Jump to content


Photo

moto Gp, torque pulsing and traction


  • Please log in to reply
202 replies to this topic

#1 mariner

mariner
  • Member

  • 1,407 posts
  • Joined: January 07

Posted 06 April 2012 - 11:26

To my shame I haven't spent much time looking at Moto GP bike technology .. until now.

skimming a book on Moto Gp technology I found out its a lot more than just engine tuning and dampers as I had thought. They run "twist by wire" throttles to limit torque and they have played with setting up strange engne firing orders to simulate a big twiin torque pulses to the rear tyre in 4 and 5 cylinder bikes.

The logic, apparently, is that if you give the rear tyre a big spike of torque then let it relax with no torque you can improve traction out of corners , one of a bikes weakest areas. This doesn't seem intuitively logical but as they all tried it there must be something in it.

Initially I thought the idea of rapidly loading and unloading the tyre with longitudinal force would just cause tyre surface break-away and wheelspin, then I began to think of the ultimate traction situation, a Top Fuel dragster. TF cars do not spin the wheels , at least not if they want low ET's. The tyre is hooked up at the start and it just wrinkles as the 6- 8,000 bhp is put into it. That gives 0 - 100 mph in under one second - maybe 5 g traction.

Now people marvel at the TF 500c.i. engienes doing "8,000 rpm" but what is more true is how few revs they actually do in a run. The std. rear tyre is 36" diameter and 114" circumference or 9.4 ft per rev. So with the ( original )1320 ft run its just 140 tyre rotations end to end. All the TF cars run a 3.20 rear end and no gears so the engine does 3.2*140 = 450 revs in the entire run. This is about 900 actual firing impulses to get from 0 to 320 mph in 4.5 seconds which says something for the power of each cylinder . One firing drives the 2,400lb car forward by 1.5 ft.

Now, going back to the Moto GP theory , if the TF tyre circumferance is 9.4 ft it gets just six ( 9.4/1.5) power pulses per rev. Given the huge power per cylinder and allowing for cylinder pressure rise and crank angles etc I have to believe that the actual thrust on the TF tyre is very non linear.Sso, like the Moto Gp theory, maximum traction ( 5g) is obtained whilst pulsing the tyre with torque and not very linear force.

A long intro. to my question - does the idea of getting the most traction from tyre by pulsing in a sine wave of torque have any theoretical basis and has anybody ever analysed or quantified it.

What puzzles me is that I have never heard F1 try it but as they will spend mega-money on any tiny but legal gain I would have expected thm to do so if the theory works?

Advertisement

#2 saudoso

saudoso
  • Member

  • 4,743 posts
  • Joined: March 04

Posted 06 April 2012 - 11:56

You have something like the reverse of it going on with ABS, when you let the tyres go just above traction, sense it and pull back a bit.

#3 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 06 April 2012 - 12:04

I think your sine wave is not very appropriate, it would be more like 'spike-nothing-nothing'. Back in the days of big-bang two-strokes, I've heard an explanation that it was mainly because this way power was delivered heating up only outmost layers of the tyre, which would cool off before reaching the next corner, whereas normal firing order presented the tyre with constant load, causing it to overheat the core of the carcas which was bad. A bike racing-mechanic friend of mine is not very convinced by that explanation, and IIRC is leaning towards the explanation that it was efficient way of getting the power the wanted but with less (average?) torque*.

As for MotoGP engines, you're probably thinking of recent Yamaha cross-plane I4 engine with uneven firing order (not bing-bang type, but still it had a 'spike' and a 'flat'). I don't think it was traction related experiment, and I would venture a guess that they felt that it was worth to have a go on account of (I think it was significantly) reduced inertia of engine internal components. Indeed it does seem logical, because it would produce engine with more usable power/torque (less would be used for accelerating engine components), but I don't think it has brought them much... I seem to remember that they reverted back to flat-plane crankshaft engine the next season.

* this may sound a bit odd... but I remember that even in recent years Ducati seem to have suffered from 'enough power, but too much torque' problem

#4 Tony Matthews

Tony Matthews
  • Member

  • 17,502 posts
  • Joined: September 08

Posted 06 April 2012 - 12:13

Didn't Honda try the 'big bang' firing order in a CART engine or two?

#5 desmo

desmo
  • Tech Forum Host

  • 13,216 posts
  • Joined: January 00

Posted 06 April 2012 - 14:15

AFAIK, this all stems from the Harley-Davidson XR-750 dirt tracker with its idiosyncratic firing phasing caused by its 45 degree included v-angle and forked connecting rod's unbreakable stranglehold on the AMA Class C championship back in the '70-'80s. My assumption has always been that it is to do with the power pulses interacting with a torsional tire mode but your guess is as good (or better) than mine. If this is the case I'd expect the optimal power phasing/firing order to be dependent on tire carcass construction and aspect ratio.

