top of page


These eclecticisms are taken verbatum or closely paraphrased from motorcycle engineering texts or articles studied over the years. The literature will be accredited in the same manner; see below. It will be my endeavor to update /stories/sightings.html frequently. Starred sentences are followed by annotations.

Motorcycle Engineering; Philip E. Irving Temple Publishing 1961 328 pgs.


Wheels & Frame:


Gyroscopic forces increase with the square of the rim velocity.


The weight of a thin-walled tube increases almost in direct proportion to its diameter, its torsional strenght increases as the cube of the diameter, and its deflection (for any given twisting moment applied) decreases as the fourth power.


Engine Design


*...the rate of pure radiation (as opposed to the transfer of heat to the air in contact with the fins) increases as the 4th power of the absolute temperature. *...when a body is very hot in relation to its surrondings the radiation varies as the difference between the 4th powers of the absolute temperatures concerned. When the difference in temperatures is only a few degrees, Newton's law of cooling applies & the rate of heat-loss is then directly proportional to the temperature difference between the body and its surroundings. Note: Radiation is the transfer of heat by electromagnetic waves (mostly infra red); can occur in vacuum. Radiant heat transfer equals area x emissivity x Stefan- Boltzman constant x absolute temperature to the 4th power. Emissivity varies from 0-1 with 0 being an ideal insulator, and 1 an ideal radiator or black body. Absolute temperature is taken on the Kelvin scale: 0 K = -273.15 C = -459 F.


...the smaller the surface area of the combustion chamber, the better. The intake valve is automatically kept cool by the ingoing charge, to which it must offer the least obstruction, because for every pound difference between the pressure in the cylinder at the start of combustion and that of the surrounding air there will be a reduction of 1/5 in power output. On the exhaust side, the bulk of the burnt gas is expelled by its own pressure, and a pound or so per square inch on the exhaust stroke will reduce the power by only 1 percent. metal surfaces dissipate heat in two ways--by radiation and by convection. In the radiation process, heat is transmitted through the surrounding air without warming it up.

The rate of emission from a polished surface is ~ 1/10 that from the same surface covered with a thin film of lamp-black, and the emissivity of a dull cast-aluminum surface is increased about 10% by a thin coat of black paint.


Radiation takes place only at right angles to the surface.


Under normal conditions radiation accounts for only 1/6 to 1/10 of the total heat loss & cannot be greatly increased by the addition of fins, whose only effect is to increase the effective radiation area from that of the base head and barrel to that of the general outside shape enclosed by the fin tips.


*Convection consists of the direct transference of heat to the layer of air in immediate contact with the surface; this layer then expands, becomes less dense and, under natural conditions, rises & is replaced by cold air which is heated in its turn. Note: Convective heat transfer is very complex, that is, no simple equation describes it. Experimental findings are: 1. the heat current due to convection is directly proportional to the surface area--hence the reason for large surface areas of radiators (convectors) & cooling fins; 2. the viscosity of fluids slows natural convection near a stationary surface, giving a surface film that on a vertical surface typically has about the same insulating value as 1.3 cm of plywood (R value = 0.7)--forced convection decreases the thickness of this film, thus increasing the rate of heat transfer; 3. the heat current due to convection is found to be approximately proportional to the 5/4 power of the temperature difference between the surface & the main body of fluid. Air is an extremely poor conductor of heat, its resistance being some 3,500 times that of iron.


The greater the surface in contact with air, the greater the heat-dissipation will be, provided that the air movement is sufficient.


...the fins on a vertical barrel should be horizontal...V-twins & singles with sloping cylinders that form an angle of inclination of up to 25 degrees from the vertical detracts little, if at all, from the effectiveness of the fins...directionof the head fins should be either horizontal, or vertical & parallel to the center-plane. They should be disposed in the manner best calculated to allow moving air to penetrate right down to the fin-roots & to eliminate pockets of stagnant air which, if left undistrubed, act as insulators...a bank of vertical fore & aft fins situated between valves--bad in two repects--ventilation is bound to be poor and deeping the fins makes matters worse; differential expansion of the hot head-metal between the valve seats and the cooler fin tips sets up stresses which have been known to create valve-seat distortion and subsequent gas leakage.


*Aluminum has 2/5 the density cast-iron but with double the thermal conductivity. Note: Conductive heat transfer occurs within or between two bodies in contact. Conductive heat current = kA (Th-Tc/L); where k is thermal conductivity, A is the cross-sectional area perpendicular to the direction of transfer, Th is the Kelvin temperature of the hot region, Tc the Kelvin temperature of the cool region, and L the length in meters of the heat transfer-path. R value = L/k, is also known as thermal resistance.


