In this part, I will consider what kinds of engines exist, and what characteristics they must have in order to be useful as airship engines. When I move on to specifics, it will be with respect to internal combustion engines; steam propulsion will be covered in Part Three and more exotic systems in Part Four.
While the focus of this article is on airship propulsion, much of what is said here about engines is applicable to other vehicles.
In selecting an engine for an airship we need to consider initial cost, operating cost, and performance. The most important operating costs are fuel and maintenance. Fuel costs are dependent on what fuels the engine can use, the cost and energy content of the fuel, and the efficiency with which that energy can be converted into propulsive power. Maintenance costs tend to increase with increasing operating pressures and temperatures.
With regard to performance, the most important parameters are probably the power/weight ratio, the power/volume ratio, and the sensitivity of power to altitude. (Fuel efficiency also affects stamina and range.)
The Engine Menagerie
An engine, in essence, is a device for taking the chemical energy of fuel and using it to do work. While the concept is simple, there are a bewildering variety of implementations.
The first great divide is between external and internal combustion engines. With an external combustion engine, the working fluid is heated in a boiler by burning fuel, and fed to the engine to do work. The working fluid and the combustion gases never mix together.
The obvious example is the steam engine powerplant, in which the fuel is used to boil water, and the steam moves the pistons (or turbine) of the engine. Steam engines may further be characterized as condensing (the spent steam is condensed into a liquid water and returned to the boiler) or non-condensing (the spent steam is exhausted into the atmosphere). Steam locomotives had non-condensing engines, and thus required water towers along their route. Clearly, steam engines for airships will need to be of the condensing type.
Because steam engines use water as the working fluid, they have two important operational limitations. First, heat must be used to vaporize liquid, and the heat so used doesn’t generate power. Secondly, water freezes at a temperature that is encountered in the normal working environment and it may be necessary to supply additional heat to the water tank to keep it liquid.
The Stirling engine is an external combustion engine in which the working fluid is always in a gaseous state.
Some stationary powerplants make use of binary vapor cycles, in which a low temperature working fluid cycle is coupled with a high temperature one, e.g. water-mercury.
In an internal combustion engine (ICE), there is no separate working fluid; the expanding combustion gases do the work.
The internal combustion may be intermittent (as in gasoline and diesel engines) or continuous (as in most jet engines). Intermittent combustion engines may be further divided into those with two- or four-stroke cycles (other cycles are unusual). Typically, two-stroke engines have higher power-weight ratios, and four-strokes better fuel efficiency, but this depends very much on the design.
Alternatively, internal combustion engines may be classified by ignition system: those in which the fuel-air mixture is ignited by a spark (as in a conventional gasoline engine), by compression (as in a diesel engine), or by contact with a hot metal surface (as, at least initially, in a hot bulb engine).
Another possible categorization relates to how the working fluid does work, and it is between reciprocating piston, and pistonless, engines. With a piston engine, the expanding fluid/gas is inside a cylinder, and on one “stroke” they press against a piston, forcing it to the bottom. On another stroke, they leave the chamber, and the piston returns to its original position. The back-and-forth (reciprocating) movement of the piston is mechanically converted to a rotating movement, whereby it turns a crankshaft. Note that both external and internal combustion engines may be piston engines.
Piston engines may further be classified as counterflow or uniflow. In the first case, the working fluid leaves the cylinder the same way it came in; in the second, it enters at the ends and leaves at the center.
Piston engines usually have more than one cylinder, in which case they may be classified according to how those cylinders are arranged. The most common arrangements are straight (cylinders in single file), flat (two banks of cylinders, directly opposite each other), V (two banks of cylinders, at an angle to each other), and radial (an odd number of cylinders, arranged radially around the crankshaft). The flat (horizontally opposed) configuration is probably the most common for general aviation aircraft, because of its relatively low drag. (FTGU 41). In contrast, radials have been criticized for causing higher drag, because of their girth. But bear in mind that while, on aircraft, the engine accounts for a substantial percentage of the frontal surface area, that’s not likely to be true for airships.
In the above piston configurations, while the different piston/cylinder combinations all act mechanically in cooperation, there is no fluid connection between the cylinders, they are “parallel.”
However, cylinders may also be connected in series, so the exhaust gas from one (high-pressure) cylinder becomes the working gas for another (low pressure). This is called compounding. While it is most commonly found in steam engines, there were some (disappointing) attempts to apply it to internal combustion engines. (Self).
In pistonless engines, an expanding working fluid, or, more commonly, combustion gases, may be used to rotate a turbine. Both external (think steam turbine locomotives and stationary powerplants) and internal combustion engines can include a turbine, but all of the pistonless aircraft engines feature internal combustion.
In a turboprop, the turbine drives a propeller. In a turbojet, the gases escaping the turbine provide jet propulsion, and the turbine is used to compress the air fed into the turbine. In a turbofan, there is a fan “upstream” and a second turbine “downstream” of the core engine (the latter being similar to the engine of a turbojet). Some of the air leaving the fan bypasses the core engine.
