[quote=big howe;2549792]I'm not sure how the above quote has come across but it was no way meant to be anything but complimentry.

I apologise to Shawn if he thought it was meant in other way.

Its like taking your 14 sec car to the track and having some fun and driving home is totally different from turning up with the same car that now runs 8s , but its somethng you always dreamed off.

 

You've lost me on that one

 

Paul, I think you're making the assumption that your trap speeds on those occasions are 1) not being affect by other variables and 2) are representative of all situations.

As I've pointed out with both third party documented examples as well as my own experiences, trap speeds can vary due to traction, as well as a variety of other factors. I don't think these variances can be argued, and as such, throw a monkey wrench into calculations of absolute power. Note, this does not preclude calculating a rough relative power change, but absolute is much more difficult.

 

Looking at your examples from post #266 in this thread, I can see where the change in trap speeds (although still proportionally small) are coming from. Losing 25 pounds of unsprung rotating mass will increase trap speeds, since it has the effect of increasing horsepower at the wheels. A chassis dyno will show the same increase.

With your other examples, you're talking about a high 9-second/low 10-second car, with trap speeds in the 135-140 MPH range. If you have a car with so much power that it's still pulling *hard* through the traps, then yes, a traction problem at the start of the run can effect the trap speeds (although the before and after speeds in your examples were still within 1.5% of each other).

But for cars that aren't doing the quarter-mile that quickly, if you look at a plot of their speed versus distance down the track, you'll find that they're not accelerating very quickly at the end of the run (the speed-versus-distance curve has almost flattened out), so if you have a traction problem at the start of the run and are only "hooked up" for 1250 feet or 1200 feet rather than the entire 1320 feet, it simply doesn't make much difference in your trap speed.

 

Nothing was changed from run to run, I've got a NX / FJO nitrous controller if this sees an afr which you can set i.e 13.5 for .2 secs (whatever time you want) it'll switch the nitrous system off.

All runs were with the bottle at 950 psi.

 

We must remember that we're talking about mostly 14 sec MINI's some will be in the 13s and most will run between 96-105 with the majority in the 96-100 range.

The Mini can be surprisingly consistant espically if you can stage at the same IAT

 

Hi Scott, you must have misread my post. The reduction in tire weight was on the non-driven wheels. Of course the lower MoI and rolling resistance were responsible for the improvement in trap speed, but that would _not_ show up on a chassis dyno. This is my point. There are many other factors that affect trap speed, including traction, that are not taken into account in simple first order, straight line approximations such as those being discussed here.

Furthermore, while the difference may have only been 1.5% in trap speed on some of the other examples, run the difference through the standard power calculators out there and you'll see they return differences in hp between 3 and 5%. On the 130 mph trap speed cars that can equate to 30-40 hp predicted change even though real hp didn't change at all. On a 100 mph Mini, a 1 mph change in trap speeds equates to a predicted 5-8 hp change in power depending on weight. Yet, slow shifting, changes in atmospheric conditions, or changes in tire pressure can all result in changes in trap speed larger than 1 mph. Paul also makes an inadvertent point about IAT. While we can apply atmospheric corrections to try and equalize times from day to day and track to track, the advent of modern computer controlled engine management systems means another variable is introduced. For example, while we can apply a density correction for temperature, what is the ECU doing if IATs are 60F vs. 100F? I don't happen to have the IAT ignition correction curves for a Mini at hand, but I can be pretty sure that the ECU is pulling out timing at 100F vs. 60F. Yet, this will not be reflected in the atmo corrections. So, if you tuned your car at 80F inlet before and after a mod (exhaust?), but you tested before the mod on a 60F day, and after on a 100F day, attempting to apply simple atmo corrections to trap speed would lead you to believe that you didn't gain as much power as suggested, or even lost power.

Look, the point is this - the model people want to use breaks down when tested under a variety of conditions. This is simple fact. The examples provided show that. The model's estimates may seem close over a narrow range of trap speeds, but that isn't because the model is highly accurate, its because the model happens to briefly converge with the far more complex equation that represents reality. For those with any background in mathematics, think of it as a tangential plane intersecting a complex surface whose equation includes variables for traction, weather, power curve, gearing, shift time, weight, rotating inertia, wind speed/direction, etc.

While the changes in trap speeds for your typical Mini may be small based upon traction, they are there and they confound the simple models being discussed. Simply changing to 4 lightweight wheels and tires will produce a bigger effect on trap speed than the chassis dyno results would predict, or the simple power vs. weight model would predict. In the case of the chassis dyno, it only measures the improvement from two of the wheels, and the improvement is for a single range of acceleration rates. For the power to weight model, the estimate does not consider MoI which is a far more critical factor.

