Mark - I disagree with your logic.
Consider:
- What is the most survivable location on the plane? Looking at crumpled up RVs (upside down or not) the intersection of the VS and HS seem to always remain mostly intact.
- The “antenna must be vertical” argument. How will you know the orientation of the antenna after a crash? Also consider the antenna at 406mhz ~3/4 of a wavelength doing burst communication to a satellite - direction unknown other that above the horizon. In other words the vertical orientation may have marginal benefit at 121.5mhz, but none at 406.
The RV-10, RV-8A and RV-8 all have the ACK ELT antenna under the empennage fairing, base mounted to the last top skin bulkhead. On the RV-10 the owner inadvertently set it off - it got immediate attention.
Carl
Respectfully, Carl, your statement fails to address some of the more serious implications of an "under the emp fairing" installation.
One of the greatest challenges in ELT antenna placement is, as you have acknowledged, finding a location which is both functional and crash-worthy. The area around the intersection of horizontal stabilizer, vertical stabilizer and tail cone does indeed seem to provide one of the more robust mounting locations. The amount of structure required to support the loads imposed by those tail surfaces is considerable, thus likely to survive a crash.
BUT... when we bury an antenna within metal its radiation pattern can change quite significantly. As I pointed out previously, setting the ELT off accidentally in the rather pristine environs of the airport is anything but a real-world test. Try it at the bottom of a mountain ravine with iron-laden granite and huge coniferous trees surrounding the aircraft. Now we're talking real world. All of that "stuff" around the airplane sucks up, bends, reflects, re-radiates etc etc etc. That's why the 406MHz signal is a thumping 5 Watts as compared to the typical 100 milliwatts of the 121.5MHz signal. In designing a better mousetrap we recognized that too many times that puny 121.5MHz signal just didn't make it out of the crash site.
Still, we need that puny 121.5MHz signal for the "last mile" people to find us via brute-force ground search. How well is that signal going to get out of the mangled metal at the empennage fairing? How true will be the radiation pattern? Both of these, as just two variables, make the jobs of those "last mile" searchers anything from more difficult to darned-near impossible with an under-the-emp-fairing installation.
While it is your prerogative to disagree with my logic I might suggest that disagreeing with the logic of a large team of technical experts may be inadvisable. That large team of technical experts would be the folks who designed the Minimum Operational Performance Specification for the ELT - the MOPS is one of the fundamental documents which eventually drives the performance definition which ultimately makes its way to us in the form of TSO C126 and its follow-on variants. Those technical experts aren't just the avionics weenies that build ELTs - they're also the folks who build the satellites that listen for the ELTs, the folks who run the Search and Rescue organisations which go searching for those ELT signals in the most horrific weather conditions, not to mention the NTSB folks who bring with them their sheaves of accident reports which are intended to better inform future equipment designs.
With this in mind I will suggest you are well within your right to disagree with my logic. After all, what can I possibly know? By the same token, it would behoove us to pay heed to the logic of experts as it is manifested in TSO C126.
If we wish to really examine logic I would ask you to perform some introspection. Your logic about the installation of your own particular ELT antennas doesn't stand up in light of the ultimate test. Have you crashed any of those aircraft to determine how well those ELT antenna mounting locations work in the real world? Once you've got this empirical test data in hand let's rejoin this conversation. Until then, let's rely on the guidance of the expert logic expressed in TSO C126.