The advantages of a NA engine?

[Erhan]

Originally Posted by mkoesel

It boils down to what Bruce said earlier: the car with more power at a given speed wins.

In your example, the two cars will have the same power at a given speed because their power curves are the same except one is multiplied by two (on the RPM axis), while at the same time their gearing differs by exactly a factor of two.

Remember, while gearing multiplies torque, "power at the wheels" is the same as power at the engine, less parasitic losses of course. IOW, gearing does not multiply power like it does torque.

I see what you're saying. Both cars will be producing same power at the same speed. Then once again, high revving engine is better argument is not valid...

I'm confused though, why the power is not multiplied by the gearing?

[pbonsalb]

While it is great to consider the gearing, because it does multiply torque, the spreadsheeters and simulators are still missing some of the equation. The drive of the car does not stop at the differential. My car has tires attached to the differential.

We have already learned that in reality the M3 torque is about 10% less than the 335i torque (because the N54 is underrated from the factory), which drops that 25% gearing advantage to maybe 15%. Let's factor in the tire diameters as well since they affect torque multiplication. The 255/35/18s fitted to the 335i are about 5% smaller than the 265/40/18 fitted to the M3. This has the effect of reducing the gearing -- like lowering (higher numerical) the rear axle ratio -- or multiplying torque. So now we are down to a 10% torque advantage for the M3 . . . starting from 3900 rpm. I am learning a lot from this spreadsheet simulation stuff.

[bruce.augenstein@comcast.]

Originally Posted by mkoesel

Ok, fair enough then. Thanks for the correction.

So, let me see how that effects my example. So, then, we start with two identical M3s. We then modify one of them with a power adder such that it now makes as much power in second gear at 25mph (@ whatever RPM that happens to be - I don't know off the top of my head, and it would differ between 6MT to DCT anyway) as the still-stock one does in 1st gear at 25mph (again, whatever RPM that happens to be - it will definitely be a higher RPM than the other car though). Now, if we put them next to each other traveling at 25mph, and punch it, they will accelerate at the same speed. Well, this is true only for the initial instant the throttle is pushed. After that, presumably, the power curves no longer mirror each other @ the respective engine RPMs so one car will start to pull. But if the curves somehow matched up identically throughout the rest of the rev range, until the car in first gear hits redline, then indeed they would be neck and neck the whole time to that point.

This I am more apt to accept. It does make you shake your head a little though.

From my point of view, the concepts of torque and horsepower tend to be confusing, particularly to a driver. Any car will pull hardest at its torque peak in any given gear, but at its power peak at any given speed.

Counterintuitive.

[bruce.augenstein@comcast.]

Originally Posted by Sedan_Clan

How much "power" does a car really need to make a great daily driver? <--That was a rhetorical question! Like I stated in an earlier post, claiming one is better than the other sounds silly to me. Both the 335i and the M3 have MORE than enough power (...MORE power than a car needs).

Pretty much anyone on this forum isn't thinking in terms of "need", but in terms of "more is better, and too much is just enough".

[bruce.augenstein@comcast.]

Originally Posted by erhanh

I have one question. We always say high revving engine is better. This way we can have shorter gear ratios and take advantage of torque multiplication.

BUT, let say we have two cars with perfectly flat torque curves. Car1 has 200lb-ft torque, 305hp, and revvs to 8000rpm. Car2 on the other hand has 400lb-ft torque, 305hp, and revvs to 4000rpm. For the sake of example, assume the engines idle at 0rpm.

The cars weigh same, use same tires etc. They have the same gearbox, but the differentials are 2x different (or differentials are same but the gear ratios are 2x different) such that they put the same torque to the wheels.

Are these cars will have the same acceleration?

In a sort of theoretical way, the two cars will be even in acceleration, having flat torque curves and making the same power - assuming the cars weigh the same.

As a practical matter, the 4000 rpm car will be quicker in terms of acceleration, because it will have less rotational inertia to contend with.

If you remember Sir Isaac's theories, you know that nothing will move, speed up or slow down without some force being applied. Car 1 has to speed up its engine and associated parts by twice as many rpm per mph gained as car 2, and since these parts don't like to do that, car 1 loses twice as much power to overcome rotational inertia as car 2, leaving less power to actually accelerate the rest of the car's weight. This of course assumes identical rotational inertia for both cars.

