Today turned out sunny and warm so I was able to work with this problem child a bit.
I disassembled the support arms and moved the feed arm up and down and to both sides. I was able to gain a db or so with this so I shimmed things to stay at this orientation. The biggest change came when I swapped the older Avenger DRO LNB for a GeoSat Pro S1Pll LNB. That move gained a little over 2 DB on 11960 and close to three DB on 12076 and 11760. I haven't tried moving it to any other birds yet but given it does this well on SES2 it should perform close to this benchmark on any other location that I have decent LOS. LOIF frequency error was less than 100Khz for the SL1 but I forgot to check the Avenger DRO. Just for grins, I checked polarization with my SA, and my best quality was almost 2 degrees higher than the spec at 7.3 degrees. Until I check this against the HD90 I'm inclined to believe it's as good as it gets with this one. The one thing I did forget to do was move the LNB around the focal point by hand to see if any noticeable change could be detected but the feed arm support test should have been very close to what that would have shown.

Thanks to everyone who took the time to participate in this "project" Hopefully I can repay the favor in the future

Thanks
Wt