Ultimate Power: Just How Much Power Can Be Squeezed Out Of A Road Car Engine?
02-13-2003, 08:21 AM
#21
Registered User
I don't know Luis, I might take that bet.
For those of us stuck with the English units, 120 nm/liter is about 90 lbs-ft per liter. We've been doing a lot of testing on the K20A and K20A2 engines from Honda lately.
In completely stock format, the K20A2 produces about 125 lbs-ft of peak torque on my dyno (the S2000 produces 135-138 lbs-ft stock). Using OEM Honda parts (intake manifold and cams) and then adding simple bolt-ons (intake, header, exhaust, TB) and reprogramming the ECU torque rises to 165 lbs-ft peak at the wheels. Even if you assume a fixed torque loss (which is my preferred method), that would put the lightly modified car at 91 lbs-ft/liter.
All this on 91 octane fuel and no internal modifications (stock compression, etc.). I suspect that with headwork and a little more compression (still streetable), we'll see that number climb to between 95-100 lbs-ft/liter.
While this does not fit your unmodified car criteria, the ease with which this engine hits those output levels leads me to believe that with the incremental improvements that are made over time, we might just hit them with an OEM car. I'd definitely take the bet if you gave it 10 years. 5 is right on the cusp. Of course, the big problem is so few makers are pushing the limits with NA engines. VW/Audi, Mitsubishi, DC and Subaru are all pushing forced induction. Nissan, Chevy and Ford seem to be pursuing displacement first, specific output second. Outside of Honda, BMW, Ferrari and sometimes Porsche, there isn't a huge movement on the NA front.
UL
For those of us stuck with the English units, 120 nm/liter is about 90 lbs-ft per liter. We've been doing a lot of testing on the K20A and K20A2 engines from Honda lately.
In completely stock format, the K20A2 produces about 125 lbs-ft of peak torque on my dyno (the S2000 produces 135-138 lbs-ft stock). Using OEM Honda parts (intake manifold and cams) and then adding simple bolt-ons (intake, header, exhaust, TB) and reprogramming the ECU torque rises to 165 lbs-ft peak at the wheels. Even if you assume a fixed torque loss (which is my preferred method), that would put the lightly modified car at 91 lbs-ft/liter.
All this on 91 octane fuel and no internal modifications (stock compression, etc.). I suspect that with headwork and a little more compression (still streetable), we'll see that number climb to between 95-100 lbs-ft/liter.
While this does not fit your unmodified car criteria, the ease with which this engine hits those output levels leads me to believe that with the incremental improvements that are made over time, we might just hit them with an OEM car. I'd definitely take the bet if you gave it 10 years. 5 is right on the cusp. Of course, the big problem is so few makers are pushing the limits with NA engines. VW/Audi, Mitsubishi, DC and Subaru are all pushing forced induction. Nissan, Chevy and Ford seem to be pursuing displacement first, specific output second. Outside of Honda, BMW, Ferrari and sometimes Porsche, there isn't a huge movement on the NA front.
UL
02-13-2003, 10:49 AM
#22
Registered User
all this talk about hp/liter and torque/liter is making my corn-hole ache. I am much more interested in the whole car as a system. HP/Weight and Engine weight/HP means more to me. Too me the efficency of the engine is how much power it makes versus the amount of weight it adds to the system. Also, the way the weight of engine is distributed is important too. yes Dohc high technology engines are great and everything, but there is something to be said for simple, compact, low center of gravity, low weight to HP engine. A push rod motor is smaller and lighter (usually) then the same displacement motor that uses over head cam technology. A push rod motor also usually leavs a smaller foot print for the same displacement also. I have nothing against new technology, but a lot of people who say DOHC is better because it revs higher are looking at engines that rev to 12k and beyond. There are many, many pushrod motors out there running up to 8k and beyond which is more then enough for most street cars (not stock but modified). That car I posted above runs to 11k actually.
The S2000 motor and Z06 motor make about the same HP/Weight of the engine, but which do you think responds to mods better?
So if you are worried about the 130lbs of the engine more then making 100's of more HP...maybe and engine like the F20C is the engine for you, but if you are willing to cut weight somewhere else a 150 extra engine lbs might be worth almost 2x the HP. Since we don't have too many 1200lbs cars with 12k redlines running on the streets yet...these HP/liter, torque/liter arguments are purely academic.
Most of this stuff is marketing hype and has nothing to do with the real world. Having a larger displacement motor that doesn't have to work very hard to make 400hp is much more useable then a strung out 2 liter boosting to make the same HP. Check out how the imports did vs mumfords viper. The weather was hot and the conditions were hostile and he owned them. 3 or the cars couldn't even finish. Strung out street cars make very little sense. If we are talking 1300 lbs cars that need 900 HP...that's a different story. Now we need small and light engines...so they have to rev like crazy, but I don't own that car and neither does anyone else spouting off HP/liter numbers.
The S2000 motor and Z06 motor make about the same HP/Weight of the engine, but which do you think responds to mods better?
So if you are worried about the 130lbs of the engine more then making 100's of more HP...maybe and engine like the F20C is the engine for you, but if you are willing to cut weight somewhere else a 150 extra engine lbs might be worth almost 2x the HP. Since we don't have too many 1200lbs cars with 12k redlines running on the streets yet...these HP/liter, torque/liter arguments are purely academic.Most of this stuff is marketing hype and has nothing to do with the real world. Having a larger displacement motor that doesn't have to work very hard to make 400hp is much more useable then a strung out 2 liter boosting to make the same HP. Check out how the imports did vs mumfords viper. The weather was hot and the conditions were hostile and he owned them. 3 or the cars couldn't even finish. Strung out street cars make very little sense. If we are talking 1300 lbs cars that need 900 HP...that's a different story. Now we need small and light engines...so they have to rev like crazy, but I don't own that car and neither does anyone else spouting off HP/liter numbers.