#6 carlt

carlt
  • Member

  • 1,089 posts
  • Joined: June 09

Posted 06 April 2012 - 14:22

The wrinkle tyres reduce in diameter on launch [ when they wrinkle ]
and expand by a surprising amount at speed [ this is the cars 'gearing' ]
I don't know what the actual figures for expansion are - but when tubbing the arches on a saloon drag car running wrinkle tyres , a good few inches have to be allowed for expansion

This will alter your maths a bit ?

#7 munks

munks
  • Member

  • 350 posts
  • Joined: January 03

Posted 06 April 2012 - 16:33

Three things:

1) I don't have any data or experience to back it up, but with all the elasticities in the driveline, I would expect the engine pulses to be smoothed out significantly, especially by the time they reach the ground through the tire sidewall. Is this not the case?

2) I had the same assumption as desmo, which is that it had to do with power pulses interacting with a torsional tire mode.

3) Most of you have probably seen those super slow-mos of the dragster tires wrinkle up, so I presume you noticed that they do some serious stick-slip. I'm actually not sure off-hand if this has to do with power pulses, or just the typical nature rubber grip. Or a combination. But it certainly works out well for them, considering the almost unbelievable results. Why, though? I'm not sure. It could be heat, or maybe it's the intermittent 'cleaning' of the track by one part of the tire, allowing better static grip by another part, or both.

#8 mariner

mariner
  • Member

  • 1,407 posts
  • Joined: January 07

Posted 06 April 2012 - 18:24

To answer Carl's point the 450 revs for the 1320 ft represents 6,000 rpm for a 4.5 second run so I think the 900 power pulses are in the ball park regardless of variation over the run.

I don't know exactly how torque pulses might help traction but "tyre shake" is a big problem in Top Fuel and can make the driver unconscious so there are some major transient forces in the tyres beyond just normal weight.

If you can time the torque peaks from the engine every 1.5 ft of tyre rotation to coincide with a downward component of the tyre internal forces then it becomes better htan a zero sum game on traction

#9 hogits2

hogits2
  • Member

  • 68 posts
  • Joined: March 10

Posted 06 April 2012 - 18:30

It will be interesting to see if the electric race bikes use an 'interrupted' (don't mention gap) power supply to enhance traction.

#10 hogits2

hogits2
  • Member

  • 68 posts
  • Joined: March 10

Posted 06 April 2012 - 18:38

.... 450 revs .. ..900 power pulses


Possibly 1800 pulses?

Unless these are 'big bang' engines.

#11 Grumbles

Grumbles
  • Member

  • 326 posts
  • Joined: September 09

Posted 06 April 2012 - 23:37

AFAIK, this all stems from the Harley-Davidson XR-750 dirt tracker with its idiosyncratic firing phasing caused by its 45 degree included v-angle and forked connecting rod's unbreakable stranglehold on the AMA Class C championship back in the '70-'80s. My assumption has always been that it is to do with the power pulses interacting with a torsional tire mode but your guess is as good (or better) than mine. If this is the case I'd expect the optimal power phasing/firing order to be dependent on tire carcass construction and aspect ratio.


Interesting that it works on dirt as well. Anyone remember the old Yankee 500 bikes? These were basically a pair of Ossa 250 cylinders grafted onto a twin cylinder bottom end. It was found that traction on dirt was drastically improved when they were phased to fire simultaneously. At the time I thought that was completely counter-intuitive - and still do.


#12 bigleagueslider

bigleagueslider
  • Member

  • 894 posts
  • Joined: March 11

Posted 07 April 2012 - 01:53

mariner,

With regards to rear tire traction, road race motorcycles present a different situation than most race cars. Back in the 80's, GP bike riders like Kenny Roberts and Freddy Spencer who started out racing dirt tracks, showed that the fastest way around a road course was to slide the rear end in the turns. Sliding the bike's rear end in turns requires precise throttle control, and getting the rear tire to break loose and start sliding using the throttle is made easier if the engine is "torquey". It takes much more torque to get the tire to initially break loose than it does to keep it spinning. Thus the engine guys intentionally design engines with firing orders that produce big torque spikes, making it easier for the riders to modulate spin of the rear tire using the throttle.

It's not just road race motorcycles that use rear tire sliding to go fast. Think about rally cars or sprint cars. Or flat rack and speedway bikes.

slider

#13 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 07 April 2012 - 02:51

This is a fascinating subject.
If what Slider says is right - here's another aspect. If the torque at the rear tyre was completely smooth, sliding the rear would be a precipitous process - once the traction threshold is exceeded and the rear starts to slide, the friction coefficient starts to reduce, reducing the torque applied by the road to the rear wheel, which then allows the rear wheel to spin even faster and so on. It needs the engine torque to fall rapidly with increasing revs or very quick throttle reaction from the rider or both.

On the other hand, if the torque at the rear tyre is not smooth, the traction threshhold might be exceeded only briefly, causing a "very temporary" slide which is almost instantly "caught" when the torque is reduced between "bangs".

Perhaps this mechanism gives the rider some extra warning of the imminent slide allowing him to more easily apply the throttle control required?

