A pushrod is a slender rigid rod that transmits axial compressive force between two pivoting or sliding elements — most often between a cam follower and a rocker arm in an engine, or between a control horn and a servo in an aircraft or RC linkage. It solves the problem of moving force across a distance where a direct lever or shaft will not fit. The rod takes load only along its length, so geometry stays simple and parts count stays low. You see it in every Chevrolet small-block V8 valve train and in the elevator linkage of a Cessna 172.
Operating Principle of the Pushrod
A pushrod works on one principle: a straight rod loaded along its axis is stiff and light. Push one end, the other end moves the same distance, minus a tiny amount of elastic compression. In an OHV engine pushrod the cam lifts the lifter, the lifter pushes the rod up, the rod pushes the rocker arm, and the rocker pivots to open the valve. In an RC aircraft control linkage the servo arm pushes one end of the pushrod, the rod transmits that motion straight to the control horn on the elevator or rudder. Same physics, different scale.
The rod itself does almost no work — it just carries load. The geometry around it does the work. That's why pushrod design is mostly about two things: keeping the rod from buckling under peak compressive load, and keeping the ball-and-socket ends aligned so the rod stays in pure axial loading. If a 7.40 mm OD steel pushrod in a 6,500 RPM small-block sees side load because the rocker arm geometry is off, you get pushrod deflection, accelerated tip wear, and eventually a bent rod sitting in the lifter valley. Euler buckling is the failure mode you size against — the longer and thinner the rod, the lower the critical load before it folds.
If the pushrod length is wrong by even 0.030 inches in a high-RPM engine, the rocker arm geometry shifts, the valve tip wear pattern goes off-centre, and lifter preload moves outside spec. Builders use adjustable-length checking pushrods to find the exact length before ordering the final set. On the RC side, a control linkage with too much slop in the clevis ends gives you flutter on the elevator and a sloppy stick feel — the rod is doing its job, but the ends are not. Tolerance at the ends matters more than tolerance in the middle.
Key Components
- Rod shaft: The straight section that carries axial load. Diameter typically 5/16" to 7/16" in automotive applications, 2 mm to 4 mm in RC. Material is chrome-moly 4130 steel for engines, music wire or carbon fibre for RC. Wall thickness on tubular pushrods is usually 0.080" to 0.135" — thicker walls add stiffness against buckling without much weight penalty.
- Ball end (lifter side): Hardened steel ball, typically 5/16" radius, that seats in the lifter cup. Hardness should be 58-62 HRC. A soft ball will brinell into the cup within a few thousand miles and the rod will run noisy and short.
- Cup end (rocker side): Concave seat that mates with the rocker arm adjuster ball. Surface finish must be Ra 0.4 µm or better — a rough cup chews through the adjuster ball and dumps metal into the oil.
- Clevis or threaded end (linkage applications): On RC and full-scale aircraft control linkages, the rod terminates in a threaded clevis or ball link. Thread engagement should be at least 8 turns for safety. Lock nuts are mandatory — vibration backs out unsecured threads in flight hours, not weeks.
- Guide plate or tube (engine applications): On non-rail-rocker engines, a stamped steel guide plate keeps the upper end of the pushrod centred. Slot width typically 0.020" larger than rod OD. Without it, side load from the rocker tip walking off-centre will bend the rod.
Who Uses the Pushrod
Pushrods show up anywhere you need to transmit linear force between two pivoting points along a straight line. The geometry is simple, the parts are cheap, and when sized right the failure rate is near zero over millions of cycles. Below are five real-world applications where the pushrod is the right answer.
- Automotive engines: Chevrolet LS3 6.2L V8 — uses 7.40" long, 5/16" OD chrome-moly hardened pushrods to drive the rocker arms from the in-block camshaft.
- RC aircraft: E-flite Apprentice STS trainer — 2 mm carbon fibre pushrods with threaded steel clevis ends connect the elevator and rudder servos to their control horns.
- General aviation: Lycoming O-360 flat-four aircraft engine — aluminium pushrods inside enclosed steel tubes drive the overhead valves on each cylinder.
- Motorcycles: Harley-Davidson Milwaukee-Eight 114 — adjustable-length pushrods sit in chromed external tubes, allowing valve lash to be set without removing the heads.
- Industrial diesel: Cummins 5.9L 12-valve B-series engine — heavy-duty 3/8" OD pushrods drive the valve train at sustained loads on commercial trucks and irrigation pumpsets.
- Tractor and ag equipment: John Deere 4020 tractor — original 4.020"-bore inline-six uses solid steel pushrods rated for the engine's 6,500-hour rebuild interval.
