Why do airplanes ride so bad (on the runway)?

Factual Questions
1

I was thinking of this last week at the end of a lengthy business trip, when I wheeled my rented car from the hotel to Pittsburgh airport, and marveled again at the smooth, comfortable ride (it was a new Chrysler Sebring, and it really soaked up the bumps). Then I climbed aboard my flight (in this case an EMB-145 regional jet), and as we made the lengthy taxi out to the runway, I had occasion to note, as I always do, that while on the ground the thing rode like an overloaded dump truck. The plane clanked and shuddered over what seemed to be almost invisible defects in the surface, with the shock from every single unevenness transferred right to my bottom. I bet if the famous princess of the fairy tale were aboard, and the plane rolled over a pea, that she would have noticed it.

OK, I know that ground travel is a relatively small proportion of a plane’s usage and perhaps a smooth runway ride is not a priority, but why do planes ride so bad in the taxi? It would seem to me that there must be a lot of travel built into the landing gear to allow it to soak up landing loads, so it’s not immediately apparent. As a follow-up question, do aircraft designers consider the amount of ground travel (and consequent shock loads/vibration) when determining the useful lifespan of an airframe?

2

My guess is that the suspension has to be really stiff to support a multi-ton aircraft hitting the runway at >100 miles an hour.

3

That, and the fact that car suspensions are very heavy and complex. In an airplane, you want light and simple. The airplane spends only a small part of its time on the runway and taxiways, and it’s going very slow for most of it, and the runways are very smooth compared to typical roads. This allows the airplane maker to save weight and cost by making the landing gear simpler.

In my old Grumman, the shock absorpers in the suspension were the tires, and the spring steel landing gear legs. That’s it. In our old Mooney, the ‘suspension’, such that it was, was a stack of rubber biscuits on each gear leg. Other aircraft have Oleos, which are basically an oil-filled shock absorber.

Another wrinkle in larger airplanes is the need to fold the landing gear and stow it away. Think of how complex it would be if you had a sophisticated suspension that you had to fold up and deploy every time you used it. And how much greater the chance of a gear failure would be.

Given all that, providing a slightly smoother taxi experience is not even close to being worth the additional cost and weight and risk.

4

Former airline pilot …

The travel range of landing gear is less than you’d think, on the order of 2 feet for a large Boeing and more like 12 inches tops for an EMB-145.

To meet regualtory requirements, the gear has to absorb the equivalent of the airplane falling straight down at ~15 mph. The gear cannot bottom out; all that energy has to be absorbed within the travel.

The result is the gear has a very stiff spring rate. That means small lumps & bumps are transmitted almost undamped into the fuselage.
Ref your question about ground loads being a factor in design life, almost certainly not. Turbulence loads & touchdown loads are much larger and more frequent than the comparatively mild rolling along the runway/taxiway loads.

5

You’d be right (sort of) but it is far more complex than just spring stiffness in a commercial airliner. Airliner landing gear, like the suspension in a modern car, has two components in the longitudinal (to the strut) direction: a spring, which provides a constant stiffness that increases resistance in proportion deflection, and a dashpot or damper, which has a variable resistance based upon some proportion (not necessarily a linear function) to the speed of suspension travel. On most modern cars these are combined together into a single part called a strut, which is what keeps the chassis from collapsing upon the axles.

Large aircraft landing struts have a long travel, and the dampers are typically very compliant at low speeds, which allows this springy, bobbing travel when driving across the tarmac, but provides adequate stiffness at higher strut travel speeds to resist aircraft landing loads without overtravel or destabilizing the aircraft. The cost of this is heavier, longer landing gear, and as Sam Stone points out, this is a parasitic cost in terms of the aircraft’s primary functional mode of locomotion, which is to fly. Larger airliners to have a fairly sophisticated landing gear, but for something like and Embraer, there’s definitely a limit on how much weight, and thus how much strut travel, is permissible. (There are also other limits based upon the loading envelope and ground handling characteristics, but we’ll ignore those for the purposes of this discussion.) It’s also the case that the unsprung weight–that not supported on springs–of the landing gear is pretty light. This is good from a handling performance standpoint (hence why high performance cars like to use aluminum wheels, and other lightweight components) but it also means that it doesn’t serve as much of an inertial damper to loads transmitted to the airframe.