Too many variables, to simple an approach. And Paul, your nitrous system is hardly the only variable in play now, you know that as well as I do. There are hundreds of variables in play.

 

That's a good point - the Fox/Lucius/Hale/Huntington formulae are very tightly-correlated with reality, but as you said, only over a fairly narrow range. The Huntington formula, for instance, only used samples from runs with trap speeds in the 70-115 MPH range. And since the Fox formula was devised in the 60's, I imagine Fox wasn't using very many sub-12-second cars in his data set either.

 

You're right Shawn in so many ways, but if there is a group of you like we had at British Mini Showdown, Mini In The Park and The BMW Show and we weighed them and dragged them, I think you will some relationship form.

You guys in Ca will always have a higher IAT than us due to you still being in shorts now and we've got multi layers of cloths on.

Though I still believe in some ways its a leveler for a Mini customer looking for horsepower, its something you can compare to magazines around the world when they right an article about the latest EVO, 911, M3 etc etc there's usually going to be a 1/4 mile ET and Speed in the spec.

Also I've seen and spoke to other tuners with engine dyno's where they can se a gain on the dyno but it doesn't show on the track. for whatever reason.

Which has just reminded me when we had a similar discussion somewhere there is a EVO magazine report on a Ferrari where they had the engine on there engine dyno and when you used the formula it came out spot on.

I'll see if I can dig it out.

Did you have time to see if you had a stock GP dyno sheet?

 

Shawn with all due respect I have been drag racing for 25 yrs and have studied and logged most of my runs in my MINI. Your Motor Trend example is a brillant one for your argument, but were you driving? If not how do you know the throttle position was at 100% going over the line?

What more can I say it run 115mph 3 times back to back on the same day in the same weather conditions, but the 3 60fts are all different due to the nitrous switching off the first two times, so the car died in the 60ft. The only thing I changed was the time on the nitrous controller to see anything leaner than 13.5 as when I looked at the data log I sussed out was going on.

I cant find my other table but here is some early testing from beginning off last year with a stock head, all times were done with MT 24.5 x 8.5-15 slicks and small burnout.

I realise now that I should of logged start and finish line IAT, espically when you see how high the IAT go in a burnout

 

Here's all the runs at MITP 08 by new MINI's, I've sorted them in mph and highlighted a few in red. Most of them where there is a 1 sec difference would be where they haven't turn the traction control off or just blowing the tires away.

If they all had shiftlights and slicks I reckon most of them would be quite consistant.

 

Here's another one, Hyper Active was a dragstrip virgin at the BMW show, this car is stock as a rock apart from a Milltek catback

 

Hey Paul, no disrespect on my part and you've been very cool throughout the discussion. I don't recall you asking for a stock GP dyno, but no, I do not have one.

In the case of the Motor Trend argument, I'm pretty sure they weren't at full throttle until at least the 1/8th mile, but that was precisely my point. You're perfectly entitled to try and model your results and obtain some semblance of predictability. I'm just saying that the model doesn't hold across a wide spectrum of cars/tracks/conditions. It just isn't complex enough to model all the necessary variables.

I'm a regular EVO reader and I think I recall the piece you're talking about. I wrote them a couple of rather data laden letters back when they began dyno testing and publishing estimated crank power instead of what was measured at the wheels. I never received a reply, but given their regular disdain for all things from the US (except the Mustang for some reason , I wasn't surprised. Chassis dynos measure power at the wheels or hubs, and that is what should be published. Anything else is conjecture.

BTW, on Ferraris, they seem to be one of the more inconsistent brands out there. We did a Ferrari dyno day that showed variances of up to 10% between the same year/make/model of car. And the new 430 Scuderia seems vastly more powerful than the older basic 430s. We saw a whopping 50 hp difference between a Scud and the base 430 on the dyno. Given that the Scud is trapping 5-7 mph in the quarter, that might not be unreasonable even if it is only rated 27 hp higher.

Thank you for including your data. It is an admirable practice to keep detailed notes such as this. Something you might want to examine is the average deviation between 1/8 mile traps and the average vs. the average deviation between 1/4 mile traps and the average. You'll find that the 1/8th mile typically exhibits a higher average deviation. Stew on that for awhile. Next consider the daily air temperature profile as well as the track temp profile and examine your runs in that context.

 

No problem Shawn, I'm as passionate about 1/4 mile times as you are your dyno figures.