The Formula Behind the Pushrod
The single most important calculation for any pushrod is the critical buckling load — the axial force at which the rod folds in the middle instead of transmitting force cleanly to the rocker. At the low end of the typical operating range (low-RPM, low-spring-pressure engines or light RC linkages) the actual load is well below critical and you can run a thin rod safely. At the high end (high-RPM race engines with heavy valve springs) peak compressive load can spike to 4-5× the static spring load due to inertia, and a rod sized only for static load will buckle. The sweet spot is sizing so peak dynamic load stays below 40% of the Euler critical buckling load — that gives you safety margin for wear, side load, and the occasional valve float event.
Variables
| Symbol | Meaning | Unit (SI) | Unit (Imperial) |
|---|---|---|---|
| Pcr | Critical buckling load — axial compressive force at which the rod becomes unstable | N | lbf |
| E | Young's modulus of the rod material (≈ 200 GPa for steel, 70 GPa for aluminium, 130 GPa for carbon fibre) | Pa | psi |
| I | Second moment of area of the rod cross-section. For a solid round rod, I = π × d<sup>4</sup> / 64. For a tube, I = π × (D<sup>4</sup> − d<sup>4</sup>) / 64. | m<sup>4</sup> | in<sup>4</sup> |
| K | Effective length factor — 1.0 for pinned-pinned ends (ball-and-cup pushrods), 0.5 for fixed-fixed | dimensionless | dimensionless |
| L | Free length of the rod between supports | m | in |
Worked Example: Pushrod in a Chevrolet LS-based race engine build
A circle-track engine builder in Mooresville is sizing pushrods for a 427-cubic-inch LS-based race engine that will see sustained 7,800 RPM operation. The valve springs produce 650 lbf at peak lift. The pushrod is 7.400" long, 5/16" OD (0.3125"), 0.080" wall, made from 4130 chrome-moly with E = 30 × 10<sup>6</sup> psi. The builder wants to confirm the rod has enough margin against buckling across the engine's operating range from 3,000 RPM (warm-up) to 8,200 RPM (over-rev event).
Given
- L = 7.400 in
- DOD = 0.3125 in
- twall = 0.080 in
- E = 30 × 10<sup>6</sup> psi
- K = 1.0 dimensionless
- Fspring,peak = 650 lbf
Solution
Step 1 — calculate the second moment of area I for the tube. ID = 0.3125 − (2 × 0.080) = 0.1525".
Step 2 — calculate the Euler critical buckling load at nominal geometry:
Step 3 — compare against actual peak dynamic load across the RPM range. At 3,000 RPM (low end) inertial load on the valve train is small, peak compressive force on the rod is roughly 1.2 × spring load = 780 lbf. The rod is loaded to 33% of Pcr — comfortable, no side-load risk.
Step 4 — at nominal 7,800 RPM, valve train acceleration peaks during the closing ramp. Empirical multiplier for an LS-style train at this RPM is 2.8×.
That's getting tight. Industry practice is to keep peak dynamic load under 40% of Pcr for long-haul race use. At 76% you are inside Euler's elastic limit but you have no margin for wear, oil-temperature softening of E, or a valve float event.
Step 5 — at the high end, an 8,200 RPM over-rev pushes the inertial multiplier to roughly 3.3×, peak load to 2,145 lbf — 90% of critical. One bent pushrod and the engine eats itself.
Result
Nominal critical buckling load is 2,390 lbf for this 7. 400" × 5/16" 0.080-wall 4130 pushrod. At 3,000 RPM the rod runs at 33% of critical — bulletproof. At 7,800 RPM it sits at 76%, which works for short stints but leaves no safety margin for the inevitable. At 8,200 RPM it climbs to 90% — one missed shift and the rod bends. The builder should step up to 3/8" OD with 0.105" wall, which roughly doubles I and pushes P<sub>cr</sub> past 5,000 lbf. If a measured pushrod failure happens earlier than this calculation predicts, check three things: (1) rocker arm geometry pushing the rod tip off-axis adds bending stress that this Euler calculation does not capture, (2) a worn lifter cup lets the lower ball end shift, increasing effective L, and (3) oil starvation softens the rod tip hardness below 58 HRC, the ball mushrooms, and the rod runs longer than spec — all three move you toward buckling well before the static number says you should be safe.
