Re: Strange 727 Flight

From:         rdd@rascal.ics.utexas.edu (Robert Dorsett)
Date:         28 Jun 93 22:35:04 PDT
References:   1
Followups:    1
Next article
View raw article
  or MIME structure


In article <airliners.1993.471@ohare.Chicago.COM> weiss@hurricane.SEAS.UCLA.EDU
(Michael Weiss) wrote:
> Well, it sounds like particularly heavy traffic in BOS.  Often what will 
> happen
> is planes will be stacked like a wedding cake on approach durring heavy
> traffic.  That is, the aircraft will be put in circles, each separated
> vertically by 1000' (FAA regs require either 1000' vertical separation or 1
> mile horizontal separation at all times).  As each plane at the bottom of the
> stack lands, the aircraft above are allowed to descend 1000', and would be
> asked to do so quickly.  Consequently, while holding, the attitude would be 5
> deg up, and while descending would be 5 deg down.

I don't think that's it.  

It's ATC's responsibility to maintain separation in IFR: they'll clear 
aircraft below before they ask aircraft above to descend.  For them to clear 
multiple aircraft simultaneously, they'd have to synchronize all their rates 
of descent, which might not be possible (and I don't believe I've ever heard 
of it).  Maybe one of the air traffic controllers subscribing to the group
could elaborate on this further.

Also:

The 727 is a flying brick (apologies to F-4 fans): if you dropped the nose 
ten degrees while holding, you could easily punch 4000 fpm (or even peg the
IVSI at 6000), I would think.  This creates other problems, and isn't really
desirable in a terminal environment.  

It sounds more like porpoising: improper technique (hands-on or autopilot), a
faulty autopilot, or some kind of a control system failure.


> Adding to this, aircraft would be asked to land quickly when they are at the
> bottom of the stack, so a relatively high landing speed would not be unheard
> of.  I believe that there were some weather delays on June 13th in the
> midsection of the US, from where many connecting flights originate.
> Consequently, there would be a higher traffic flow during the late afternoon.
> I'm speculating here, but it sounds reasonable.

Airline pilots are trained to "fly by the numbers."  Professional pilots will
tend to politely decline ATC requests to "expedite" approaches, and will tend 
to fly approaches safely, how they are trained to.  If they don't land "on 
the numbers" (vis. VRef), they won't make legal stop distances, will land 
long, tend to balloon, etc.  Thus creating a thoroughly unsafe situation.

It is ATC's responsibility to adapt to pilot requirements for safe flight, 
NOT the pilot's responsibility to keep things "orderly" for ATC.  It's best 
when both interests can be served harmoniously, but the buck stops at the 
pilot, always (FAR 91.3).  

Even if the pilot "cooperates" and keeps speed up through the initial 
approach, or while maneuvering, I don't think the 727 would have *any* 
trouble slowing down to the proper reference airspeeds, so at the point the 
passenger would become aware of speed differences (last 1000' of flight), 
it should be well into a standard profile.

Possible explanations for the high speed are that it was a hot day, the 
airplane was unusually heavy (fuel, passengers, cargo), that there were high 
winds aloft (there was a 70-knot wind at 2000' here in Austin, recently, even 
though ground winds were 15 knots), or even that it had a bit of a tail-wind 
on final (I've noticed that controllers won't switch the runway if there's 
a light tail-wind following frontal passage, until traffic dies down).  Or 
maybe a variety of optical illusions: whizzing through scud, for instance, 
can give an illusion of speed.

Perhaps a combination of all these effects.

For reference, a 727-200 at 110,000 lbs has a VRef of 108 knots; at 155,000 
lbs (max. landing weight) 133 knots; to this you can add wind + 1/2 gust, to 
a maximum of 20 knots.  Approach maneuvering speed, flaps up, is 200 knots.  
All standard conditions.




Cheers,


--
Robert Dorsett
Internet: rdd@rascal.ics.utexas.edu
UUCP: ...cs.utexas.edu!rascal.ics.utexas.edu!rdd