Those “invisible” flaws in the surface of a taxiway aren’t so invisible, either. There’s probably a permissible maximum, but some taxiways are pretty rough. The actual runway, however, has definite limits on flatness, crowning, washboarding, and other defects, and even at very high takeoff speeds the ride is pretty smooth. (It doesn’t hurt that the faster you go, the more weight is supported by the wings.) It all looks small from above, but the flaws are real New York City potholes from the surface.

To answer the second question, the loads seen during taxiing are insiginificant compared to those that the airframe sees in flight and during landing. For the airframe, the life limit is typically based upon fatique in the wing joints and/or vertical stabilizer, and cracking in the aluminum skin of the wings. (Yes, there are often cracks in the wings of a used airframe, and they’re allowed to grow to a specified length before repair. This is a very well understood phenomenon, and at least in the U.S. and Europe there are very stringent requirements on inspection and repair of airframes, so the failure of a commercial aircraft from airframe structural failure is almost unheard of, short of some other aggravating circumstances.) Taxiing is such a short part of the process, and the loads on the wings–which are based strictly on their own weight–is much lower, albeit in the opposite direction of flight dynamics–that they aren’t really an issue. It may seem rough to you, but taxiing loads are quite a bit less than loads from even moderate turbulence.

Stranger

6

Imagine the equivalent speed bump to this hit and tell me how the Chrysler Sebring would handle it.

7

Put me down as skeptical here - I’m a pilot, and have never heard of this.

I would have said that pretty much any visible crack in a structural component would render a plane unairworthy. (In view of how readily cracks propagate in aluminum, how can it be safe to postpone a repair?) I’ll concede that there might be a few unusual cases where this is done, but surely it can’t be a common policy?

8

Some cracks are okay, but not in anything structural, so far as I know. Cracks in some fairings and even windshields can be stop-drilled and made serviceable.

9

I’m both a pilot and an aerospace engineer, and it’s true. Details vary by model and usage, of course. If you made every airplane strong enough never to propagate a defect into a crack anywhere, it would never get off the ground.

A factor that hasn’t been mentioned yet is the small *size * of aircraft tires. A typical light aircraft has 12 inch OD rubber (on 6-inch rims), and you can feel every little bump or pit they roll over. The contact footprint of an airliner tire, in proportion to its weight, is pretty small compared to that of any car (no, I don’t have numbers handy, but just look at one and it’s clear) - that makes it difficult for the tires to attenuate any shock loads without transmitting them into the gear.

Plus, the main landing gear is normally placed just slightly aft of the CG, in the interest of keeping the airspeed required to rotate during takeoff low, and in minimizing sudden pitchdown on landing. There is little moment arm to attenuate an impact load from the gear, and little structure to dampen it before it reaches the passenger compartment, so you just have to live with it.

I might add that the 727 is (was) noted for smoothness of landings and taxiing, due to the travel of its trailing-link suspension. It was tricky to land, though, because the mains were unusually far aft of the CG - when entering the flare, the nose had to go down, not up, or the mains would slam into the runway and be followed by a fierce pitchdown onto the nose gear. A friend who used to be a 727 captain has told me how much trouble it was for any pilot new to it to get used to that, and it was easy to forget once learned, too.

10

Some very interesting answers here. I know that some (relatively expensive) cars have spring/shock combinations where the rate can be varied to alter the ride and handling, and I was wondering why planes might not have something similar (soft for taxying, harder for touchdown, for example). I suppose weight, cost and reliability would be the issues. Thanks.

11

I’ll certainly agree with this. But I question whether any crack in any meaningful part of a wing can be accepted.

12

Can you give an example?