...a matter of vital importance in valve gear is weight reduction, since at high speeds the inertial forces generated will amount to several hundred times the weight of the componets concerned.

A push-rod is, in effect, a long slender column loaded in compression & it has a tendency to fail by buckling which increases as the square of its length.


For good valve-action is to be obtained...the axis of the rocker-bearing must lie in a plane which is square to the axis of the valve, & it is difficult to adhere to this conditions if the valves have an included angle of 70 degrees or more, without extreme amount of splay in the push-rods...splay cross-loads the rocker bearings.


...the effective weight of a rocker can be taken as being 1/3 of its actual weight, as the heavey central portion moves at a slower speed than the ends.


Rotational forces on big-end bearings have the effect of causing the rollers to press outwards and, if no cage is fitted, the inner-roller pressure is cumulative and the *outermost pair rub against each other with a severity which increases as the square of the crankshaft (pin) speed. *outermost pair--can be pictured as the two bearings directly in line with the central axis of the con-rod at TDC.


The connecting rod can be considered as swinging about the gudgeon (wrist) pin, its motion being opposite to that of the crankpin at top-dead-center, but in the same direction at bottom-dead-center. Since the maximum angular motion of the rod occurs at these two points, and with usual proportions of rod-length to crank-radius is equal to 1/4 the crankshaft speed, the actual instantaneous rotational speed of the big-end at say 8,000 crank rpm is 10,000 rpm at TDC & 6,000 at BDC.


It is a well established fact that if any caged ball or roller bearing fitted to a high-speed spindle is run full of oil, it will become very hot through the churning action set-up in the lubricant.


If the oil-supply is accidently cut off, the bearing is momentarily still full of oil, which rapidly heats up due to frictional heat plus churning heat. At around 300 degrees Celsius (depending on its composition) the oil will start to vaporize off, at 400C the rollers will commence to bond themselves to the aluminum cage, after which complete seizure follows immediately.

Steel bearing silver-plate the cage, as this treatment (used, on some aero-engine components) is excellent for preventing seizure between two steel surfaces.


In the early days of the dry-sump system, trouble with faulty scavenging was encountered due to the pump having to handle a great mass of froth instead of liquid.


If circulated at a rate of 30-40 gallons per hour through external pipes to a dark-colored tank, oil can be used to remove a lot of heat & can exercise a great effect in keeping the piston cool.


While oil is good at collecting heat, it is very bad at getting rid of it again, because the layer directly in contact with a cooling surface incresaes in viscosity & simply stays there, acting as an insulator & effectively prevents heat being dissipated from the hotter oil in the interior...sump ribs placed on areas against which hot oil is violently thrown by rotational action can be made to radiate a lot of heat...polished cases cut down the heat-radiating ability to a fraction of what it would be if the metal wer left "as cast".


Cylinder base studs, especially in a high-compression engine, should be secured into the metal for a distance at least twice their diameter, and the bossed should not be over-hung from the parent metal.


The run-out at the end of a thread cut with a self-opening die-head exercises a powerful wedging action if the stud is tightened on this portion, and the boss may split


Metal Coefficient of Thermal Expansion Comment


Magnesium 0.000028" per Celsius degree low-silicon Al alloy 0.000022" per Celsius degree w/ Al alloy, when cold a 2.5 in. dia. bearing must be fitted 0.003"-0.004" tight to prevent loosening at operating temperatures


Bronze 0.000018" per Celsius degree because of its expansion bronze can be used as oil-feed


Steel 0.000012" per Celsius degree


Crankcases...should always have the bolt-bosses blended well into the casting walls with generous fillets, and extending if possible to the full width available. Cracks are very likely to develop adjacent to the holes if the bosses are narrower than the main walls and thus place the neighboring metal in a combined state of stress-shear, bending, & tensile loads. engines are much more sensitive to ignition timing than they are to valve timing.


From the viewpoint of thermal efficiency, it would be best if ignition started at t.d.c. and was completed in one or two dergees; but this is not possible or, even desirable because the pressure-rise would be so rapid that extremely rough running would result. It is necesssry, therefore, on this count alone, to fire the charge many degrees before t.d.c., so that the pressure can build up fairly gradually to a maximum of 3-4 times the compression pressure a little after t.d.c., when the piston has started to descend. The high pressure so generated before t.d.c. then constitutes a waste of power and, if the ignition-timing is too far advanced, the useful power developed will be reduced, even though the engine may not give any audible sign of distress by "knocking" or "pinking".