Then there’s the Wankel rotary engine—”rotary” not because of the arrangement of its pistons, but because the gases turn a rotor. This rotor is triangular, but with curved sides, mounted eccentrically on the driveshaft. Unlike the turboprop, it features intermittent combustion.
Yet another “rotary” engine was the 1908 Gnome. The engine rotated around a fixed crankshaft. This sounds crazy, but rotary engines powered “more than 80 percent of the airplanes built in World War I before 1917.” The principal disadvantage of the rotating engine design was that the high centrifugal forces limited the engine speed (rpm) and thus the design power. ( Anderson 152ff).
Internal Combustion Engine Performance
The power output of an internal combustion piston engine is a function of three groups of factors: (1) the engine design (number of cylinders, cylinder arrangement, piston diameter and stroke length, compression ratio, etc.), (2) the choice of fuel, and (3) the settings of the engine controls. The thrust of an aircraft engine is further dependent on the propeller design (number of blades, diameter, etc.) and the propeller rpm and blade pitch.
The engine displacement—the nominal volume of the “breath” of fuel-air mixture it could take in one intake stroke—is the product of the number of cylinders, the length of the stroke, and the surface area of the piston.
The brake (crankshaft output) horsepower of the engine may be measured with a dynamometer. If this brake horsepower is divided by the engine displacement and the engine speed (the number of revolutions per minute of the crankshaft, itself proportional to the average piston speed), we get what’s called the brake mean effective pressure (BMEP). The BMEP is affected by the compression ratio, the energy density of the fuel, intake air temperature, the richness of the fuel-air mixture, friction losses, etc. If these are all held constant, then the mean effective pressure will be proportional to the intake manifold pressure. An engine will have a maximum allowable BMEP, and this varies with engine speed.
We need to distinguish between power and torque. Torque is the rotating force that the engine applies to the crankshaft. When you tighten or loosen a bolt with a wrench, you are applying torque—the force you place on the end of the wrench, multiplied by the length of the wrench. Torque causes the speed of rotation (engine speed) to increase until the resistance (which is from the propeller and shaft) equals the torque applied by the engine. Power is the product of the torque and the engine speed. (And thus torque is proportional to BMEP.)
Generally speaking, obtaining high torque requires a high displacement. The engine displacement is the total volume swept by all the pistons, i.e., the number of pistons times the piston bore cross-sectional area times the stroke length. Unfortunately, a large displacement usually makes for a large and heavy engine.
An engine will always have its maximum torque at a lower engine speed (rpm) than its maximum power, and the range of rpms between the two is called the power band.
An engine can have a high torque over a narrow rpm band, or a lesser torque over a wider band. With auto engines, the wide band is preferred, and the maximum torque is usually achieved at a substantially lower engine speed (rpm) than the point of maximum power. This permits adjustment to a wide range of driving conditions (trudging uphill, coasting downhill, etc.).
Aircraft engines are designed to operate primarily in a narrow engine speed band and hence have a torque peak at close to the normal cruise setting, and remaining constant for higher rpm. (EPI). Typically, they generate high torque at low rpm (< 2500rpm typically) because of propeller limitations.
Excessive manifold pressure stresses the pistons and cylinders, and may cause detonation of the fuel-air mixture (we want it to burn, not explode). It also increases cylinder head temperatures and thereby reduces the ability of the elements to withstand the stresses placed upon them.
Power increases with engine rpm, but only up to a point; for high rpm the manifold pressure has to be limited. Also, friction increases with engine speed, so the gap between indicated and “brake” horsepower widens. Consequently, maximum power is at less than maximum rpm.
Altitude Effects on Internal Combustion Engine Performance
A normally aspirated engine is one in which the pressure of the intake air equals the ambient pressure—which decreases with altitude, reducing the power output for a given engine speed.
A turbocharger uses the engine exhaust (otherwise wasted in a piston engine) to drive a turbine, which powers a compressor, which increases the pressure of the intake air. The speed of the turbine, and hence the degree of compression, is proportional to the difference in pressure between the exhaust gas and the outside air. Thus, the turbocharger compensates for the pressure drop up to a critical altitude. The power cost is relatively small (the exhaust stroke must push against back-pressure created by the turbocharger), but the element does add weight to the powerplant. The turbocharger’s compressor is exposed to hot (2,200–1500oF) exhaust gases and turns at very high rpm (20,000–30,000), so we need a high-temperature, high-strength metal. Early turbochargers used nickel-chromium steel alloys. (Howard-White 233–4).
Turbochargers come in two varieties. Turbo-normalized turbochargers seek only to raise the intake pressure to normal sea level pressure. Ground-boost turbochargers increase it further, to enhance performance even at sea level.
A supercharger is a compressor powered directly by the engine output (crankshaft). Generally speaking, superchargers have the advantages of less complexity, shorter lag times and greater reliability, and turbochargers of greater fuel efficiency and (with qualifications beyond the scope of this article) higher maximum power output. (aerosuperchargers.com).
A downside of compression is that it raises the temperature of the intake air, which reduces its density. While the net effect of turbocharging or supercharging is still an increase, you can counteract the temperature effect by adding an “intercooler” or “aftercooler.” This is an unpowered heat exchanger of high surface area, transferring the heat to the atmosphere or a liquid coolant. It adds to the size and weight of the powerplant. The intercooler was proposed by Ricardo in 1914 and incorporated into the RHA engine in 1918.
