Since we all know that PfT and our departed collegue, Turbofan read theses threads, and given that I am banned on PfT,
here is another link, this one from Harris, they could peruse concerning surge suppression devices. These are just a tad more useful and complex than Tf's simple zener and transitor ccts.
Of particular interest is the SP72x line of data bus surge suppressors.One might notice that the device simply protects from voltages over V+ or under V- from remaining on the data line. This of course means that any surge over V+ would keep the data line high, well in fact it would keep that data line no higher than V+. What PfT might ignore is that the V+ may also be much higher than it is supposed to be or that V- may well be higher or lower than it normal potential in the event of a violent aircraft crash. There are many ways then that this suppression will fail to guard in such a situation. These devices are designed to guard against surges and spikes that the aircraft might reasonably be expected to encounter in its normal operation. This would include spikes and surges due to lightning strikes on the aircraft, over volatges due to faulty power supplies etc. One can guard against such surges with inductive and capacitive ccts to shunt voltage and current away from sensitive devices along ground and power buses. Such surges would cause the crowbar ccts in the FDR to suspend operation and completely isolate itself from the rest of the aircraft until normal power values were restored. During such time the FDR records but there is no data. This though takes time to implement and the data just prior to the isloatiing of the FDR may be suspect. Of course in a crash there will not be time for an isolation from the aircraft systems.
You also cannot use inductive or capacitive devices on a data bus because they will corrupt the shape of data pulses. All you can do to a data bus is limit the voltages to within a diode drop value of the nominal 'high' and 'low' values. For instance, if a data 'high' is 5 volts, surge suppressors on a data bus (or control, register, interrupt, address) will not stop a surge from leaving causing a 'high' to be present forever.
However the main thing is that the aircraft would be considered to be intact. There would be no consideration for all the combinations and permutations of gound, data line and power line spikes and surges that could accompany a violent end to the aircraft. That is because it probably cannot be done and if it could it would be prohibative in terms of weight, size and cost per unit.
The fact that FDR's have lost data in the past even in aircraft that were intact but had mechanical control problems and the information supplied by the poster known here as MacgyverS2000 illustrates that it is hardly the great mystery that PfT attempts to make it.
TF and PfT would demand an exacting, detailed description of how this particular FDR lost or corrupted several seconds of data. On the other hand they believe that the aircraft flew over the SE corner of the Navy Annex, banked left to take it north of the Citgo while desending to tree top level near the Citgo, which is itself downhill from the Annex, leveling out both the bank and the desent and then pulling up and over the Pentagon at the (generally recognized) impact point. The last part, the pulling up and over, supposedly hidden from the view of onlookers by the fireball.
The irony is that to date no one at PfT or CiT, despite their claims to expertise, has bothered to do any calculations that would illustrate that such an aircraft can perform this manouvering. They demand exacting details of how the FDR lost data but will not apply the same level of detail to themselves.
