Yet another long winded post from me! So, Let's play!

Here's a simple graph. The blue line represents torque.

http://gmthunder.com/tino/torque.jpg

Tech A says: Torque is equal across the band, therefore an engine exhibiting
this torque graph will accelerate just as well from 2000 - 3600 RPM, as it will
from 4000 - 5600 RPM.

Tech B says: Not true. The torque curve represents volumetric efficiency of the
motor. This motor happens to produce the same VE across the entire spectrum,
however, there is more average torque from 4000 -5600 RPM, than between 2000 and 3600 RPM.

Tech B also says, "If you don't believe me, calculate the horsepower at any
RPM and compare the values.

Since Horsepower is a measurement and not a force, we can use the HP
curve to show where the most torque is available (average)."

>>>>>>>>>>>>>>>>>

2000 RPM / 60 seconds = 33.33 strokes per second (one cylinder)

1 / 33.33 strokes per second = 0.030 mseconds per stroke


4000 / 60 seconds = 66.67 strokes per second (one cylinder)

1 / 66.67 = 0.015 mseconds per stroke

At 4000 RPM, there are twice as many power strokes per second which
produces more energy in the same amount of time (over one minute, or one second).

From those calculations we can infer that a torque curve which tapers off
at a rate of 300 ft./lbs. @ 1000 RPM to 150 ft./lbs. @ 6000 RPM will continue
to accelerate across the RPM spectrum according to the HP curve.

I like Tech B...does anyone else like Tech B, or have anything else to add?

 

Both A and B need to think about this a little more. As speed goes up, the force needed to accelerate further goes up and the rate of acceleration goes down even though the force remains constant. This is mostly due to aerodynamic drag, though other sorts of friction play into it. In a frictionless model, at less than relativistic speeds, with a constant force, there will be constant acceleration. The simplest model I can think of to illustrate this is an object falling in a vacuum. It continues to accelerate at the same rate as a constant force (gravity) acts on it.

Here's the best explanation I have seen of these concepts: "Any given car, in any given gear, will accelerate at a rate that *exactly* matches its torque curve (allowing for increased air and rolling resistance as speeds climb). Another way of saying this is that a car will accelerate hardest at its torque peak in any given gear, and will not accelerate as hard below that peak, or above it. Torque is the only thing that a driver feels, and horsepower is just sort of an esoteric measurement in that context. 300 foot pounds of torque will accelerate you just as hard at 2000 rpm as it would if you were making that torque at 4000 rpm in the same gear, yet, per the formula, the horsepower would be *double* at 4000 rpm. Therefore, horsepower isn't particularly meaningful from a driver's perspective, and the two numbers only get friendly at 5252 rpm, where horsepower and torque always come out the same." So, A is closer than B but still off the mark.

The quote is from http://www.4x4abc.com/jeep101/torque.html

Rich

 

Interesting article.

How about the torque average? Tech B would be closer than Tech A?

The acceleration explanation makes perfect sense however.

EDIT - After reading a little deeper, I'm not so sure I agree with some of the
tests in that write up.

Accelerating the car at Horsepower Peak? Isn't that like saying, "go run as
fast as you possibly can...now run faster?"

Not very fair to the motor when it's already beyond peak power?

Now that I think about it, if I'm cruising at 30 MPH and nail the throttle, I'm
pretty sure it would snap me back more than rolling at 20 MPH and slamming
the accelerator?

My torque peak is at 2500 RPM

http://www.gmthunder.com/tino/dynograph.jpg

Hmmmmmm....time to grab the keys and try it.

 

Yep, what Rich said.

A couple of points, Tino:

Very few internal combustion engines have a torque "curve" shaped like this, as you know.

Acceleraton gs in a given gear with this torque curve would be constant if drag & driveline friction were constant. They aren't so gs would decrease as the vehicle accelerated. If you remove aero drag and just count driveline friction, thrust force at the drive tires would be constant (in one gear) MINUS the driveline friction increases.

Now if this hypothetical engine was accelerating a vehicle from one speed to another and that speed range wasn't within the rpm limits of one gear, max accelaration would involve multliple gears as well as available traction....but that's a much more complex issue, as you know.

