The question I haven't seen asked is why the turbo car has as lousy of transient response as it does.
There are plenty of real-life vehicles producing similar power with far more tractable torque curves and throttle response...at least it seems so to me.
There are plenty of real-life vehicles producing similar power with far more tractable torque curves and throttle response...at least it seems so to me.
Hehe, teams orders carried out overtly may not be allowed, but thats hardly the end of all team orders. Just the really tacky ones lacking subtlety. 
Hey, I have a gun in the drawer!
Not really, just some hunting rifles and shotguns. Remember that not all Americans are against sensible gun control...
Hey, I have a gun in the drawer!
Not really, just some hunting rifles and shotguns. Remember that not all Americans are against sensible gun control...
I still stand by the Riemann sum comment, because its one of the only actual practical uses I've ever seen for Riemann sums.
Yeah, so I'm definitely NOT a NASCAR fan, but this is just wrong.
You need to do just that, except with 800bhp and a live rear axle and unsettling transitions from asphalt to concrete, all with 42 other cars nearby. Read: Bristol/Martinsville.
You know how sometimes watching college basketball is more fun than watching pro ball because the kids make mistakes every once in a while?
Well, thats why I go to Watkins Glen for the Cup/Busch races whenever I get the chance. Although ten or twelve guys have the chops, its hilarious to watch the other thirty throw 3400/3600lb autos around left and right with no real clue how to do it properly. And it always points out that the northeast indeed DOES have its fair share of 'necks and hicks.
Its quite good fun, really. Big party.
Well, thats why I go to Watkins Glen for the Cup/Busch races whenever I get the chance. Although ten or twelve guys have the chops, its hilarious to watch the other thirty throw 3400/3600lb autos around left and right with no real clue how to do it properly. And it always points out that the northeast indeed DOES have its fair share of 'necks and hicks.
Its quite good fun, really. Big party.
This is a confusing statement. If you want to make the gearing work best to get from a to b, you gear the vehicle to have the maximum area under the engine's power curve for the event...maximizing the area under the engine's torque curve might come close to coinciding with this, but in many cases won't.
If you have any transmission, you maximize the area under the power curve for the time spent going from A to B. You are confused by thinking that a CVT is anything more than an optimized multi-speed transmission.
You believe that optimizing the area under the torque curve is optimal. I don't. I'd like to provide an example as proof.
Lets say we have a modern engine tuned with electronic turbo wastegate to have a flat torque peak from 2000 to 6000 rpms. After 6000 rpms, the torque decreases 10 percent by 7000 rpms.
Your theory would show the car to accelerate fastest from A to B if shifted between 6000 and 2000 rpms, where the area under the torque curve is greatest. However, the car will actually get from A to B faster if you don't upshift until some point past 6000 rpms. Also, if you have enough ratios, it will get there much quicker if you don't make it shift all the way back to 2000 rpms. A gearbox spending time on a torque plateau between 5000-6000rpms will get there faster than one on the same plateau between 2000-6000, regardless of "area under torque curve". The important quantity thus becomes more obviously the power generated over time.
I'm going to leave the actual math as an exercise to the reader (mainly because I don't have Excel on this computer, and my Excel macros aren't playing nice with OpenOffice), but there are numerous programs on the web that will allow you to simulate this and bear this out. Importantly, by shifting after the torque peak of an engine with a wide, flat torque curve, you don't maximize the area under the torque curve. You also don't necessarily go slower, you may go faster.
No, but if you let go of it, it goes further.
The danger here is that you risk missing the forest for the trees. You don't actually care that much about instantaneous accelerations unless you are trying to write a traction control system. You care about the average acceleration over time, and optimizing it. You do this by optimizing the torque delivered to the wheels over time. Your stumbling block is understanding that this is different than the torque delivered by the engine over time.
I already did a graph for you using your equations of power = force / time, and it quite clearly showed that the peak wheel torque and the peak wheel force (and thus the peak acceleration) occurs at peak torque for a fixed ratio when derrived from both torque AND power.