#14 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 07 April 2012 - 03:35

WRT drag racing. Perhaps the advantage of the wrinkle tyre is to allow a much wider range of slip ratios without too great a loss of traction.

For those unfamiliar with the concepts.
1. Slip ratio is the rotational speed of the wheel divided by the speed it would be doing if there was no torque applied (no slip).
2. Race tyres provide maximum acceleration (grip) when the slip ratio is maintained in a fairly narrow range. A typical range might be 1.0 to 1.15. Traction falls off on either side of this range, ie too much slip or not enough slip are both bad news.

Clearly the ultimate Top Fuel car would have electronic traction control which could control driveline torque and maintain the slip ratio at exactly the right value. (The rules do not permit traction control.) Perhaps what is happening during the "stick-slip" operation we have all witnessed is this:

- During the "stick" phase, while the drivetrain is "winding up" the sidewalls of the tyre like a giant spring, the slip ratio is increasing gradually across the "peak traction" region providing near-maximum traction.
- During the "slip" phase, the sidewall spring is rapidly unwound, the tyre surface rapidy skids across the road at less than optimum traction but because of the much higher slip ratio during this phase, a large rotationof the tread belt occurs in a very short time ie a very small corresponding rotation of the drivetrain ie a very large slip ratio.

When you combine the two processes on a displacement basis, the average slip ratio is high because of the large displacement of the tread belt during the slip phase.

When you add the two processes together on a time basis, the average traction is also high because the time spent during the "stick" phase is greater than the time spent during the "slip" phase. The average traction obtained will be significantly higher than would be the case for constant slip at the average slip ratio.

Perhaps this principle also applies for "big bang" torque impulses?

#15 bigleagueslider

bigleagueslider
  • Member

  • 894 posts
  • Joined: March 11

Posted 07 April 2012 - 06:06

This is a fascinating subject.
If what Slider says is right - here's another aspect. If the torque at the rear tyre was completely smooth, sliding the rear would be a precipitous process - once the traction threshold is exceeded and the rear starts to slide, the friction coefficient starts to reduce, reducing the torque applied by the road to the rear wheel, which then allows the rear wheel to spin even faster and so on. It needs the engine torque to fall rapidly with increasing revs or very quick throttle reaction from the rider or both.

On the other hand, if the torque at the rear tyre is not smooth, the traction threshhold might be exceeded only briefly, causing a "very temporary" slide which is almost instantly "caught" when the torque is reduced between "bangs".

Perhaps this mechanism gives the rider some extra warning of the imminent slide allowing him to more easily apply the throttle control required?


gruntguru,

Can't personally say whether it's true or not. It's just what I've been told. I don't have the stones to test the theory on a racetrack.

Here's a cool video though:

slider

#16 MatsNorway

MatsNorway
  • Member

  • 2,087 posts
  • Joined: December 09

Posted 07 April 2012 - 08:48

TF cars do not spin the wheels , at least not if they want low ET's. The tyre is hooked up at the start and it just wrinkles as the 6- 8,000 bhp is put into it. That gives 0 - 100 mph in under one second - maybe 5 g traction.


Not entirely correct. If the tyre hooks up to much it will slip and grip and slip and you get tire shake.
www.youtube.com/watch?feature=player_detailpage&v=XttOe4rerLo#t=6s

Optimum grip is not at zero wheelspin either its at 4-10% or something.

The "racier" TCs got that programmed into them.

www.fasterandfaster.net/2011/10/2012-kawasaki-ninja-zx-10r-gets-racier.html
"The quickest acceleration requires a certain amount of wheel slippage, so to optimize traction, S-KTRC actually allows for optimum wheelspin.’"
http://www.sportride...ol/viewall.html

#17 MatsNorway

MatsNorway
  • Member

  • 2,087 posts
  • Joined: December 09

Posted 07 April 2012 - 08:58

Personally i think its related to something like engine rotational inertia.. or how wide the powerband is.

Engines with little i consider "nevrotic" and nervous. too responsive if you like.

In RC racing its very common to program the ESC to be more "sluggish" less punch and so on. Would not be surpriced to se similar features on the E bikes.

http://www.rctech.ne...ntrol-best.html

Edited by MatsNorway, 07 April 2012 - 09:17.


#18 mariner

mariner
  • Member

  • 1,407 posts
  • Joined: January 07

Posted 07 April 2012 - 09:09

yes, hogits, you are right 1800 pulses per run, so much for my arithmetic skills! Hopefully it doesn't invalidate asking the question though.

One thing I didn't mention , which complicates the power pulse argument on a dragster, is that they use a slipping clutch to bring the wheel speed up to the engine speed in the first part of the run. Don't know if cylinder pulses could get through the clutch slip.

The mention of dirt bikes makes me wonder if the Sprint car community have ever played with firing order. As far as I know they have not

#19 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 07 April 2012 - 10:33

One thing I didn't mention , which complicates the power pulse argument on a dragster, is that they use a slipping clutch to bring the wheel speed up to the engine speed in the first part of the run.