Isn't average torque for this engine from 1000-6000 (or is that 60000 ) 300 lb-ft, the same as the max torque?

Obviously the hp changes with rpm. Note that it is a straight line.

If you want max acceleration from this vehicle, choose a lot of trans ratios close together and keep the engine in it's upper range. Why? One reason is higher rpm for a given vehicle mph = more gear ratio needed and therefore more torque (and thrust) at the tires. Wouldn't a CVT work nice here?

FWIW, OEM vehicles with flat torque curves work fairly well with transmissions with wide gear splits. About the best you'll find (NA) is 90% of peak torque from converter stall to WOT shift point. Yeah, it feels like a jet aircraft accelerating down the runway...very even push in each gear.

 

No, I've found the secret that all F1 teams are searching for. My street motor
actually turns to 60,000 RPM and has an even torque

I don't know, and that's what I'm trying to understand.


Obviously, if we drop the clutch at 4000 RPM, this hypothetical car with a good hook will take off harder than at 2000 RPM, even though torque is
EVEN.

If we look at my chassis dyno graph, the torque rolls off, yet the launch kicks
me back much more if leaving the line at 4000 RPM, rather than 2500 RPM:

http://gmthunder.com/tino/dynograph.jpg

So...how would I explain shift points, and launch RPM? I've been taught
to use the HP curve?

 

Yeah, as your last paragraph says the feel is very similar. Jets feel that way because even though drag increases so rapidly as the speed rises, the engines make more hp at higher air speed due to the "ram effect". So, they have a pretty constant or even increasing rate of acceleration at these relatively low speeds.

Interestingly, percieved acceleration is greater when an engine is "peaky". I am sure many of you have experienced this if you have driven a wide variety of vehicles. It's not only the rate of acceleration you feel, it's the rate of increase in the rate of acceleration. One of the fastest feeling vehicles I ever owned was a Honda 650 Turbo motorcycle. Damn, I should have kept that bike. One of the first fuel injected mtorocycles. Anyway, as a 650 twin in a heavy bike, it didn't accelerate very quickly until it built boost and there was a fair amount of turbo lag. But when the boost came on and the rate of acceleration steadily climbed as the speed increased you got a real "jet plane" feeling which was quite exhilarating. I replace it with another Honda - 1000cc V-4 sportbike which was a fair bit faster. It had a very flat torque curve with a relatively constant acceleration rate and felt slow in comparison to the 650.

I prefer a somewhat "peaky" or "cammy" engine as far as feel goes.

Rich

 

That's because of the stored energy in the rotating parts. With the hypo car, as with any car, one of the reasons to launch high is that it allows you to use energy "stored" in the form of kinetic energy to move the car. Ideal stall speed for a converter, traction issues aside, is near peak torque.

This gets to the issue of flywheel weight. It's a trade off. A heavier flywheel allows you to launch harder but will accelerate slower. What is ideal will depend on the particualr combo. These effects are relatively small. But in general, a street car or any car with a manny tranny is going to benefit from a relatively lighter flywheel due to the wide rpm range the motor will see. A drag car with an automatic and a race converter operates over a relatively narrow rpm range and might be a better candidate for more mass in the flywheel as the rpm range is so much narrower. The flywheel, of course, has another major function which is to prevent stalling and improve smoothness at low engine speeds and damp vibration at higher speeds. More mass helps in that context.

Rich

 

With a straight line torque curve, avg torque = max torque. Over any small interval avg = (min + max)/2. Avg =(300+300)/2 = 300, right? Dyno software will sometimes average torque (or hp) over 100 rpm chunks, or whatever they get in their sampling rate. If you consistently use 100 rpm intervals you can compare engines fairly well. See Engine Masters scoring.



(oops, Rich beat me to this!)
Launch is different from steady state acceleration. If your tires hook, the energy stored in the engine/flywheel combination at 4000 is dumped into the driveline. If I recall, the energy is proportional to the square of the rpm, so 4k is 4 times as much as 2K rpm. IMO, we should ignore launch to analyze the flat "curve" you originally proposed.
There's a ton of discussion on launch, shift points, hp vs. torque etc, ad nauseum here and on other forums. Study them for a while. I'm not sure what all the spikes are in your dyno curve, but it ain't engine output!

