1966 Mainly to produce more power to maintain performance at the new higher weights of bikes with electric start, rear suspension, etc., the Shovelhead engine is introduced, produced 1966-85. Sales increase 26 percent, to 36,310 units.
First-year Shovelheads were given aluminum versions of Sportster iron heads (long fore-and-aft fins, straighter ports giving 10 percent more power—the so-called “Power-Pac.” These heads have less deep combustion chambers and their valve included angle drops to 78.5 degrees (intake is 40.25 degrees from centerline, exhaust is 38.25 degrees). The importance of this is that the shallower the combustion chamber, the less surface area it has through which combustion heat can enter the head. Also, as compression ratio rose from the 6.0:1 range of the original “E” to higher numbers such as 8.0:1, piston domes had to grow taller, both interfering with combustion and gaining in crown surface area, causing pistons to run hotter. The shallower chamber helped cooling and worked better at higher compression ratios than did the old 90-degree valve-angle chamber.
Rocker arms pivots were now made as part of the bolted-on rocker-box casting. The 1966-68 models still had generators rather than alternators. Linkert DC carb (M has round floatbowl, DC has square-ish bowl), 8.0:1 compression. All-chain drive, separate four-speed gearbox. Iron Sportster heads, however, had quite long exhaust ports. Front cylinder’s port turns about 107 degrees and the rear about 80 degrees in this casting. With regard to these turns in the head casting, recall that the late Kenny Augustine (who updated the ports on Harley’s eternal XR-750 dirt-track engine), said, “You can make air go fast, and you can make it go around corners, but you can’t do both at the same time.” Valve sizes are 1.94-inch intake/1.75-inch exhaust. The flow value of straight ports was still of lower importance than convenience in carburetor and exhaust pipe location.
Oil could pool in shovelheads rather than return to the crankcase, then leak past the valve guides to produce exhaust smoking. Cylinders had 10 fins each.
Something to think about here is what is known as “wet-sumping.” If for any reason oil thrown off moving parts accumulates in the crankcase rather than being promptly picked up by the scavenge pump and returned to the oil tank, that oil can cause all kinds of problems. On the flathead KR race bike, Dick O’Brien once said, “When it’d start wet-sumping on the Daytona banking, you could see it. It was like a big invisible hand came down and just held the bike back. Slowed right down.”
The un-returned oil, being violently sheared between spinning crank and crankcase, acted like a brake, and oil temperature shot up as speed came down. Wet-sumping is a potential problem in any engine that, like a Harley-Davidson, has a close fit between crank wheels and case. Many a manufacturer has in the past reduced flywheel diameter in hope of controlling this problem.
1969 Harley-Davidson seeks to solve its financial problems by merger with AMF. Generator output was marginal. Tillotson D carburetor (this is a floatless carburetor whose fuel pressure was regulated by a polymer/fabric diaphragm).
Read about the history of Harley-Davidson's Knucklehead V-twin here:
1970 Kick start is deleted on FL. Alternators from here onward located on the left-hand crankcase, concentric with the crank, behind the primary drive in a redesigned case. The external “timer” is gone – new cone-style timing cover. (“cone motors” came after “generator shovels”). A single bolt retains header pipe to port instead of the previous, always loosening threads in the head. Tillotson carb.
1971 Zenith Bendix carb. Extending the car industry’s “factory hot-rod” concept, Willie G. Davidson now began to offer an approach to a “factory chopper”: the FX-1200 SuperGlide. Style and individuality for sale. This in effect co-opted the chopper movement but brought the look to a wider audience.
1972 Hand-shift option is dropped.
1973 Disc brake on FX, XL 1973-91. Oil pump is still twin gear type.
1974 The first fuel crisis hits, and octane number and consistency of gasolines drops. The resulting chronic engine knock caused overheating. With cylinder heads running hotter again, their aluminum material expanded more, scrubbing gaskets into failure and stretching and loosening hold-down bolts. The natural result was leakage and even outright gasket failure. None of this was good for Harley-Davidson’s decades-old reputation.
Crankshaft weight is 35.5 pounds, flywheels are 8.5 inches in diameter, 4 inches wide. Connecting rods have 7.4375-inch eye-to-eye length for a rod ratio of 1.87. Rod big-ends have separate, hardened, pressed-in steel races, allowing five honed oversizes (with oversize rollers).
1976 AMF understood profit and loss but not the engineering need to keep a product equal to its operating conditions. Crank out more units! Production is 48,000, but this was a time of production speed-up, declining quality, and poor employee relations. As early as this year, management knew it needed a new, more capable engine to recoup their declining reputation.
In the US auto industry at this time it was considered more desirable to keep production lines rolling and correct any quality problems later through “rework”—in effect, manufacturing the product twice. This worked in the postwar boom years, but as competition from Europe and Asia developed, it became a serious handicap.
Read about the histroy of Harley-Davidson's Panhead V-twin here:
1978 In response to emissions-driven mixture lean-out (no more cooling effect from running a bit rich) and weight growth, the Shovelhead increases displacement with 3.5 x 4.25 inches = 81.8 cu. in. Since the beginning of time, air-cooled engines had been run a bit rich to help with cooling. When this source of cooling was denied by regulation, and riders had to open throttles more to accelerate, engines ran hotter, often being driven into detonation.
Why worry about engines running hot? Aren’t they “heat engines,” after all? The hotter an engine becomes, the more density its intake air loses by the time it gets into the cylinders. Torque drops in direct proportion to air density loss. Race tuners knew that while a best-power mixture gave more power for a lap or two, as soon as an engine reached operating temperature this heat/density effect would cause the loss of several horsepower. Betting in the long run, then, to run a bit rich, losing less power to overheating. Overheating at low speed can lead to cylinder distortion (wind cools the front of the cylinder more than it does the rear), poor ring seal, smoking, and elevated oil consumption.
