by Chirag Asaravala. Drawings adapted from Crane Cams.

By the time you get done reading this tech article you may be disappointed with your conventional roller rocker arms. Don't be too discouraged though, a good set of stud mount roller rockers are fine for most applications. However, if you've got a competition oriented motor with super high spring rates, a .600" lift or greater solid roller cam, and power that goes from 5000 on up, then the information presented here is precisely for you. Most power mongers are well aware of the hierarchy of rocker arm technology. Roller-tip arms are better than stock sled style. Full rollers are even better and the most common choice amongst enthusiasts. Shaft mount rockers are the cream of the crop and often selected by the most serious of racers. However, not all shaft rockers are the same and whether you are in the market for shaft rockers or not, you'll find the latest technology to be eye opening.

The challenge with pushrod engines is overcoming the limitations in valve speed due to the inefficiencies of the mechanical linkage that is the lifter, pushrod, and rocker arm. Overhead camshaft (OHC) engines have set the benchmark. By locating the camshaft over the valves, the valves respond directly to the cam lobe profile. As a result, valve speed is much quicker and lobe profiles can be very aggressive as there are none of the geometrical limitations resulting from lifter or rocker angles. The lifter design in a pushrod engine offers up some challenges. Flat tappet lifters are quick off the base-circle of the camshaft, but the valve lift rate is limited by the profile of the cam lobe, which must be gradual in order to prevent the tappet from digging in. Roller rockers, due to their weight, are slow off the base circle but their infinite contact area against the cam lobe offers the ability to design very aggressive lift rates.

Crane Cams has spent considerable effort in developing technologies which bring pushrod valve trains even closer in performance to the OHC precedent. Their latest shaft rocker systems for small block Fords incorporate two major advances in rocker arm ratio and bearing design which result in significant power and durability over conventional roller rockers. Let's take a detailed look at how these two enhancements work.

Rocker Arms: Translation versus Multiplication
The most basic function of a rocker arm is to translate the motion of the camshaft to the valve. Early engineers were content to leave it at that, disregarding the notion that a rocker arm could also serve to enhance valve activity. Crane's founder, Harvey Crane, was also of the opinion that the rocker arm should only be translation device, and all valve timing activity is ground into the cam. This thinking didn't change in the automotive industry until well after WWII, when automotive engineering began to pick up again. Many of the advancements in rocker arm technology, including the two presented here, are the brainchildren of Ralph Johnson, legendary engineer for Smokey Yunick, GM, Holley and Ford. In fact, even Ford guys should appreciate that Ralph Johnson was on the original small block Chevy design team at GM, and it was his father's company that supplied the original stamped steel rocker arms for those engines.

The idea of using a rocker arm as a multiplication device is quite simple. By acting as a lever the mechanical advantage serves to multiply the lift at the cam lobe to increase the overall valve lift or distance the valve opens off the seat. The amount of multiplication is termed the rocker arm ratio. A rocker arm ratio of 1.6:1, for instance, would indicate the arm moves the valve 1.6 times that of the lobe lift at any given point. If the cam specifications indicate a max lobe lift of .350", a 1.6:1 rocker arm would result in the valve moving .560" off the valve seat.

If you look at any cam card you will see listed the maximum valve or lobe lift. We deliberately used that same figure in our example above to set up another important point - rocker arms are not really a fixed ratio. The rocker arm ratio is theoretically calculated by dividing the perpendicular distances from roller-tip centerline to fulcrum and pushrod-cup centerline to fulcrum (see right.) However because these points move in an arc, as the pushrod rises and valve drops, the rocker arm ratio also varies. Early factory rocker arm designs would start off the valve at significantly less ratio than maximum. For instance, a 1.6:1 OEM rocker arm would start off as low at 1.47 and approach 1.6 at max lobe lift. It is likely this gradual multiplication was largely designed to keep the early two piece valves from breaking by slamming closed too quickly.

QuickLift™
The geometry of the rocker arm determines the rate at which a rocker arm ratio is achieved. By changing the location of the pushrod seat relative to the center of rotation of the rocker arm, the rocker ratio can be achieve in one of three way. It can gradually approach the maximum ratio, stay fairly constant, or, as in the case of Crane's QuickLift design, the ratio can start off high and then tapers down to the final ratio.

