Notices
993 Turbo Forum 1995-1998
Sponsored by:
Sponsored by:

Ultimate hp test 100-300kph

Thread Tools
 
Search this Thread
 
Old 11-13-2006, 06:14 PM
  #76  
Matt Wofson
1st Gear
 
Matt Wofson's Avatar
 
Join Date: Nov 2006
Posts: 1
Likes: 0
Received 0 Likes on 0 Posts
Default

I have been following this forum with some interest but have not posted on it yet. I have a masters degree in mechanical engineering and am studying for my phd in England.

Much of what has been said above is correct, the best way to measure power reading from an engine is a dyno cell, especially with more powerful cars, as on a rolling road there are simply too many factors that will affect the reading, slip betweent he tyre and the rollers, false air flow through coolers, higher temperatures in the engine..but in the dyno cell ALL of these factors can be accuratly controlled and the only variable, assuming the dyno is not faulty, is the engine being tested. I will however agree that it may be 'quicker' to use a rolling road, but any figure measured this way is an indication of the power but by no means the actual power being produced.

It is interesting what Basal Skull has said about gearshifting and staying in positive 'G'. However i am afraid it is incorrect. The easiest way to approach this problem is to use Newtons Laws

F=mA

Force=Mass*Acceleration

Obviously the mass is constant (ignoring burnt fuel as the effect will be negligible) so we can remove it from the equation for our purpose. Now, when you change gear you remove ALL driving FORCE from the car, unless you can change from one gear to another simultaneously, which from your above posts you seem to agree is impossible ~0.15-0.3 seconds) Now this means the only FORCES acting on the car will be aerodynamic drag and friction (friction in the axles, wheels.....) Therefore the resultant force on the vehicle will be opposing the direction of movement of the vehicle. Therefore the vehicle will ACCELERATE, but in a NEGATIVE direction, ie SLOW DOWN.

Now, the above chart shows that the acceleration never goes below zero, seemingly showing that the car continues to accelerate, however this is misleading. This is most likely due to sample rate of the system. I am afraid it is IMPOSSIBLE for the car to continue to accerlerate with no force.

The non constant acceleration is simply due to a non constant force, but when the forse is removed, the acceleration will become NEGATIVE, F=mA. If this were not the case, you would have invented a perpetual motion machine and would be VERY VERY wealthy indeed =D

The post about inertia is interesting..
The only effect the inertia will have upon the vehicle is to reduce the rate of deceleration, think about a heavy ball and a light ball, the heavy ball is harder to start moving but also harder to stop moving, but it will have no effect on the sign of the acceleration. I belive this is what was said before =D

There is a case perhaps for the drivetrain to 'store' a very very small amount of energy, suah as a spring would, but in my proffessional opinion this would be such a small amount of energy, ~10kW, that it would have no effect at all
Old 11-13-2006, 08:46 PM
  #77  
Jean
Addict
Lifetime Rennlist
Member

 
Jean's Avatar
 
Join Date: Apr 2002
Posts: 5,451
Received 176 Likes on 106 Posts
Default

Welcome Matt, another 993TT owner?

I agree with Basal Skull. The reason I see continued positive Gs while shifting is because of the inertia of the car that is sustained for a few fractions of a second while doing shifts without engaging the clutch that's the beauty of a sequential gearbox.

Wheel thrust is still there! On a clutch based dog box, if you shift fast enough with hardly lifting off, you will see the same thing. Notice how TB's shifts with stock gearbox are at the limit of Gs, almost staying positive, which is perfect shifting. DKnebes did that as well, it is feasible.

What is more important than the mass circulating at a certain speed is that you have wheel thrust and momentum carrying a moment of inertia that keeps going for a short period of time after you lift off.

There is no error in the readings, I have multiple runs that show it, some of the Gs do fall below zero when I lift-off a bit more, in this case my foot was still at 90% throttle while shifting, that's why.

Give it another fraction of a second, and my Gs would have moved to negative territory. Remember that they are falling, you were running at 0.6Gs and they fell to almost zero in about 0.1s., so you are decelerating, or in fact "accelerating less" until you reach zero and then negative Gs. The higher in the RPM range that you shift, the more flywheel inertia you will be carrying and the longer your Gs will remain positive. A stock heavy dual mass flywheel helps here, it also gives you the fastest drag time speeds

One should not look at acceleration as such, but to wheel thrust, which is what is pulling you forward.