When to Use a Pushrod and When Not To
Pushrods are not the only way to transmit motion from a cam to a valve or from a servo to a control surface. The two main alternatives are overhead cam (OHC) valve trains in engines and direct-drive servos in RC and robotic linkages. Each has a place — here is how they compare on the dimensions that actually matter when you're choosing.
| Property | Pushrod | Overhead Cam (OHC) | Direct-drive servo / linear actuator |
|---|---|---|---|
| Maximum useful RPM | 8,500 RPM (race), 6,500 RPM typical | 12,000+ RPM | N/A — non-rotating |
| Cost (per valve, automotive) | $8-25 per pushrod | $200+ per cam tower assembly | $15-150 per servo |
| Packaging height (engine) | Compact — cam in block, low deck height | Tall — cam(s) above heads, adds 75-150 mm | N/A |
| Failure mode | Buckling under peak load, tip wear | Belt/chain failure, cam bearing wear | Servo gear strip, electronic failure |
| Reliability (cycles to failure) | 100M+ cycles when sized correctly | 200M+ cycles with proper belt service | 1-10M cycles depending on servo grade |
| Design complexity | Low — 1 cam, simple block geometry | High — multiple cams, drive system, tensioners | Medium — needs control electronics |
| Best fit | Truck engines, V-twins, RC linkages, aircraft controls | High-RPM performance and economy car engines | Robotics, automation, RC control surfaces |
Frequently Asked Questions About Pushrod
Valve train inertia. A pushrod system has more reciprocating mass — lifter, pushrod, rocker arm — that the spring has to control. At RPM above roughly 7,500 the spring runs out of authority to keep the lifter on the cam lobe and you get valve float, where the valve loses contact with the cam profile and bounces.
An OHC engine eliminates the pushrod and lifter from the moving mass, so the spring has less work to do at the same RPM. That's why a 7,000 RPM redline is comfortable on a Chevy small-block but a Honda K20 spins to 8,500 without breathing hard. Fix it on a pushrod engine by going to lighter components — titanium retainers, hollow-stem valves, and thinner-wall pushrods — but you'll never match an OHC layout above 9,000 RPM.
Yes, and here's why. Pushrod length sets where the rocker tip contacts the valve stem. Off by 0.040" and the rocker tip rolls across the valve tip during the lift cycle instead of staying centred. You'll see a wear pattern that walks toward one edge of the valve tip within a few thousand miles, accelerated valve guide wear, and noisy operation.
The fix is an adjustable checking pushrod. Set it to your measured stack and pick a production length that puts the rocker sweep pattern centred on the valve tip — a 0.060"-wide band centred across the tip is what you want to see when you pull the valve cover after a few engine cycles.
Tubular every time, unless cost is the only consideration. A 5/16" OD tubular pushrod with 0.080" wall has roughly 90% of the buckling resistance of a solid rod of the same diameter at about 60% of the weight. Less reciprocating mass means less spring load needed, which means less cam wear and less power lost to the valve train.
Solid pushrods only make sense in two cases — extreme nitrous or boost applications where peak loads exceed 1,800 lbf on a small-OD rod, and budget rebuilds where you're keeping stock springs and stock RPM. For a typical 5,500 RPM street engine, a quality 4130 chrome-moly tubular set in the right length is the correct answer.
Almost certainly slop at the clevis ends, not the rod itself. Plastic clevises and ball links wear at the pin holes after repeated cycling. A new clevis has maybe 0.05 mm of free play at the joint. After 30 flights of vibration and stick input that can grow to 0.3-0.5 mm, which translates to 1-2° of free play at the elevator surface — enough to start aerodynamic flutter at higher airspeeds.
Diagnostic check: hold the servo arm fixed, wiggle the elevator surface by hand. Any perceptible movement before the rod starts pulling on the servo means worn clevises. Replace both ends, not just one. Metal ball links last 5-10× longer than plastic clevises and are worth the upgrade on anything bigger than a park flyer.
Thermal expansion of the block, head, and pushrod, all at different rates. An aluminium head on an iron block expands roughly 0.012" over the head's 4" thickness when going from 70°F to 200°F. The steel pushrod expands maybe 0.005" over 7.4" length. Net effect — the pushrod stack changes by 0.005-0.010" hot versus cold.
That's why valve lash is always set hot on solid-lifter engines, and why hydraulic lifters exist — they self-compensate for that thermal change. If you're checking pushrod length cold with a checking tool, target the upper end of the rocker sweep pattern range, because it will move as the engine warms.
You can, but only within limits. The longer pushrod approach works for compensating up to about 0.030" of head mill. Beyond that, the rocker arm pivot point sits in the wrong place relative to the valve tip arc and you lose effective rocker ratio — a 1.7:1 rocker can drop to an effective 1.65:1, which costs lift and changes the cam's effective profile.
For mills above 0.030", shim the rocker stand back to nominal and use a stock-length pushrod. The geometry stays correct and the rocker ratio stays where the cam grinder designed it. On LS engines specifically, ARP makes shim kits in 0.010" increments for exactly this reason.
References & Further Reading
- Wikipedia contributors. Pushrod. Wikipedia
Building or designing a mechanism like this?
Explore the precision-engineered motion control hardware used by mechanical engineers, makers, and product designers.
