Some technical considerations related to Spot tracking performance

In another thread today, the issue of Spot tracking performance came
up. In particular, it was pointed out that several pilots at the
ongoing Senior's contest were experiencing long periods of missed
tracking reports. This is an area that I have personally investigated
recently. I gave a presentation at a recent Arizona Soaring
Association meeting on the subject. As Tuno pointed out in that other
thread, my presentation is available for download at my Wing Rigger
web site at this page:

http://www.wingrigger.com/wingrigger5_005.htm

Unfortunately though, that presentation does not inform well without a
lot of words to go with it. With that in mind, I will summarize some
of the key points here. I did a bit of analysis and a lot of long
duration experiments to arrive at some conclusions which I think are
pertinent to our application in sailplanes.

First, I think it is useful to understand that Spot is using the
Globalstar constellation of 40 satellites orbiting in low Earth orbit
- like 800 miles up. At any given point on the earth's surface at any
given time, there will likely be two or three satellites in view and
sometimes only one. Statistically, you must expect that the
satellites that you need at any given time are going to be very near
to the horizon. It would actually be quite rare to have a Globalstar
satellite in a 45° cone overhead for example. If you want to have
really good Spot performance, you must be thinking foremost about
having a good line of sight to the horizon -- the whole horizon. This
is not easy to achieve in a glider cockpit. The unit should be
located high in the cockpit in order to have a good view of most of
the horizon.

It is also useful to understand a little bit about the antenna in the
Spot unit. It is a patch antenna oriented horizontally as the unit
sits flat on a table. This antenna transmits preferentially into its
upper hemisphere and only weakly in the downward hemisphere
direction. In long duration ground-based testing on top of the gazebo
in the back yard of my house, I was able to demonstrate about 99%
throughput performance with the unit oriented flat, logo side up.
However, with the Spot unit turned upside down, throughput performance
drops to about 65%. This gives an indication of the relative
performance through the back lobes of the antenna. Ideally, you would
always want to keep your Spot unit perfectly horizontal.

Of course, gliders don't stay horizontal -- you have to bank them a
good portion of the time. If you think about a Spot unit banked at
45°, what that really means is that about half of the horizon will be
visible through the main lobe of the Spot patch antenna and about half
of the horizon is viewed through its back lobes. Furthermore, when
banked, a good portion of that weak back lobe energy will need to be
transmitted through portions of the glider structure which will
attenuate the RF energy even more. Carbon structures are much worse
than fiberglass in that regard.

This would certainly lead one to expect that a properly oriented and
well-placed Spot unit would perform much better while cruising as
opposed to tharmalling. To investigate that premise I analyzed two of
my recent five hour flights. I used SeeYou to examine whether or not
my glider was cruising or circling at Spot track transmission time
stamps (which occur every 10 minutes) as well as whether I was
cruising or circling at missing transmission times. That analysis
showed that for my mounting position, I am getting 98% message
throughput performance in cruise and 80% throughput while banked in
climb. For a typical flight of 70% cruise, my aggregate throughput
performance is about 93%. My mounting location is high behind the
headrest in my glider. This was the best location that I could come
up with from the point of view of track mode throughput.

My mounting position does not give me access to the unit during
flight; but that suits me fine. My concept is to turn it on and
forget it. I have better things to worry about than pushing Spot
buttons while I'm flying. I realize that many users are considering
that it could be important to have the Spot unit mounted on their
parachutes, and it might be. However, I don't suspect that there is
any good place to put the gadget on a parachute without significantly
diminishing the track mode throughput performance. I say this for
three reasons: it will be very hard to maintain the Spot unit
horizontally while attached to a parachute; spot unit will be
necessarily lower in the glider cockpit so that there will be more
airplane structured to attenuate signal; and the human body is a very
good absorber of RF energy at Spot's 1.6 GHz transmit frequency, so
close to your body will mean that a good portion of the horizon would
become blocked out.

As I weigh the likelihoods and the relative benefits of having spot
with me during a parachute jump as compared to the benefits of having
strong tracking mode performance, the latter wins hands-down. Having
reliable tracking mode performance means that I can expect a real ELT-
like benefit of having a feasible area to search for my glider should
the day ever come that I'm not heard from. Actual ELTs don't work
most of the time in crashes whereas Spot really will work. However,
if I were having 30 minute Spot gaps, then the potential search area
would become hopelessly large. It is primarily for that reason that I
consider it to be especially important to have good track mode
performance.