Where does the force go?
A rod end is a chain, and the load passes through every link in it. In a typical clevis-mounted joint, the force goes from the machine into the rod end's eye, through the ball and into the shank, down the thread, and out through the pin that holds the rod end to its bracket. The chain is only as strong as its weakest link.
This is the first point to get right. A bearing's load rating covers the bearing itself. The pin, the bolts and the housing sit outside that rating and have to be checked separately. A rod end can have a large static rating and still fail early if the pin under it is undersized. Treat each part of the chain as its own calculation.
What are the two ratings, and which one do you need?
A static rating and a dynamic one. They answer different questions.
The static rating is about strength when the joint is not moving. The static load capacity in a catalogue table is a guide to the mechanical strength of the rod end, not a working load. Two static figures are in common use, defined as NHBB defines them. The limit load is the load that gives a stated permanent set, the point where the joint starts to deform. The ultimate load is the load that can be applied without fracturing the ball, race or eye. It is usually, but not always, 1.5 times the limit load.
The dynamic rating is about movement. For rod ends, some makers define the dynamic capacity as the upper load at which the ball can swing without seizing; for spherical plain bearings, the dynamic rating C is mainly an input to a life calculation (K02-Q1-03). Check which definition your catalogue uses. A rod end that only holds a fixed position needs the static check. A rod end that swings or turns under significant load, speed or cycle count needs both. (K02-Q1-04 covers life under oscillation.)
Do not mix them. An ultimate or proof load is a test threshold, not an operating load. If you compare two catalogues, check which definition each uses.
Which safety factor?
The safety factor depends on how the load acts. Lower limits for the static safety factor, as THK tabulates them:
| Type of load | Lower limit of safety factor |
|---|---|
| Constant load in a constant direction | 2 to 3 |
| Fluctuating load in a constant direction | 3 to 5 |
| Load in varying directions | 5 to 8 |
The rule is: the permissible load must not exceed the static capacity divided by the safety factor.
The reason the numbers rise with the complexity of the load is simple. A load that changes direction reverses the stress in the race and the pin each cycle, so fatigue can govern as well as overload. A load that is steady in one direction does not reverse. Worked example, with illustrative numbers: a rod end sees 10 kN, and the load changes direction every cycle. With a factor of 5 to 8, the static capacity you need is 50 to 80 kN.
What is the real peak load?
Most sizing errors come from using the normal working load, not the peak. The design margin depends on shock, cycling, uncertainty in the real load, and the consequences of failure. A jam, an impact, a hard stop or a sudden reversal produces a short spike that can be several times the steady load.
For a hydraulic or pneumatic actuator, one practical check is the maximum force the cylinder can produce, not the force the work needs. It is pressure multiplied by piston area. A 100 mm bore at 10 bar (1 MPa) produces about 7.85 kN on the extend stroke. If the relief valve is set higher, or the cylinder can be driven from a full-pressure line, the rod end has to be rated for that force. This is derived from basic hydraulics, not from a catalogue, and it is often the number that decides the size.
Write down three loads: the normal working load, the peak load and the abnormal case (a jam or an emergency stop). Size to the highest one that can really happen, and say which one you used.
What else limits the rod end?
More than the bearing. Picture the rod end as a clock face with the shank pointing down at 6 o'clock. Five things limit it:
- The pin through the ball bore, in double shear. The load is shared between two shear planes, so the shear stress is 2F ÷ (π × d²). For a 10 kN load on a 12 mm pin, that is 44 MPa. The same load on a 10 mm pin gives 64 MPa. Pin choice is covered in K05-Q1-05.
- The bearing itself, meaning the race material or the self-lubricating liner (Q1-04).
- The eye in hoop tension, mostly at the 3 and 9 o'clock positions.
- The shank stress area, which depends on whether the end is male or female. For a male end it is about the minor diameter squared times π/4. For a female end it is the ring between the outer shank and the thread's major diameter (Q1-03).
- The transition between the threaded shank and the eye.
In typical test failures, most rod ends fail in tension at about 4 to 8 o'clock on the eye. Simple formulas do not cover special body shapes, thin race sections, hardness variation, lubrication holes and grooves. For special parts, ask the maker.
A practical warning: a rod end with enough radial capacity can still sit in a weak assembly. The pin may bend, the clevis may spread, the thread may not be engaged far enough, or the shank may be side-loaded every cycle.
What about axial and off-axis loads?
Rod ends are built for loads along the shank centreline. An angled or off-axis force adds bending stress to the shank and the thread. Keep the load in line where you can, or allow for the bending.
Axial capacity depends on three things: how the bearing is retained in the eye, the axial capacity of the bearing element, and any bending moment on the rod end. The axial rating is a separate number in the catalogue, and it is not the radial rating. Ask the maker for the axial figure of the exact part.
When does the dynamic rating matter?
When the joint moves while it carries load. A rod end that swings through a small angle under a steady load may need only the static check. One that rotates or swings under high load, at speed, or for many cycles needs the dynamic capacity, and it needs a life estimate on top (K02-Q1-04).
Reversing load is the hardest case. Each reversal drives the ball through its clearance, so wear builds faster and fatigue, not only overload, can end the life. This is why the safety factor for varying direction is 5 to 8 and not 2 to 3.
What do you send the maker?
Send the peak and normal radial loads, the axial loads, whether the direction reverses, the swing angle and how often it swings, the temperature range, how the joint is lubricated, and what happens if it fails. That set of facts is what lets the maker check the part against the real chain, not just the bore.
Some catalogues and online tools use a different method: rated load ÷ (operating load × shock multiplier), with the multiplier and the pass threshold chosen by the user. That is not the table above, so state which method you used.
Mistakes to avoid
- Sizing to the normal load and ignoring the cylinder's peak force.
- Reading the ultimate load as the working load. It is a test threshold.
- Checking the bearing and not the pin, the eye and the shank.
- Using the same safety factor for every load type. A steady load and a reversing load are not the same.
- Comparing ratings from two catalogues without checking whether they use the same definition.
How do you size a rod end for the peak load on a vehicle link?
The peak load decides the size, and on a vehicle the peak is far from the normal load. A jump landing, a wheel dropping into a rut or a kerb strike drives the wheel up hard, and braking or traction can load a locating link much harder, or in the opposite sense (tension instead of compression). A locating link (toe, trailing arm, panhard, four-link) can therefore reverse, which is the "varying direction" case above; a pivot or pushrod that mainly carries the vehicle's weight is loaded one way, with the peaks on top.
An illustrative estimate: a corner carrying 375 kg has a static wheel load of 375 × 9.81 = 3.7 kN. A landing that briefly takes the wheel to 3 g gives about 11.04 kN at the tyre. If the link carries 1.5 times the wheel force, its load is 11.04 × 1.5 ≈ 16.6 kN, about four and a half times the static wheel load, at each end. The 3 g and the 1.5 are placeholders; real figures come from the suspension geometry and, better, from logged data.
Two cautions. The safety factor table above is a machinery table; a design factor for a suspension or steering link is for the vehicle designer or the series rules to set, so take no number from here. And size the pin, bracket and weld for the same peak, because they usually fail before a correctly chosen rod end does (Q1-05, K05-Q1-05).
Useful tools on this site are the motion-ratio and link-load calculator and the rod end safety factor calculator. For the vehicle application, see off-road rod ends.