Engine Efficiency, Overall Efficiency, and Fuel Consumption
To compute fuel consumption, we must consider both propulsive efficiency (see part 1) and engine efficiency (for a steam propulsion, I mean the powerplant efficiency, the combined efficiency of the boiler and the steam engine proper). This is the ratio of the engine power output to the potential power output of the fuel at the rate it is being burnt.
The overall efficiency of the airship propulsive machinery—its effectiveness at converting fuel into motion—is the product of the engine efficiency and the propulsive efficiency.
The total fuel (mass or volume) consumed on a route segment is the propulsive work done, divided by the product of the overall efficiency and energy density (energy/mass or volume) of the fuel. Thus, it’s proportional to the square of the airspeed, and to the distance traveled.
Engines are frequently compared on the basis of “brake specific fuel consumption” (the rate of fuel consumption divided by the engine power output), but this varies according to conditions. Naturally aspirated piston engines have BSFC of 0.45–0.5 lh/hp-hr, whereas supercharging raises BSFC to 0.5–0.6, and turbocharging to 0.6–0.65 (Li).
The fuel consumption per hour is the BSFC times the current brake horsepower output of the engine, and the fuel consumption per unit ground distance traveled is the BSFC times the brake horsepower, divided by the ground speed (which, for still air flying, equals the air speed).
Engine efficiency equals one divided by the product of the BSFC and the energy density of the fuel. The engine efficiency can be broken down into a series of individual efficiency factors, such as thermal, volumetric and mechanical efficiency.
The ultimate limit on the efficiency of an engine is thermodynamic in nature, and that determines its thermal efficiency. For a gasoline engine, the theoretical thermal efficiency depends on the compression ratio, with a value of 57.5% for an 8.5:1 Lycoming O-320-D2J engine. However, power is dribbled away in other ways (for a breakdown, see Lowry, 124–6), so that the “total” efficiency for that engine is 33.3%.
Power and Weight
The power-powerplant weight ratio (P/W) is significant for any vehicle powerplant, but most of all for aircraft, and least for surface ships. The weight tolerance of airships falls somewhere in between. Airships generate lift statically, and thus have the advantage over aircraft that they don’t need power to generate lift (unless they are using hot air or hot gas lift). Aircraft generate lift dynamically and their engines must overcome the corresponding induced drag as well as propel the aircraft forward.
Nonetheless, according to Muller, “the chief factor which delayed for so long the development of the Zeppelin ships was the weight and limited power of the engines available.”
Bear in mind that even with pure hydrogen, every additional 71 pounds of engine requires another 1,000 cubic feet of lift gas to support it. And that additional lift gas volume requires enlarging the envelope, which in turn requires providing a suitable support structure . . . and more lift gas to support its weight. If the lift gas is hot air, then each 1,000 cubic feet provides only about 20 pounds of lift.
Applied power is force times speed and thus P/W varies with speed. Generally speaking, the ratios are quoted for either maximum (rated) power or a typical power setting.
The higher the P/W the better, all else (e.g., fuel consumption) being equal. (If P/W is expressed in horsepower/pound or kilowatts/kilogram, the value is usually less than one. And using horsepower/ton or kilowatts/tonne, as is done for locomotives, seems strange when applied to smaller vehicles. Hence, some people prefer to use the reverse ratio (W/P), for which lower is better. That’s what I do below, in pounds/hp. )
What can reasonably be achieved, either with converted auto engines or downtime-builds, may be estimated from historical auto and early aircraft engine data (Table G1). Judging from the historical airship engine data (Tables A1, A2), a W/P of five or less is sufficient for practical use.
A 1929 airship design guide assumes that “the weight of the power plant and its cars to be 8 lb./HP and the weight of the fuel and fuel system to be 0.6 lb. per horsepower hour.” (Burgess/TN194, 4).
Gasoline. Looking at early aircraft gasoline-fueled internal combustion engines, the Wright Brothers’ 1903 engine weighed 16 [11.2] pounds/horsepower. The 1905 water-cooled Antoinette dropped this down to 2.2 [2.9], but was unreliable. ( Anderson 152; bracketed values, Taylor Table 1). A 1909 model of the air-cooled Gnome rotary weighed 165 pounds and yielded 50 horsepower, which was quite respectable (153). A few years later, water-cooled engines made a comeback, with an in-line configuration; the Wright 1913 engine was 5 pounds/hp, and the 1915 Mercedes was 3.43 (155). Other noteworthy aircraft engines included the 1922 Curtiss D-12 (325 hp @ 2.16 pounds/hp)(156), and the 1917 Liberty V-12 (400 hp @2.04 pounds/hp)(157). The latter remained an important engine even in the Thirties. Also note the Hispano-Suiza “A” military aero engine of 1914 (3.17 lb/hp)(Gunston 75). By 1930, engine designers had reduced the W/P below two (1930 Rolls Royce Kestrel VI, 1.77) and by 1936 the Rolls Royce Merlin I was at 1.28 (NASA). It’s possible for a gasoline piston engine to break the 1 pound/hp barrier (Pratt & Whitney 28-cylinder supercharged R-4360 used in the B-50 Superfortress was 0.91) but it’s not easy.
