Erbman's Engine Emporium, Part IV

Russ Erb

Originally published March 1994

Erbman's Engine Emporium, Part I
Erbman's Engine Emporium, Part II
Erbman's Engine Emporium, Part III
Piston Speed
Bore and Stroke
Compression Ratio
Fuel Injection
A Little Interesting History Speculation
The Emporium Is Closing

So now that we are all self-proclaimed experts in reciprocating engine design, lets look at some actual engine data and see how the engine design factors affect the engine performance. Table 1 is an accumulation of data for many different types of engines, including World War II V-12s, radials, current horizontally opposed engines, auto engines, and even a chainsaw and motorcycle engine.

Go to Top

Piston Speed

We have seen previously that piston speed is proportional to volumetric efficiency up to a limit, with higher piston speeds allowing higher torque. Near the far right side of Table 1, the piston speeds are listed. Several interesting trends can be noted. The three highest piston speeds at rated power listed are the Chevy 350 V-8 (58.96 ft/sec), the Allison V-1710-F (50 ft/sec), and the Packard V-1650-1 Merlin (50 ft/sec). This seems to be the upper limit for piston speed. Also note that to keep up with these piston speeds, these cylinders each have 4 valves; 2 intake and 2 exhaust valves. While this increases the complexity of the engine, it also allows more area in the ports. Consider two equally sized valves in a cylinder (See Figure 1). Draw one large circle, then draw the two largest equally sized non-overlapping circles inside this circle that you can. Each of the inner circles (valves) has half the radius of the large circle, and thus has .25 the area of the large circle. Now draw four equally sized circles in a square formation tangent to each other and tangent to the inside of the large circle. The radius of the small circles (valves) is .414 the radius of the big circle. The area of two of these small circles is .34 the area of the large circle, an increase in area of 9% the area of the large circle, or a 36% increase in valve opening size. I have not heard of anyone using more than 4 valves per cylinder.

Comparison of 2 vs 4 valves per cylinder

Figure 1. Cylinders with 2 and 4 valves

The highest piston speed for an engine with only two valves per cylinder is 49.09 ft/sec for the very large radials.

So if auto engines produce their maximum horsepower at 3500 to 6000 RPM, while aircraft engines produce their maximum horsepower at about 2700 RPM, are auto engines running faster? Not when you look at the piston speed. The piston speeds for auto engines are comparable to the piston speeds for aircraft engines. So will the auto engine conversion wear out faster than an aircraft engine? It depends on which part you are talking about. The pistons/cylinders? Probably not. The piston speeds are comparable. Shafts and bearings? Probably, since the rotation speeds are much higher. Even so, consider this: To drive a car 100,000 miles at 50 mph would take 2000 hours, typical TBO for a Lycoming or Continental. On the other hand, the car engine in this case is at a much lower power setting, so this may not be a fair comparison.

Consider the Rotax 582UL. This engine produces the same horsepower as a Continental A-65, but has only 1/5 the displacement. How can this be? Remember the horsepower is made up of torque multiplied by RPM. The Rotax 582UL must turn at 6200 RPM, while the Continental A-65 only turns at 2300 RPM. Additionally, the Rotax is a 2-stroke engine, so it is producing a power stroke on each rotation rather than every other stroke. If you multiply the number of power strokes per minute times the displacement for each of these engines, the Rotax actually pumps slightly more air per minute than the Continental. The Rotax produces more of its horsepower from rpm than torque. The down side of this is the Rotax must be geared down to turn a propeller at an appropriate speed. Even though the Rotax is running at 6200 RPM, its stroke is so short that the piston speed is only 43.4 ft/sec, which is comparable with other engines listed. The TBO on the Rotax is only about 400 hours, which is much smaller than 2000 hours for a typical Lycoming or Continental, but this is offset by a significantly cheaper price per overhaul. The accumulated cost by 2000 hours for both types of engines is roughly equal.

Also interesting to note is the chainsaw and motorcycle engines listed. While these turn at 11000 and 8500 rpm respectively, the piston speeds are only 42 and 54 ft/sec. These engines are also comparable to the other engines listed.

Go to Top

Bore and Stroke

As we saw in an earlier article, the stroke/bore ratio does not affect the torque or horsepower. Even so, we can note some interesting trends. The stroke/bore ratios range from 0.72 to 1.12. The 0.72 belongs to a Ford V-8, and the second highest stroke/bore ratio (1.11) belongs to the Packard Merlin V-12. There does not seem to be a definite trend between engine type and stroke/bore ratio.

What is interesting is that the Allison, Packard, Pratt & Whitney, Wright, and Lycoming engines use larger cylinders than Continental and auto engines. For instance, for an O-360 class engine, the Lycoming has 4 cylinders, while the Continental uses 6. It can be shown that using more, smaller cylinders will result in a lighter engine. However, more cylinders means more complexity, and may result in an overall larger engine. The Rotax 582UL uses 2 small, fast cylinders for the same horsepower as the Continental A-65, which uses 4 larger, slower cylinders. However, the Rotax requires a reduction gear, which adds extra weight. I don't have any figures to compare the fuel consumption.

Go to Top


For high rpm engines and high horsepower engines, a reduction gear must be used. Looking at Table 1, the fastest propeller rpm numbers are 2700 - 2800 rpm. Why is this? One of the everlasting laws of propeller design is that for high efficiencies, the propeller tips must remain subsonic. At 2700 rpm, the tips of a 72 inch propeller would be moving at 0.76 mach at sea level on a standard day. At tips speeds much above this, shock waves will start to form, the torque required to drive the propeller increases significantly, and the propeller efficiency drops. (Not to mention the noise increase--I have heard the reason that T-6s are so loud on takeoff is the propeller tips are supersonic, producing shock waves.)