What really gets me is when we hit the 2nd kit with Jeffs it might pick up 3mph but never translates to ET even though he would of hit after the 330ft mark. He's only just fitted a data logger for the last two meetings, I've learn't a lot already and found a major problem before the last run of the year.

I'll stew on the last bit, learning can be fun even at my age

 

Scott, you might take a look at this site if you haven't already. It does a good job explaining all the models:

http://www.stealth316.com/2-calc-hp-et-mph.htm

Interestingly, none of them seem to have applied atmospheric or elevation correction factors in developing the model (Lucius in particular says he did not). And, AFAIK, all used published hp numbers which we know are not always accurate (there are certain cars that make more or less than advertised and these days the number that make more seems to be on the rise). Finally, since there are few if any FWD cars in the published data with trap speeds of 100 mph or higher, the traction breakdown issue is not addressed.

Good stuff guys.

 

I think Patrick Hales programs will let you program weather in as well.

I personally like to use my own weather in a 10lb bottle

I believe as we collect more data from runs by people we'll see a pattern form

Jan dont skip read this bit;

Fox notes that the first two variables, weight and power, are the primary influence on MPH making it easy to use a constant power approximation to determine a formulaic relationship. It is the variability of the remaining variables and their influence that makes it much harder to determine ET based soley on vehicle weight and engine power. What this means is: look at MPH as a determination of engine power rather than ET.

 

Now you're into the "looking for the finish line racing" Shawn, is there any formula which you can apply to that?

I've seen some stuff that in theory should work but when you take it to the track, it doesn't.

Doing laps is surely harder to quantify if the extra hp / trq you've gained on the dyno can actiually show itself without some serious data logging.

I remember when I was talking to Jan on the phone when he was testing the cam for the first time and you saying "I'm always happy to trade 4hp up top for all that extra trq you've just gained across the rpm"

 

Some more Dynapack questions;

When you set the machine to run a Cooper S do you set a set amount of secs or from say 3k to 8k ?

Why does sometimes the two lines never start togeather, you always think that they should start at the same point on the graph.

 

Hi Paul, the key entry points are starting rpm, finish rpm and total run time. Sometimes we'll start runs later and reduce the run time accordingly if we don't feel there is anything to be found down low and there is on sense loading up the motor unnecessarily.

The point with the link to the model derivations is to show the relatively shaky foundation upon which they are based. Developing a best fit model using a least squares approach using 1/4 mile data from magazines, especially magazines which do not correct for elevation or weather (like R&T or Hot Rod) is a very sketchy way to obtain power numbers. Even if they did correct them (ala Car and Driver), we know that turbo car correction factors are really screwy and rarely accurate. Furthermore, they are trying to correlate them with published hp numbers at the crank, assuming both manufacturer accuracy in advertising (which we know is not the case) as well as a constant loss factor in the driveline (which we also know not to be true).

I've used some trial copies of several cool pieces of software for simulating road course lap times. I know the most advanced pieces used by the F1 teams are very accurate, but the stuff I've used has so many variables that at best you can hope to tune it to match your known times at a venue and then see what changes are predicted for a change in tire, or power, etc. But you're basically hoping that your tuning of the model didn't screw up something that will mess up your prediction. Again, like most models the best you can hope for is some accuracy of prediction on relative times from a baseline, not absolute measures. I've also learned never to underestimate aero forces.

 

nice to see old myths are being drawn away

 

Not really, I still rather use a dragstrip as a power comparison to using dynos as a comparison

 

thanks for making it crystal clear....you would rather add more variables into the mix instead of mitigating them to get accurate data

GOT IT

 

Not really when you're at the track with 20 Minis on the same day everybody has the same variables, weather etc.

Also have you ever seen a European hold a NHRA world record, NO

So in theory the USA has way more time to advance because you have mores seat time availble to you.

Also things what you might see as an improvement sometimes doesn't translate to the track, thats documented by the greats in drag racing, not sure about road racing.

We've provided documented evedience that we can adopt a formula to show horsepower which is more accurate than you telling someone in Texas to multiply by 12%

Like I've said earlier you've raised the bar with Bat Raggers and Big Howe's numbers now you have to live with them as thats what evryone expects from the same sum of parts were ever they are in the world which on some dynos isn't going to happen.

I still haven't seen a drag racing calculation which is 12% out on a Mini.

Until you put a Mini on Churches and document your runs, you haven't really got an arguement.

Theory is one thing proving is another

 

I guess you really haven't learned a thing in this whole thread....

oh well......

 

On the contray, I've learnt a lot more but have you learnt anything ?