From: Tom Cloutier [tomasc@pacbell.net] Sent: Monday, January 22, 2001 3:41 PM To: 928 Subject: [928] Comprehensive Dyno Testing Report (the good one I hope!) I have reformatted the tables so that they are more legible and added some explanatory comments. Hope this helps. Louie Ott and I have been working together recently to test several different exhaust configurations. Louie did the dyno testing with his 1990 GT on a Dyno Jet in accord with the DEVEK protocol. I have done the data compilation using Excel. This is an excellent way to compare the averages of three dyno runs for multiple configurations. The data used came directly from the raw data print out for each dyno test. I calculated the areas under the curves in order to quantify the performance of the car over the entire speed range of the test, not merely peak values. We compared total (average) torque and power as well as the time required to accelerate to a particular rpm. Louie put forth a Herculean effort to conduct most of the tests on a single day. A basic question we wanted to answer was how much performance loss is associated with the stock exhaust system, including catalytic converters. And how much loss is attributable to the mufflers and pipes, and to the cats? To answer these questions, Louie established a stock baseline. Then, on another day, he ran tests with no plumbing connected after the cats and then with 2.25” pipes replacing the cats. Finally, on a third day he conducted the major portion of the tests. We also wanted to know how Louie’s X crossover and my venturi merge collector systems compared to the cat bypass systems. We didn’t have a cat bypass, so we substituted the stock head pipes with the H crossover and 2.25” pipes in place of the cats. We’re both of the opinion that the stock H crossover is more efficient than that in the commercially available cat bypass systems. We also wanted to know if there was any performance gain after installing a K&N air filter. The tests were done without a tailpipe or muffler to eliminate the effects they might have. Appropriate length and diameter pipes were added as necessary to maintain approximately equal lengths for each configuration. Among the configurations tested were: 1. Stock (baseline) 2. Stock cats without muffler or tailpipe 3. Stock head pipes with H crossover with 2.25" pipes replacing cats 4. Dual 2.5" head pipes, 4" X crossover, dual 2.5" pipes out, no muffler or tailpipe 5. Burns Stainless type venturi merge collector with two 2.5" head pipes, 3" venturi, 3.5" single pipe out without muffler or tailpipe 6. 4” X with full exhaust and K&N air filter 7. 4” X crossover with Louie’s full dual Bullet muffler exhaust system 8. 3” venturi with straight 3.5” tailpipe and Magnaflow muffler at end Louie utilizes the X-type crossovers in his dual exhaust systems while I use the Burns Stainless venturi merge collectors in my Avenger and Scavenger systems. Results Stock Exhaust System (Complete): Maximum loss of 32 lb-ft and 28 hp. Average loss of 20 lb-ft and 18 hp compared to stock H. This compares the dead stock configuration to the stock H with no cats or mufflers. Stock muffler & pipes: Average loss of 11 lb-ft and 10 hp. This is the contribution by tailpipes and mufflers to total loss. Stock dual cats: Average loss of 9 lb-ft and 8 hp. Cat contribution to total loss. K&N air filter: Loss of 3 lb-ft of torque and 3 hp compared to stock filter. Cat bypass vs 4” X: The X crossover gains 14 lb-ft of torque and 11 hp more than stock H. Cat bypass vs 3” venturi: The venturi gains 19 lb-ft and 16 hp more than the stock H. Using the stock baseline as a basis, the average performance gain for each configuration is tabulated below. _______________Average__Average__Average__%Gain Over__%Gain Over __________________TQ____%Gain______HP______Baseline____Stock H Baseline__________266___Basis______227______Basis__________ Stock Cats, Open__277___4.0%_______237______3.9%___________ Stock H, Open_____286___7.1%_______245______7.2%_______2.9% 4” X, Open________295___9.7%_______252______9.9%_______3.2% 3” Venturi, Open__296__10.1%_______255_____10.7%_______4.2% 4” X, Full________299__11.0%_______255_____10.9%_______3.5% 3” Venturi, Full__297__10.3%_______254_____10.4%___________ Stock Air Filter__293___9.2%_______250______9.2%___________ K&N Air Filter____290___8.1%_______247______8.1%___________ Time to reach given RPM for the various configurations tested. ________________Average Time per_______Elapsed Time _________________100 RPM (sec)_____2700 to 6400 RPM (sec) Stock Cats___________0.216______________8.01 Stock H, Open________0.199______________7.54 4” X, Open___________0.187______________6.94 3” Venturi, Open_____0.186______________6.89 4” X, Full___________0.185______________6.85 3” Venturi, Full_____0.187______________6.90 Stock Air Filter_____0.188______________6.94 K&N Air Filter_______0.190______________7.03 Even though a muffler may be a “non-restrictive” type, it and the associated plumbing often increase the back pressure in the system. We found that the addition of after-market tailpipes and mufflers increased performance. You may have noticed that the venturi gave slightly more performance than the X crossover in the tests with open exhaust systems, but slightly less with tailpipes and mufflers attached. The test tailpipe and muffler used with the venturi was simply a straight piece of 3.5” pipe with a straight-through Magnaflow muffler attached to the end. We both feel this configuration was freer flowing than the dual pipes and mufflers used with the X crossover and, with a little more back pressure, would have given better results. Although back pressure increases pumping losses, and eliminating it should increase performance, this can also, “alter the flow characteristics of the whole system so that the all-important port pressure at exhaust TDC is increased; as the power loss from the impeded scavenge was greater than the gain from the open exhaust, the result was less power.” (Scientific Design of Exhaust & Intake Systems; Phillip H. Smith and John C. Morrison). I know this statement will be met with some objections and if there is a better explanation than Smith and Morrison’s, please let us know. Louie and I have invested quite a bit of time and money in these tests and hope the results are as illuminating for you as they have been for us. There are about 7000 data and calculated values in this study! I’ll post the dyno charts on my website some time this week (www.thepowerbroker.net). I can email the spreadsheet to anyone who wants it. We’d welcome your input, questions, comments, etc. Tom Cloutier & Louie Ott tomasc@pacbell.net LouisOtt@cs.com