Good point. Certainly affects the number of engine revolutions required for a pass. AFAIK its full throttle all the way and the revs are dictated by the torque level set by the clutch. The clutch torque begins relatively low (lack of aero DF means less grip) so the revs are high - up near the power peak. As the car builds speed and DF the clutch torque setting increases and the engine speed drops. When it gets to the torque peak, the clutch is locked and the car accelerates to the power peak (and beyond?) with no clutch slip. Interestinglythe high revs at the start-line do have an advantage - DF courtesy of thrust from the headers is maximised by the high gas flow. This is reputed to be 1000 lbs or more.

Don't know if cylinder pulses could get through the clutch slip.

Another good point. Cylinder pulses cannot get through a slipping clutch so the torque delivery will be smooth.

Advertisement

#20 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 07 April 2012 - 11:27

Posted Image

If you could see and handle a Top Fuel or Funny Car tire demounted and uninflated, it has very little carcass structure, almost like an innner tube -- you can easily wad it up with your hands. It's like a big, semi-inflated beach ball, no kidding. So when it is mounted and running on the car, its width, diameter, aspect ratio, etc are determined primarily by wheel speed due to centrifugal / centripetal /angular (choose your argument) forces. This is easy to see in videos: in burnouts or at the finish line, the tire is very tall (also narrow) due to centrifugal force. This variation in tire diameter with wheel speed also serves as the vehicle's only transmission as it were.

Since the tire is distinctly lacking in sufficient structure of its own, a significant amount of longitudinal slip aka wheelspin aka wheel speed is required at low speeds to keep the tire's carcass from wadding up against the traction forces at the pavement. In terms of longitudinal slip, in the range of 15 to 30 percent.

This is what tire shake is, essentially: the wheel speed falls too low relative to the grip forces, the tire wads up against the pavement (just as when you wad it up with your hands, see above) and then, while absurdly distorted, loads and unloads in a series of violent oscillations. If you have ever seen it in person or experienced it, TF/FC tire shake is probably the most violent dynamic in motor sport that does not involve hitting a wall. Drivers are knocked out as their helmets are battered against the roll tubes or they simply black out from the extreme vibration.

#21 Canuck

Canuck
  • Member

  • 1,687 posts
  • Joined: March 05

Posted 07 April 2012 - 14:45

WRT drag racing. Perhaps the advantage of the wrinkle tyre is to allow a much wider range of slip ratios without too great a loss of traction.

For those unfamiliar with the concepts.
1. Slip ratio is the rotational speed of the wheel divided by the speed it would be doing if there was no torque applied (no slip).

If there is no torque (force x angular velocity) applied, wouldn't the angular velocity and thus wheel speed be 0?

#22 NTSOS

NTSOS
  • Member

  • 692 posts
  • Joined: February 05

Posted 07 April 2012 - 15:31

Posted Image

If you could see and handle a Top Fuel or Funny Car tire demounted and uninflated, it has very little carcass structure, almost like an innner tube -- you can easily wad it up with your hands. It's like a big, semi-inflated beach ball, no kidding. So when it is mounted and running on the car, its width, diameter, aspect ratio, etc are determined primarily by wheel speed due to centrifugal / centripetal /angular (choose your argument) forces. This is easy to see in videos: in burnouts or at the finish line, the tire is very tall (also narrow) due to centrifugal force. This variation in tire diameter with wheel speed also serves as the vehicle's only transmission as it were.

Since the tire is distinctly lacking in sufficient structure of its own, a significant amount of longitudinal slip aka wheelspin aka wheel speed is required at low speeds to keep the tire's carcass from wadding up against the traction forces at the pavement. In terms of longitudinal slip, in the range of 15 to 30 percent.

This is what tire shake is, essentially: the wheel speed falls too low relative to the grip forces, the tire wads up against the pavement (just as when you wad it up with your hands, see above) and then, while absurdly distorted, loads and unloads in a series of violent oscillations. If you have ever seen it in person or experienced it, TF/FC tire shake is probably the most violent dynamic in motor sport that does not involve hitting a wall. Drivers are knocked out as their helmets are battered against the roll tubes or they simply black out from the extreme vibration.


Yes sir......that is exactly how Eric Medlen died in 2007 as the result of extreme violent tire shake!

Eric Medlen

John

#23 phoenix101

phoenix101
  • Member

  • 268 posts
  • Joined: September 11

Posted 07 April 2012 - 16:35

A long intro. to my question - does the idea of getting the most traction from tyre by pulsing in a sine wave of torque have any theoretical basis and has anybody ever analysed or quantified it.


The theory has no basis. I'm not really sure how it started or why, but motorcycle racing engine developers have known the secret for some time. Yamaha finally made it public leading up to the release of the new crossplane crank R1.



Basically, a naturally balanced engine is better than a 180 crank with balancing components. This only tells half the story, though. Torque delivery to the rear wheel is superior, but the Yamaha engine also supposedly torques the front wheel off the ground. Lofting the front wheel reduces the rate of acceleration as the rider backs off or wheelie-control kicks in. Yamaha crossplane race bikes have wheelie control issues. In a longitudinal layout, the engine would torque severely to the right or left.