You might consider this in terms of the M3. It is very stiffly geared, but BMW has gone to great lengths to reduce/minimize its engine and transmission rotational inertia, so it's a very free revver anyway.

[Erhan]

Originally Posted by bruce.augenstein@comcast.

In a sort of theoretical way, the two cars will be even in acceleration, having flat torque curves and making the same power - assuming the cars weigh the same.

As a practical matter, the 4000 rpm car will be quicker in terms of acceleration, because it will have less rotational inertia to contend with.

If you remember Sir Isaac's theories, you know that nothing will move, speed up or slow down without some force being applied. Car 1 has to speed up its engine and associated parts by twice as many rpm per mph gained as car 2, and since these parts don't like to do that, car 1 loses twice as much power to overcome rotational inertia as car 2, leaving less power to actually accelerate the rest of the car's weight.

You might consider this in terms of the M3. It is very stiffly geared, but BMW has gone to great lengths to reduce/minimize its engine and transmission rotational inertia, so it's a very free revver anyway.

A key item here is that, in the real world, the 8000 rpm car would be making a ton more power than the 4000 rpm car, each with the same torque.

Thanks for the response.

So if the engines in my example are real, can we still say high revving engines are better because they have the advantage of gearing? This is what confuses me.

[swamp2]

Yes, without considering drivetrain inertia or complex loss effects, the cars will indeed accelerate identically. It is a very useful thought experiment.

As a side note on the drivetrain inertia: Using percentage drivetrain losses which CarTest does, it computes the losses to be identical for two such vehicles (when plotted vs. road speed, not vs. rpm). I'm also fairly sure that CarTest includes the effects of drive train inertia and if so it is more or less "noise" between these simulations despite the huge factor of 2 in FD ratio.

[swamp2]

Originally Posted by pbonsalb

While it is great to consider the gearing, because it does multiply torque, the spreadsheeters and simulators are still missing some of the equation. The drive of the car does not stop at the differential. My car has tires attached to the differential.

Tires are part of the torque multiplication system for sure and mixja's effoert would be improved by using wheel radius to convert from torque to force. The corrections would be pretty small, probably 2% max for the cars considered.

Originally Posted by pbonsalb

We have already learned that in reality the M3 torque is about 10% less than the 335i torque (because the N54 is underrated from the factory), which drops that 25% gearing advantage to maybe 15%. Let's factor in the tire diameters as well since they affect torque multiplication. The 255/35/18s fitted to the 335i are about 5% smaller than the 265/40/18 fitted to the M3. This has the effect of reducing the gearing -- like lowering (higher numerical) the rear axle ratio -- or multiplying torque. So now we are down to a 10% torque advantage for the M3 . . . starting from 3900 rpm. I am learning a lot from this spreadsheet simulation stuff.

Yes, correct again.

If you would just also take the simple shortcut to hp and not focus so much on either torque or torque to the wheels you would have it made.

[bruce.augenstein@comcast.]

Originally Posted by swamp2

Yes, without considering drivetrain inertia or complex loss effects, the cars will indeed accelerate identically. It is a very useful thought experiment.

As a side note on the drivetrain inertia: Using percentage drivetrain losses which CarTest does, it computes the losses to be identical for two such vehicles (when plotted vs. road speed, not vs. rpm). I'm also fairly sure that CarTest includes the effects of drive train inertia and if so it is more or less "noise" between these simulations despite the huge factor of 2 in FD ratio.

There is some other function within CarTest which handles rotational inertia. Either that or it's a piece of crap, which I don't believe.

If you remember previous conversations, the smallest amount of acceleration "loss" I found in first gear compared to second gear was 12%, on a car with a three to two ratio between first and second gears, showing .50 G max acceleration in second gear, and .66 G in first. This was on an '85 Vette with an undersized trans (Borg-Warner Super T10) and an 18 pound flywheel. Every other car I tested (and there were at least a couple of dozen) showed a larger loss than that.

Rotational inertia is real, and significant. I know you and lucid never bothered to check Gillespie's book, but you should have.