Thankfully, we don't have single speed car transmissions. We are offered a number of ratios to use. You admit that the CVT goes from A to B fastest when used at peak power, but refuse to recognize the same relationship in a 6MT. (a lower resolution CVT, as Ball Bearing Turbo correctly put it)
Since you bring up the calculus, if you determined the perfect peak power gearing from A to B for a CVT, the optimal gear ratios for a 5MT would be the interval points on the CVT ratio curve for a midpoint Riemann sum where n was the number of ratios.
Road and racecars alike almost never have those gears, because other concerns are more important than a particular performance from a to b.
Quite correct. You need to admit to yourself (hey, publically if you want, I don't care) that optimizing the area under the wheel torque curve doesn't necessarily mean optimizing the area under the engine torque curve. Particularly if you have something with a broad and flat torque curve, such as a modern turbocharged gas engine or an electric motor or a gas turbine.
Yes, the torque curve is the useful bit of info. Sometimes they are hard to come by.
This is why people are being confused when they state that "peak power" is worthless. On the contrary, peak power tells you more than peak torque. If you have the opportunity to grab a valid torque curve, you are completely set. The confusion sets in when you misinterpret the use of that curve and make dangerous assumptions about areas under curves.
Oh, race cars eh? Well, I turbo'd a track day car once, crashed it, now I'm going to do so again this summer. Unless I had a lot more money, this discussion becomes rather moot because I don't have the money for bespoke gear clusters.
I would hope everyone involved would learn something. I learned my roommate has a copy of Excel to let me borrow. Amongst other goodies.
Yet even in its awful, unrefined form it won a race, IIRC.
Also, I do believe they were using either a locked center or locked front diff. Obviously that would explain the unpleasant handling bits. To say that modern traction control and modern differentials offer much more is an understatement.
Whether or not "clever" 4wd will be faster depends entirely on the set of regulations. Mandated low power makes it less viable, mandated small grooved tires makes it more viable. It would be rather fun to see. Its really unfortunate that the golden days of competition for the technical sake of it have been more or less replaced by spectator sport with controlled costs and huge rules packages.
The point I was making is that it matters not if you are using a CVT or MT. If you are running at WOT and accelerating, you seek to run the engine at peak power. Hopefully the dudes that designed your gearing made that easy.
Yes. Thankfully, we have gears so we can run the engine at peak power rather than peak torque, maximizing the area under the curve of acceleration vs. time.
Depends on what you are trying to prove with it. It does prove what you stated above. It does NOT prove:
Which is from experience quite wrong...I'd have to have awfully wide ratios for me to settle with being in a gear lugging the motor that far below the power peak. In practice, that wouldn't happen in fourth gear. Second gear, yes, as the ratio is very wide between first and second in street cars.
I know you've studied the calculus, so I don't understand how you can make such an obviously false statement.
Point of the calculus is to make an infinitely large number of infinitely small things the same as something singular and variable. Area under a curve, for instance, or volume of a solid of revolution.
You made this up. You can run a CVT at peak torque. Sometimes, you do. Particularly when trying to save fuel, as BSFC usually hits its lowest point around the peak torque figure.
When trying to get someplace in a hurry, you run the CVT at peak torque. For exactly the same reasons you run the infinite-number-of-ratios MT at peak power.
Thank you very much for providing the solid math that I've been avoiding typing up because it shouldn't be necessary really to illustrate a simple concept.
I do believe, however, that while its obvious that torque and power are directly related when one has a dyno sheet to work with and an application in mind, power figures are vastly more useful for making cursory comparisons than any simple torque figure can be. It is my contention that this is why we have the unit power, because it offers a convenient and effective means of measuring work done over time.
When it comes down to it, no one really cares how much torque you have at any given moment when looking at the big picture. This is why AFAIK every thing under the sun that produces power is rated in units of power, not units of torque or force.
When you calculate the resistance of a hull in water at 15 knots, you go to the engine supplier with an estimate of how much horsepower you need, not how much torque you need. Same for building your train, plane, automobile, lawn mower, and everything else on earth. The torque figures and curves are needed for more subtle reasons, like making parts with appropriate duty cycles (re: speed of parts) and gearing that is efficient and won't break. For the race car...those are all secondary or tertiary considerations.
Wow. Forest for the trees.
It takes a really cursory understanding of physics to understand that running the prime mover at peak power is going to get the car down the track the quickest, fastest, whatever. There is no need for spreadsheets and tractive effort curves to make this simple statement. Peak torque, indeed, has NOTHING TO DO WITH IT.
