Re: Wing vs. tail-mounted engines?

Date:         16 May 2001 17:45:30 
From:         Pete Mellor <pm@csr.city.ac.uk>
References:   1
View raw article
  or MIME structure

> In article <airliners.2001.149@ditka.Chicago.COM>,
> John Liebson  <jliebson@cybermesa.com> asked
> >What about wing-mounted-engine aircraft that _also_ have tail-mounted
> >fuel tanks?
>
> Not nice:
> * fuel pipes in the aft fuselage (--> big fire risk in crashes)
> * fuel tanks far from center of gravity --> big trim change as you burn
>   that fuel --> hurts aerodynamic efficiency (fighting that trim change
>   will cost extra drag)

Exactly the opposite, in fact.  This misrepresents the purpose of
the tail tank, which is not to store masses of fuel, but precisely
to effect a trim change.  (In fact, the tail tank is often referred
to as the "trim tank".)

This is used on long-haul aircraft for which efficient cruising at
high altitude is desirable.  Examples are A300, A310, A330, A340
and (I think) most Boeings in the medium to long haul range.
(Note that it does *not* apply to the A320, A319, A321, and similar,
which are short haul.)

As an example of the size of the tail tank, that on the A330 holds
about 7000 units compared to around 25000 in the wing and centre.
(I am going by memory from the last time I rode in the jump seat
of an A330 several years ago.  I think the units the crew quoted
were gallons.  If I'm wrong, please correct me.)

During cruise, fuel is pumped between the tail and wing/centre
tanks in order to move the CoG as close to the Centre of Lift as
possible while still retaining marginal static stability.
The result is that elevator trim and hence drag is reduced to the
minimum compatible with safe cruising.  Therefore fuel burn is
reduced with a significant cost saving.

On modern aircraft, this transfer is (surprise! :-) controlled by
computer.  A similar feature exists on Concorde, and as it goes
supersonic, the centre of lift changes, hence the fraction of the
fuel load required in the tail changes also, but the transfer is
performed under the manual control of the flight engineer.

The disadvantage of the system is that the aircraft *must* be
controlled by the autopilot while in this trimmed-back
configuration.  The reason is fairly obvious.  The aircraft is
up in coffin corner where the stall speed approaches the cruising
speed *and* it is only marginally stable.  The suddent disconnection
of AP can therefore lead to a spectacular upset.  This is thought
to have been a factor in the crash in Russia a few years back
(the A300 or 310 in which it was suspected that the captain's
teenage son was at the controls).  I think there is now some
doubt about the "kid at the controls" theory, but even an
experienced pilot has great difficulty retaining control if the
AP is unexpectedly disconnected.  (A Lufthansa pilot was quoted
to that effect in Der Spiegel at the time of the Russian crash.)

I would be very interested to hear any other thoughts/information
on this topic.

Pete
----------------------------------------------------------------
Peter Mellor, Centre for Software Reliability, City University,
Northampton Square, London EC1V 0HB
Tel.: +44 (0)20 7040 8422  ) NOTE: Code recently changed from
Fax.: +44 (0)20 7040 8585  )       7477 to 7040
e-mail: Pete Mellor <p.mellor@csr.city.ac.uk>
----------------------------------------------------------------