Also, what the video doesn't say, imo, is that balancing the engine with the firing order essentially reduces the rate at which the engine can accelerate. In F1 they didn't use balanced engines b/c they needed the acceleration and they had traction control to sort the issue. Eventually, 180-degree cranks were mandated to protect the sound, imo. In MotoGP, the rate of acceleration is controlled by the bike's propensity to wheelie. Screamer config in a racing motorcycle basically creates a machine that wheelies and slides all over the track. Big Bang configurations are more beneficial.




#24 MatsNorway

MatsNorway
  • Member

  • 2,087 posts
  • Joined: December 09

Posted 07 April 2012 - 17:30

Are we supposed to buy that if the engine rotates the same way as before and delivers the same power to the wheel as before it now somehow a magical force is pulling it down for faster acc?

The myth about pulses and power delivery started with Ducati twins..

Edited by MatsNorway, 07 April 2012 - 17:33.


#25 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 07 April 2012 - 18:48

Interesting that it works on dirt as well. Anyone remember the old Yankee 500 bikes? These were basically a pair of Ossa 250 cylinders grafted onto a twin cylinder bottom end. It was found that traction on dirt was drastically improved when they were phased to fire simultaneously. At the time I thought that was completely counter-intuitive - and still do.


Actually, to me it doesn't sound surprising at all- I think that other considerations might be more relevant for gravel and snow (loose surfaces)... Just as locked wheels when braking might shorten the braking distance (wasn't it an argument against ABS when it was introduced- that on loose surfaces braking distances were longer?), I would suspect similar thing happening with 'torque/power spikes' which might cause instant and momentary wheelspin and sweep the looser topmost layer of the surface...

#26 phoenix101

phoenix101
  • Member

  • 268 posts
  • Joined: September 11

Posted 07 April 2012 - 21:23

Are we supposed to buy that if the engine rotates the same way as before and delivers the same power to the wheel as before it now somehow a magical force is pulling it down for faster acc?


No. That's not how it works at all.


#27 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 08 April 2012 - 00:21

If there is no torque (force x angular velocity) applied, wouldn't the angular velocity and thus wheel speed be 0?

Torque is force x lever arm not force x angular velocity.
TF does burnout - lots of torque applied - wheelspeed much greater than roadspeed = large slip ratio.
TF finishes burnout and continues to roll down the track - zero torque applied - wheelspeed equals roadspeed - slip ratio = 1.0

#28 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 08 April 2012 - 00:32

Also, what the video doesn't say, imo, is that balancing the engine with the firing order essentially reduces the rate at which the engine can accelerate. In F1 they didn't use balanced engines b/c they needed the acceleration and they had traction control to sort the issue. Eventually, 180-degree cranks were mandated to protect the sound, imo.

Large fluctuations in "inertial torque" (actually fluctuations in crankshaft velocity) are a feature of in-line 4cyl where all four pistons stop simultaneously, transferring all their momentum to the crankshaft every TDC and BDC. This phenomenon is reduced or not present at all in engines with more cylinders eg V8 F1 engines.

#29 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 08 April 2012 - 01:35

Or cross-plane I4 engines (as I've alredyy guessed that such matters were more influential for Yamaha's choice than any grip considerations)...

#30 mariner

mariner
  • Member

  • 1,407 posts
  • Joined: January 07

Posted 08 April 2012 - 10:52

Thinking a bit more about the dragster launch and tyre distortion the whole thing is really puzzling, I don't know if engine pulses play a part or not but everything happens in so short a distance.

Assuming my arithmetic is right this time (!) 0 - 100mph in one second occupies only 60 - 70 ft of track or 8 to 10 tyre revolutions.

that is assuming 0 - 100 in 1 second gives an average of 50mph ( not quite right I know ). 50 mph is 73 ft/ second so the acceleration to 100 mph takes just 70 ft. With a 9.5 ft tyre circumference that just 8 tyre rotations , excl. slip, to move 2,400lb at 5G.

If the 15 - 30% slip is true it is still only 9 to 10 tyre rotations. I appreciate that the wall wrinkling reduces efective diameter a bit but it is all quite extraordinary in terms of force application and result.

If you assume the effective tyre radius is , say 1.25 ft , and not the full 1.5 ft then , I think, you need 3,000 ft lb torque at the axle for 1 G. That means 4,700 ft lb from the engine to get the required 5g acceleration. ( 3,000/3.2*5).

IF I have got the calculations right this time that suggests that 5,000+ bhp for a TF engine is real or 0 - 100 mph in one second isn't real.

Now that is all simple calculation BUT there is no allowance for the controlled clutch slip they use. So I am puzzled - if my calcs. are right you need all 5,000 bhp to do 0 100 in 70 ft AND you need clutch slip to keep th engine at 6,000 rpm so how does it all really work?