Wheel Thrust (lbs) = Torque at the wheels/ (tyre radius/25.4/12)
Torque at the wheel = Engine torque x diff ratio x gear ratio x drivetrain efficiency
1 G= Wheel Thrust/mass (lbs)

Therefore all the factors above, are involved while calculating long Gs (acceleration) and 1 G = 9.81m/s/s

This is what is driving your car forward, not the engine torque alone, wheel thrust takes the gearing, differential, driveline losses, mass and rotational momentum into consideration.

That is also why engine torque curve alone or RPM drops does not mean much when analyzing optimum shifting, it is a common misconception that one should shift between peak torque and peak HP....you need to shift when your thrust at the wheels in the lower gear becomes lesser than the one you would have after you shift to the next gear at the RPM at which the revs will have dropped.

I have seen the same thing when hitting the rev limiter, the car continues its forward acceleration momentum for a fraction of a second, it does not drop negative immediately. It does make you loose time however.

The chart below with stock gearing (but 7K RPMS) shows clearly that the optimum shifting point for a 993TT is almost at the RPM rev. limiter, the thrust lines never cross each other, if they did, it would have indicated that staying longer in gear is making you loose forward Gs.

Check out the thrust output at the wheels as a result of the gears, 6800ft./lbs.!!

Old 11-14-2006, 05:18 AM
  #78  
935racer
Intermediate
 
935racer's Avatar
 
Join Date: Apr 2006
Location: Cambridge, UK
Posts: 41
Likes: 0
Received 0 Likes on 0 Posts
Default

Wow, I have really started something! Acropora and Matt Wofson have to be correct. One of the fundamental rules of the world is that we cannot have a pertpetual motion machine and acceleration with no nett applied force is just that. I originally asked the question because a lot of people seem to believe it to be possible. It is the same rule for rev limiters. If your limiter absolutely cuts at, say, 10000 revs per minute the engine cannot go fasterthan this. The reason is that every movement requires a force and ifthere is no force then the body will stay in equilibrium. Of course, in my example there is plenty of internal friction in the engine (and drag etc etc) so it will slow down.
The only reasonable explanation for Jeans G force to stay positive during a gear change (and I have seen plenty of other traces showing the same phenomina) is the sample rate. Matt Wofson said the same and it appears he is much better qualified than me (I never stayed for a PhD - I wanted to earn some money!!). I do not know the sample rate and discrimination rate of the system Jean has used but it is very likely that it is "too slow". This is not a criticism of the system as in 99% of cases this does not matter one jot and nor does it alter the overall acceleration. It is just that once one "homes in" on a particular bit of data then great consideration has to be taken in considering the validity of the datafor the particular interval that you are looking at.
Anyway, all too complicated at this time in the morning! Anyone willing to share the secrets of how to post pictures?Also, anyone got any experience of ball bearing turbos?
Cheers,
Richard.
Old 11-14-2006, 06:20 AM
  #79  
Kiko
Racer
 
Kiko's Avatar
 
Join Date: Mar 2005
Location: Porto/Portugal
Posts: 394
Likes: 0
Received 0 Likes on 0 Posts
Default

A bit off-topic but 10 years ago when I started racing had the chance to drive round Estoril with a lengendary 80's portuguese racing driver by my side correcting my niggles and I never forgot what he said to me when I was doing this lightning fast almost no lift gearchanges, "a race is never won by how fast you change but it can certainly be lost for that".

To this day I still believe that between lightning fast (TB's) gearchanges and just being swift (mechanically more sympathetic) the difference per lap is negligable. Am I wrong?
Old 11-14-2006, 07:33 AM
  #80  
935racer
Intermediate
 
935racer's Avatar
 
Join Date: Apr 2006
Location: Cambridge, UK
Posts: 41
Likes: 0
Received 0 Likes on 0 Posts
Default

Kiko, I think you are absolutely correct! A missed gearchange will, at best, cost you a second or two and that could mean the race. At worst it costs the engine and box and that means many missed races. Lightening fast changes are all very good (and I certainly try to do them!) but they are dangerous! However, when measuring acceleration runs that only last 6 seconds the difference between a change at 0.25 seconds and one at 0.75 seconds, for example, is very material. Its nearly 10% of the overall time. It would certainly take a lot of horsepower to win that back!
Its everything in moderation I guess!
Richard.
Old 11-14-2006, 08:25 PM
  #81  
Jean
Addict
Lifetime Rennlist
Member

 
Jean's Avatar
 
Join Date: Apr 2002
Posts: 5,451
Received 176 Likes on 106 Posts
Default

Originally Posted by 935racer
The only reasonable explanation for Jeans G force to stay positive during a gear change (and I have seen plenty of other traces showing the same phenomina) is the sample rate. .... I do not know the sample rate and discrimination rate of the system Jean has used but it is very likely that it is "too slow"..
Hi Richard
I sent you a private message concerning posting the pictures, if you click on the "private messages" link on the top right hand side of your screen it will take you there.

