50
$\begingroup$

According to Wikipedia (and various other online sources), the control stick on the F-16 does not move. The input values are determined by the amount of force acting on the stick, rather than the deflection angle/distance away from its neutral position like the majority of sticks and yokes. After pilots reported the design as "unnatural", the stick was modified so it would move a little.

Clearly the designers must have thought that having a very stiff control stick is advantageous. Why?

$\endgroup$
9
  • 8
    $\begingroup$ And I thought Airbus sidesticks were bad. I cannot imagine a worse way of controlling an aircraft. $\endgroup$
    – Bianfable
    Aug 30, 2019 at 9:23
  • 33
    $\begingroup$ @Bianfable, wait until they come up with a virtual joystick on a touchscreen... ;) $\endgroup$
    – bogl
    Aug 30, 2019 at 9:24
  • 1
    $\begingroup$ @bogl We've come quite advanced in material engineering that we don't have to worry about a control mechanism wearing out. Steering wheels in cars are not falling off any moment! $\endgroup$
    – kevin
    Aug 30, 2019 at 9:27
  • 1
    $\begingroup$ The short answer is because airplanes are designed by engineers, not pilots. Engineers don't always think like regular folks! ;) $\endgroup$ Aug 30, 2019 at 12:31
  • 4
    $\begingroup$ It's more of less like riding a motorcycle, that the motion of the handle bar doesn't matter (nobody pays attention, and it moves counter-intuitively) and it's more of the force that give the rider the feedback. $\endgroup$ Aug 30, 2019 at 20:18

5 Answers 5

79
$\begingroup$

The stick does not need to move in order for the pilots to sense their inputs!

Humans have very accurate force sensors in their fingers, and no direct position sensors. Proprioception provides positional information as @aroth mentions in a comment, however it is based on muscle tone sensing - force detection. The hand is at the end of a 5-DoF arm, with serial accumulation of measurement errors. Meaning that without looking, we only have a general idea where our hand is due to muscle tone sensing, but we don't need to look at our hand to know exactly how hard it is pushing at an object due to the force sensors in the skin of the fingers. For the human, force is the imminent control feedback factor.

The F-16 can do 9G in manoeuvres, and the pilot is inclined backwards with their arm on an armrest. If this arm needs to move a stick it shifts position, and during a manoeuvre can get grabbed by the load factor - precise control is much easier if the hand, wrist and arm can stay in one position.

A pilot who has flown both the F-16 and the centre stick F-18 has this to say about it:

Flying a side-stick control takes a while to get used to, but once you do, it's a joy. The conformai stick's shape feels very natural (it fits in the hand like a melted candy bar), and it allows easy access to nine of the 16 hotAS controls. Two fully adjustable forearm rests on the right cockpit bulkhead stabilize and isolate the pilot's arm and wrist, so when rattling around the cockpit during turbulence or going after the bad guy, the pilot's arm won't accidentally move and initiate unwanted control inputs.

Initially the stick was fixed completely and pilots would get comfortable with the novel concept of no stick movement pretty quickly. But there were two issues:

  1. Cross-coupling. The control input must be separated in two axes, pitch and roll, and this is difficult to separate if feedback is from force only. In order to aid this, a small bit of movement was introduced with a breakout force in the centre: upon release, the stick moves back to neutral and requires a bit of force to be deflected from there. From the article linked above:

    However, after initial F-16 flight tests, a ¼ inch of stick movement was incorporated to give a small dead band and a nominal breakout force to give better "feel" of a neutral stick because otherwise it was entirely too sensitive. The control harmony is quite good (the pressures required for pitch and roll mix well), but without the capability to physically position the stick, it's easy to contaminate roll inputs with unwanted pitch inputs, and vice versa

  2. When the sticks were fixed, F16s kept getting back with bent sticks. During chasing another plane and pulling, pulling on the stick, they pulled excessively hard while the fly-by-wire was already generating the deflection for maximum load factor. With the stick now moving a little bit, the pilot can feel that the limit is reached when hitting the travel stop at 25 lbs. At this force the acceleration is 9G and pulling harder has no additional effect. (As explained during the course of Human Factors at uni.)