One answer might be that TF engines do actually produce the 7,000 bho plus some people claim, 2,000 bhp goes into clutch slip heat (!) and the rest is the required torque.

Posted Image


Posted Image

Posted Image

Edited by mariner, 08 April 2012 - 11:37.


#31 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 08 April 2012 - 11:36

Here are some representative segment times from the most recent event at Las Vegas. Tony Schumacher, 2nd qualifying round Top Fuel (which is now 1000 ft of course)

60 ft 0.852 sec
330 ft 2.164 sec
660 ft- 3.071 sec 277.54 mph
1000 ft 3.837 sec 321.42 mph


I can attest that the absurd hp numbers bruited about -- 7000, 8000 hp, more -- are legitimate because I have seen the data from load cells on the driveshafts. More exact figures are difficult due to all the signal noise, variables, and brief event durations, but that is the genuine neighborhood.

Tire slip is highly variable in the 60 but as you can see is >100 percent at times.

http://www.youtube.c...feature=related

#32 mariner

mariner
  • Member

  • 1,407 posts
  • Joined: January 07

Posted 08 April 2012 - 12:01

Thanks, Magoo.

I think the overall acceleration works out at just under 4 G ( 3.83) for the whole run.

What is also interesting is that the top speed has hardly dropped with the cut from 1,320 ft to 1,000 ft.

Slightly defeats the original objective of the distance cut though!


#33 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 08 April 2012 - 21:21

http://www.hotrod.co...ne/viewall.html

#34 bigleagueslider

bigleagueslider
  • Member

  • 894 posts
  • Joined: March 11

Posted 09 April 2012 - 02:00

TF dragsters and road race bikes are similar in one respect and very different in others.

TF dragsters and RR bikes are similar in how they use 100% rearward weight transfer to maximize rear tire traction when accelerating in a straight line. But that is the only similarity I can think of.

RR riders slide the rear end in turns because it results in quicker cornering, and not because it results in better power transfer. There is also a difference in the amount of rear tire torque needed to initially break traction and that needed to keep the tire spinning/sliding. Using the throttle to initiate and maintain rear tire spin takes skill, but it is made easier if the engine has high peak-to-mean instantaneous torque characteristics. Breaking the tire loose requires far more torque than keeping it spinning. The high peak-to-mean torque characteristics require less precise throttle modulation.

TF dragsters aren't set up for using tire spin to run quick. Maximum power transfer is obtained by operating the tires right at their point of traction loss. They use controlled clutch slip to modulate power to the rear tires in order to prevent tire spin. They use rearward weight transfer. And they use low structural stiffness in the tire sidewalls and chassis to allow some accumulation of strain energy that would otherwise contribute to tire spin at launch.

slider

#35 gruntguru

gruntguru
  • Member

  • 5,479 posts
  • Joined: January 09

Posted 09 April 2012 - 07:30

Some timeago I created a spreadsheet to calculate acceleration graphs from basic vehicle data. The settings for the following graphs are based on approximate TF properties, in this case
Mass = 1045 kg
Power = 8,000 hp
COF = 2.6
Exhaust down thrust = 731kg (1630 lb) (I needed to set this artificially high to get the 60' times down to the right ballpark. My spreadsheet does not allow for a momentarily high COF at launch courtesy of static friction andlifting of COG as the tyres expand off the line)
CdA (lift) = 1.5
CdA (drag) = 1.45

Outputs from the spreadsheet for the above inputs are:
60' time = 0.9s @ 96 mph
1000' time = 3.7s @ 321 mph
Top speed (with optimum gearing) = 359 mph

Interesting points from the graphs.
- Car is traction limited to about the 1.5 sec, 200 feet mark and power limited above that.
- Acceleration increases during traction limited phase due to increasing DF from the wing.
- Acceleration spike at launch not shown as mentioned earlier.
- Car is till accelerating at 1.3G at the end of the run. This is in spite of the speed at 320 mph being only 40 mph shy of the 360 mph top speed.

I make no claims about the accuracy of the simulation but it does answer a lot of questions and gives a reasonable picture of general trends throughout the run.

Posted Image
By gruntguru at 2012-04-09

Posted Image
By gruntguru at 2012-04-09

Edited by gruntguru, 09 April 2012 - 07:37.


#36 MatsNorway

MatsNorway
  • Member

  • 2,087 posts
  • Joined: December 09

Posted 09 April 2012 - 10:00

peak-to-mean instantaneous torque characteristics.


Could you explain this further. Possibly with a graph?

#37 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 10 April 2012 - 10:29

Here's another interesting tire video...


http://www.youtube.c...eature=youtu.be

#38 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 10 April 2012 - 11:54

Could you explain this further. Possibly with a graph?


Hypothetically speaking, it's rather simple. Say you have peak-to-mean torque ratio of 1 (meaning constant torque), once you have reached breakaway (and here an analogy would be apt- the breakaway point is similar to coefficient of friction, where static coefficient is higher than dynamic... if you want to get something moving and then move it at constant speed, you'd have to reduce the force once it starts to move) you'd have to reduce the torque almost immediately because until you do the tyre will keep accelerating the spin (causing the loss of grip, and even major oversteer and loss of control).

Then take, for example, peak-to-mean ratio of 1.5 (most simple example would be that half a revolution is at 1.5 times the average/mean torque, and half at 0.5). So, when approaching the point of breakaway, the 'spike' would cause the tyre slip, followed by drop in torque which would allow the tyre to regain traction and the driver that 'half-a-rev' to react by closing the throttle a bit...

In this respect, as Bigleagueslider has said, throttle control would be easier, but I think it might not be so simple- in a way, I'd say big-bang engines were harder to master than 'screamers'. I have a feeling that by the end of 500cc era, the only driver that was able to constantly extract the maximum from big-bang engined bike was Doohan (and being partial to him, I'd say Biaggi, were he not a rookie at that time, would've been too), and even he reverted to screamer for his final championship (and I'd say OTTOMH prior to that he was the only one using them). I'd draw a comparison to olden days of F1 (say, late '50ies, early '60ies) when it was easier to get into a slide (whether understeer, oversteer or 4 wheel drift) and to control/maintain the slide, but only a few drivers have truly mastered it to a degree that they could a) do it constantly and consistently, b) feel their way into the slide and get the exactly right amount of slide to gain time instead of scrubbing off the speed...

#39 chunder27

chunder27
  • Member

  • 1,054 posts
  • Joined: October 11

Posted 10 April 2012 - 22:03

Doohan reverted to a screamer style engine for one reason only.

To intimidate his team-mates who were all riding big-bang engines like his and allowed to see the majority of his data. He didn't like this so pushed Honda into letting him test a 180 degree and 90 degree screamer engine after the 96 season.

With the advances in electronics from 1991 to 1997 when he used it first, the engine was found to be actually better for the tyre and Mick managed to gain an edge he was never to lose really, because the first time Criville used it he crashed and never used it again until the next year when he had to. Similarly other team mate Okada tried it a few times and was miles slower.

Electronics and tyres tamed the screamer engine and made it easier to ride, and later on from I think 99 on all NSR Honda's were screamer spec.

But Yamaha and Suzuki never tried it, I think Suzuki may have tried a screamer with Kenny Junior but I dont think it worked, but remember theri engine was twin crank and Honda's was one.

Advertisement

#40 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 10 April 2012 - 22:07

Thnx for the correction and additional info. :up:

#41 desmo

desmo
  • Tech Forum Host

  • 13,216 posts
  • Joined: January 00

Posted 10 April 2012 - 22:16

Then take, for example, peak-to-mean ratio of 1.5 (most simple example would be that half a revolution is at 1.5 times the average/mean torque, and half at 0.5). So, when approaching the point of breakaway, the 'spike' would cause the tyre slip, followed by drop in torque which would allow the tyre to regain traction and the driver that 'half-a-rev' to react by closing the throttle a bit...


Suspiciously like traction control. (OMG OMG!!!) If I remember correctly Kevin Cameron suggested that the real secret of the XR-750 wasn't the odd spacing of the power pulses but simply that the bikes were slid in cornering just over the top of a falling torque/power curve so that if geared and tuned correctly they would self-correct for too much wheelspin; passive traction control.


#42 Greg Locock

Greg Locock
  • Member

  • 4,609 posts
  • Joined: March 03

Posted 10 April 2012 - 22:41

I don't buy the original premise either, that the tire carcase somehow cools down better as a result of spikey warming rather than continuous. There may be a tiny second order effect, in that the peak temperature is a better radiator than the valley temperature is worse, but that'd be it. The reason I'm dubious s the time constants involved in heating effects - tire carcases take times of the order of multiple seconds and laps to warm up, not the interval between firing cycles.

Edited by Greg Locock, 10 April 2012 - 22:45.


#43 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 11 April 2012 - 02:08

Greg, I know this may not be relevant (or even accurate, for that matter), but before I was told that explanation was very unlikely, I tried to explain it to myself by seeing it that bigger but less frequent torque spikes would bring the same amount of heat to the tyre, but with longer periods of cooling which then might more easily be radiated from the tread before reaching carcass. Then I proceeded to do a (over)simplified calculation, which sort of *ass*umed few things like rolling circumference of 1.8m, with bike traveling at about 180 kph, and 2-stroke V4 happily revving at 9000 rpm, a screamer engine would have 'torque spike' every 80mm along the circumference of the tread, whereas the big-bang could have the spacing as large as twice as much, if it had two cylinders firing at the same time. If the contact patch length was less than that, maybe there was the benefit they were finding.

As an aside, seeing how bike tyres don't get much load in lateral direction, which is the most dominant contributor to their heating? Is it longitudinal slip or their 'vertical springing' in corners?

#44 Greg Locock

Greg Locock
  • Member

  • 4,609 posts
  • Joined: March 03

Posted 11 April 2012 - 05:16

As an aside, seeing how bike tyres don't get much load in lateral direction

That statement should cause some comment!

