My experience with this type of bolted connection came from our off road racing days. We used adhesives, lock wire, lock nuts, etc in attempts to keep fasteners from turning and what we found is the connection can and often does come loose without the hardware turning. This is because of fastener stretch or connection shift, failure. In my experience the connection fails first, starts shifting, then the fasteners start turning, so assembly to maximize the strength of the connection is the most productive preventive to failure.
I’m a big fan of loctite, have an 8.45oz bottle sitting on the bench that I use daily, but also believe it has it place and this isn’t it for two reasons. When the fasteners are glued together there is no way of confirming the clamp force of the connection, when retorqued, the wrench can click at the proper torque, but the fastener is loose. It can prevent metal to metal contact in fasteners, a prime example is if a fastener is assembled without load, like a stud. When a fastener is screwed into a hole it has play in all directions, if tightened into the hole the last thread on the stud and the first in the hole jam together, the rest of the threads are relatively centered with each other, when load is applied the stud will shift slightly towards the load, until all threads make contact. If there is adhesive, the treads are held slightly apart by it, when it fails the stud will shift and reduce the clamping force. Loctite connections must be loaded in their final configuration relatively immediately during assembly for maximum strength.
All hardware has specs, most important in this case is yield strength. When a fastener is torqued/loaded under yield it deforms/stretches in what is called elastic range, can be reused many times and will repeat the same clamping force. If torqued/loaded past yield it will suffer plastic deformation. Once this has happened the fastener will never hold it’s rated clamping force, will continue to stretch, loosen and must be replaced. IIRC the torque spec on the cone washer hardware is very close to yield, so it’s very important that they are properly torqued. Sometimes a plastic deformed fastener is easily detected with the torque wrench. When torqued a fastener should stay in place as the wrench load is removed, if the fastener moves, springs back, it often junk, seen most often on “discount tire torqued” wheel lugs. Any fastener that is bent is plastic deformed, any that have been pounded by running loose, maybe, so should be replaced.
This type of connection relies on solid metal to metal contact, if there is any rust, debris, burrs, etc, the parts wont make good contact. On the flat surfaces I use a wet stone or small file, slide it over the surface, the point is to look for/remove any burrs or high spots without changing the shape of the surface. The number one mistake that I see is paint, if the contact points are painted it will be squeezed out loosing the connection, paint after assembly.
I assemble with all moving, sliding parts (threads, washers, cone washers) lightly oiled, just a film. Torque the fasteners, then using a “soft” punch (brass, aluminum) put some pretty good whacks on the arm, then on each stud, trying to drive them in. I use a 40oz dead blow hammer and put some arm into it, enough to vibrate, settle the parts together without dents, damage, call this ringing the connection. Then retorque, if any of the fasteners move, ring and retorque again. Since I started ringing the fasteners, have had zero problems with them loosening, don’t use any adhesives and only install the studs snug. I use this method on all cone washer connections.
Put a few miles on the rig (~500) and recheck the torque, shouldn’t move. On a wheeling rig, some things should be torque checked, these are on my list along with most of the other suspension bolts, get checked a few times a year.
but more to to try and post something backed up by Toyota documentation in this thread. Re: