I was thinking about comparing 928 engine dyno graphs and guessing from them which would give the best 1/4 mile trap speed, assuming equal weight. Here's a measure that I came up with, I am sure it's not new. Compute the average hp for the rpm range [x, 1.5x], picking the x that gives the highest average hp for that range. 1.5 is a round number that is not too far from the gear ratios.

Eyeballing Ott's dyno graph, I am guesstimating a maximum of 525 average hp at 4500-6750 rpm:
http://www.performance928.com/cgi-bi...ss_parent=1125

John's twin turbo seems to be making a maximum 475 average hp at 4250-6375 rpm. (Everyone knows where that dyno graph is.)

Simard's 7 liter eyeballs to about 575 average hp at 5000-7500 rpm:
http://forums.pelicanparts.com/uploa...1248986219.jpg

Any thoughts, is there an obvious better simple measure?

 

Some really good information in this thread, a lot of data for you to play with:

https://rennlist.com/forums/928-foru...ng-thread.html

Comparison of dash mounted measuring devices versus what the "pro's" us:
https://rennlist.com/forums/928-foru...a-loggers.html

 

Some good graphs in that thread:

I am guessing 480 average hp from Murphy's graph at 4500-6750 rpm?
http://www.928motorsports.com/parts/exhaustbypass.php

This graph posted by atb looks like 450 average hp at 4000-6000 rpm:
http://forums.rennlist.com/upload/jt.methanol_copy1.jpg

Shane's graph looks like 390 average hp at 4333-6500 rpm:
http://www.erik27.com/shane/Rennlist%20post/dyno.jpg

jorj7's at about 475 average hp 4200-6300 rpm:
http://928.jorj7.com/dyno/0307-201550-01.jpg

 

Congrats to Simard. A proper Porsche racing engine. There is no substitute.

 

How about just looking at what rpm you go down to when shifting, and taking the integral from that rpm until redline? Though I guess this would maybe unfairly favor an extremely peaky engine.

 

The most obvious benefit of the turbo setup you are building is its ability to make more torque lower in the rpm's than the supercharged or even strokers can....depending on how you size the turbos etc....

John's TT was running fairly low boost....with race gas he could go MUCH higher.....John has seen as high as 630tlbs at 1 bar.....but I don't think he has done any testing past that....

 

That's almost exactly what this measure does. The gear ratios are about 1.5 or somewhat under. So if you shift at 6000, you drop down to 4000 rpm, and if you shift at 7500 you drop to 5000 rpm. Then you have to make it back to your shift point, whether that's 6000 or 7000 rpm. Averaging over all possible [x, 1.5x] rpm ranges and then picking the maximum average value is approximately equivalent to picking the optimal shift point and then integrating over the relevant rpm range.

The two principal reasons why it's an approximation is that the gear ratio of 1.5 is assumed regardless of the actual gearbox and that averaging over rpms ignores the fact that transition speed across rpms itself is a function of the hp curve. In practice, though, I am fairly confident that it's a decent approximation.

 

Yeah, I see what you're saying now. I've had a long, disappointing, day in the garage and I guess the heat hindered my ability to comprehend.

 

Why is that a benefit if an objective is to drive fast, which is the point in this thread? For driving fast, who cares how much torque you are making under your shift point rpm divided by the gear ratio?

Why would anyone want to make over 600 ft-lbs of torque with the stock manual transmission? What you describe is a control system failure.

 

I guess for a car that is not purely a track car, it can be nice, but then, these power numbers are so high it's not like it's a dog on the street. However, with something like an S2000, you'd really want more power down low as a daily driver and gladly give up some up top for it.

Also, I just saw that the point of this measure is 1/4 mile trap speed. In which case, I think you're undervaluing the time spent in first gear below the shift point by not including it at all...maybe need to come up with some sort of compensating factor. Working in 100F heat, I don't have the mental capacity left to come up with such a compensating factor though.

 

That's a very good point for ET. Glen Larson has a spreadsheet macro that takes the whole dyno curve and gear ratios. Experimenting with that spreadsheet (or commercial tools) quickly displays the importance of launch rpm, 1st gear total drive, and low rpm hp. However, the trap speed is relatively insensitive to what happens right off the line, so I think it's fine to ignore the launch.

 

Since he doesn't post anywhere, Todd's results might get overlooked sometimes in some of the comparisons. Here's the post with a dyno chart of his. Also keep in mind that he has eight forward gear ratios, so the RPM drop as a result of shifts on his car is half of what it is on other 928s. In other words, instead of figuring the average horsepower between maybe 4500-6500 RPM for a typical 928, on Todd's car it would be maybe 5500-6500 RPM.

https://rennlist.com/forums/5995687-post1.html

 

Let me be clear that Todd's projects are not being ignored at least by me, I am paying a lot of attention especially to the twin-turbo project!

The transmission point is of course valid, but I am looking to make this simple and comparable measure of engines. The real trap speeds will of course be strongly influenced by the actual box.

Todd's graph in the link gives the best average hp at 4133-6200rpm at 575 average hp, again eyeballing. That's an extreme result.

Significantly, since the car has a centrifugal supercharger, the result would be even higher if the operator would have run the engine to a higher rpm. The boost will keep building and the maximum average hp will go up if the redline is pushed further up. I'd very much like to see this dyno graph extend to 6750 rpm.

The other engine that would be fun to see revving about 1000 rpm higher would be Ott's ITB engine. His torque curve is showing no sign of letting down at his (self imposed) redline.

 

The ignition miss problems were why the engine wasn't run higher. There really wasn't any point in trying to go higher with the ignition problems that were already occuring and getting worse as the RPMs climbed through the range. It would really be nice to see a chart with the ignition and some other aspects of the car more optimized.

 

actually, the percentage of RPM drop is near 72% or 140% (.72)

the averaging of the HP curve works, but not really, when you look at the time spent at the higher RPM is higher. thats where the simulations work better, however you can approximate them by factoring in a longer time period. Using HP-seconds will give you the most accurate approximation of who will be fastest over a distance.

mk