$\endgroup$
4
  • 24
    $\begingroup$ Given you last sentence, I can't imagine the adrenaline level during dogfight: maximum load factor and the pilot pulling hard to exceed that value. $\endgroup$
    – Manu H
    Aug 30, 2019 at 11:22
  • 2
    $\begingroup$ @ManuH 25 pounds is just 11kg. It's not little, but it's very achievable. I regularly carry six-packs of mineral water (6x2 kg), so hitting the travel stop should be no issue for any decent pilot. I don't know how much extra tolerance is there, but it shoudn't be that hard to bend the stick. $\endgroup$ Aug 31, 2019 at 21:04
  • 16
    $\begingroup$ @JohnDvorak I thing you misunderstood. I meant the pilot is already experiencing a load factor of 9g and is still pulling to turn tighter. $\endgroup$
    – Manu H
    Aug 31, 2019 at 23:09
  • 7
    $\begingroup$ "Humans have very accurate force sensors in their fingers, and no direct position sensors." - Not entirely true. We have a sense of proprioception. The methods by which the body senses/reports proprioception are not dissimilar from how it senses/reports direct contact, so I'd be hesitant to dismiss the information as "very vague" (or even as less valuable/clear than touch/force feedback). Certainly I can close my eyes and have no difficulty determining how my arms, wrists, and hands are currently positioned through their entire range of motion. $\endgroup$
    – aroth
    Sep 2, 2019 at 4:41
15
$\begingroup$

With a deflection-sensing (conventional) stick, it takes a certain time to move the stick. If you're trying to do quick manoeuvres going from (say) full nose-up to full nose-down deflection, that fraction of a second can be a significant delay, especially in a fly-by-wire aircraft with powerful, fast-acting actuators. With a force-sensing stick, that delay is removed. Force-sensing sticks also cause less arm fatigue.

Others have pointed out that force-sensing sticks are also mechanically simpler, because of the lack of moving parts. This seems like a small concern, but for military aircraft, which get more frequent inspections and have a very stretched-out supply chain, removing any need for a replacement part can be a big benefit. It's the kind of design criterion where, even if it isn't really a huge factor on this aircraft, an engineer who was burned by a spares or inspection problem on a previous design might consider that a big project risk.

The mechanical simplicity can also make the extra buttons etc. on the stick more reliable, because there doesn't need to be wiring through a rotating joint, which might get trapped or wear insulation. And the fact you don't need space for the stick to move can make it easier to design other components around the stick.

As this project found, the chief downside is that force-sensing sticks are unfamiliar to most pilots, and they take a lot of getting used to. They also make training harder: with a deflection-sensing stick, it's easy to show a student pilot the appropriate stick position for a particular manoeuvre (e.g. to show the progressive pulling back in the flare). This is much harder with a force-sensing stick. Obviously, pilots won't be doing their initial training on this aircraft, but they still need to demonstrate handling characteristics etc. when familiarising on type.

$\endgroup$
0
9
$\begingroup$

The motivation is the same, albeit for stronger reasons, as that for the 1960's Citroën DS's brake button or frein champignon, which responded to foot pressure rather than foot travel. Namely, faster response, to take advantage of the 2500 psi hydraulics in this unorthodox and quite fast car. Moving your toe half an inch instead of your leg half a foot (ahem) can seriously shorten your braking distance.