Other than the 1903 Wright engine, the aircraft engines of historical importance with the highest W/Ps were the 1908 Renault (6.9), the 1909 Anzani (5.9), and the 1909 Darracq (5.06)(Taylor).
The down-time built Swarz radials in canon supposedly offer a substantially better weight/power ratio (1.83 lb/hp) than WW I military aircraft engines. I suspect that this is just a case of advertised power being a lot better than installed power, or of creative interpretation of what parts count as part of the weight of the engine, and probably also assumes gasoline fuel.
Diesel. The Napier Deltic engine on the fast patrol boat Dark Hunter (1954) had a W/P of 4.2 pounds/hp. (Flight). The 1930s Jumo 205-E two-stroke, water-cooled aero diesel engine had a dry weight of 1,260 pounds (I am not sure whether that includes the 290 pound cooling system weight), and yielded 700 hp @2600 rpm (1.8 pounds/hp). Diesel engines were used on the R101, Hindenburg and Graf Zeppelin II.
Up-Time Engine Selection and Acquisition
I suspect that the USE Navy is going to grab every high-powered diesel engine it can find, to propel those ironclads. Also, since diesel engines can run on vegetable oil, owners of cars and trucks with diesel engines are more likely to keep them intact than to sell off the engines.
On the other hand, it’s clear from canon that there are endemic shortages of gasoline. Given that there are probably several thousand cars with gasoline engines, and little in the way of paved roads for them to drive upon, I suspect that it will be relatively easy (although still expensive) to buy gasoline engines.
Automobile gasoline engines may be repurposed for airship use, as indeed some already were for the first 1632 universe aircraft. The new universe’s first aircraft, the Las Vegas Belle, has a VW engine. ( Flint, 1633, chapter 11). In 1633, Hal conducts an inventory of the autos in Grantville with a view to determining which the nascent air force should “take out options on,” as possible sources of engines for the Gustav. His initial list includes two Mazda RX-7s (with 13B powerplants, providing over 160 hp), four Saturns (2.2 liter engines), two Honda Preludes, ten or so Chevy S-10s, and a variety of makes and models with V-6 engines. ( Flint, 1633, chapter 27). Some of these engines are covered in the table below:
Table G1: Engine Power and Weight for Sample Auto Engines
Max Pwr hp@rpm*
Engines for cars mentioned in Canon
GM Iron Duke Tech IV, 2.5L, 9:1
used on the chevrolet S-10 in 1985-1993. Has fuel injection. Iron block and head.
Mazda 13B Wankel rotary (RX7 sports car)
Power, Wt for 13b-Fi. Has fuel injection, must be modified for aircraft use. Iron rotors in aluminum** housing.
4x Saturn Ecotec L61, 2.2L, 10:1
introduced in 2000. Cast aluminum cylinder head and block with cast iron cylinder liners.
Honda A18A, 1.8L
Honda Prelude, 1984-87. Carbureted. Iron block, aluminum heads.
Acura Integra, 1994-2001, listed for comparison to A18A. Aluminum block.
Popular Pre-RoF Auto Engines for Modern Experimental Aircraft
Volkswagen, 1584cc, 8:1
air-cooled. VW generically mentioned in canon. Iron cylinder, aluminum cylinder head and piston, magnesium (Elektron) crankcase.
Subaru EA81, 1781cc
liquid-cooled. 3.2–4 gal/hr @75% power. Aluminum blocks, cylinders.
Other Popular Auto Engines, Pre-RoF
Chevrolet small block 350 L-48,
1967–1980. Iron block, cylinders, head.
*rated power on gasoline, will get less on M85.
** reference to aluminum or magnesium means to alloys, not elemental metal!
In selecting engines, Hal’s stated concerns are raw power (and thus, in turn, compression ratio), the power-to-weight ratio, the availability of spare parts (tending to disqualify one-of-a-kind engines), and compatibility (immediate or by conversion) with an alcohol-rich fuel mixture.
Hal is quite happy to find auto engines that offer a power-weight (P/W) ratio greater than 0.35 (i.e., W/P under 2.86). I am not sure if that’s quoted on the basis of gasoline or M85 (see below); the latter provides less “oomph!”
Increasing the compression ratio potentially increases available power, without a change in weight. Doing this requires increasing the bore or stroke, or changing the piston shape (domed is better than flat-top). You can increase stroke, for example, by milling the heads. Increasing the power-to-weight ratio is also possible if there are unnecessary accessories that can be stripped away.
Up-Time Engine Conversion for Alternative Fuels
Canon suggests that at least through 1635, there is a shortage of gasoline, and the newly built aircraft are running on an alcohol-gasoline blend. The alcohol may be methanol or ethanol, and the most popular blends are M85 (85% methanol) and E85 (85% ethanol).
