Part 1: Applying road race technology to a street/strip motor
Ray T. Bohacz - July 09, 2014 08:28 AM
The 383 has a removable sleeve Race Demon carburetor (with the ability to flow 775, 875 and 975 cfm) that was set to flow 675 cfm.
The L-98 cylinder heads had a nice bowl blend and valve job but other than that were stock. The intake and exhaust runners were not touched except for a gasket match.
The engine block is a production Corvette design that was used from 1984 until 1991. It is nothing special.
The bores looked very good though there was a normal amount of wear. The block will be made 0.040-inch over (4.040 inches) since it is already 0.030-inch over-sized.
The main bearings looked good and were still in specification. This engine was making just shy of double the factory horsepower and was raced for 10 years.
The Callies Stealth crankshaft measured up fine and will be used along with the old block.
The 383 enjoyed 6.00-inch Oliver forged connecting rods that are still in excellent shape but will not be used due to the lofty rpm Tony is hoping for.
The engine showed no signs of excessive wear. It could have run hard for many years yet.
The color on the piston crown was nice and even which means the flame travel and combustion event were excellent. When the flame travel is not even, the piston will be darker in areas.
The second compression ring did show wear for the amount of abuse the Chevy took but the mill still made almost 450 hp on the dyno.
The only real problem was the bolts from the timing gear wore into the cover. It appears that the shop that built the engine did not check the gap and the bolts clearanced themselves.
The gear showed no sign of wear on the Mallory distributor and will be re-used.
As eclectic as the enthusiast community is, no one can deny that the standard performance benchmark is the quarter-mile.
But, a drag race engine is designed to make substantial power in a tight rpm window and does not have to last for tens of thousands of miles. A street car needs to have a broad power range and live under grueling heat and other real-world conditions. We propose the majority of enthusiasts would be better served looking at road race engine theory instead of quarter-mile time slips for their street/strip power plant.
RaceKrafters Automotive Machine in Lancaster, Pennsylvania, is building a road race engine that would be a perfect fit for a street/strip car. The car is owned by Tony Cocurullo and is an old Speed Vision Challenge Corvette that competes in events sanctioned by the Sport Vintage Racing Association and the Historic Sports Car Racing LTD.
The current engine is a 383 Chevrolet with a single-plane intake manifold and one carburetor. The cylinder heads are L-98 aluminum castings with some minor port work and the camshaft is a hydraulic roller. The 3.750-inch stroke crankshaft is connected to the piston with a 6.00-inch long connecting rod. The compression ratio is just shy of 11:1. The engine has performed well for many years but it lost gumption before 6,000 rpm.
We will explore the engine in its current state and discuss what will be done to increase power while meeting other requirements. Part 2 will reveal the machine work, engine assembly and how the parts choices were made. In the final installment (Part 3), a final tune on RaceKrafter’s chassis dyno will be included with tuning changes from the engine dyno to the chassis dyno. It will also assign a loss for the power consumed through the drivetrain. A less power-hungry driveline is just as beneficial as installing a more powerful engine.
RaceKrafters wanted to baseline the current engine. After the dyno session, the engine would be disassembled and checked for wear along with flow testing of the cylinder heads. Though most components would not be reused, examination of all of the parts would reveal potential problem areas. The flow bench testing would be particularly telling since we would be able to see where the air throughput was right now. Over the years I’ve used a very simple but rudimentary equation that can predict horsepower from airflow. It is:
Air flow @ 28 in/water x 0.257 x # of cylinders = HP
The cam/rocker combination yielded a total intake valve lift just shy of 0.600-inch so we used that point from the flow bench test. At 0.600-inch lift with a radius inlet on the port to direct the flow and limit shear, the L-98 cylinder head with 2.00-inch intake valves and the bowls blended flowed 217 cfm.
The cylinder head already started to stall (port flow dropping as the valve lift is increased) since at 0.500-inch the head moved 226 cfm. For the equation to have the least amount of error you need to use the value closest to the actual valve lift. Thus the equation became:
217 X 0.257 X 8 = 446.152 potential horsepower from air flow. We now had a baseline number.
Though the engine was dyno tested before we obtained the flow data, it is interesting to note that at 5,600 rpm (peak power) the little 383 produced 446.20 hp — within 0.05 hp of the air flow estimate calculation! Keep in mind that the calculation is not normally that accurate. On an engine of less than 750 hp, it is usually within three to four percent either way if you have accurate flow bench data.
A 383 is a 350 with a stroked crankshaft. The current engine enjoyed a 6.00-inch long connecting rod in lieu of the standard 350 Chevrolet 5.700-inch long rod. A standard 350 Chevrolet is a 4.00-inch bore with a 3.480-inch stroke crankshaft. To make it a 383 the bore is 0.030-inch over for a total of 4.030 inches with a 3.750-inch stroke crankshaft. This stroke was originally the domain of a production 400 cubic inch small block but with a larger bore.