The largest engine shown with direct drive to the propeller is rated at 450 horsepower. For high horsepower engines, such as the Pratt & Whitney R-4360, the propeller rpm is as low as 1012 rpm. To absorb this much power, large diameter propellers are required. As the diameter of the propeller goes up, the limiting rpm of the propeller drops to keep the propeller tips subsonic. This is why high horsepower engines are geared to such low propeller rpms.

Go to Top

Compression Ratio

Does a higher compression ratio increase the horsepower output of the engine? Yes, because the thermal efficiency is increased. Compare the Lycoming O-320-A2B and the Lycoming O-320-B2C. The only apparent difference between these engines is upping the compression ratio from 7 to 8.5. Since the bore and stroke are identical between engines, this was presumably done by reducing the volume of the combustion chamber above the piston at top dead center. This results in a 10 horsepower gain. The down side of this is the fuel required changes from 80/87 to 100LL.

Also note that the World War II engines all have compression ratios between 6 and 7. All of these engines were equipped with gear driven superchargers, and generally were used with external turbo-superchargers, hence the lower compression ratio. After 1935, 100 octane fuel was available in commercial quantities and was being used by the military in World War II. The higher octane rating allowed use of supercharging without needing to decrease the compression ratio any more.

Go to Top

Fuel Injection

Looking at the Lycoming and Continental engines, the larger engines are fuel injected (as indicated by IO in the designation instead of just O). While this will increase volumetric efficiency, it may also be that fuel injection is simpler than designing a large carburetor that would be required to supply fuel to this large of an engine.

Go to Top


As previously discussed, when supercharging is added to an engine, the compression ratio normally will be reduced. Three examples of this are shown in the table: Lycoming GO-480 to IGSO-480, Lycoming IO-540 to IGSO-540, and Continental IO-520 to TSIO-520.

Go to Top

A Little Interesting History Speculation

Close inspection of the engine parameters, primarily those built during World War II, leads to some interesting speculation and meaningless facts. The stroke and bore of these engines were designed before the metric system was in vogue, and before decimal measurements were popular either. I say this because the bore and stroke measurements are all even inches or end in multiples of 1/16, 1/8, 1/4, or 1/2 inch. Interesting, but not real important.

At the beginning of World War II, many smaller radials were available, but there was an immediate need for larger, more powerful engines. A quick way to get an engine of double the horsepower would be simply to stick two existing engines together to make a two row radial. The following speculations are based simply on the bore and stroke dimensions:

To reduce the number of parts required to build numerous sizes of engines, Pratt & Whitney only had 3 sizes of cylinders (bore): 5.1875, 5.5, and 5.75 inches. Wright used 2 sizes of cylinders: 5 and 6.125 inches.

Go to Top

The Emporium Is Closing

Well, this brings to an end this series of articles. If you started reading this to figure out how to set you timing or change you spark plugs, then you're probably still looking. Hopefully you have gained some insight into the engineering principles of how your engine is designed, and some of the design changes required for converting auto engines for aircraft use.

Go to Top

Table 1. Engine Parameters
ManufacturerTypePrimary UseCoolingNumber of CylindersValves per CylinderBore (inches)Stroke (inches)Stroke/Bore RatioDisplacement (cubic inches)Compression RatioMax HorsepowerRPM for Max HorsepowerGearingPiston SpeedPropeller RPM
Pratt & WhitneyR-985AircraftAir925.18755.18751.0098564502300133.142300
Pratt & WhitneyR-1340AircraftAir925.755.751.001344660022500.6735.931507
Pratt & WhitneyR-1830AircraftAir1425.55.51.0018306.7120027000.5641.251512
Pratt & WhitneyR-2000AircraftAir1425.755.50.9520006.5145027000.541.251350
Pratt & WhitneyR-2800AircraftAir1825.7561.0428046.65200027000.5451350
Pratt & WhitneyR-4360AircraftAir2825.7561.0443637350027000.375451012
Rotax582ULAircraftLiquid22 cycle2.992.520.8435.445.756662000.38843.42405
McCullochM2-10ChainsawAir12 cycle1.751.3750.783.37.16.511000---42.01---
ChevyCorvair SuperchargedAutoAir623.442.940.851648.251804000---32.67---
Chevy230 I-6AutoLiquid623.753.250.862308.51404400---39.72---
Ford289 V-8AutoLiquid8242.880.7228910.52716000---48---
Ford2.5L I-4AutoLiquid423.683.30.891539.7904400---40.33---
Ford3.0L V-6AutoLiquid623.53.140.891839.11404800---41.87---
AMC232 I-6AutoLiquid623.753.50.9323281003600---35---
AMC258 I-6AutoLiquid623.753.8951.0325881103500---37.86---
Ford4.9L I-6AutoLiquid6243.980.9930013.5145------------
Ford5.0L V-8AutoLiquid82430.75302---185------------
Ford5.8L V-8AutoLiquid8243.50.8735112.9210------------
Ford7.6L V-8AutoLiquid824.363.850.8846010.8230------------
Chevy350 V-8AutoLiquid8243.480.873509.52504400---42.53---
Chevy350 V-8AutoLiquid843.93.660.93350113755800---58.96---

Erbman's Engine Emporium, Part I

Erbman's Engine Emporium, Part II

Erbman's Engine Emporium, Part III

Go to Top

EAA Chapter 1000 Home Page
E-Mail: Web Site Director Russ Erb at
Contents of The Leading Edge and these web pages are the viewpoints of the authors. No claim is made and no liability is assumed, expressed or implied as to the technical accuracy or safety of the material presented. The viewpoints expressed are not necessarily those of Chapter 1000 or the Experimental Aircraft Association.
Revised -- 3 August 2002