In article , Don Johnstone
writes
Ian
Surely you are not telling us all that the GPS accurately
recorded your true height but the barometric logger
got it wrong. This cannot be, and a valid trace as
well, this is blasphemy, the ICG sacred cow, the barometric
height widget has failed, the world will end.
There will always be occasional failures in systems needed for flight
verification. Cameras, barographs, GPS, pilots and OOs sometimes fail
to record what they are supposed to (I deliberately include human
factors as well as technical ones!).
The questions are, how often do problems occur, can they be easily seen
for what they are, and what "checks and balances" are there to prevent
false data from being used in validated claims.
In more detail:
(1) Is the failure obvious or is it difficult to detect?
In this case what seems to be a failure in pressure altitude recording
is easy to pick up. A gross error is easy to see, a small one is
difficult. I understand that this particular anomaly was due to the use
of a high input voltage to the recorder unit (up to 16.5 volts). The
anomaly is understood to have ceased when a 12 volt DC input was used.
Correspondence is taking place with the manufacturer and an announcement
giving more detail may be made shortly. Meanwhile users are advised to
be cautious in using battery inputs other than the large range of 12V
sealed lead-acid "gel" calls that are available worldwide.
and,
(2) Independent sources of data.
The merit of being able to compare the figures from two independent
altitude sensors (GPS above ellipsoid and ICAO ISA pressure altitude) is
clearly shown in the example. Relying on only one system, either
pressure or GPS, would lessen the chance of any anomalies being
detected.
and,
(3) Rate of anomalies found in IGC files.
How many significant anomalies are found in proportion to the total
number of IGC files looked at? Particularly, what is the "anomaly
rate" of electronic pressure altitude transducers and recording in IGC
files, compared to the recording of GPS altitudes in IGC files? In my
experience of analysing thousands of IGC files over the last 10 years,
the ratio is of the order of 1 to 10 in favour of pressure altitude
figures.
In other words, anomalies do occur, of course, but the large majority of
anomalies in terms of altitude figures in IGC files are in GPS altitude
compared to baro. Baro altitude is generally very reliably recorded,
GPS altitude in IGC files has a higher anomaly rate with unexplained
"spikes" both up and down and unlocks both with and without simultaneous
lat/long recording.
----------------------
Finally, pilots are advised to use high voltage inputs with caution
where equipment designed primarily for 12 volts DC is concerned. As
indicated above, up to 16.5 volts seems to have been used in the case
concerned and almost certainly caused this problem.
Please report any other suspected anomalies by email to me, including
the IGC file concerned as an attachment. The situation can then be
looked at and we can correspond with the manufacturer if this seems to
be required.
--
Ian Strachan
Chairman IGC GFA Committee