$\endgroup$
4
  • $\begingroup$ ALL brakes respond to pressure rather than travel. The only difference was shorter travel, which indeed can lead to faster response. But the motivation for aircraft is deeper than that. $\endgroup$
    – Zeus
    Feb 26, 2020 at 0:04
  • $\begingroup$ Of course. Even bicycle brakes apply more force with more pressure, obeying Hooke's law. For aviation, which deeper motivation do you mean? $\endgroup$ Feb 26, 2020 at 2:01
  • 1
    $\begingroup$ Dynamics and kinematics of flight control is more complex. (For brakes, the effect simply depends on the pads pressure). Traditionally (mechanically), actuation is by position, but the effect strongly depends on airspeed. In reversible control (light airplanes), stiffnes changes accordingly, which gives an effect of force control. But with irreversible control this effect is lost, and one needs to design workarounds like variable control stiffener or ratio. Instead, why not just sense force? (Why force and not position is covered in Koyovis' answer; we humans are just better at sensing it). $\endgroup$
    – Zeus
    Feb 26, 2020 at 3:09
  • 2
    $\begingroup$ I’ve owned a Citroen GS when in uni, was fascinated by the demo of hydraulic suspension. It worked great, and at traffic lights could keep entertained by moving the car up & down. But once when driving late at night to a town 300 km away the hydraulic fluid warning light came on, and the car barely made it, was pretty comcerned and no petrol stations were open. An important lesson on passive safety of systems, the hydraulic system powered the brakes as well. $\endgroup$
    – Koyovis
    Nov 26, 2021 at 23:21
5
$\begingroup$

It does move. The F-16 Side Stick Controller has always measured displacement. The YF-16 prototype had a much smaller displacement than production F-16s, but it still moved.

The small throw of the stick allows a much smaller controller that saves weight over direct linkages. Also, stabilizing the arm on the rests and flying with the wrists requires less elbow room in the tight cockpit of what was originally a light daylight fighter. Ergonomically it reduces fatigue by employing smaller muscle groups during basic fighter maneuvers. Designwise, there is stiffer structure on the side of the cockpit than up from the floor in the center console to attach it.

The ultimate loads required to break a side stick are over 300 lbs. No one has ever broken one inflight.

As an engineering principle, you can’t measure force. You measure displacement against a linear spring force - like a fish scale. The side stick uses multiple linear transducers in two axes to measure displacement. Always has.

$\endgroup$
3
  • $\begingroup$ You can measure force, this is how control loading systems in flight simulators work. Strain gauges measure force from the stretch in the metal they’re attached to - which does affirm your statement, granted. $\endgroup$
    – Koyovis
    Nov 26, 2021 at 23:25
  • 2
    $\begingroup$ You aren’t measuring force directly, you are measuring displacement against a linear spring force; whether it is a spring or a bending beam. It is a calculation. Even with a strain gage, the beam has to bend. “Stretching” is displacement. Strain gages are used in simulators and gaming but they are not used to fly a plane. $\endgroup$ Nov 28, 2021 at 14:53
  • $\begingroup$ It sounds like you might have been involved in the design of the aircraft. Thank you sir! The F-16 was a joy to fly. My only complaint was the AOA limiter. I think it was 25 degrees, which is not nearly enough in a slow speed fight. I suppose it's a moot point now with modern weapons, though. $\endgroup$ Oct 25, 2022 at 18:28
1
$\begingroup$

The F-16 was the one of the first production "fly-by-wire" planes without direct mechanical linkage between cockpit controls and aerodynamic controls. The development of electrical-output pressure sensors made it possible to do things a different way.

Why have the weight, mechanical complexity, cost and maintenance overhead of hinges and stuff when you don't need them?

Turned out some movement is required to provide pilot feedback, but the principle remains.

$\endgroup$
2
  • 2
    $\begingroup$ By no means was F-16 the first FBW. Also, its FBW (as well as pressure sensors) were originally analog. Finally, FBW can be made position-sensing with just the same weight and mechanical savings (e.g. Concorde - which also predates F-16), and the question was why even attempt to do force-sensing. $\endgroup$
    – Zeus
    Feb 25, 2020 at 23:59
  • $\begingroup$ Thanks, well spotted. Now added/corrected. But Concorde is a bad comparison and the savings on the stick and rudder pivot mechanisms remain. $\endgroup$ Feb 26, 2020 at 10:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .