Perhaps there's a little production issue with my version of the Sony 55-210mm f/4.5-6.3 OSS?

The dogs and cars I've been photographing are less than sharp where I think they should be and I was beginning to think Sony had done a terrible job making their 55-210mm f/4.5-6.3 OSS E-mount optic, but I wasn't sure the problem was with the resolution.

When I'd taken a recent look at the lens in comparing it with other optics I noticed the edges of the comparison setup were sharper than the center.  This got me to thinking.  Before I could do much about it we went to the US for six weeks to take care of family business and to attend our son's wedding.

Once home I used the "force" (Google) to research back-focus issues with this particular lens.  Over on DPReview I found someone talking about their experience with seemingly the very same issue.  What struck me were the responses where people said the phenomenon was impossible.  The thinking was that since the Sony A6000 has on sensor phase detect that the problem had to be with the user.  Said another way, PDAF supposedly _ensured_ accurate AF and there was no way it could fail.  Evidence led me to believe the problem was not mine, but lay with the lens, PDAF or no.

When I photographed cars and dogs I took to using the center AF point and kept that point right over the thing I wanted in focus.  Yet that part of the scene was seldom (never?) in focus, whereas the area a few feet (about a meter, maybe two meters) behind the intended focus point was always tack sharp.

At that point I decided to investigate the issue further.  I set up a simple test.

• Sony A6000 with phase detect (PDAF) as well as contrast detect (CDAF) AF enabled
• Sony NEX-5T with PDAF enabled, and then retested with CDAF only (in case the problem really was with PDAF)
• Sony 55-210mm f/4.5-6.3 OSS SEL at various focal lengths and apertures
• Test target 1: Trees in front of a brick building at around 30 meters (give or take)
• Test target 2: Edge of a building with foreground and background sloping away from me

Here is test target 1.  As you can see, the tree has a bit of depth to it -

Here is test target 2.  This test concentrated on corner of the transition from lighter plaster to rock area on the right -

Here is the test comparison.  I share the f/8 results here as the f/6.3 and f/11 results are identical.  Open the link and locate the full resolution version of the image to verify differences between the various panels - 


In each image I ensured that the AF box(es) that lit were directly over the center of the scene.  With this in mind, what I see is that the camera did not matter and neither did the AF mode (PDAF or CDAF), the lens clearly focuses behind the point that the AF system tells me is being selected.  As confirmation of the condition and that I'm not loosing my mind, when I focus manually the intended focus point is achieved and the lens is actually acceptably sharp.  Lastly, the phenomenon is experienced starting around 135mm.  Focal lengths less than 135mm seem to focus accurately and correctly.
My copy of the lens is not behaving the way I think it should (understatement, I know).  So, I've written Sony France to ask if it might be repairable for something less than the cost of a used copy (which I of course would carefully test prior to purchase) off leboncoin.  I was in a little snarky mood when I wrote them so we'll see what they come back with, if anything.  
I'd purchased this lens as a used item at the Bievre camera swap and it's well beyond it's warranty period.  Sony really doesn't owe me anything.  Though I would much prefer a lens that can AF correctly, right?  I might have to go buy another and jeter this one into the recycling bin.