The information that the program
uses on how each parameter is stored in the subframes is mostly from
the
partial data frame layout in this file.
There is no guarantee that this partial data frame layout is correct
since airlines can and do use modifications of the standard data frame
layouts. I am not as certain that I have been able to correctly decode
the
parameters as I was for the UAL93 FDR Decoder program, since I have
little data to correlate against.
For the numeric values, I used the same formulas to convert the
binary values to engineering units as used in the UAL93 FDR Decoder
program where
these appeared to work.
There are many other parameters listed in the NTSB's American Airlines
Flight 77 FDR Report (available
here) that the
program currently doesn't handle. If
there are any that you are interested in
tell
me (Warren Stutt) about them and I'll consider including them.
ALTITUDE (1013.25mB)
Although the partial data frame layout referred to above
indicates that bits 3 and 4 of the coarse data word are used, they
appeared to be the copies of bits 10 and 11 of the fine data word and
when I used them, produced values that were far too high unless both
bits were zero. Therefore, I changed the program to not use them. Bear
in mind that these values are uncorrected pressure altitude values that
presume that the air pressure is 1013.25mB and will be different from
the true altitude if the air pressure was not that value when the
altitude values were recorded.
COMPUTED AIRSPEED
This shows some values of 384 whereas all the values
should be 0 or close to it since the plane should be stationery when
the engines and the FDR are started. Perhaps the most significant bits
in the partial data frame layout referred to above should not be used
or perhaps this parameter is just plain wrong.
GPS
I guessed that a value of 1 for this single bit parameter
meant the GPS was working and a value of 0 meant that the GPS was not
working.
PRES POSN
Although the partial data frame layout referred to above
indicates that bit 12 of the Least Significant (LS) data is used, when
the NTSB used this bit to generate their CSV files, it produced PRES
POSN LONG (DEG) values that were out by about 0.35 degrees. This
program and Readout2 provided by
UnderTow
ignore this bit and produce more accurate values.
RADIO HEIGHT F/O
I couldn't produce sensible values for this parameter so
I removed it from the program. The other RADIO HEIGHT parameter values
looked good.
Block Position
Bit Offset
Flight Stream Number
These values are calculated from the Bit Position value
and can be used to check the order in which the subframes appear in the
output file. The fact that there are two flight streams of
data that are recorded simultaneously makes it difficult to be certain
of the order in which the
subframes were recorded. As far as I know, frames of data
are alternately recorded in flight streams 0 and 1. If two subframes
have the same flight stream number then they will appear in the output
file in the order they were recorded. If two subframes
have different flight stream numbers and different Block Position
values then I am fairly certain that the
subframe with the higher Block Position value will have been the later
one to be recorded and will appear later in the output file. If two
subframes have the different flight stream numbers and the same Block
Position
values, then the one with the higher Bit Offset value probably but not
definitely will have been the later
one to be recorded and will appear later in the output file.
Go back to the
AAL77 FDR Partial
Decoder page.