The actual amount of advance is determined most importantly by: the shape of the combustion chamber, the compression ratio, and the analysis of the fuel. Other factors are the amount of charge pre-heating (or cooling if alcohol is used), the location of hot areas, and efficiency of scavenging of combustion products. In general, anything which raises the compression pressure and/or temperature tends toward the need for less advance, anything which has the opposite effect tends toward more advance...the final setting can only be obtained by testing...the higher the power per c.c., the more accurate must be the timing.


Engine starting the engine will kick back at full advance, and when pulling hard at slow or medium speed less advance is required than at open throttle at high speed, because the mixture has a longer time in which to burn completely. [Mechanical] Automatic Timing Devices (ATD) need to operate only from 0 to 2,500 or 3,000 r.p.m. [I have found it necessary to bring the r.p.m. to nearly 6,000 to fully advance a /2 BMW. This is, however, in agreement with the figure given in the factory service manual.] From then on to maximum revs in a normal, well-designed engine, the advance appears to be independent of speed, because the higher this is, the greater the mixture turbulence becomes, and the rate of combustion automatically speeds-up in unison with an increase in r.p.m.--a fact which permits some racing engines to be run with fixed ignition timing...for touring, the automatic advance is superior to manual...and has the additional merit that the spark intensity is constant whatever the amount of advance. It would still be better with a vacuum control operated by manifold pressure, but this device cannot be applied to a magneto and does not work well with a single cylinder's variable suction when applied to coil ignition.


Mixture strength...between the plug points, which may be 0.018" apart--must be within the limits of 16:1 to 12:1 by weight if gasoline is the fuel, or about 8:1 if alcohol is used...this applies to the amount of fuel present in the form of vapor. If a flooded engine does not start at once, it will be reluctant to do so at all, because the previously unvaporized fractions have been given time to vaporize and the mixture the becomes too rich to maintain combustion, even if it is initated by the spark. ...generally speaking, the plug is best placed at the apex of the angle made by the offset intake and exhaust ports, rather than in the space between them. ...excessively domed pistons, especially in small bores...the plug is severely masked by the dome just when combustion is occuring...the spread of flame (flame travel) is severely hampered by the piston-crown and power is lost...increasing the advance only compounds the problem...a better idea is to make a local flat or depression in the crown adjacent to the plug or use an asymmetrical crown.


Dual plugs...firing both plugs at the correct instant in relation to each other is also a problem...optimum results are obtained with simultaneous firing, or that one plug should "lead" the other by one or two degrees because of their positions in the combustion chamber, but the advance will be less (often by 5-6 degrees) than that required with a single plug. If the spark at either plug occurs a little too soon, the spark at the other will not do anything much and the engine will then be running with single ingnition at less than optimum advance and consequently at reduced power. ...if absolute synchronization is required, use a double-ended coil or a magneto, such as the rotary-induction Lucas with a divided high-tension coil and two plug leads. ...if a phase difference is necessary two contact breakers must be employed regardless of whether they operate coils or are built into the magneto. ...coil vs. magneto: Ignition wise, the coil gives its highest voltage at low speeds, and at high speeds eventually reaches a point, measured in sparks/minute, when the secondary voltage becomes too low and misfiring sets in. The magneto, owing to the fact that it generates its own primary current which rises with speed, gives a higher voltage at high speeds but this does not confer any real advantage over the coil, so long as the latter is being worked within its rated capacity.


Magneto Ignition: parallel twin four-stroke--the magneto gives two sparks per revolution so all that is needed is a two-lobe breaker cam, a slip-ring, and two high-tension leads, however, considerable precision in manufacture is required to ensure that the sparks occur exactly at 180 degrees. With a V-Twin, it is not possible to use the same basic idea just with the cam lobes altered in angular position to suit the cylinder angle, because the sparks produced are only at maximum intensity over narrow angles. If the lobe positions are altered much from 180 degrees, one or the other of the two sparks would be ineffective; to cure this, the pole-pieces are "staggered", an expedient which produces sparks of nearly equal intensity at one cam-ring position, but at any other position one spark becomes weaker than the other. It still may be good enough to fire the charge, but generally speaking a V-Twin magneto is not good with manual advance, especially starting from cold when one needs the most retard and two best possible sparks. For that reason, better starting is obtained with an automatic advance mechanism which only alters the position of the contact-breaker with relation to the crankshaft, but not the contact-breaker in relation to the pole-pieces, as the manual advance does.

Petrol, in the form of vapour, combines completely with air when in the proportions o f 15 parts of air to 1 part of petrol--by weight, not by volume.

bottom of page