[Erhan]

So all in all, can we still say high revving engine is better due to gearing?

[bruce.augenstein@comcast.]

Originally Posted by erhanh

So all in all, can we still say high revving engine is better due to gearing?

No.

High revving engines tend to make more power per liter than low revving engines, because if you're a manufacturer, why bother with forged rods and other expensive materials, etc., just to be able to safely rev higher.

You rev higher so you can make more power per liter. Then, as in the M3, you gear the car aggressively so it's easy for the driver to get to automotive nirvana on the right hand side of the tach.

The pinch point on naturally aspirated engines is torque related. At current levels of technology, about 90 pound feet of torque per liter is about as good as it gets, so if you want more power, you've got to make that torque higher up in the rev band. Power equals torque times rpm.

Otherwise, overall gearing is not hugely important - as we've dragged some folks (kicking and screaming in this string) into the realization that it's power and weight that essentially count at any given instant.

So, high revving engines are better because they make more power for a given engine size - and more power makes you go faster.

On the other hand, the quote you've read in this string that says "It's better to make torque at high rpm than at low rpm, because you can take advantage of gearing" is also true. It's just another way of saying more power is better, and that you can apply that power thru aggressive gearing that let's the engine easily get up there where max power is.

[pbonsalb]

Originally Posted by swamp2

Tires are part of the torque multiplication system for sure and mixja's effoert would be improved by using wheel radius to convert from torque to force. The corrections would be pretty small, probably 2% max for the cars considered.



Yes, correct again.

If you would just also take the simple shortcut to hp and not focus so much on either torque or torque to the wheels you would have it made.

Try 5%. Do the math for the 18 inch wheels and tires used on the M3 versus the 335i. It is 5%, not 2%.

Now that we have the M3 torque advantage down to 10%, let's think about what the torque is being asked to move since that should be a consideration as well. The E90M3 weighs about 100 lbs more than the E90 335i, which is about 3%. The weight affects the torque to weight ratio, obviously. Let's lop that 3% off the M3's 10% advantage. We are now down to a 7% real world torque advantage after considering ALL the factors, and that is from 3900 rpm, which is about 2000 rpm higher than the 335i torque peak.

It is easy to see why the 335i makes for a great daily driver from off idle to about 4000 rpm and why it will often outsquirt the M3 around town in daily driving -- the mighty M3 needs some space to catch up. Drag racing off the line or speed contests from its torque peak certainly favor the M3, but I don't daily drive drag race or daily drive around at 3900 rpm.

Again, I like my M3. Got six sets of shelves in it from Lowes last night. Had to make two trips, though. A wagon version would have been awesome.

With my new ARC8s in 10 x 9.5 and 10 x 10.5, I will be running 265/35/18 and 285/35/18. The rears work out to a slight gearing improvement over the stock 265/40/18. This change is more like the 2% you were guesstimating for the 335i versus the M3 stock 18 inch tire sizes. I will add this 2% to my 7% M3 gearing advantage on my spreadsheet simulator.

[mkoesel]

Originally Posted by pbonsalb

Try 5%. Do the math for the 18 inch wheels and tires used on the M3 versus the 335i. It is 5%, not 2%.

Now that we have the M3 torque advantage down to 10%, let's think about what the torque is being asked to move since that should be a consideration as well. The E90M3 weighs about 100 lbs more than the E90 335i, which is about 3%. The weight affects the torque to weight ratio, obviously. Let's lop that 3% off the M3's 10% advantage. We are now down to a 7% real world torque advantage after considering ALL the factors, and that is from 3900 rpm, which is about 2000 rpm higher than the 335i torque peak.

...

But you can't just add up percents like that - it doesn't make sense. They are multiplicative. For example, 10% off 100 is 90. But if you knock off another 5%, that brings it down to 95.5, not 95. The difference looks subtle here, but if you apply them repeatedly as you have in an additive rather than multiplicative way, you end up with a huge error margin.

Just sayin. Now, I get the impression you are math-averse by nature and so I suspect your answer to this will be "Whatever, geek!".

[pbonsalb]

Originally Posted by bruce.augenstein@comcast.