Alcohol has a lower energy density than gasoline, so the fuel-air ratio fed to the engine must be richer. Adapting an auto engine to run on alcohol is much easier if the engine is carbureted. It’s then a matter of increasing the diameter of the main jet orifice of the carburetor, and it may also be necessary or at least desirable to fiddle with the idle mixture screw, the power valve, the accelerator pump, the choke system, the ignition timing, and the compression ratio (increasing it is desirable). (Mother’s).
The last year in which new American cars were carbureted was 1990. Carburetors were replaced by fuel injectors, which atomize and inject the fuel, rather than letting it get sucked in by the passing intake air. By 1990, fuel injection systems were predominantly under electronic control. This potentially has some consequences for converting post-1990 cars to use of alcohol-rich fuel.
If you are very lucky, the car is a flexible-fuel vehicle that can run upon, and automatically detect, E85 or M85 fuel. While such vehicles were first sold in the USA in 1993, it’s not very likely that we’ll find any in Grantville.
Secondly, we may be able to modify the EFI to accommodate E85 or M85. This means, at a minimum, reprogramming the EFI, and you most likely will need a larger fuel injector (to accommodate a larger fuel-air ratio), and modifications in the working range of the lambda (oxygen) sensor. Please note that alcohol is highly corrosive and so may require substantial changes to the fuel system, e.g., replacing existing components with stainless steel ones, and use of special oils. (Joseph).
Finally, there is the option of performing an EFI-to-carburetor conversion. This has been done with various “pony” cars, e.g., Mustangs, Camaros, and Firebirds, but you would need to be able to salvage a compatible carburetor, intake manifold, distributor, fuel pump, etc. from another car and of course you would have to have a good reason to bother.
Most high-powered auto engines are going to develop maximum power at engine speeds in excess of 5,000 rpm. Aircraft propellers must be operated at a low enough rpm so the propeller tips aren’t moving at transonic (over 0.85 Mach) speed (which would greatly increase drag). This depends on the size of the propeller and its pitch, but with typical aircraft propellers (diameters ranging up to 51 inches), exceeding 3100 rpm would be unusual. Hence, when auto engines are converted for aircraft use, you usually need a propeller speed reduction unit (PSRU) of some kind. Since airships are likely to use larger propellers than aircraft, the reduction ratio for them will be higher.
Modern auto engines are usually liquid-cooled, so either you need to provide a water cooling system, or modify them to accommodate air cooling. (Water cooling would increase weight, see below.)
On U.S. roads, auto engines spend most of their time at low loads (perhaps 20 hp at highway speed), leading some authorities to question their suitability for aircraft (which are mostly operated at 75% power). Conversion advocates argue that modern auto engines are “routinely tested during development at full power . . . for periods of up to 1200 hours”, routinely cruised in Europe at speeds “50–100% higher than what we see in North America,” and sometimes have “slightly modified marine” counterparts that “operate in the same kind of high continuous power and rpm environment as an aircraft engine.” (sdsefi.com).
Auto engines are not the only gasoline engines in Grantville. In Gannon, “Upward Mobility” (Ring of Fire 3), Estuban Miro uses lawnmower engines to power a hot air-based airship (Swordfish); its first test flight is in April 1635.
New Internal Combustion Engines for Airships
New internal combustion engines are clearly going to be developed for aircraft use and the R&D will naturally also produce engines suitable for airships. Indeed, new aircraft IC engines have already appeared in canon. In spring 1635, Swartz Aviation produced four seven-cylinder, 120 horsepower, air-cooled radial engines, weighing 220 pounds each. (Huff and Goodlett, “The Spark of Inspiration,” Grantville Gazette 13, and Cooper, “Story Time Frames.”)
An obvious advantage of the IC engine is that would-be designers have a thousand or more up-time automobile, truck, motorcycle, lawnmower, snow blower, etc. engines to examine, and a large number of up-timers (including of course the mechanics at the various auto dealerships) with expertise in the repair of IC engines.
Powerplant Weight Reduction
It is possible that the W/P ratio of new construction powerplants (steam or IC) can be improved by substitution of superior materials, e.g., one with a higher strength/weight ratio than traditional materials.
Please note that this article is looking only at the issue of where these materials could be used in an engine, and how much weight they would save—not when in the new timeline they would actually be available at a reasonable price. For detailed analysis of issues relating to aluminum production in the 1632 universe, see Cooper, “Aluminum: Will O’ the Wisp?” (Grantville Gazette 8). Magnesium and titanium production are touched upon briefly in Cooper, “Industrial Alchemy, Part 2: Inorganic Chemical Bestiary” (Grantville Gazette 25).
There are three factors we are balancing: weight, cost and performance. Our goal is to reduce weight without substantially impairing performance and without too great a cost. Table M1 compares the densities (weight/volume) of the traditional metals iron and copper with titanium, aluminum and magnesium. Alloys, of course, will be of densities that are the averages of the densities of the component elements (table M1)
Table M1 also provides cost data, but this must be taken with a very large grain of salt. Not only does volume of production make a difference (a custom piece would typically cost double or triple), the economy of the new universe will differ greatly from both that of 1630 and 2000.