As far as the acceleration vs. speed while shifting, not trying to prove a point, but just for the sakes of information, here is my take again.

Concerning my device, it samples position and speed 100 times per second, the accuracy level is 0.005G. I have calculated several times the relationship between acceleration, speed and distance with mathematical formulas, and the accuracy is spot on.

My observations with more than a hundred runs with different cars, speeds, slopes etc, are that cars carry a momentum shortly until after the clutch has been engaged. During that time, the car is still accelerating and gaining speed, however at a rapidly decreasing rate. The same happens when you hit the rev limiter, engine speed stops but Inertia keeps the car going very briefly. I have charts of that if you would like to see them.

Here is a close up shot at my datalog. The lower graph is a close up of the upper one where I have highlighted with a blue circle. The crosses are at the same position on both graphs and is the point where the shift takes place, the shift is not where the speed line becomes horizontal. Speed still climbs while shifting by about 1mph, or 1/10th of a second before flattening, due to the dropping Gs. The sampling being so fast, it creates the paraisitc spikes, I normally smoothen them out at 0.2seconds, but I am trying to show the fast sampling rate, which is not necessarily a better thing in fact.

I also have charts showing that with a slower sequential shift, the speed drops (instead of flat) during the shift (due to drag, rolling resistence etc) and the Gs fall to negative territory.

The above is the same reason why your datalogs also show positive Gs, speed is in fact still climbing shortly after shifting.

I think this should answer your question.



And here are the details of speed vs Gs and time. Notice how speed always follows the trend in Gs, it does not happen simulatneously due to the rolling mass.



Concerning Ball Bearing turbos, you might want to talk to Todd Knighton, he has a lot of experience with them, I have the older generation Garretts on mine.. Rennlister "Ngoldrich" on this forum uses BB turbos on his 993GT2 for racing too.

I realize I have totally hijacked the thread and apologize to all.
Old 11-15-2006, 05:04 AM
  #82  
Ruxpin
AutoX
 
Ruxpin's Avatar
 
Join Date: Dec 2005
Posts: 14
Likes: 0
Received 0 Likes on 0 Posts
Default

Jean, this is all very interesting

It's been a while since MOD and I studied mechanics together, however one thing from your graphs confuses me.

After the shift point shown (.21s) the lateral g's measured decrease - suggesting a decrease in acceleration (Still acceleration as positive g's). However, the gradient (i.e. acceleration) of the speed/time plot remains unchanged. The gradient appears constant between .0 and .28s on the graph.

This may well be because the scales used to plot the data are hiding the truth and in fact the gradient does decrease in line with the falling g's

Or is there another explanation??? Is it just the rolling mass you mention?

I find the idea of the car continuing to accelerate during the gear change hard to understand/accept, however, it's very difficult to disagree with real data as shown.

I'm confused..... but intrigued

Rich
Old 11-15-2006, 05:32 AM
  #83  
NineMeister
Addict
Rennlist
Site Sponsor

 
NineMeister's Avatar
 
Join Date: Oct 2002
Location: Cheshire, England
Posts: 4,447
Received 194 Likes on 97 Posts
Default

I am afraid that I would have ot agree with the other engineers in that, at any speed, if you hit neutral even for just 0.15seconds the acceleration force of the engine reduces to zero and the only force remaining on the car is negative acceleration from drag. The Inertia of the system is the mass of the car x its velocity, it will keep the car rolling until the drag slows it down to a stop but it can not make it faster without the influence of an external force.

Jean, assuming that you are not testing your car downhill (that's a joke!) I suspect that what you have found is either the limitation of resolution of the AX22 or possibly a problem in the way you mount the meter in the car, maybe a mount that flexes under acceleration, storing energy which is released back into the meter when you lift off? If you are using the racelogic screen mount, try rigidly fixing it to the shell instead and see if the result differs in any way.