Shovelhead cylinders are still cast-iron, whose poor heat conductivity led to higher piston temperature. The smaller piston clearance can safely be made, the better the cooling effect of piston-to-cylinder contact, so now expansion-reducing steel struts were cast into the pistons.
Intake manifold formerly had a small-lip O-ring seal, now replaced by a flat band with no lip. Breakerless electronic ignition. Compression is 8.0:1, claimed output 65 horsepower. The 80-inch had smoother, easier power—not more power. This was the era of bad and variable gasoline, as mandated reductions in use of lead antiknock additive use reduced fuel octane number. From 1977-80 there were possible detonation and overheating, valve and guide problems. Compression was now excessive for engine operating temp and declining fuel quality. A warranty study conducted in summer 1980 led to plans to design a next-series engine.
As the Shovelhead design neared the end of its run, the factory produced countermeasure parts intended to control oil consumption, and led to increased piston ring and valve life. Pistons were given sophisticated shape that assumed the right contour when hot. The shovelhead combustion chamber worked best on the higher-octane gasolines of the period before lead reduction (pre-1977). Engines could tolerate higher-than-stock compression ratio in those days, but not once lead abatement began. In this period Jerry Branch began offering installation of hard (more tolerant of low-lead fuel) valve seats. Lead in gasoline had had the side benefit of acting as an anti-seize between valve and seat, preventing micro-welding and plucking as valves opened and closed.
Usual crankshaft construction since early days was five-piece: two cast-iron flywheels, joined by a taper-and-nut-retained steel crankpin, and with separate taper-and-nut steel mainshafts. Every manufacturer using cast-iron flywheels, whose users ran their bikes hard, had experience of mainshafts and crankpins loosening in the fairly soft iron flywheels. The XR-750 dirt-track racing engine in about 1971 was given steel flywheels, each forged in one piece with its mainshaft—the strongest construction short of switching to a one-piece crankshaft (which is just what old-time drag racer the late Clem Johnson did when he couldn’t stop his Vincent dragster’s flywheels from scissoring at 7500 rpm).
A common figure for crank balance is 60 percent (crank counterweights balance 100 percent of rotating weight, but only 60 percent of reciprocating weight). Automotive balancers sometimes use the 52 percent that is usual for V8 engines. The reason for using more than 52 percent in bike engines is that rider and passenger feel up-and-down vibration strongly, so it makes sense to move some of that shaking into the fore-and-aft direction by over-balancing.
1980 Belt final drive is adopted. The FLT swingarm bolts to the new five-speed transmission. There is a spin-on oil filter. Rubber-mounted engine and new five-speed gearbox (FLT) are rigidly bolted together. Reverse gear option deleted (previously you could choose either a 4-speed or 3-speed-plus-reverse, for sidecar). A tooth-belt primary drive was produced for a time. The vibration-absorbing and crank-mounted “compensator,” with rubber block compliant members, could deteriorate with heat, requiring the blocks to be replaced.
1981 The Motor Company buys itself back from AMF. Larger alternator output. Exhaust cross-over pipe adopted. A new oil pump, delivering one-third more oil than the original Shovelhead cast-iron pump, is adopted. As part of immediate damage control, compression ratio is reduced from 8.0:1 to 7.4:1 to prevent detonation on the new lower-quality gasolines. Improved and longer valve guides, Kayline seals, and extra oil drain lines from the rocker boxes were released in “oil control kits.” It was essential for survival to recover the company’s good name with a fresh design.
1982 Much has been said about Harley-Davidson’s revolutionary adoption of modern manufacturing concepts such as just-in-time (or “MAN”—Materials As Needed) and statistical process control, but the fact is that this was desperate self-defense. In this year, Tom Gelb’s savings in inventory cost through just-in-time manufacturing saved the company enough to offset the year’s operating loss. In traditional manufacturing, thousands of necessary parts and assemblies are ordered ahead of time and warehoused until needed, tying up both capital and floorspace. If the design of related parts is changed in the meantime, the warehoused parts may no longer fit, requiring very expensive emergency changes (something familiar in Detroit). But with just-in-time assembly, parts are delivered as-needed on a daily basis by suppliers. This takes cooperation and planning but saves real money.
People have called such methods “Japanese” because they were enthusiastically adopted in Japan after World War II, often in consultation with Dr. W. Edwards Deming. In fact, such methods were common in US industry during the war, but were set aside as “unnecessary” during postwar US prosperity. Statistical process control, for example, was worked out by Dr. Walter Shewhart at Bell Lab in the 1920s—in the good old USA.
Harley-Davidson used these methods to such powerful effect that they became the leading example of their success in US industry and taught others to use them.
A Y-manifold was necessary to allow single-carburetor operation, but there had always been problems with sealing to the two cylinders. In an attempt to make a permanent seal, builders would assemble the heads onto the cylinders with fasteners finger-tight, then tighten the manifold bolts, and only after that was done, torque the head fasteners. Why didn’t this work as hoped? Engines flex in operation; the cylinders of giant radial aircraft engines tilted sideways .020 inch on their power strokes, tearing up anything more rigid than flexible manifold connections.
A 32mm Keihin butterfly carb is fitted on Super Glide. Weight 620 pounds, wet.
A 1982 shovelhead was dynoed by Cycle magazine to the tune of 51 horsepower at 5000 rpm, peak torque 62 lb.-ft. at 4000.