There are all types of debate about the relationship between the connecting rod length and the stroke. It is defined as the rod/stroke ratio. It is the rod length divided by the stroke. As the rod length increases with a fixed stroke, the ratio goes numerically higher. The benefit of a higher rod/stroke ratio being the reduced angularity of the connecting rod as it pushes the piston against the bore during thrust.
Modifying the rod/stroke ratio has an impact on the frictional losses in the engine. With a 6.00-inch long rod and a 3.750-inch stroke crankshaft, the ratio would be 1.6:1.
A 350 with the same rod length would enjoy a ratio of 1.72:1. In contrast, a 383 with a 5.700-inch long connecting rod would have a ratio of 1.52:1.
A benefit of a higher numerical rod ratio is that the angularity and frictional losses are diminished. An aside to that is the piston dwells longer at TDC, so higher cylinder pressure can be built as the flame expands before the crankshaft rotates. Many believe that this increases engine torque and adds octane tolerance since the smaller combustion region created by the dwelling piston keeps detonation at bay.
The flip side is that the piston dwell time limits the engine’s ability to pump air into the cylinders which, in theory limits volumetric efficiency, the amount of cylinder fill. For this reason, most drag race engines will run a lower numerical rod ratio (often mated with a very short stroke) for the improved pumping action. They will accept the angularity and thrust side bore wear since the engine only runs for a short while and is frequently rebuilt. In a street/strip application the attributes that are advantageous to a road race engine will shine through. Higher torque, less wear, decreased detonation and reduced frictional losses are all good.
A few concerns
A concern when trying to build power and rpm with a 383 in any application is the bore size. A much better combination would be a 377 cubic inch engine. This is a 400 bore (4.125 inches) + 0.030 (4.155 inches) and a 350 stroke (3.48-inch) crankshaft with a 6.00-inch long rod. This would yield a rod ratio of 1.72 but more important, the large bore would unshroud the intake valve and allow for a better airflow. The 383 with a 4.030-inch bore has a good deal of shrouding and the valve needs to go much higher in lift before it un-shrouds. The same cylinder head on a 4.155-inch bore would be unshrouded and would flow more air. For this reason, when testing a cylinder head on a flow bench, the adapter needs to mimic the size of the engine bore.
Another concern was the overall height of the engine in the car. With the hood on, there is only about one-inch clearance to the top of the air cleaner assembly. This is with an Edelbrock Victor Jr. and the current deck height block and cylinder head port configuration. This will limit the choice of cylinder heads since a better flowing design would have a raised runner and that would not fit. Thus, RaceKrafters would need to work within the confines of the small bore and reduced overall height of the engine while still trying to meet Tony’s goals of 7,200 rpm and at least 50 horsepower more than the engine has with no less torque. This would equate to 500 hp and 505 lbs-ft, as a minimum, all while being very tractable and durable.
RaceKrafters has their work cut out for them. The 383 will require a good deal of engineering and pulling out the stops on all the machine work as we’ll see in Part 2.
Flow bench test of L-98 cylinder head cfm
Lift Intake Exhaust
.100 58 53
.200 117 101
.300 165 133
.400 206 155
.500 226 168
.600 217 175
.700 214 178
.800 213 179
Dyno results of the 383
RPM TQ HP BSFC
3000 439.6 251.1 0.41
3100 438.5 258.8 0.40
3200 437.8 266.7 0.39
3300 433.9 272.6 0.39
3400 431.0 279.0 0.40
3500 429.2 286.1 0.40
3600 428.4 293.6 0.40
3700 429.6 302.6 0.39
3800 434.9 314.7 0.38
3900 442.7 328.7 0.39
4000 448.4 341.5 0.38
4100 454.3 354.7 0.40
4200 458.5 366.7 0.38
4300 455.9 373.3 0.38
4400 453.8 380.2 0.39
4500 449.7 385.3 0.41
4600 446.2 390.8 0.43
4700 442.9 396.3 0.43
4800 442.5 404.4 0.43
4900 441.5 411.9 0.43
5000 442.3 421.1 0.42
5100 440.2 427.5 0.42
5200 437.9 433.5 0.42
5300 434.8 438.7 0.43
5400 430.1 442.2 0.43
5500 424.8 444.9 0.43
5600 418.4 446.2 0.43
5700 410.6 445.7 0.43
5800 402.6 444.6 0.45
5900 393.9 442.5 0.46
6000 379.0 443.1 0.44
Average from 3,300 and 5,600 rpm: 439.67 lbs-ft, 372.36 hp
For Your Information:
RaceKrafters Automotive Machine, (717) 399-8780