Would it be possible for one airliner to tow another?

Question

Comedian/podcaster/QI elf Dan Schreiber asked on Twitter, as proxy for his 4-year old

if it would ever be possible to hook a plane (like a 737) onto the back of another plane, like a train carriage, and then have both take off? I know the answer is no, but just wanted to double check with you, Twitter

and continued

I knew about gliders...Just wondering about the Boeings...in case some one comes in with an XKCD-style answer

For the purposes of focusing our question, I'd like to ask:

can a jet airliner TOW (not piggyback / eat) another airliner from takeoff to landing?

It is pre-noted that this is not a good idea, voids the warranty, etc. However, laying out considerations / difficulties / show stoppers, with particular attention to modern flavors of the 737, or noting which if any airliners would be preferable tow-ers or tow-ees, will be appreciated.

Answer 1

Pic source

Yes it should be possible for an airliner to tow another airliner. In WW2 gliders with paratroopers on board were towed to their drop zones, the technology has been around for a while.

Another interesting story can be found in this link, about capt. Robbie Risner using his F-86 Sabre to push his wing mate's F-86 into friendly airspace during the Korean war. A fighter jet pushing another fighter jet.

Risner told Logan to shut down his engine, now almost out of fuel. Then he gently inserted the upper lip of his air intake into the tailpipe of Logan’s F-86....Miraculously, Risner nudged Joe Logan’s F-86 all the way to Cho Do, maintaining an airspeed of 190 knots and enough altitude to stay out of range of automatic weapons.

One airliner pulling another one would indeed take structural preparation. The wing/body intersection is the strong point of the plane, dimensioned to absorb the forces from the jet engines which will now be provided by the tow ropes. If the tow ropes are attached at this point pitch moments will be produced into both the towed (nose up) and the towing (nose down) aeroplane, which receives an automatic trim correction for the engine thrust.

Issues to tackle:

1. Stability of the towee. The towed plane will need to deal with interference and stability issues, and will need to have its flight controls powered. The B737 does have (very heavy) manual backup modes for the elevators & ailerons, the secondary flight controls require electrical/hydraulic powering. From the APU or from a RAT (which the B737 does not have). The batteries provide 60 mins of backup power, but powered flight controls are essential.

   Longitudinal stability should be OK, since the plane is built to deal with the thrust of the turbofan engines. Directional stability must be improved by leading the tow rope through a support fixed at the node wheel structure. The rope won't always pull straight ahead like the engines do, and this degree of freedom needs to be eliminated as much as possible.

   

2. Wake turbulence. The towed plane will need to evade wake turbulence from the towing plane, as depicted above. So the towed aeroplane can take off first, and remain above the vortex field - or fly below the wake field as in this video. As @Neil_UK mentions in a comment:

   Wake turbulence is not so much a problem, as a feature, for sail-plane towing. Once reasonably competent at straightforward tow-launches, the next exercise is to 'box the tug', that is, flying in, under, over, left and right of the tug plane's wake turbulence. – Neil_UK

   And on the strength of the wake field, from the FAA site

   Weight, speed, wingspan, and shape of the generating aircraft's wing all govern the strength of the vortex. The vortex characteristics of any given aircraft can also be changed by extension of flaps or other wing configuring devices. However, the vortex strength from an aircraft increases proportionately to an increase in operating weight or a decrease in aircraft speed. Since the turbulence from a “dirty” aircraft configuration hastens wake decay, the greatest vortex strength occurs when the generating aircraft is HEAVY, CLEAN, and SLOW.

   So the towing plane can reduce the wake vortex by eliminating the payload, extending its flaps, and speed up to max climb speed (point 3. below)

3. Optimum range. The towing plane needs to provide thrust to overcome drag of two planes. The towed plane has its own wings and provides its own lift, so we'll only need to consider increased drag on the towing plane.

   In the Breguet range equation $\frac{V}{sfc} \cdot \frac{L}{D} \cdot log \frac{W_i}{W_f}$ the towed/towing plane combination has twice drag D, and the corresponding speed for maximum range will be lower than for a single plane.

One thing is clear: it can be done, because it has been demonstrated already: the Eclipse project mentioned in @ErinAnne's comment, a Starlifter towing a Delta Dart on a 1000 ft rope. This report mentions how the attachment point on the Dart was constructed at the nose, how the towing took place successfully (the Dart pilot took his hands off of the controls), and how the wake turbulence introduced some shakiness so stay either below or above the wake field.

Answer 2

Here's a quick take that I will attempt to evolve into a better, more-XKCD-style answer over time:

I believe it's possible, but planning is required.

Trying to do this with fully-loaded planes isn't a good idea; you are more likely to succeed with reduced fuel (perhaps no fuel in the towed plane) and cargo.

This gives you excess thrust available in the towing plane, and every pound less of lift saves some fraction of a pound in towing force (I'm having trouble finding 737-specific numbers right now but Lift / Drag ratios of 15 to 20 aren't uncommon for an airliner and would change over the members of the 737 family).

The towing plane will have to generate both the thrust it needs and the thrust the towed plane would need, plus some additional from the towing bridle not being perfectly straight and taut. For 737s specifically this gets a little iffy just based on the ratio between maximum takeoff weight (MTOW) and empty weight; I don't see any model of the 737 with that ratio as high as 2.0. Perhaps you'd need to reduce the weight of the towed aircraft even more than "empty" (i.e. by removing engines).

There is every reason to believe that the towed aircraft could be rigged in such a way to receive the force, because the engine pylons and wing boxes already have to resist that force and more. Presumably we will be towing at speeds slower than cruising speed to buy ourselves less overall drag and better Lift / Drag ratios. Rigging the towing aircraft may be more complicated; my first guess would be attaching cables to the wing box at the wing roots, but that's a guess.

It will help if the towed aircraft can remain under control to maintain a good following attitude, use additional lift controls like slats and flaps during appropriate phases of the flight, retract and extend the landing gear, etc. It is not clear to me at this time if a towed 737 would have to have a running engine to do so, or if an APU would suffice, or if newer models (or other aircraft like a 787) could run their controls entirely from an electrical connection from the towing aircraft (which could end up being one of the costliest things to engineer in this whole proposal).

I still don't see any reason to believe that particularly long runways should be necessary. Though a 737+737 combo will likely not be able to operate out of all the runways a single 737 would, part of the point of long runways is to be able to abort takeoffs on the ground and we'll assume that among the operating controls that the towed aircraft has, it has working brakes.

Answer 3

It is surely possible in theory, but not that much in practice - at least not without some serious custom engineering.

One big problem is going to be find a point strong enough to attach the towing cable on both planes. Planes that are made to tow / be towed have special strong points.

Also do not forget that when towing, the towed plane usually releases and it lands on his own. This means that you will also need a cable release system and the towed plane will have to have everything required for autonomous flight and landing.

And as someone remarked in the comments, the wake turbulence is going to be extremely, probably prohibitively bad. It is already a problem for much smaller towing aircraft when launching gliders. However if the cable is long enough it should be possible to fly slightly above or slightly below the worst turbulence while keeping the tow angle close to zero.

Towing is very widely used for launching sail-planes and some heavier hang-gliding wings.