At the apex the tire is all Fy

I think the contact patch work (Fx and Fy) are the dominant generators of heat, but that is just a guess.

#45 Magoo

Magoo
  • Member

  • 2,554 posts
  • Joined: October 10

Posted 11 April 2012 - 06:55

I can't be the only one who suspects that the whole issue is ultimately baloney. The tire cooling theory is difficult to take with a straight face, but then every other explanation is equally dubious.

One can listen to the pronouncements of various experts, especially of the Internet variety, only to discover they don't have their basic facts straight, like who was running what crank phasing where. And the terms "big bang," "long bang," "screamer," and so on aren't used with any particular rigor or even consistency. Blah blah blah blah. It's like a perfect storm of defective info. At some point you start to wonder if bad info is the basis of the controversy.

There is one person in a unique position to have an opinion: Jeremy Burgess, engineer for both Doohan and Rossi. How interesting that the man who might be most qualified to have an opinion has the least to say about it. He did once famously offer this view: that MotoGP is 20 percent bike and 80 percent rider. What part of the 20 percent is crank phasing, he's never specified.

I don't know any more about it than anybody and a lot less than most, but I wouldn't be surprised if we don't eventually learn the whole thing comes down to individual rider comfort, feedback, and confidence and there are no tangible benefits that can be materially validated. That is, it's completely subjective.

This occurred to me a few weeks ago while talking to Travis Pastrana about a totally different but oddly related topic, cars vs. bikes and the challenges of adapting to each. As he sees it, "Car is man and machine. On a bike you are part of the machine."

#46 Lee Nicolle

Lee Nicolle
  • Member

  • 6,250 posts
  • Joined: July 08

Posted 11 April 2012 - 07:36

yes, hogits, you are right 1800 pulses per run, so much for my arithmetic skills! Hopefully it doesn't invalidate asking the question though.

One thing I didn't mention , which complicates the power pulse argument on a dragster, is that they use a slipping clutch to bring the wheel speed up to the engine speed in the first part of the run. Don't know if cylinder pulses could get through the clutch slip.

The mention of dirt bikes makes me wonder if the Sprint car community have ever played with firing order. As far as I know they have not

Not as far as I know either. Though in years past they did use reverse rotation engines. I have not heard of any for a good while so I guess they are not used anymore.

#47 Lee Nicolle

Lee Nicolle
  • Member

  • 6,250 posts
  • Joined: July 08

Posted 11 April 2012 - 07:54

My brother has worked in the motorcycle industry for a long time.He has said to me that a lot of the engine combos really only suited some riders. Doohan always slid bikes everywhere, especially back in his Superbike days in Oz. He did calm it down on GP bikes but was always more tail out than most. The 'big bang' engines suited him more than most, as well as having the best range of tyres also. For Mick part of his success was Jeremy Burgess though, it would have been harder without a wizard engineer.
As others have said the electronics have largely alleviated any real gains on any style of engine. They are all good, lets get the electronics working properly, and get the best rider on the bike.
Bikes really will put more emphasis on rider than what F1 does. Not many average [top echolon] riders win Championships unlike in cars where quite a few have through best equipment and a sensible continually scoring points has won titles.

#48 Wolf

Wolf
  • Member

  • 7,881 posts
  • Joined: June 00

Posted 11 April 2012 - 11:36

That statement should cause some comment!

At the apex the tire is all Fy

I think the contact patch work (Fx and Fy) are the dominant generators of heat, but that is just a guess.


Greg, I sort of forgot to make a distinction between contact patch and the wheel. But during steady cornering the wheel will be subjected to very little lateral force and overturning momentum (those two being almost exclusively caused by contact patch migration and CoG movement caused by the driver*)- so when cornering the wheel will be subjected to normal force increased by 41% (if it was cornering at 1g) and I thought that this constant oscillations in normal force should cause both the tyre and the air inside it to heat up.

* if driver kept his butt firmly planted on the seat, the wheel would 'see' lateral force pointed in the 'wrong' direction

#49 NTSOS

NTSOS
  • Member

  • 692 posts
  • Joined: February 05

Posted 11 April 2012 - 14:32

A street legal XR in the UK.

Street Legal XR

....and a XR track demo.

Interesting exhaust note!

John

#50 Greg Locock

Greg Locock
  • Member

  • 4,609 posts
  • Joined: March 03

Posted 11 April 2012 - 22:34

Greg, I sort of forgot to make a distinction between contact patch and the wheel. But during steady cornering the wheel will be subjected to very little lateral force and overturning momentum (those two being almost exclusively caused by contact patch migration and CoG movement caused by the driver*)- so when cornering the wheel will be subjected to normal force increased by 41% (if it was cornering at 1g) and I thought that this constant oscillations in normal force should cause both the tyre and the air inside it to heat up.

* if driver kept his butt firmly planted on the seat, the wheel would 'see' lateral force pointed in the 'wrong' direction

I think I see what you mean, but the lateral force is still generated tangential to the tire isn't it?