[Mike Stenhouse]

Mike Stenhouse here. Answers to questions about aluminum engines and horsepower ratings. A bit long, but good questions deserve good answers.

I have an all-aluminum 351 Windsor with 4.125 bore by 4.000 stroke for 427 cubic inches. It was built by RDI, the FRPP distributor in Cornelius, NC, here in the heart of NASCAR country. Four bolt mains. Forged crank, rods, and pistons. Ported Victor Jr. heads. Hydraulic roller cam. Performer RPM manifold. Holley 750 carb. It was built as a dual purpose engine - equally at home on the track and the street. I drive it on both a lot. It is a gem.

I am running a Tremec TKO II transmission (0.82 fifth gear) and a 3.08 rear end with 275/60-15 Goodyears on the rear. This tall gearing gives me 52 mph in first, 85 in second, and 126 in third at the 6500 rpm redline. It will go 169 in fourth, but I won’t. I like to breath. Fifth is totally out of the question.

The advantages of aluminum are lighter weight and a resulting better weight distribution.
With the all-aluminum Windsor, my car weighs 2400 pounds ready to rock and roll. This is lighter than the new Mazda Miata, for goodness sake. And 55% of the weight is on the rear wheels. This thing hooks and goes like a rocket ship. Great fun. I have started giving out diapers to passengers on demo rides. It is also reasonably happy in traffic (when it cannot be avoided) and really loves back road blasting in fourth.

We ran my current engine on the engine dyno in an installed configuration (see below) - full side pipes and accessories. It made 480 net horsepower. Given the installation losses, this should be around 550 gross horsepower – adequate (as they say) for a 2400 pound car.

Aluminum has a number of other advantages. It be repaired easier than cast iron in case of (gasp) failure. The block can be welded. The cylinders use sleeves which can be removed and replaced, so it can be rebuilt forever without running out of bore. Bore damage can also be easily repaired. The RDI block is Siamesed, which allows a bore up to 4.185 inches. A 4.125 inch bore gives 427 cubic inches with a 4.000 inch stroke. The longer 4.100 and 4.200 strokes used with the 4.030 inch bore push the rings up on the piston and require a shorter piston to clear the crank. Both of these are reliability issues.

Aluminum is not a strong or as rigid as cast iron. An aluminum block must be designed from scratch as an aluminum block to compensate. Aluminum is 1/3 the weight of cast iron, but a good aluminum block is around 1/2 the weight of a cast iron block, so clearly it has more metal in it. The RDI block is designed from scratch as an aluminum block, which is why I used it. I have heard from engine builders that the Dart aluminum block is a derivative of their cast iron block. Whether this is still true or not, I don’t know. If anybody does, chime in. If it is, it would give me pause.

As to horsepower, there are three ratings that are commonly used.

Gross horsepower is engine dyno horsepower without accessories, with a bell mouth on the carb (no air cleaner), and with open headers. Gross horsepower was used by auto manufacturers up until about 1972.

Net horsepower is installed horsepower with accessories hooked up and functional, full air intake system in place, and full exhaust system in place. It is the standard used by the auto industry since about 1973. Typical installation losses for a clean installation run 15% to 18%. Cobra installations with headers, free flow side pipes, non-restrictive air cleaners, and minimal accessories can be as low as 12%. The 427 Cobras of the 1960's had the highly restrictive Snelling & Helling air cleaners and cast iron log exhaust manifolds with tiny mufflers. Installation losses were a staggering 40% or so.

Rear wheel horsepower is measured on a chassis dyno and takes into account drivetrain losses. The NASCAR guys who dyno the exact same setup on the engine dyno and chassis dyno tell me that the drivetrain loss (net horsepower to rear wheel horsepower) is about 10% to 11%.

Most crate engines and custom built engines are rated in gross horsepower since the engine builder has no control over installation losses and because their test facilities use bell mouths, no accessories, and open exhaust.

The loss from gross horsepower to rear whell horsepower can vary from 22% for a very clean installation to 50% for a very bad one.

Hope all this is helpful. If you got this far congratulations.