Jeebus...how many times do I have to point that out? Its like I'm writing into a void. Electric motors (usually) have peak torque at 0 rpm. The power peak is anywhere from a few to several tens of thousands of rpm later. The torque peak has zero to do with the power output of the motor. It has zero to do with how fast the car moves from any to point to any other point.
You are making this all up in your .confused head. Colcob already gave you the hint, and you didn't take it. Torque at the wheels, guy.
If you actually believe running the motor at peak torque is the best way to pass someone, you need to take a step back and look at some simple logic.
-At 40mph, the engine running at peak torque with a total gear ratio of 5:1 makes 5x peak torque at the wheels.
-If you downshift to nearer peak power and change the ratio to 7.5:1, you need to be making less than 2/3rds peak torque at your power peak for the car to accelerate slower than it would at peak engine torque. Very few cars make less than two thirds peak torque at the power peak. Reference widely available dyno evidence. If you make 80% of peak torque at the power peak, you now have 6x the peak torque at the wheels. Gee, that might be quicker to get around the "lorry" or whatever you call them.
Duh. Now downshift to a gear right at peak power and gun it. Bingo---bigger g-meter reading.
Who's books? Seriously, what the heck are you talking about? I don't know of a single quality race/automotive engineering book that claims that a torque figure is more relevant. Several of them extol the virtues of a smooth and tractable torque curve, but no right minded racer is ever going to sacrifice power for torque. If you can increase specific power output at the cost of the specific torque, you do it.
The torque curve is solely relevant for the purpose of developing the appropriate gearing, and making the driver happy with transient responses. The only thing relevant to getting from a to b faster than the other guy's is the peak power.
If your racing series limits what you use for gearing, the shape of the torque curve may gain more importance. That is an arbitrary bit of importance, however. As noted, airplanes fly perfectly well with internal combustion engines making 500ft-lbs of torque, but even better with a PT6 developing 50.
This is just woefully untrue.
Some rather obvious (tired, even) examples come to mind:
-I have an electric motor here capable of peak torque greater than an F1 engine. It produces less than 5HP.
-I have seen hydraulic wheel motors capable of similar feats.
-In terms of more plausable stuff in the automotive realm, the peak torque of the 3.0 liter Vulcan in my van is 165 ft-lbs. Same as the motor in my Mazda. The motor in my Mazda produces an extra 70 horsepower. It gets from A to B much quicker. Yes, in the same vehicle with the same gearing, the Vulcan is capable of the same peak instantaneous acceleration figure. Speaking of dumb figures, there is one right there. The only thing that matters is the average acceleration over time. High peak accelerations just break things and break loose tires.
-Importantly, the reverse of the non-auto examples above do not demonstrate a situation in which a power figure is "worthless". The torque output of the power turbine of a PT6A is rather tiny, but gear the thing from 20,000+ rpms to 2100 rpms at a propeller, and you've got more than a 1000ft-lbs and hundreds of horsepower. No matter what speed the engine moves at, you can find a way to make use of the power it creates.
In summary, a 300ft-lb torque electric motor might chirp the tires, but might not get you anywhere quickly, while a 40ft-lb turbine engine might get you there faster than the 200ft-lb car engine.
There are some really good reasons why you do, as you say, need a torque curve to look at. However, for a cursory inspection or simple matchup, the peak power figure will be vastly more meaningful than the peak torque figure. This is why nearly everything on earth used to provide motivation to twist anything has a power output specified and printed on it, not a torque output.
Maybe the reason why this isn't so obvious to you is that you haven't driven a 1970's Cadillac equipped with a emissions regged big block. Uh, 400ft-lbs of torque, check. Dead slow, check. Less power than a new Camry, check.
I think you are quite correct, Bob, although the inaccurate shortcuts are quite useful when trying to determine what size motor you need to run the lawn mower, or roughly how many horsepower you'll need to drop in the 2800lb auto to run the quarter mile in 13's. For relatively slow cars, the simple three/four variable quarter mile calculators can be accurate within +/- 10%, which is damned impressive considering the small number of variables used. Stuff gets twisted when you get to cars with enough power (meh, torque over a bunch of time) to spin the rubber bits in bigger gears.
Last edited by skiingman, .
Cars that want to get from point a to point b in the minimum amount of time always do so with the engine running at its maximum power, not its maximum torque. These cars have appropriate gearing.
This is commonly accomplished in drag cars by the use of various forms of loose gearing. Drag racing torque converters for automatic transmissions seek to bring the engine to peak power as quickly as possible. In true drag only applications, the point is to have flash stall at or very near peak power, not peak torque. The drag racers then further make use of the change in tire diameter that comes along with thin sidewall slicks, minimizing the number of ratio changes necessary and thereby maximizing time spent at peak power.
Really fast drag applications simply run the engine up to peak power and then slip the clutches just enough to keep it there all the way down the track. The power lost via this inefficient method of delivering energy to the road is far less than that lost by having fixed gearing that requires time consuming shifting and running the engine at points other than peak power.
BBT is correct that power is what determines the time from a to b, and given the appropriate gearing torque is completely meaningless. This is why all rudimentary estimates of time to distance and top speeds utilize power figures, not torque figures or curves.
There really is a huge amount to learn about tires. It seems that the tire industry is one of the most tight-lipped of automotive industries. Probably the best way to get a really great knowledge of how these things work is to get a job in the industry.
I find tires to be fascinating, and would definitely like to learn more about how they work.
I find tires to be fascinating, and would definitely like to learn more about how they work.
There are several factors affecting choice of profile for tires. You meant to say the open wheel cars have big tires and relatively small wheels.
Low profile tires are by no means better in longitudinal accelerations than higher profile ones. In fact, the reverse is usually true.
Reasons to use lower profile tires:
-Heavy weight of car makes developing an appropriately compliant yet also laterally stiff sidewall very difficult at large profiles. The race cars on low profile tires are to my knowledge all quite heavy.
-Larger diameter tire/wheel assemblies increase in weight in a non-linear fashion. Importantly, large wheels combined with low profile tires are far lighter than the opposite combo. Heavy rotating things are bad for race cars, as are heavy unsprung things.
-Large diameter wheel/tires add more frontal area to the vehicle.
-Heavy cars need large diameter brakes, which demands a large diameter wheel.
-edit: Another important one: Tires are relatively undamped compared to race cars. Bigger profile tire means more undamped suspension travel. Perhaps not as big of an issue for high downforce lightweight open wheel cars as it is for big saloons that find proper damping absolutely critical to achieving good grip.
Open wheel cars don't have these problems. Light cars with carbon brakes don't need big brake diameter. The tires can achieve massive lateral grip without rolling over, but remember that the lateral load on an F1 tire at 3g is still less than a big saloon at 1.2 or 1.4g
Watch more closely and often. They light up the rears in an attempt to heat the tires. When actually launching the car down the track, lighting the tires is a bad, bad idea. When used appropriately, the tire maintains a relatively low profile until centripetal acceleration stretches the thin sidewall as the car reaches the big end of the track. This has a beneficial constant change in gear ratio but also a detrimental effect on traction. The downsides to the thin wall are nothing compared to the advantage of slingshot launch.
If the tires get tall at the little end of the track, there is wheelspin and very little traction, and if not quickly corrected, a ugly mess.
As to the times, try 4 seconds. I've got a stockish front-driver that runs 14's...
The effects Vain describes are well known but perhaps not well understood outside of proprietary tire test data. Very, very fast superbikes often show a rear wheel speed much higher than a front wheel speed, despite no obvious (and catastrophic at those speeds) signs of wheelspin. The drag racing tire with its thin sidewall displays in a overall sense the opposite phenom because its diameter grows so wildly.
I think Vain's analysis is spot on. There is a point between pure "static" friction and pure "dynamic" friction that tires can sustain the largest forces in. That point is not easily defined and may well change rather dramatically on different surfaces and under different conditions.
Ever see F1 races where someone's launch control manages to pull four or five whole carlengths on the field in moments? If it were simple, no one would be doing that. A drag racer can tell you that the power required to pull several carlengths on a field of other racers vastly exceeds the power differential between the different teams. It has to come down to more subtle (but incredibly important) differences between teams.
The fact that who gets the best launch (amongst folks with similar fuel loads) seems to change very frequently suggest just how poorly understood or implemented these features are.