I just checked the previous data that you posted and it is interesting that Toby's runs show negative acceleration as expected, so maybe it is just the response speed of the Racelogic G-sensor. Are there any electronic engineers out there who have taken it to pieces yet and found the specification of the sensor??
Old 11-15-2006, 06:21 AM
  #84  
Miles965uk
Burning Brakes
 
Miles965uk's Avatar
 
Join Date: Jan 2004
Location: Switzerland
Posts: 959
Likes: 0
Received 2 Likes on 2 Posts
Default

Richard, great to see you on here! Im a big fan and have had a few chats with your mechanics in the pits at brands hatch... Would love to know more about the car & engine, maybe on a seperate thread when you come more familar with rennlist.

My 2 cents on this G force whilst in neutral...Physics say it simply cannot keep accelerating when in neutral. With the momentum that exists after the shift etc it will at best keep the car at the same velocity (assume no drag) so a=0 at best. Theres no way the velocity can increase without the force being applied.

Perhaps we can look at a velocity vs g-force plot to see if velocity is actually increasing when you change gear?

Miles
Old 11-15-2006, 07:11 AM
  #85  
935racer
Intermediate
 
935racer's Avatar
 
Join Date: Apr 2006
Location: Cambridge, UK
Posts: 41
Likes: 0
Received 0 Likes on 0 Posts
Default

Miles, I agree with you! In fact, whilst changing gear the only forces acting on the car are mechanical drag and aero drag. Thus the car will slow down - no doubt about it! My Motec traces can show a positive G during a gearchange but its all down to the sample rate and the settling time of the accelerometer. If you configure it to look at longitudinal G "all the time" and use a v. expensive accelerometer then the picture is reversed. Significant negative G as soon as the clutch is dipped!
Lets face it, if our cars accelerated during a gearchange we would have built a perpetual motion machine. Now, were is that patent attorney!!!!!
Those are the facts, however I can see why people can think otherwise. They have a valid data trace that shows positive G all the time. You just need to think about what is happening and then the truth becomes obvious! As Colin said above its probably one of the limits of the measuring instrument.
Cheers,
Richard.
Old 11-15-2006, 07:23 AM
  #86  
JBL930
Not Forgotten
 
JBL930's Avatar
 
Join Date: Apr 2006
Location: South East, UK
Posts: 1,215
Likes: 0
Received 0 Likes on 0 Posts
Default

I'm no engineer but my thinking seams to point to there being some stored energy in the car, the centrifugal energy in the wheels, however small compared to the drag, would cause some acceleration? Even if it was only for a split second
Old 11-15-2006, 07:27 AM
  #87  
Jean
Addict
Lifetime Rennlist
Member

 
Jean's Avatar
 
Join Date: Apr 2002
Posts: 5,451
Received 176 Likes on 106 Posts
Default

First let me say that I agree with all that is begin said by Richard and others, after installing my sequential gearbox, it has taken me a lot of time and analysis to get to this point just because I studied the motion dynamics formulas and they did not make sense compared to what I was seeing.

Originally Posted by Ruxpin
After the shift point shown (.21s) the lateral g's measured decrease - suggesting a decrease in acceleration (Still acceleration as positive g's). However, the gradient (i.e. acceleration) of the speed/time plot remains unchanged. The gradient appears constant between .0 and .28s on the graph.
.....Rich
Hi Rich, these are long Gs, lateral Gs are used to determine grip usually. Acceleration is m/s square, so it is how much distance the car gains in a second, per second. Maybe this is why you see that. The sampling is too quick (several times per second) , it needs to be smoothened to get more valid data.
Originally Posted by NineMeister
... if you hit neutral even for just 0.15seconds the acceleration force of the engine reduces to zero and the only force remaining on the car is negative acceleration from drag.
Colin, I never get in neutral, I have a sequential gearbox, always in gear. All the charts that I have back when I was using the normal gearbox are showing negative Gs, most of my charts with sequential gearbox show it too, however with the right shifts, I get curves that keep the Gs positive, like the one posted above. I guess I will have to post a video.
Originally Posted by NineMeister
... Jean, assuming that you are not testing your car downhill (that's a joke!) I suspect that what you have found is either the limitation of resolution of the AX22 or possibly a problem in the way you mount the meter in the car
I am open to share my raw data to be checked, I am more critical of my runs and my tuner than anyone else I test my data on excel with mathematical formulas for slope, Gs, speed, distance everytime I do a run, I have also tested throroughly on the track, comparing different lines, G forces long and lateral, braking points, etc. The device is very well secured, otherwise the slightest move will throw a strange long G on the chart. You would be surprised to see how many runs looking perfectly good had in fact some small issues.
Originally Posted by NineMeister
... I just checked the previous data that you posted and it is interesting that Toby's runs show negative acceleration as expected, so maybe it is just the response speed of the Racelogic G-sensor. Are there any electronic engineers out there who have taken it to pieces yet and found the specification of the sensor??
Colin, TB's data is done with the AX22, as are all the runs that I have. I don't know what else to post so that it is believed before assuming the tool is not accurate. Maybe TB can do some runs with the Driftbox and the AX22 inside the car and show us how they compare. If all the dataloggers including Motec (which will also show speed picking up briefly after shifting, while Gs are dropping) are wrong, then we are maybe not there yet technologically or the physics theory of rotational mass has a few more angles to it that we don't know of ourselves.

