Thanks all for your help! So I went back to test the lens again just to be sure.
Lens set to f2.8 and with a subject 16 ft away in live view x10 in the centre.
It is definitely not a large DOF issue because the critical focus point is reasonably sharp on the
subject. I can easily centre the focus without a doubt.
The problem arises when I try to pull focus by turning the ring in one direction
but I passed the critical focus point a little bit then I
tried to turn the ring in the reverse direction to re-obtain critical focus.
The more I do this the more I find the distance readout (during critical focus) all over the place
from 4ft to 5ft to 6 ft to 8ft to 10ft.
hit focus in one direction (readout 6ft) -> passed focus a little (may be half an inch rotation on the ring) -> reverse rotation ->
hit focus again in reverse direction (readout 8ft) -> passed focus a little -> reverse rotation -> hit focus again (readout 9ft) etc
So far, I think Logan's theory makes most sense: (also Jim) (Many thanks!)
"sounds like backlash to me. can you rotate the focus ring without moving the lens elements, if so, by how much? a helicoid(?) and a ballscrew are basically the same thing, i wonder if you could tighten something up inside the lens to fix it, maybe machine some new pins or whatever rides in the helicoid (i know its not really a helicoid but ive seen it called that i think?)"
Yes, I can rotate the focus ring a little (about 1/4 inch on the ring) but the lens element is not changed. (verified
by live view) This happens at the onset when I reverse the focus ring rotation after achieving critical focus.
Hysteresis is a term used in Electrical Engineering... but the wiki explanation seems appropriate:http://en.wikipedia.org/wiki/Hysteresis
"Hysteresis is the dependence of the output of a system not only on its current input, but also on its history of past inputs. The dependence arises because the history affects the value of an internal state. To predict its future outputs, either its internal state or its history must be known. If a given input alternately increases and decreases, a typical mark of hysteresis is that the output forms a loop as in the figure."
My 1st copy was much worst in "backlash" (much larger "play"). This makes me wonder if this is a
common problem to this lens. Has anyone else seen this problem b4 on this lens?