Table M1: Metal Density and Specific Cost
Relative Costs/Weight (by Source)
C Steel (*A36)
Gray cast iron
Performance is the trickiest aspect of all. That’s because an engineering material has a multitude of mechanical and thermal properties, and which are relevant, and how much, is both dependent on how it is to be used, and somewhat subjective.
Many of these properties are expressed on a geometric basis, per unit thickness, area or volume. A substitute may be inferior in say yield strength (pounds/square inch) but comparable or better if that characteristics is divided by weight—implying that you can use more of the substitute, to compensate, and still come out ahead on a weight basis. (Kevorkijan).
We need to briefly mention how parts are made, because it affects performance. Casting involves pouring the molten alloy into a mold and then allowing it to cool and solidify. This is the best way to make complex shapes cheaply. Machining (drilling holes, grinding down surfaces, etc.) may be needed to fine-tune the part.
The forging process starts with a cast ingot, although alloys intended for shaping a material without melting it (by forging, rolling, drawing, extruding, etc.) are called wrought alloys. Forging shapes metal by pressure; this can be done with a manual hammer and anvil or with a press or a power hammer. This produces a metal product which has a grain structure which confers greater strength in the direction the part is worked. Forging tends to be more expensive than casting, and is generally limited to relative small, simple forms.
Pure aluminum is weak (1600 psi yield strength), so usually we are using alloys (and possibly heat treatment) to increase strength. The alloys have high thermal conductivity, and the strength and corrosion resistance depend on the alloying element (these relationships will probably need to be rediscovered in the new universe). Both casting and wrought alloys are known.
Pure magnesium is also weak (3000 psi) and therefore is usually encountered in engineering as a casting or wrought alloy. These typically have a higher specific strength (strength/weight) than aluminum alloys, but also lower stiffness, thermal conductivity, and melting point. (Gaines). It may be cast and machined faster than aluminum. Molten magnesium must be protected from oxygen or it will oxidize and explode. It is easy to machine, but can catch fire if cut dry. Magnesium is resistant to corrosion by fresh water but readily attacked by sals, acids and alkalis.
Titanium is mostly used in unalloyed form. Its principal advantages are high specific strength at high temperature, and corrosion resistance.
Materials Selection for Internal Combustion Powerplants
We look at materials criteria for the various parts of the IC engine (Raman, Chapter 28), and consider whether lightweight metals can satisfy them. The heaviest component is the engine block (~25%)(Nguyen), followed by the crankshaft and cylinder head. Those would be followed by the oil pan, and the intake and exhaust manifolds. The pistons and the piston and connecting rods are a very small portion of the total weight. (Oreo; Lycan).
The engine block must be castable (because of its complex shape) and machinable (to take all the holes). In autos, it’s not seriously stressed, and gray cast iron is commonplace. In aircraft, to reduce weight it may be cast or (for more strength) forged aluminum. Similar materials choices are appropriate for the crankcase.
In the cylinder barrels, we want high thermal conductivity (heat dissipation), thermal shock resistance (resist thermal cycle), abrasion resistance (against piston), and corrosion resistance (hot gas exposure). Gray cast iron is still often used. Cast aluminum is great from a heat transfer standpoint, but the problem is poor abrasion resistance. The latter is especially problematic if the contact is with a material of similar hardness. If the piston is aluminum, it’s possible to use a cast iron sleeve or to chrome plate the piston to alter the relative hardness.
For the pistons, high thermal conductivity lowers the operating temperature. Since power is proportional to piston speed, low density improves performance, not just W/P ratio. Both of these factors favor aluminum over cast iron, and for aircraft, forged is preferred over cast, to improve strength.
Despite aluminum’s advantages in heat dissipation, aircraft designers were still concerned about how strong aluminum alloys like duralumin (3.5–4.5% Cu, 0.4–0.8% Mg; modern “2017-T4” adds 0.4–1% Mn, 0.2–0.8% Si) were at high temperatures. The first alloy to satisfy this requirement was the WW I vintage Y alloy (“LM4”; 4% Cu, 2% Ni, 1.5% Mg), which could be used at 250oC. It became popular for high duty pistons. By the Thirties it was replaced with the stronger Hiduminium (RR series) alloys; RR56 (2% Cu, 1.3% Ni, 0.8% Mg, 1.4% Fe, 0.1% Ti, 0.7% Si) for forged cylinder heads, RR59 for forged pistons, and RR53 for cast parts. If compositions for these alloys exist in Grantville Literature, it’s in an old (at least pre-1970) engineering handbook, like Judge’s Engineering Materials (1943). I think this would be a big stretch.
The Marks Handbook (1987) recommends alloy 222.0 (formerly, “122”) for high-temperature piston use, and provides its nominal elemental composition (9.2–10.7% Cu, 2% Si, 1.5% Fe, 0.5% Mn, 0.15–0.35% Mg, 0.5% Ni, 0.8% Zn, 0.25% Ti). Unfortunately, it doesn’t define the heat treatment protocols for the cited tempers (T2, T61, T65, T551) of that alloy, except that T2 means “annealed.”
It should be noted that just knowing the temper designation isn’t enough. For example, “T4” means solution heat-treat and then naturally age. But it doesn’t tell you the heat treat temperature, or the soak time, which can vary depending on which alloy is receiving the T4 temper. Likewise, for precipitation heat treatment, the temper doesn’t tell you the exact temperature or aging time.
And for every alloy, there are going to be other little process refinements that we have to learn by trial and error. For example, a 1922 article says, “In preparing the Y alloy certain precautions are necessary. The magnesium must be added just before pouring and it is best to add it in the form of fairly large pieces of pure metal which should be immediately pushed below the surface of the aluminum so they can dissolve coming in contact with the air. If added at an earlier stage of the melting operation, a considerable proportion of the magnesium is likely to be lost . . . .”
We still have the problem of low wear resistance, and thus gray cast iron is favored for large bore engines. Another problem with aluminum is its high coefficient of thermal expansion, which creates problems with achieving a good piston-cylinder fit under a wide range of operating conditions. This led to adoption of high silicon alloys of aluminum; unfortunately these are more brittle and more difficult to machine. (Wikipedia/Hypereutectic Pistons).
In 1998, aluminum-beryllium pistons enjoyed a brief vogue in Formula One racing; come 2000, metal matrix composites (aluminum-ceramic) made their appearance. (Bamsey).
The crankshaft has a complex shape, and requires both strength and fatigue resistance. It is typically cast iron, cast steel or forged steel, depending on stress. The piston rods and connecting rods also require strength; steel is favored but wrought aluminum isn’t unheard of.
Titanium connecting rods are found in high-end sports cars, racers, and aircraft.
For the intake manifold, we want corrosion resistance. The exhaust manifold is exposed to higher temperatures and so we must consider not only the effect of temperature on strength but also on corrosion. While steels were traditional, in recent years reinforced plastic (Nylon 66) and even magnesium manifolds have appeared. (Wan).
Now, let’s pull back and look at the big picture. Of course, there are no guarantees that history will repeat itself . . . .
Aluminum adoption came early. Looking at aircraft engines, the Wright Brother’s rather anemic (12 hp) 1903 engine had an aluminum (alloy #12, 92% Al, 8% Cu) crankshaft and gray cast iron cylinders. ( Taylor ). The 1916 Hispano-Suiza water-cooled V-6 had an aluminum cylinder head with steel cylinder sleeves. (MoAS). The 825 pound Liberty engine used 225 pounds (27%) aluminum; if this were replaced with cast iron, the engine weight would have been 1160 pounds. (Jehle).
Early automobiles also made significant use of aluminum. In 1911, a four cylinder Ford “Super T” with aluminum block and pistons was used in racing. ( Bryan 125) The 1916 Marmon Model 34B Speedster was entirely aluminum over ash and this may be the first commercial aluminum engine. (Adler 153). However, aluminum (except for pistons) has received only limited acceptance by the automotive industry, partly because of price and partly because of durability issues (overheating, gasket failure—Vartabedian). Even in 2000, cast iron was the primary material for automotive IC engines.
Replacing the heaviest cast iron or steel parts with an expensive light metal has the downside that it substantially increases costs, too. That, apparently, is why aluminum blocks were slow catching on. (Wan).
The other light metals were of minimal importance for engine design as of RoF. The air-cooled VW Beetle did have a magnesium crankcase, but pre-RoF use was primarily outside the engine: “Mg does not have the same creep resistance as does Al, and therefore it is unlikely to be used for the two most massive and critical housings in the engine: the block and the head.”
Magnesium is most likely to be used in vehicles in areas where operating temperatures are not elevated, such as the transmission. (Gaines). Note that the available magnesium casting alloys didn’t maintain strength at temperatures expected during operation. The breakthrough was the development of AMC-SC1 in 2003. It isn’t likely that scientists in the new universe will stumble upon it serendipitously; it contains the rare earth elements lanthanum and cerium. (Nguyen).
Titanium has been used just in small stuff; connecting rods, rocker arms, wrist pins, valves, camshafts, etc. The largest titanium engine component is probably an exhaust.
Propulsion System Controls
An aircraft will have means, manual or automatic, for regulating the fuel-air mixture (by the mixture control, which regulates the flow of fuel into the carburetor) and the engine manifold pressure (by the throttle, which regulates the flow of the combustion mixture into the cylinder).
The fuel/air ratio must be 1:20 to 1:8 by weight (assuming typical aviation gasoline) to even be combustible. The stoichiometric ratio (theoretically resulting in complete combustion) varies with the fuel but is assumed to be 1:14.7. Best power is achieved at around 1:14, and best economy at about 1:18. Rich mixtures are used for takeoff. (Mullender; FTGU 50). At altitude, to compensate for the reduced air density, you reduce the fuel/air ratio, and this results in some reduction in fuel consumption. ( Langley ).
With a fixed pitch propeller, the engine speed is controlled by the throttle, via the engine manifold pressure (the pressure of the fuel-air mixture in the inlet manifold). ( Gardner ). As long as the fuel and fuel-air mixture don’t change, for each throttle setting, there is a corresponding engine speed and output power.