13

I’m sure that’s right. No doubt the manufacturers’ response to complaints of rough taxiing would be that the taxiways should be smooth.

14

Metallurgical Engineer checking in. (At least that’s what I studied. Not doing that now.)
We need to understand something about cracking. Any structure that is subjected to cyclic loading needs to be designed to withstand cracking. This is true even if those loads are well below the material’s yield stress. What happens is this:
When a structure is laoded there are physical regions in the structure where the stresses are concentrated. Typically at sharp corners or regions designed with a small radius. (Ever wonder why aeroplane windows are rounded?) At these stress concentration points a small crack may propagate a microscopic amount. This can occur even when the load stresses are ignificantly less than the yield stress of the material.
Now, planes are made from aluminium whichg has a face-centred-cubic crystaline structure. (Compared with steel which has a body-centred cubic structure.) One of the peculiar features of FCC metals is that there is no threshhold below which a crack won’t propagate. In theory this means that you could load and unload a plane wing with a grain of sand and given enough cycles you will observe a visible crack. The situation will not be resolved merely by making the wing’s structure bigger.

The normal practice is to design wings to still function correctly when there is a crack up to a certain size, and include in the design a maintenence schedule that stipulates repair when the crack has reached a certain length.
The only other alternative is to build the plane from a BCC metal, which would involve increased cost or weight.

15

Look for “hydropneumatic suspension” and you’ll find that there are several cars–not all of them expensive–have offered this, including Citroën, Peugeot, British Leyland, and Mercedes-Benz. (Okay, the Mercedes was the top of the line 450 SEL 6.9, which I can’t desist from pointing out was the car featured in Ronin, being driven by Jean Reno in that awesome chase scene in Nice.) This requires a hydraulic pump with fittings and lines, though; not something you want to have hanging off of your landing gear, and certainly not for the short time spent taxiing. As Xema notes, aircraft manufacturers would argue that they’re not building over the road vehicles and that the taxiway should be graded to an acceptable level.

Facinating discussion on the metallurgical point of view. Not being a materials person myself (I know just enough to keep myself out of trouble, and know when to hand off to a proper metallurgist or polymer composite expert), I tend to come at it from the mechanics point of view. Aluminum is a material with no set yield point or fatigue strength; its “yield strength” is defined by a proof load limit to a given strain (typically 0.1% or 0.2%) and its capability to resist repeated, cyclic loading is based on the safe-life for the expected lifetime (i.e. the cycle load limit at which it is well below ultimate failure). That is, aluminum (and as you point out, other FCC materials) are in a state of progressive failure, and you have to design such that the cummulation of the loads is substantially less (generally by a couple orders of magnitude) than that necessary to achive catastrophic structural failure in the member as designed. Cracking is a particular issue; it’s really hard to prevent microcrack formation in any cyclically-loaded part, and with aluminum, it is important to not just select the strongest material, but one that will maintain ductility over the expected load range and life so it isn’t work-harded to a brittle failure state where crack propagation accelerates (literally) exponentially. Most wing surfaces and ductile joints are designed using a more compliant grade of aluminum than the high tensile structural grades. Even then, aluminum can be a tricky material, and when you’re stressing it highly, over a long lifespan, and under uncertain environmental conditions, visual and ND inspection are vital in any critical failure point article.

Stranger

16

Plenty, along with the full theory, in the USAF Handbook for Damage Tolerant Design, if you want to dig in. It’s pretty readable as engineering texts go, actually.

17

Part of it is that aircraft have three point suspension. This is the minimum required for stability, and weight considerations dictate that few aircraft will have more than this.

With four wheels, one can unweight as it goes over a hole, or bear aditional weight while two (the adjacent ones) unweight as one goes over a bump.

With three wheels, all the wheels bear about the same weight all the time. (at very low speeds anyway, so ignoring suspension dynamics) That means the aircraft must tip to follow the terrain under it.

This is very noticeable the first time one rides a motorcycle with a sidecar. It “feels” even more unstable than it is.