Last edited by skiingman, .
Clutch start interlocks=required by FMVSS in the United States since before I was born or so.
OEMs do sometimes have cancel buttons for this "feature" on 4x4 type vehicles.
It'd be interesting to know when this feature became required.
OEMs do sometimes have cancel buttons for this "feature" on 4x4 type vehicles.
It'd be interesting to know when this feature became required.
Reminds me that the vehicle I am most certain had the biggest "lightening" effect past ideal slip angle was a go-kart that probably had 10 degrees of caster, or something insane like that. Very noticeable jacking effect as well, which LFS simulates and few other "sims" seem to.
Bullshit.
Sorry, thats not how it works in real life.
I don't know how many times other people have to post that in most cars in most situations, the wheel doesn't magically go light when the front wheels go past the knee in the traction curve.
The steering may feel "lighter" but its more like the feel doesn't get any heavier and these type of very subtle hints do show up in LFS.
Last edited by skiingman, .
Please tell me how.
/very interested laptop user.
I was using a WR set for a FE track with the XFR the other day, on BL1.
Where the big camber change is in the second turn after the rear straight, I was flipping the thing over just by smoothly driving through the corner without even touching a kerb. Whoa!
Where the big camber change is in the second turn after the rear straight, I was flipping the thing over just by smoothly driving through the corner without even touching a kerb. Whoa!
Technologies like DSG are referred to by the SAE and professionals as AMTs, or automated manual transmissions.
An automatic transmission uses an internal mechanical design of various sort to shift internally without intervention from shift rods hooked to humans or computers. Slushboxes have more and more intervention in their shifting process from the outside world, and in fact begin to look more and more like manual transmissions as time marches on.
SMG, DSG, etc are AMTs.
If you'd ever driven a AMT vehicle for any length of time, you'd understand they have limits. DSG is a phenomenal technology that sidesteps many of these limits, but not all of them. Nor is the consumer-ready form of DSG anywhere close to proven in continued high performance use.
Also, you make the contention that a AMT equipped vehicle can be quicker around a track than a MT counterpart. I've seen very little evidence of this being true in road cars. It is undoubtedly true in race designs where the AMTs can shift an order of magnitude faster than the street legal counterparts, but thats an entirely different kettle of fish.
Last edited by skiingman, .
Not if the automatic is actually an automatic, and not a computer shifted manual box.
The point is that usually grocery getter engines (BMW M engines are hopped up grocery getters, not race hardware) are designed to live for a very long time despite a good bit of abuse and what not. Thus the ability to carefully add a couple bar of boost and have them live for a good while.
The E46 M3 motor was an example of the engineers getting it just a bit wrong. It took them quite a long time to figure out what was going on, which actually involved more than one part and more than one oversight. In the meantime, people were putting holes in blocks with a few thousand miles on them. Similar oversights in a Chevy Ecotech driven by grandmothers might have only shown up after the warranty period expired.
As noted before, sometimes banging against the rev-limiter will cause engine damage unrelated to the stresses of the engine speed. For instance, fuel-cut limiters causing detonation, or spark-cut/retard limiters allowing fuel to cook the exhaust valves. Running any motor against the limiter at WOT is a risky proposition, sometimes its a downright dangerous thing to do.
The E46 M3 motor was an example of the engineers getting it just a bit wrong. It took them quite a long time to figure out what was going on, which actually involved more than one part and more than one oversight. In the meantime, people were putting holes in blocks with a few thousand miles on them. Similar oversights in a Chevy Ecotech driven by grandmothers might have only shown up after the warranty period expired.
As noted before, sometimes banging against the rev-limiter will cause engine damage unrelated to the stresses of the engine speed. For instance, fuel-cut limiters causing detonation, or spark-cut/retard limiters allowing fuel to cook the exhaust valves. Running any motor against the limiter at WOT is a risky proposition, sometimes its a downright dangerous thing to do.
And herein you highlight the difference between product liability attidudes in the US and Europe.
A few stupid people leaving their kids in the car with the keys (or whatever) ruins it for the rest of us.
1988 Honda without clutch interlock=not how it came from the factory.
Its somewhere in the FMVSS, and its been there for a long time.
FGED GREDG RDFGDR GSFDG