Tailoring valve performance into the rocker arm offers advantages over utilizing only camshaft lobe profiles. For one, as mentioned earlier, there are limitations presented by the lifter type. While roller lifters offer the ability to use aggressive lobes, there is still a limit to how fast the lifter can move. "In the end, the engine only responds to the valve, it doesn't care whether the added lift, duration or valve acceleration is due to the cam or the rocker arm." says Mark Campbell, Vice President of R&D at Crane Cams. By moving the pushrod seat down further in relation to the center of rotation, the pushrod cup travels upward and outward along its arc. (See right.) The result is very quick valve opening, and at a much higher initial ratio. For example, a 1.60:1 Crane shaft rocker arm with QuickLift technology will come off the seat at 1.72:1 up until .200" valve lift. Then it will taper down to 1.62:1 and maintain this until the valve returns to 0.200" and the ratio is back to 1.72:1 on the seat. This does result in closing the valve harder against the seat, however with aftermarket one piece valves and high quality valvetrain hardware, this is not an issue. The benefit of a quick opening is two fold. First, there is as much as six degrees more duration in the low lift range. Secondly, this is achieved with the same seat-to-seat timing. The intake valve, for instance, is opening more and thus enabling greater cylinder fill, and then closing fast. This is not only measurable with a degree wheel and dial indicator but also on the dyno, where roller rockers with QuickLift put out 15-18 more horsepower than rocker arms which works up towards the advertised ratio at maximum lift.

Bearing Design
Conventional roller rocker arms have two predominant bearing designs, caged bearings and "full complement" roller bearings. Both designs are subject to two factors; internal friction and bearing inertia. In a caged bearing design the friction is between the roller and the cage. In a full complement bearing, where there is no cage, the needles move in the same direction but the contacting surfaces turn opposite of each other, creating friction. Friction increases with bearing speed and is also proportional to spring load. The stiffer the valve springs and the higher you spin the motor, the greater the resulting friction and heat, in the oil. In fact, compared to the Polymer Composite Matrix bearing design, Crane Cams has measured a full 100°F temperature increase in the oil around needle-bearing rocker arms.

Bearing inertia is the other factor that robs power and generates tremendous friction at high engine speeds. Because a rocker arm does not rotate a full 360 degrees, but rather oscillates in a limited motion, the bearings must turn one direction as the valve is opening, then stop and reverse direction as the valve is closing. This not only contributes to the friction and parasitic power loss, but results in compromised durability. Bearings are best suited when the entire surface area wears evenly. Due to the limited rotation of a roller rocker the bearings wear unevenly. This results in an oval shape which can lead to rocker arm failure. A full set of 16 V-8 rocker arms has 552 needle bearings, so it is easy to appreciate the magnitude of the problem.

Polymer-Matrix Composite Bearings
Crane's polymer-matrix composite (PMC) bearing prototype was actually first run back in 1992 at Daytona in Sterling Marlins car. While even today Nascar teams change out Jesel shaft rockers every race or two, the PMC bearings showed unprecedented durability, running 15,000 race miles with minimal wear. The PMC bearing resembles a simple bushing. However its construction is quite sophisticated. The steel ring consists of a bronze overlay. The bronze surface serves to anchor the proprietary polymer-matrix compound. Since the bearing itself is relatively thin compared to a roller needle or caged bearing, the shaft which it rides on can be proportionally larger. In this case it is 5/8", which is as much as 3/16" larger than traditional shaft diameters. The larger shaft results in greater rigidity and virtually eliminates deflection of the rocker arm.

Bearing Designs From left to right, PMC bearing, full-complement roller bearings, and caged roller bearings. The thin walled PMC bearing enables a larger shaft and eliminates point loading since the load is spread across the full circumference of the shaft. The full complement needle bearings generate high internal friction and significant point loading on the shaft. Roller rockers with caged bearings create the most load due to the least surface area contacting the shaft.

The clear advantage to the PMC bearing is in its elimination of internal friction and bearing inertia (see above). There is no component within the bearing which must stop and reverse direction. Furthermore, the load on the shaft is spread evenly around the circumference of the bearing, whereas in a roller bearing design, loading occurs at points around the shaft. Point loading increases shaft wear and friction.