KLR

In 944 Spec and Spec Miata, we often wonder about the impact of HP on lap times. If I got a few more horsepower, how much faster would I go? If I’m down 5hp (or 3%) on someone, what does it cost me?

Well, I managed to conduct a great experiment last weekend at Gingerman in my 944 Spec car. Same driver, same car, same setup, same tires, pretty consistent weather, and a measureable loss in horsepower throughout the weekend due to worsening loss of compression past the rings. The only difference was the temperature between runs and any variability in my driving. I don't think either were significant factors.

Below is a screen shot of TraqMate’s imputed RWHP plots from fast laps in three of my sessions over the course of the weekend and one control line from a healthy car (Joel Karns running a new Bennington Motorsports motor that has been tested on their engine dyno; I collected several of his runs throughout the weekend to control for weather).

The lines, from top to bottom, are:

Turquoise line on top = Joel Karns (138hp, didn't vary noticeably with the weather; his plots were virtually identical all weekend and the lap times were 1:51 to low 1:52s)

Green = Neal Agran (me) in Friday afternoon test and tune (1:49.3 lap, 128.9hp, sunny and 75 degrees)

Red = Agran in Sunday morning qualifying (1:49.1 lap, 125hp, clear and 65 degrees)

Black = Agran in Sunday race #3 (1:50.2 lap, 115hp, sunny and perhaps 80 degrees).

The weight of Karns’ car and mine is set to 2630, which is very close to what the cars actually weighed. Changing the weight setting in TraqMate within reason doesn't have much effect anyway.

These were all clean laps without benefit of drafting, etc. Other laps in the sessions are similar in lap time and HP plot. There are no outliers here. The track record for 944 Spec on Gingerman is 1:49.7 and was set in similar conditions, so these lap times are competitive (official records can only be set in a race). My driving across sessions was very consistent and cornering speeds captured by TraqMate back this up. The differences in lap time are mostly coming from the straights, with the exception of some contribution of the 65 degree AM weather to cornering grip in the red session.

Whether one believes that absolute power levels can be determined from this or not, the relative position of the lines and the magnitude of the difference on the straights is clear. The point estimates for horsepower are calculated using a combination of the instructions in the TraqMate manual and the application of some coaching on this that we received from Ron (of TraqMate) at Putnam last month.

I picked a flat spot on the straight between turns 10 and 11 and advanced the vertical line to where the HP numbers match the known dyno number for one of the cars. The cars are at 5400-5500 RPM in 4th here, which is not quite the power peak of 5800 RPM, and I have no idea what this software actually does to generate these lines beyond the generalities in the TraqMate manual. I’m no engineer – I just use the stuff the way I’m told to use it.

However, these point estimates do seem to be bang on correct. Karns' car would dyno on a Dynojet at around 138hp based on the engine dyno for this motor and Bennington’s prior experience with their rental car. Mine actually did dyno at 128hp and a fraction on a Dynojet before this season started. We thought that was due to a failing alternator, the need for some fresh tuneup parts, and perhaps the need for some AFM tuning. Guess not. Rectifying all of the above never seemed to change the peak power, but the AFM tuning did increase torque.

The 125hp and 115hp seem right relative to the Spec Miatas I observed in those sessions. In the 115hp session, I actually lost ground to an SM backmarker in a straight line. He should be dynoing in the low one-teens and weighs about 12% less.

I’ll leave it as an exercise for the reader to decide what conclusions to draw. My three observations are that:

1.) Tires, handling/suspension, and the loose nut behind the wheel have a lot more to do with lap times than does HP. A few horsepower will not take lap times from mid-pack to the front of the grid.

2.) However, this lap time comparison does not take into account the effect of HP on your ability to pass in a competitive situation. On a track with an 80mph average speed, 1 second comes to ~117 feet lost exclusively on the straights. Given that the difference between making a pass and tucking back in comes down to a few feet (the difference between bumper overlap and being front wheel to door), it matters a lot. Even being down a few horsepower means tenths, which might translate into 30-40 feet per lap. It makes passing brutally difficult against a driver of equal skill in otherwise equal cars. Don’t ask me how I know.

3.) It’s pretty flat here in the Midwest. You might see different results at a track with significant elevation changes.

ProCoach

Great post. Key in your test results success was Ron's input AND that the test was done on a flat, level area AND in an area requiring full throttle acceleration.

It's a great tool and I got similar results in 2-Liter Sports Cars and Historic Grand Prix cars a few weekends ago at Barber.

Sterling Doc

Great stuff Neal! Your consistency at track record pace makes this very useful/relavent.

Really fascinating. This makes it clear, visually, why a HP deficit means less to lap times than to passing/racing. The difference in area under the curve from the low to high HP is fairly small in reference to the whole time span. You can compensate for the power deficit in what are normally part throttle situations by using more throttle, so it's not until your beyond 90% throttle that a difference occurs. However, all of the laptime lost is concentrated in the straights, which makes for a bigger difference during there than the lap time delta would indicate. So it feels, and is different than just driving at 9/10ths, or a lower skill level. This is also where passing opportunities are capitalized on. We knew this going in, but it's neat to see it played out visually, and it's also why HP is more important than generating better laptimes. Losing 1/2 second a lap can be made up by a small increment of talent, but making a pass stick is relatively harder, given the same HP deficiency.

This is similar to the inverse situation I had when I lost my brake booster racing you at Autobahn. I could turn competitive lap times, but making a pass stick was nigh impossible with the finess lost having to stand on the pedal, even with my 4 HP advantage (which was subtle, but noticable).

Now we've got to see if we can tease out differences from torque curve variations, set up changes.
Sharing TM data in a Spec class is very interesting, and benefits all involved to some degree.

Van

That air resistance is a bitch.

himself

This is particularly interesting, as NASA has proposed using TM data to verify HP in addition to (or to replace) the dyno. Neat.

Thanks for posting. Get more HP.

-td

wanna911

My Traqmate has my hp down almost exactly. It's pretty accurate for me. But you have to make sure you put in some good drag for aero components and enter all of the data as best possible.

Sterling Doc

Interestingly, Ron at Traqmate advised us to leave all the settings at the defaults, and not mess with them, even the there is a setting for the 944's aero (default is a Miata).

While the Traqumate is very interesting for comparisons such as these, there is still a lot to be iron out for rules enforcement at Spec level in low-HP cars. Right now there is a fair amount of subjectvity involved - throwing out outliers from drafting or variations in conditions, etc. You can get a good idea of trends and a gestalt feeling on who's making more power. But for rules enforcement, you need more objectivity and consistancy than we can get, yet, at least for low HP cars, where 5 HP is a siginificant difference.

Heck, even dynos are an issue here. My car had more than 10 HP variation among 10 dyno pulls done back to back. The low numbers were with cooler oil temp (starting about 150-160 degrees). This high numbers were with 210 oil temps. Suprising how much difference this made, as we let the car warm up quite a while before putting it on the dyno. Must be the 20/50 oil!

wanna911

I tried leaving the settings alone, didn't work for me.

Sterling Doc

Yes, you can twirl the dials and get most any HP rating you want out of the system. Part of collecting this data is to eventually establish a protocol than can objectively spit out reliable numbers, but were a ways away from being able to do that at our level, now. It is the holy grail of rules enforcemnent - to be able to do it, on track, live, and during the racing, so we'll keep working on this.

PedroNole

Cue Kibort in 5..4...3....

Gary R.

I agree, but take two EQUAL drivers in identically set up cars and give one 10% more HP and see who ends up ahead.

bobt993

Beetlejuice!!!!!!!

IcemanG17

great data...... I look at race car performance as a triangle..... HP-braking-handling on all sides with weight in the middle......in reality HP is the bottom side of the triangle and much shorter than either handling or braking....since its the least important.....

Of course increasing HP is a 100whp car by 10whp will make a measurable improvement vs increasing 10whp in a 500whp car.... HP really is diminishing returns..... Lets say a 200whp car turns a 2:00 lap at a given track...without changing anything else it becomes a 300whp car.....it might gain 5 seconds to 1:55......but another 100whp might only gain 1-2 seconds and it keeps dropping from there.....

Here is a video of my two 928 race cars, both driven by me at the same track to nearly identical lap times.....1 is 200whp @ 2800lbs and the other is 300whp @ 2850lbs with HUGE tires....

http://www.vimeo.com/15164600

The "estate" which was built for lemons racing is on the left....the "widow" is on the right

Veloce Raptor

el oh el

38D

If I read your post correct, you lost 13.9hp which equated to 0.9 seconds. That's a little over a tenth for every 2 hp. That seems significant to me.