If you really want to get folks all riled up again, take the position that it is a force at the tire contact patch and the reaction force in the opposite direction at the chassis that accelerates a vehicle. That force is torque at the drive wheel/rolling tire radius. Throw in the rate of application of torque, and stand back.

On second thought, don't, we've all been thru that many times.

 

Those spikes would be the tranny shifting down (TH-350) during the sample.

Thanks for clearing this up.

EDIT:

Shift points using HP peak + a few hundred RPM seems to work best,
"straddle the curve" as they call it. I've been able to get repeated best
1/4 mile results with this method.

Luckily, my G-tech accelerometer is on the way and I'll be able to test these
points of discussion once I learn how to use the device!

Now you know the inspiration for my post!

 

i thoroughly enjoy reading responses to what is essentially the old hp vs tq question.
i will add my damn near irrevelant observations as always.

stored energy is a topic---the exact same SBC will be a totally different animal with say, a heavy steel flywheel for improved launch on a drag car compared to a flywheel setup for the 30+ shifts at a track like sonoma.
a beefier torque curve is necessary in some applications, while peak power is what counts in others.obviously.
i am of the belief that one can go too far in either direction ,though. I have always thought some engine builders like Lingenfelter[HOW DARE some peon disdain the mighty John L!!!!-no disrepect intended] take torque too far.honestly, one can only blow the tires off so much. example "magazine" car builders are putting twin turbo SBCs in cars now routinely----tell me how your ETs do with 900 ft/lbs going to 20 in 285 radials???
anyway it has always been a delicate balance dictated by application--i wish i could find a superram to borrow to illustrate that even a mundane SBC like mine will respond to higher hp/revs than overall tq peak, especially in a severely traction limited application,as almost all street cars are.
I have a trans with THE worst rpm drop of all, and i have noticeable increase in ET by shifting past hp peak in 1/2 to drop back into the HP section of the dyno graph, as opposed to shifting early to get into the meat of the tq band.
im not saying put a 283 with 11 deg heads and 5.14s i your car, but i dont think its that critical.
take this for what its not worth.

 

Well, I hate HP vs. Torque threads and this wasn't supposed to be "one of those'.

This was more of a...torque vs. torque discussion.

Once I get the g-tech hooked up, I'll post some data and you can all pick it
apart.

I still swear, that after punching my 1996 V6 3.8L STOCK Camaro in 2nd gear,
or first gear...I feel more "g's" at 4000-5000 RPM, than at 2000-3000 RPM.

NOt really sure of the torque peak, but I would imagine it's within 2K- 3K RPM?!

It also seems that I can complete that sprint in less time from 4K to 5K RPM.

Time to put the SOTP dyno away, and use an accelerometer to train my brain.

(FYI:This is a 5 speed transmission)

 

If you can record data from the G-tech, do a number of pulls in each gear recording each separately. Plot the acceleration (longitudinal) g vs. mph, convert the mph to rpm for each gear and compare to your dyno torque curves.

Overlay the plots, and where the longiitudinal gs in 3rd (for example) equal the gs in 2nd, that's a good shift point. It may well be a different rpm for each gear. If you do this in mph it will be easier, and you could choose mph shift points.

FWIW: The development of racing engines has always held to the dictum of the great automotive engineer Ferdinand Porsche that "The perfect race car crosses the finish line in first place and then falls to pieces."

Dr. Porsche didn't have to run under today's F1 rules.

Dr. Porsche probably had trouble keeping drivers if he related this dictum to them.

 

I never liked Porsche (the car), but I thought that quote was pretty cool.

Maybe I'll go back to my old signature:

"Welcome to the Internet, where everyone's a mechanic and has a ten second car."

I will certainly test your points when acquiring the data. I'll take some
video as well as the G-tech data.

The G-tech will allow me to download to computer recording g's, RPM,
and allow overlays too!

http://gmthunder.com/tino/gtechpic.jpg
http://gmthunder.com/tino/passRR.jpg

 

this was YOU!
i have quoted this many times, and love it.


oh and my car is almost 10 seconds................................

 

Hell yes - the original baby!

I guess if I can't be known for engine smarts, a signature can make me famous!