On the other hand, the quote you've read in this string that says "It's better to make torque at high rpm than at low rpm, because you can take advantage of gearing" is also true. It's just another way of saying more power is better, and that you can apply that power thru aggressive gearing that let's the engine easily get up there where max power is.

"Better" is a subjective term. Not everyone wants the motor with 400 hp and 300 lbs of torque and low rear gears that have it running at high rpm on the highway and around town, sucking gas. But yes, it can be done and is the choice for naturally aspirated racing engines. In daily driving, not everyone finds it to be the "better" choice. Gearing has its pluses and its minuses. Drag racing is such a small part of driving for most of us. I spend more time filling up my car at the gas pump than I do drag racing. I might even spend as much time filling up my car each week, if I fill it twice a week, as I do driving it at its 3900 rpm torque peak or higher.

[Erhan]

Originally Posted by bruce.augenstein@comcast.

No.

High revving engines tend to make more power per liter than low revving engines, because if you're a manufacturer, why bother with forged rods and other expensive materials, etc., just to be able to safely rev higher.

You rev higher so you can make more power per liter. Then, as in the M3, you gear the car aggressively so it's easy for the driver to get to automotive nirvana on the right hand side of the tach.

The pinch point on naturally aspirated engines is torque related. At current levels of technology, about 90 pound feet of torque per liter is about as good as it gets, so if you want more power, you've got to make that torque higher up in the rev band. Power equals torque times rpm.

Otherwise, overall gearing is not hugely important - as we've dragged some folks (kicking and screaming in this string) into the realization that it's power and weight that essentially count at any given instant.

So, high revving engines are better because they make more power for a given engine size - and more power makes you go faster.

On the other hand, the quote you've read in this string that says "It's better to make torque at high rpm than at low rpm, because you can take advantage of gearing" is also true. It's just another way of saying more power is better, and that you can apply that power thru aggressive gearing that let's the engine easily get up there where max power is.

Thanks for the answer. So the part I made bold in the quote is the key I think. I believe your reference number, 90ft-lb/liter is for NA engines. So if we can make that 180lb-ft/liter for an FI engine, we can live with a low revving engine, since we'll have power due to torque anyways. Although 180lb-ft/liter sounds very aggressive and may not be practical to build.

[mkoesel]

Originally Posted by erhanh

Thanks for the answer. So the part I made bold in the quote is the key I think. I believe your reference number, 90ft-lb/liter is for NA engines. So if we can make that 180lb-ft/liter for an FI engine, we can live with a low revving engine, since we'll have power due to torque anyways. Although 180lb-ft/liter sounds very aggressive and may not be practical to build.

You don't need that much though because in all practicality the ceiling RPM for useable power with an F/I engine is not going to be half the RPM of a high revving naturally aspirated engine like it was in your example.

In other words, take your example, and make the redline RPM for the F/I engine at 6000 RPM instead of 4000 RPM. Now you can use more aggressive gearing (75% or the gear ratio of the N/A example, rather than 50%) and still get the same power down at the same speed as before. (I think I stated all of that correctly, but if not, you still get what I mean. )

[Dodge2Dub]

Originally Posted by erhanh

So all in all, can we still say high revving engine is better due to gearing?

I believe that we have come full circle. Is a high revving engine better due to gearing?....depends on what YOU want and how YOU like to drive.

[Erhan]

Originally Posted by mkoesel

You don't need that much though because in all practicality the ceiling RPM for useable power with an F/I engine is not going to be half the RPM of a high revving naturally aspirated engine like it was in your example.

In other words, take your example, and make the redline RPM for the F/I engine at 6000 RPM instead of 4000 RPM. Now you can use more aggressive gearing (75% or the gear ratio of the N/A example, rather than 50%) and still get the same power down at the same speed as before. (I think I stated all of that correctly, but if not, you still get what I mean. )

Oh yes you're right. 4000rpm is a little too low

[Erhan]

Originally Posted by Dodge2Dub

I believe that we have come full circle. Is a high revving engine better due to gearing?....depends on what YOU want and how YOU like to drive.

Definetly! I can see the advantege of heaving huge amount of torque with very tall gearing.

