Dynoed my 427so Cobra today.... interesting to say the least!

[ByronRACE]

Why use a chassis dyno? Confirmation of the known and/or measurement of the unknown. The former, is what causes most problems because it causes controversy. There is usually a gigantic discrepency between most engine builder figures and what is actually measured; even when accounting for drivetrain losses. I've personally dyno'd over 1600 Ford powered cars on a dynojet 248 and can count the number times on one hand that a SBF or BBF actually exceeded 1rwhp per cubic inch on pump gas without a power adder. Compare that experience with the number of engine advertisements with power figures that would EASILY exceed 1rwhp per cubic inch in a car, and you have discovered the root of the problem...marketing fudge factor; big numbers sell engines.

As far as dyno accuracy is concerned, I've compared engine dyno to chassis dyno directly, with a variety of transmissions and have a very good idea regarding 5spd drivetrain losses for various transmissions (within 5%). I've also then taken the same car from the chassis dyno to the street, and produced horsepower/torque functions using dataloggers and simple physics...mass, time, speed...and produced the same data (within 5%) of the chassis dyno.

So, for example...

If you had an engine dyno that showed 600fwhp, and put that engine into a car with say a T5 transmission and 8.8 Ford rear, and did a 4th gear pull that that read 400rwhp... and your engine builder is telling you it's because of the water pump, alternator, and drivetrain losses...there's more to the story.

600fwhp
- 13-15% for drivetrain losses (90hp max)
- 5% for max accessory losses (30hp max)
- 5% fudge factor for tire slip, calculation error, whatever (30hp max)

So worst case, the chassis dyno should read at LEAST 450rwhp for an engine advertised at 600fwhp. In reality, it should read higher because the error isn't truly cumulative as shown above; some losses are related.

Most of the time, a 600fwhp advertised engine will dyno at 400rwhp or below. And, the same 600fwhp engine as tested on your own engine dyno will read far less than 600fwhp; which is even harder to explain away. This is all too common and expected.

In your case, after reviewing the dyno graph, peak torque is at at 5200rpm and peak power has been reached by 5800rpm. As advertised, the engine was supposed to reach peak power by 6700rpm. Either the intake/exhaust geometry has changed dramatically; or the cam or cam timing has changed dramatically since it was dyno'd. If you corked the intake with air filters, it would not change the location of the peaks by nearly 1000rpm; it would just change the amplitude of the high rpm portions of the graph. Same is true for exhaust corking (mufflers). My guess is it's not the same cam, or perhaps not the same cam timing. Furthermore, changing the cam timing after achieving a peak number on an engine dyno isn't unheard of...it's done to produce more bottom-end torque to make the car idle better and be more fun to drive in the rpm band you use the most. You can make a better high rpm number by retarding the cam a bunch, but it may not idle / drive worth a damn...so you compromise. If I had to guess, something like this is what took place. Either that, or the cam is the same and they had a very different intake/exhaust system on the car when it was on the engine dyno. Race gas and timing advance would help amplitude as well, and shift the RPM slightly higher...but I wouldn't expect 1000rpm out of fuel/timing alone.

Either way, you' still have a great running engine approaching 1rwhp/cubic inch at the tires...and that's nothing to be ashamed of.