guesses on the enigines that that produced those numbers?

Sorry these were the closest match I could get for HP while tyring to get significant differences in torque using real data. Charts to follow.

 

[TexasRebel]

So 182hp, 109 ft-lb vs 176hp, 127 ft-lb vs 179hp, 190 ft-lb. All geared to reach the same top speed. Which wins, which loses, and by how much? Any guesses on the enigines that that produced those numbers?

Sorry these were the closest match I could get for HP while tyring to get significant differences in torque using real data. Charts to follow.

Where are the HP Tq peaks?

 

[Grintch]

I understand the differences of torque vs Horsepower on paper but how to apply them in the real world is a different story.

Ford offers a couple power packs. Some give low end torque boosts while the others give high end power. If my thinking is correct a drag racer would want high end power as they spend more time up top and also wouldn't the low end torque lead to more wheel spin during launching?

Whereas someone who does track days would probably prefer the torque to help push out of corners.

Is my thinking at all correct here?

The real secret to horsepower and torque improvement claims is that below 5252 rpm, the torque number is bigger. So if you make more power below 5252 you use the bigger torque value in your adds. If you make more torque you also make more power at that rpm.
Above 5252 rpm, the HP number is bigger, so that number gets quoted. Again more power at 7000 rpm is also more torque at 7000rpm (but bigger is better so the bigger number is used).

Now do you want more torque (& power) at 4000 rpm or more power (& torque) at 7000 rpm? Depends. How much time do you drive at 4000 rpm; how much at 7000 rpm? If you shift at 5000 then power at 7000 doesnt matter. If you shift at redline, then you never see 4000 rpm after you leave first gear.

 

[Grintch]

Where are the HP Tq peaks?

Good question, but I am not being tricky so about where you expect them to be based on the engine type.

182Hp [MENTION=17112]10[/MENTION],500 & 109 ft-lb @ 7500
176Hp @7500 & 127 ft-lb@7000
179Hp @5500 & 190 ft-lb @3000-3500

Sponsored