Even though I have never taken off the ring gear assembly (no reason to so far), I think I know what's going on here.
The problem is that the ring gear assembly is not sufficiently radially constrained by the "bushing" that it spins on.
Further, the needle bearings that support the saddle shaft can't have a zero-clearance fit because you have to be able to pull the shaft out.? Thus there will always be slop no matter what.? Because of the slop in the needle bearings, the saddle shaft rests on the "lower" side of the bearing.? The first time you tighten the clutch (for example, in the CWD position) the upper half of the clutch mechanism grabs the lower half of the clutch (ring gear mechanism) offset and below axis.? If there is also too much slop in the central bushing, then this will cause the "ring gear eccentricity" when the axis rotates, because the axle always rests on the lower side of the needle bearings.
In the older style Losmandy mounts, the worm assembly is on the lower side of the axis.? On the newer version it is on the upper side.? Thus, on the older style, when you spin the axis 180 degrees (or any amount) then this eccentricity "jams" the ring gear into the worm mechanism, tightening the mesh.? On the newer style, this moves the ring gear away from the worm, loosening the mesh.
As I reported for the new G11 DEC axis, it still "works fine" with a too-loose worm mesh, as long as the axis is a little unbalanced to keep the worm teeth meshed on one side always.
In any case, I always do a test of worm mesh before a long imaging run anyways, so this isn't a big issue for me.
Here's a few possible solutions I thought of:
- Loosen and tighten the clutch before a prolonged session, and this should "reset" the ring gear eccentricity to what you set the original mesh to
- Find a way to better radially constrain the ring gear mechanism, for example a radial bearing instead of that bushing.? In this case, you can have a zero-clearance fit since you don't constantly need to pull this part off
- Modify the clutch mechanism to only give torque clamping, but minimal lateral clamping, so that the two halves can slightly slide relative to each other.? (Like an oldham coupling, hurr hurr)? For example, cut away most of the center section of the clutch disk but leave the outer ring.? Or just use a lighter clutch pressure
- Come up with a mechanism that forces the saddle spindle to ride on the ring gear mechanism instead of the upper needle bearing.? For example, a piece of tape or thin sheet metal such that when you insert the saddle shaft, it fits snugly in the bore of the ring gear mechanism.? This isn't a great solution because now the ring gear bushing supports the weight of the axle.
?
