SuperTANS
I concede I'm far too old to have first hand knowledge of SuperTANS, but the question has been raised before as to what actually occurs when this switch is switched "Off" in the manner decribed by Dalek? Does TANS still accept and store input data from Doppler etc? Such a "standby" mode would explain both the "fact" data was there to be extracted post-crash and the possibility of the crew switching the unit off (a recognised technique to clear comms EMC problems in the aircraft).
Alternatively;
The Racal report is an interesting read, although of course most of it was withheld by MoD. (That is, the ROs based their verdict on information, much of which hasn't been released). What stands out to most readers is that Racal only say the inputs were working (Doppler, Air Data, GPS) but nowhere claim they were accurate. MoD clearly had better information, as they claimed both; and also that all units were serviceable.
However, in light of the conflicting evidence between AAIB and Racal over the switch, it is interesting that the released part of the report explains each function in great detail, except that of the on/off switch or the behaviour of the device when subjected to power interrupts (a common occurrence noted by Sqn Ldr Burke, but ignored by MoD). As stated before, one is struck by the single short "SuperTANS" paragraph in the AAIB - the author (Mr Cable) clearly assumed, or was told to assume, that the conflicts would reconciled elsewhere (given he was not the AAIB "lead" on TANS/Nav). One short paragraph, yet the ROs based their entire case on this area.
But what nobody menions is the Trimble 8000 GPS report, or whether its content was reconciled with that of the Racal report. This omission could be for a number of reasons. Very many assume GPS was embedded in TANS. "Learned" Lords who supported the ROs, even stated that there was no GPS at all.
But when reading both reports side by side (Windows 7, I love it) I was immediately struck by the "fact" that GPS powered down 27.6 seconds after TANS using, if I read the reports correctly, the same time baseline (i.e. that fed from TANS to the GPS internal clock upon initialisation). I suppose it is possible that TANS lost, or GPS advanced, 27.6 seconds in the 52m TANS was "fixed", but that is a lot of drift. But, from this information, it is also possible TANS was powered down some time before GPS.
The GPS wasn't entirely serviceable, carrying two known faults (actually, one fault and one defect), including one at the GPS/TANS interface. And the crew were told to ignore TANS error messages relating to GPS.
Again, I would never say this was the cause, but it is an information conflict; and conflicts must be resolved (or so it says in MoD's Safety Management regulations). Perhaps someone more up to date would care to check what I've surmised from the reports?