I hope you have better luck than I did. I upgraded from my OLD Cal Amp (I think) lnb that had a freq stability somewhere in the 750-1000 KHz range, to an 8115 a year or so ago, with the same intent. I thought that the better freq stability (I forget exactly, but I think the 8115 is around 100-150 KHz). Well, despite the fact that it was a much more stable lnb, I couldn't lock a lot of transponders that I WAS able to lock with the old Cal Amp. I think it's because the old LNB had a bit more gain, and the Norsat must get it's freq stability at the expense of a bit of gain.
Modern receivers generally have a wide lock range, so a high frequency stability LNB is normally of little value unless one is trying to lock very low SRs. I've tested entering the wrong frequencies into all of my receivers, and they can readily capture signals 5-10 MHz off. Still, I like my signals to show up at the correct frequency on my bench spectrum analyzer, so I happen to use 8115s.
What 8115s really have going for them is very low phase noise for a DRO LNB. That might help lock S2 signals better than a crummy LNB with poor phase noise specs, but I read a study recently that seems to indicate the value of 8115-level phase noise specs would be more important for 16PSK and higher.
As to the Cal Amp vs. 8115 gain issue, are there a lot of losses in your distribution chain following the LNB? If so a low gain, high quality amplifier at the LNB might help. I finally got my 3.2m up last week and put some 8115s on it. I'm guessing there's about 200' of cable between the LNB and receiver, and some switch losses along the way. Now admittedly a tweaked 3.2 has tons of signal, and without any amplifier in the chain I was seeing better SQs and SNRs than I ever had before on my 3m. However when I looked on the spectrum analyzer, I noticed the potential for a little noise combining on the receiver side. I ran out and installed a Channel Master amp near the LNBs and the SQs shot up another 4-5 points. It's going to stay.
One thing that has piqued my interest at times, but not enough to do anything, was the question of how well LNBs match a particular feed. I don't have anything to offer, but it might be the Cal Amp simply works better for that feed. I'm pretty happy with my 8115s so I don't have a big drive to do it, but I had thought about trying different LNBs on all the different kinds of feeds I have, and possibly on different dishes to see if there is any correlations I could draw. That would certainly burn up some time.
Regardless of all the above, you are correct that S2 is going to put a premium on dishes in decent condition.