A couple of things. I have an Fscan2, and I've been using it for a year, typically with very good results.
That said, you do have a couple of valid gripes about the unit. The ADC arrangement they're using is kind of Rube Goldberg-ey, and I have reservations about the DIRP methodology. My experience, however, has been that DIRPs above 100kHz are repeatable and consistent. Have a look at http://www.tookworks.com/fscan for some of the results I've gotten, as well as some software I wrote for the unit.
So, the question I have is "why doesn't the Fscan work for you?" I have come across 2 people who seem to be entirely impervious to my unit. I can DIRP and zap them all day and get no result. None. Zilch. On the other hand, I respond pretty well to it, as do most people I work with. (Note- I'm not a professional practitioner, I don't charge for doing anyting with the Fscan as I don't need the legal entanglements, nor do I have the time.) Additionally, have you tried it on anyone else? If your DIRP results are that consistent, I'd have to wonder if the unit isn't buggered. And, I have to ask the obvious- when DIRPing, you're NOT connecting the blue lead, right? (Not trying to be insulting, just thorough.)
Further DIRP discussion: Personally, I think that the differential mesurement concept is flawed, especially when taking measurements with a small delta. For example, if a microbe had a frequency signature that covered 1000Hz, and you were scanning in 100Hz steps, you would miss the overall picture of the peak because many measurements would have similar values, even though the entire response was elevated. So, for narrow deltas, I'd like to see a baseline integrated resonance protocol (which has a suitably snarky acronym) where the test frequency is compared to a baseline frequency. This can be accomplished in software when controlling the unit from a PC. I just haven't gotten around to writing it yet.