Checkboard “…interpolating the Airbus AD backwards…” absolutely not; the BAe/Honeywell work preceded the concerns on the ‘big jets’.
Links to the science relating to the TAT probe freezing are in
http://www.pprune.org/tech-log/37578...ml#post4963257 ; most of this relates to engine core freezing, but the mechanism and atmospheric conditions are similar if not identical, for certain TAT probes.
BOAC, et al, as the TAT probe freezes the indicated temperature increases toward zero - ice temperature in a static airflow; because the airflow is blocked. The atmosphere - ambient air temperature is still cold.
In engine rollback incidents the TAT increase was a preceding or confirming characteristic seen on the FDR in the vast majority of events. Conversely there were TAT rises without engine malfunction. (
See slide 8)
In the very unusual conditions associated with large Cbs (probably only those which penetrate into the troposphere), the area in and around the anvil contains large quantities of minute ice crystals and often a high super cooled water content. In these conditions the TAT probe heater is progressively overcome by the high negative heat flow – melting ice crystals and water-flow off, such that further ice crystals accumulate in the probe throat (behind the heater). The melt water acts as glue for the ice crystals until the probe becomes blocked. IIRC the effect has been demonstrated in an icing tunnel.
With respect to the AF accident, it is of interest to establish how the various sensors use ‘temperature’, and particularly where there are multiple TAT inputs, how the systems (ADC/ ADIRS, Cabin, Rudder Lim) might cope with different values.
Assuming that multiple probes do not freeze identically; there will be different TAT values.
In a dual system, comparison can provide a comparator warning (systems still operative). However, with a triple input it may be possible to vote out an erroneous value; except in this instance the odd-one-out might be the true value and the voting system may have capability to shut down a ‘good’ system i.e ADC/ADIRS, etc.
Another issue is pitot icing. If the pitot system is also susceptible to icing in the same conditions, then conventionally the pressures existing at the time of blockage will be trapped. Thus there is no change in any of the input parameters with icing – whereas TAT will change with probe icing.
However, is has been suggested that the A330 pitot systems have a small leak path (drain) which will result in a pressure reduction to ambient (decreasing or fluctuating IAS). Thereafter the issues of value comparison and shut down are as above.