Why does the length change at all?
Because the thread loses its clamping force, and then it can turn.
Tightening a thread stretches the bolt slightly and presses the parts together. That pressure is called preload. Vibration can reduce it, and the usual cause is sideways movement between the joined parts. Once the pressure is gone, the thread can turn, in either direction depending on how big the vibration is, and a nut can even work off a bolt against gravity.
A push-pull link has a second cause that a bolt often does not: the axial load itself changes sign. At each reversal the thread flanks that were carrying the load unload and the opposite flanks take over. If the jam nut's preload is smaller than the swing of the load, the flanks lift and change sides every cycle, and the thread can creep. Bending from a skewed rod end adds to it. So the preload has to be larger than the load swing, not just "tight".
How does a single jam nut work?
It presses the threaded part against the end face of the body, so the thread cannot move.
On a linkage the usual arrangement is one jam nut on each threaded end, run down by hand against the end face of the body or tube and then tightened. The nut pulls the thread flanks hard against each other and the face. That friction is the locking force, and it is the same preload described above.
What the nut does under load depends on the direction. Under a pull, the working load goes through the thread engagement inside the body, not through the nut, so the nut only holds the setting. Under a push, or a load that reverses, the end face presses on the nut and the nut shares the load, so it has to be tightened to a real preload. The same reasoning is in K11-Q1-05, together with the reason no single torque value applies. The thread must still be engaged by the full length that Q1-04 calls for, because the nut does not make up for missing engagement.
What if there is no end face to tighten against?
Then two nuts are tightened against each other, and the order matters. The usual procedure for a bolt is to put the thin nut on first and tighten it to 25 to 50% of the full torque, hold it with a spanner, then tighten the second, full-height nut to the full torque. The reason is the way the threads jam. With the thin nut next to the joint, tightening the thick nut on top relieves the thin nut's threads, then jams the bolt thread between the upper flank of the thin nut and the lower flank of the thick nut. A thread held on both sides cannot slide, and a thread that cannot slide cannot loosen.
A Junker vibration test of both arrangements, reported by Bolt Science, found that with the thin nut on top both nuts turned together and could come off completely, while with the thin nut next to the joint there was some relaxation but not much. Field practice differs, but the test result is consistent with the principle above. Tightening one nut and just putting another on top achieves little.
An adjuster is a different joint from a bolt, so the same torque numbers do not apply. Take the value from the maker of the rod end and the nut. What carries over is the principle: the two parts must press against each other.
How do you fit jam nuts on a linkage?
Set the centre-to-centre length first (Q1-04), then run each jam nut down against the face by hand and tighten it.
Match the thread hand. On a link with a right-hand thread at one end and a left-hand thread at the other, the right-hand end takes a right-hand nut and the left-hand end a left-hand nut. Left-hand nuts are normally marked; SYZ, for example, marks them with a groove or yellow zinc plating. Check for the mark before fitting.
Do not stack two thin nuts and tighten them. The short thread contact can strip part of the thread without anyone noticing, and the joint can later lose its preload. Two normal-height nuts avoid this.
What other locking methods exist?
- Thread-locking adhesive. A liquid applied to the thread that hardens and fills the gaps between threads. It resists vibration, corrosion and galling, and grades range from low to high strength. A medium-strength grade that can be undone with hand tools (Loctite 243 is one) is the kind to look for if the link will be adjusted again.
- Nuts that grip the thread. Prevailing-torque nuts, including nylon-insert nuts, have a feature that resists turning. They are not true lock nuts: in a vibration test they loosen at first and then stop at a leftover preload. They suit many jobs but are no cure-all.
- Both together. A jam nut and an adhesive can be used on the same thread. Check the adhesive maker's data for the metals, the temperature and the oils involved.
How do you keep an eye on it?
Check the set length at the intervals in the maintenance plan, and after any shock or overload. A paint or marker line across the nut and the body shows at a glance if it has turned. If the length has moved, re-set it and lock it again. A joint that is losing its preload usually gives time to find the problem, but a nut that has come right off can fail suddenly.
What is different on a vehicle that sees shock and vibration?
A steering or suspension adjuster is shaken constantly and also struck: by ruts, kerbs, rocks and landings. Both are the conditions in which a thread turns, and a link that has crept shows up as changing toe or a vehicle that wanders, which is noticed only after the length has already moved. So treat the jam nuts as part of the geometry setting. Tighten them to the preload the maker specifies (K11-Q1-05 explains why no general torque exists), keep the full thread engagement from Q1-04, and mark each nut and body with a line so a turned nut is visible at a glance. Check the marks and the length after any hard hit and when the vehicle handles differently. On a link with a right-hand thread at one end and a left-hand thread at the other, the nut hands must match, as above. The jam nuts and tie rod pages are the product pages involved.