If the aircraft has a variable pitch propeller, the propeller control controls the blade pitch, and thereby the propeller rpm and engine rpm. There are an infinite number of combinations of manifold pressure and engine rpm that result in the same power. (FTGU 66).
If the variable pitch propeller is of the “constant speed” type, and the throttle setting is changed, then the control system adjusts the blade pitch so that the propeller rpm remains constant. However, the engine power output will still change.
I want to emphasize that even though you can find curves plotting power against rpm, you don’t actually set the engine rpm directly, all you can set are the fuel/air ratio, the intake rate, and perhaps the propeller blade pitch.
Propulsion System Monitoring
Engine speed is measured with a tachometer. A mechanical tachometer may be designed so that the rotation causes a pair of weights to swing out, away from the shaft, against a resisting spring; and the angle is related to the centrifugal force on the mass, and thus on the rotational speed. This is very similar to how the governor on a Watt steam engine works, except that the tachometer doesn’t have a control function, and the governor doesn’t report the variation of speed. Bryan Donkin received a gold medal in 1810 for his tachometer, in which the “weight” was mercury in a cup. (Nicholson, Operative Mechanic 39ff).
The powerplant may be equipped with an oil pressure gauge (which measures the pressure of the lubricating oil) and a manifold pressure gauge (which measures the pressure in the engine’s intake manifold). The manifold pressure is a good indication of the “brake mean effective pressure.”
The first pressure gauge was a vertical tube mercury manometer invented by Torricelli in 1643; Huygens introduced the U-tube manometer in 1661. Liquid manometers are not suitable for aeronautic use, so the first pressure gauges of the new time line are likely to be similar to aneroid barometers.
There are several places where it can be desirable to monitor temperatures, including in the lubricating oil pump, at the cylinder head, at the manifold (intake air or air-fuel mixture), and outside the plane.
We know that Jesse’s first plane has a tachometer and an oil pressure gauge. ( Flint, 1633, Chapter 11).
In engine qualification testing, the testers predict the average flight profile (times at various engine settings) and cycle the engine through it, repeatedly. Generally speaking, there’s an expectation that engine life is greater if the power setting is lower than the maximum; higher power is the result of the parts moving at greater speeds and accelerations, and cylinder head temperatures are higher. These of course create higher mechanical and thermal stresses. Even if these don’t cause cracks, there is increased wear. Mathematical models exist in which engine life is assumed to be inversely proportional to the engine load (power setting) raised to an exponent, and the engine life for a particular flight profile is then determined according to the “linear damage rule.” (Johnson 256). Speaking in particular about small gas engines, Dempsey (16) wrote, “No engine survives long at the speed needed to develop full power.”
For airship engines, expect that there will be problems in-flight. The R38 was equipped with six Sunbeam Cossacks; it carried four spares (as well as two spare sets of reduction gear). (Robinson 34). The Hindenburg, with four Maybach engines, carried one spare. (Dick 131).
On the ZR3, an engine shut down, with burnt-out connecting rod bearings, after only 390 hours of operation. (Robinson 145). On the L59, one of the five engines failed the second day of its November 1917 Africa flight, and later that day the remaining engines overheated in the warm air and stalled. (Dick 74). The Hindenburg also experienced engine failures. (Dick 118).
Since heat stresses an engine, it’s important that the designer think about how the engine is cooled. Most auto engines are water-cooled, as were early aircraft engines; most modern light aircraft engines are air-cooled. In essence, water cooling is more efficient but adds to the weight of the powerplant—about 0.6 pounds/hp in 1926 (Wilson). It’s most likely to be seen in aircraft when the engine is of very high power, or high specific power (power/displacement).
Air is a poor conductor of heat, so we have to help it along by adding fins (more surface area for heat transfer) and baffles (to force the air around the cylinders). Air cooling obviously works best when the aircraft (or airship) is in motion. Some aircraft have cowl flaps; these are opened to expose the engine to more ventilation when it’s in danger of overheating, and then closed to reduce drag. Some engines may have fans or pumps to increase cooling airflow.
Historical Airship Engines
This information probably isn’t in Grantville Literature, but the specifications for the historical airship engines (all gas unless otherwise stated) are given below, and may be used as a guide for judging what a new time line engine should be able to achieve:
Table A1: Engines Used in U.S. Airships
|eng wt |
|fuel consum |
Packard 1A-1551, 6 cyl 6.5:1
Maybach VL-1 (1924), 12 cyl 5.3:1*
Maybach VL-2 (1928), 12 cyl 7:1*
“, reduced power
Maybach MbIVa, 6.7:1*
|Liberty, 12 cyl|
Table A2: Additional German Airship Engines
|eng wt |
Daimler LOF 6 Diesel, takeoff power
” cruising power
(1) Robinson 210, Dick 194. (2) Robinson 221 n. 50. (3) Vissering 61. (4) Dick 5 (5) Dick 195.
(6) 2522 pounds/engine, plus 401/engine cooling system, per Burgess.
While this part has given particular consideration to the internal combustion engine as an airship powerplant, canon has already made it clear that other means of propulsion will be put to the test. I will look at steam propulsion in Part Three.
To be continued . . .