[swamp2]

Originally Posted by pbonsalb

Try 5%. Do the math for the 18 inch wheels and tires used on the M3 versus the 335i. It is 5%, not 2%.

It's not me who needs to do that math, its you who needs to read. I said 2% on for the selection of cars in mixja's post. I explicitly agreed with you about 5% for the 335i.

Originally Posted by pbonsalb

Now that we have the M3 torque advantage down to 10%, let's think about what the torque is being asked to move since that should be a consideration as well. The E90M3 weighs about 100 lbs more than the E90 335i, which is about 3%. The weight affects the torque to weight ratio, obviously. Let's lop that 3% off the M3's 10% advantage. We are now down to a 7% real world torque advantage after considering ALL the factors, and that is from 3900 rpm, which is about 2000 rpm higher than the 335i torque peak.

It is easy to see why the 335i makes for a great daily driver from off idle to about 4000 rpm and why it will often outsquirt the M3 around town in daily driving -- the mighty M3 needs some space to catch up. Drag racing off the line or speed contests from its torque peak certainly favor the M3, but I don't daily drive drag race or daily drive around at 3900 rpm.

Yes, this is more or less correct. As long as you want to continue to obsess about engine torque (prior) and then torque to the wheel (later) and now finally torque to the wheels per weight (now). Once you simply graduate to power to weight you'll be getting there. Assuming the 335i makes 320 hp. The numbers are:

335i: 11.2 lb/hp
M3: 8.9 lb/hp

That is a 26% worse value for the 335i. That is why when you choose the right gear and use the car the M3 stomps the 335i from and to any speed and in distance at a given time. If you like the flexiblity of good acceleration in a high gear at low speed and very low rpm that's fine, it is a preference.

You are still missing something about tire size as well. Why don't you simply put a set of RC sized tires on the rear of your car. Torque multiplication is HUGE. For essentially free you would have the highest wheel torque car on the planet. Even if traction was not an issue there is a reason why this does not magically make a car faster by the increased percentage of torque multiplication. This is exactly why you are also confused about FD ratio modifications as well. You must consider what happens not at an instant but ACROSS multiple gears.

[swamp2]

Originally Posted by bruce.augenstein@comcast.

There is some other function within CarTest which handles rotational inertia. Either that or it's a piece of crap, which I don't believe.

If you remember previous conversations, the smallest amount of acceleration "loss" I found in first gear compared to second gear was 12%, on a car with a three to two ratio between first and second gears, showing .50 G max acceleration in second gear, and .66 G in first. This was on an '85 Vette with an undersized trans (Borg-Warner Super T10) and an 18 pound flywheel. Every other car I tested (and there were at least a couple of dozen) showed a larger loss than that.

Rotational inertia is real, and significant. I know you and lucid never bothered to check Gillespie's book, but you should have.

I'm pretty sure the author of CarTest confirmed with me that he included drivetrain inertial effects. I never said it is not important. It clearly is, it translates along with the car and it also takes extra energy on top of that to spin up all rotating components. I just doubted certain details of the math relationships with you.

Other than knowing gear ratios there is obviously no way of knowing how large or what materials (i.e. densities) or sized all of the relevant drivetrain components are. I checked Gillespie, he simply substitutes the so called "mass factor" for not knowing the details about each cars drivetrain. The mass factor is an approximation and certainly does not take into account actual details of the driveline. The mass factor in 1st gear between the 335i and M3 does vary by about 18%. I'll try to check CarTest to see if this "mass factor" is used as part of its acceleration calculation by specifically comparing the M3 and 335i.

[swamp2]

Originally Posted by bruce.augenstein@comcast.

Otherwise, overall gearing is not hugely important - as we've dragged some folks (kicking and screaming in this string) into the realization that it's power and weight that essentially count at any given instant.

I always like to add this clarification to that comment:

For actual production cars with gears and FD ratios that are fairly well matched/optimized to the particular cars engine and redline this is true. However, gearing is very important in theory and the physics of car performance. A big change in a cars gearing after already "optimized" will result in a huge performance change. We are just lucky because cars with very different engines are already matched to their drivetrains. That is the only reason gearing appears transparent in car comparisons and why we can shortcut straight to power to weight ratio.