The key here are 2 things, first, I have a sequential gearbox, I am always in gear, the second thing is that while shifting, the car is seeing Gs dropping, but in the meantime the car is still gaining speed through momentum. I know the formulas involved I have researched this at length and the ones I posted above on an earlier post explain what this is due to.

In any case, this has been a good discussion, my intention is just to share my observations about shift points, gearboxes aerodynamics etc. which I find to be an interesting subject, so with the data posted, everyone should go with their own outcomes, in the meantime, looking forward to see more runs
Old 11-15-2006, 07:30 AM
  #88  
Jean
Addict
Lifetime Rennlist
Member

 
Jean's Avatar
 
Join Date: Apr 2002
Posts: 5,451
Received 176 Likes on 106 Posts
Default

Richard just saw your post.. I agree with the negative Gs, however on a gearbox with clutch I only have them positive on my seuqential as stated. Also very interesting what you are saying about Motec, I will ask a friend of mine to show us his datalogs from David Saelens or Bergemeister in the Cup car to see what it shows us with the sequential gearbox.
Best
Jean
Old 11-15-2006, 07:33 AM
  #89  
Kiko
Racer
 
Kiko's Avatar
 
Join Date: Mar 2005
Location: Porto/Portugal
Posts: 394
Likes: 0
Received 0 Likes on 0 Posts
Default

Originally Posted by Miles965uk
Richard, great to see you on here! Im a big fan and have had a few chats with your mechanics in the pits at brands hatch... Would love to know more about the car & engine, maybe on a seperate thread when you come more familar with rennlist.

My 2 cents on this G force whilst in neutral...Physics say it simply cannot keep accelerating when in neutral. With the momentum that exists after the shift etc it will at best keep the car at the same velocity (assume no drag) so a=0 at best. Theres no way the velocity can increase without the force being applied.

Perhaps we can look at a velocity vs g-force plot to see if velocity is actually increasing when you change gear?

Miles
In cars yes, as soon as you dip the clutch you will have negative acceleration due to drag but look for instance at a bullet, whereas the force is applied up until the bullet exits the barrel but the speed at 1m from the barrel is considerably lower than at say 10m. Go figure.

What I really think it would be nice to know (and test) is in a lap what is the difference between swift shifting and lightning quick shift. I'm convinced the difference is VERY, VERY small even difficult to prove since the natural variations from lap to lap is greater than the shifting.
Old 11-15-2006, 07:47 AM
  #90  
935racer
Intermediate
 
935racer's Avatar
 
Join Date: Apr 2006
Location: Cambridge, UK
Posts: 41
Likes: 0
Received 0 Likes on 0 Posts
Default

Kiko, Interesting point about the bullet. Luckily, I am no expert on guns but this can perhaps be explained by the fact that the expanding gases are still exerting a force on the bullet AFTER it leaves the barrel. This is easily explained by the viscosity of air and the massivly expanding gases from the barrel. The surrounding air simply cannot letthe hot gases expand quick enough dueto the airs viscosity and thus there is still a net force exerted on the bullet.
I really dont know but things dont accelerate unless they are subject to a net force.
Richard.


Quick Reply: Ultimate hp test 100-300kph



All times are GMT -3. The time now is 08:14 AM.