What is the fuel bias?
A planning and dispatching software that one uses can have limitations
based on the fact that plots provided by the aircraft manufacturer have
to be translated into functions by using discrete values read from those
plots. If the functions are linear (linear dependency on the plot) than
the function fitting is quite precise, but more often than not, the
dependencies are not linear, so fitting algorithms can have errors which
will translate into different values found during flight for the output
parameters given by the software.
Another source of differences for actual flight parameters as opposed to
computed ones are the actual environment that an operator uses the
plane. For example, the manufacturer provided plots may be given for ISA
conditions, while the operator constantly flies the plane in a high
temperatures-low density area. This will again introduce some errors
into the computed values that may have to be corrected by the
dispatcher. And this problem is more acute in the virtual world as
developers usually learn as they go along.
One of these values is the fuel burn rate. The manufacturer (real world
plane) provided extensive plots describing the fuel burn of an engine in
their aircraft configuration and the model developer (virtual world
plane) did his/her best to make the engine follow the manufacturer's
performance plots, but as I said above, there can be differences that
will arise from the actual translation of plots into continuous
functions or from flying in ISA and flying in Congo all the time. Or it
may just be a feature of the plane that drastically changes the fuel
burn: i.e. the B737-300's auto-flight in turbulent weather that
increases the ordered speed thus decreasing the climb rate =>
increasing climb time => burning more fuel. Whatever the case, you
will see that the fuel used can differ from expected computed values.
So how does one compute this correction value?
PFPX and fuel bias
PFPX, dear PFPX, offers simmers a way to account for these discrepancies
by imploying an algorithm to compute the fuel bias for all aircraft in
the database. It is as simple as reading some in-flight parameters and
inserting them into PFPX. It will consequently provide the fuel bias and
a drag index correction. But to have a coherent computation, the values
need to be taken in stable flight as in the cruise part of a flight.
The values needed by PFPX to compute the fuel bias are:
- flight level/altitude
- true airspeed
- ground speed
- Mach number/outside air temperature
- aircraft weight
- actual fuel consumption
- (cost index)
Outside Air Temperature
The fuel bias calculation needs the outside air temperature to be known,
but in a jet that is not really available because of the high mach
numbers which in turn give way to air friction that heats up the air
around a jet liner. So what we know in a jet is the total air
temperature that is not that helpful. That is why PFPX provides
alternative ways of entering/finding the outside air temperature. The
most convenient way to insert the outside air temperature without ever
knowing it is by means of ISA deviation and mach number. If you know the
altitude, the true airspeed (easily read as all jetliners have TAS
indications) and the mach number you can derive the outside air
temperature directly in PFPX Bias window. First you insert the
true airspeed and the altitude, which will make PFPX compute a mach
number in ISA conditions at that altitude. But you know the actual mach
number from the jetliner's PFD so what you need to do in the PFPX window
is to modify the ISA deviation temperature window until the computed
mach number matches your actual mach number. That way you will have a
good estimation of the actual atmospheric conditions for your fuel bias
calculation.
That is why I wrote "outside air temperature/mach number" in the
requirements list - GA can give you OAT while for jets it's better to
read the mach number and derive the OAT.
TAS, GS and drag index correction
So as you can see from the above list, one needs not just the true
airspeed but also the ground speed. This is for wind calculation.
Normally, if there would be no wind, the aircraft would always have a GS
equal to the TAS. Knowing the two values will tell the algorithm what
kind of longitudinal wind did the plane encounter (headwind or tailwind
component). While this parameter doesn't change the fuel bias in any
way, it does help PFPX compute something called drag index correction
for the actual flying conditions. This is the same as the fuel bias
correction in the sense that a positive value will tell the algorithm
that the plane is dragging more so more engine power is required to
maintain a certain speed so the actual fuel consumption shown is now
higher than normal.
Actual fuel consumption
This is pretty straight forward, you read the current fuel burn rate
from the screens or dials. One thing I want to point out in here is that
you should make sure you understand the unit of measurement. For
example, in IXEG's B737-300, the fuel burn is ALWAYS given in thousands
of kilograms per hour for each engine, even if you have the fuel
quantity in pounds!!! So please be sure you always know what unit you
have the value in and what unit is required in PFPX window. Use Google
to convert.
Cost Index
Some jets employ the use of the Cost Index concept (all turbo-fan
jetliners) which is not the scope of this post to explain, but is
suffice to say that CI is inversely proportional with the fuel burn so
that is why PFPX needs to know what CI did the pilot actually request
from the FCS before doing any computation regarding fuel burn rates. If
the aircraft doesn't employ the CI logic, you won't be asked for it.
Conclusion
So if you enter all required input PFPX will come up with a fuel bias
value and a drag correction value. If you apply this, all future fuel
calculations will have these corrections in for this particular
aircraft. While the correction values are usually low (|5%| to |8%|) I
found that the IXEG model needs a larger correction (+16% to +22%).
Nevertheless, after correcting the fuel bias I get more consistent fuel
values when comparing PFPX flight plans with in-flight values. But one
thing to point out is that a new re-computation of the fuel bias value
may be needed at extreme atmospheric changes, so I would advise anyone
to recompute once in the summer and once in the winter, just to cover
all bases.