02-13-2003, 12:16 PM
#24
Registered User
Originally posted by Luis
Read the Gerhard Richter interview. A lighter engine is potentially more immediate in it's response.
Read the Gerhard Richter interview. A lighter engine is potentially more immediate in it's response.
02-13-2003, 12:46 PM
#25
Registered User
Luis,
Here is the torque curve. This one is in lbs-ft at the axle hubs, so you'll have to divide by the "ratio" listed in the bottom of the image (which is the combined gear ratio and final drive multiplier). Since then, with a bit more tuning the torque has come up a bit.
As you can see, the torque climbs pretty substantially after VTEC (5200 rpm), but you're still making 85% of peak from 3000 to 9000 rpm. The ability to dynamically alter cam timing is key. In stock form it makes for a pretty flat torque curve, but with the RSX-R parts, you get a bigger peak.
UL
Here is the torque curve. This one is in lbs-ft at the axle hubs, so you'll have to divide by the "ratio" listed in the bottom of the image (which is the combined gear ratio and final drive multiplier). Since then, with a bit more tuning the torque has come up a bit.
As you can see, the torque climbs pretty substantially after VTEC (5200 rpm), but you're still making 85% of peak from 3000 to 9000 rpm. The ability to dynamically alter cam timing is key. In stock form it makes for a pretty flat torque curve, but with the RSX-R parts, you get a bigger peak.
UL
02-13-2003, 12:57 PM
#26
Registered User
s2kpdx01,
I'm in general agreement with you about power/weight and power/volume being more important from an overall car development perspective. But, when talking engine technology, looking at efficiencies is a useful way to compare how well an engine extracts energy.
On the topic of weight, you're quite correct in your LS6 comparison. But, you forgot the cascade effect. You can't just increase the size of the engine, you've got to increase the size of many other things too. Chevy has done a great job of keeping weight down in the Vette, in no small part due to composites, but a Vette convertible is quite a bit heavier than an S2K (for many reasons). I use the convertible because a hardtop should inherently be lighter.
I've often considered transplating a J32 into the S2K for fun. I haven't been able to get exact weight specs, but it appears its about 60-70 lbs heavier (engine only). That would move the weight bias forward from 50/50 to about 52/48. Not bad. Adding in an LS6 (assuming you didn't need to add the heavier tranny), would move that bias forward to about 54/46. Still not a bad balance, but you are getting away from the ideal and you'll have to compensate with handling setup changes.
Another thing to consider is that larger engines inherently produce more wheel torque at lower rpm ranges. Generally this is considered a good thing. But having driven some C5 Vettes on track now, I can tell you it makes the driving experience a little more difficult. I now know why they have a stability control system - the throttle is a weapon. For a good driver that's a useful tool, but for most, the extra torque hit can be difficult to manage, and not of much use in lower speed corners. Honda's FWD roots make this an even bigger issue. I'd always wondered why Honda kept torque numbers down (especially seeing what we can achieve with simple tuning). Well, I drove the RSX whose power curves are posted above on track. It was a handful. Despite an LSD and semi-race tires, wheelspin was a problem everywhere in the first two gears, and even occasionally in 3rd. The extra power was difficult to use anywhere but in a straight line. It certainly wasn't any help on corner exit.
The integrated approach leads to some interesting compromises, and I still find myself realizing where they occur and why. I suspect that will continue as long as I keep getting into new cars and driving them. Drivability, especially for OEMs but also for race vehicles (witness Honda's CART motors and the new RC211V engine), is often an overriding factor.
UL
I'm in general agreement with you about power/weight and power/volume being more important from an overall car development perspective. But, when talking engine technology, looking at efficiencies is a useful way to compare how well an engine extracts energy.
On the topic of weight, you're quite correct in your LS6 comparison. But, you forgot the cascade effect. You can't just increase the size of the engine, you've got to increase the size of many other things too. Chevy has done a great job of keeping weight down in the Vette, in no small part due to composites, but a Vette convertible is quite a bit heavier than an S2K (for many reasons). I use the convertible because a hardtop should inherently be lighter.
I've often considered transplating a J32 into the S2K for fun. I haven't been able to get exact weight specs, but it appears its about 60-70 lbs heavier (engine only). That would move the weight bias forward from 50/50 to about 52/48. Not bad. Adding in an LS6 (assuming you didn't need to add the heavier tranny), would move that bias forward to about 54/46. Still not a bad balance, but you are getting away from the ideal and you'll have to compensate with handling setup changes.
Another thing to consider is that larger engines inherently produce more wheel torque at lower rpm ranges. Generally this is considered a good thing. But having driven some C5 Vettes on track now, I can tell you it makes the driving experience a little more difficult. I now know why they have a stability control system - the throttle is a weapon. For a good driver that's a useful tool, but for most, the extra torque hit can be difficult to manage, and not of much use in lower speed corners. Honda's FWD roots make this an even bigger issue. I'd always wondered why Honda kept torque numbers down (especially seeing what we can achieve with simple tuning). Well, I drove the RSX whose power curves are posted above on track. It was a handful. Despite an LSD and semi-race tires, wheelspin was a problem everywhere in the first two gears, and even occasionally in 3rd. The extra power was difficult to use anywhere but in a straight line. It certainly wasn't any help on corner exit.
The integrated approach leads to some interesting compromises, and I still find myself realizing where they occur and why. I suspect that will continue as long as I keep getting into new cars and driving them. Drivability, especially for OEMs but also for race vehicles (witness Honda's CART motors and the new RC211V engine), is often an overriding factor.
UL
Thread
Thread Starter
Forum
Replies
Last Post