7 ms·

I saw a talk a long time ago about the structural aspects of runway design. The most interested fact I remember was that the stresses on the runway generated by

by jld 8mo ago

I saw a talk a long time ago about the structural aspects of runway design. The most interested fact I remember was that the stresses on the runway generated by departures was higher than those of arrivals, as departures repeatedly stress the same part of the runway, while jets land on a much more distributed area of the runway. Plus jets weigh a lot less at arrival than at departure.

WalterBright 8mo ago
When I worked at Boeing, I talked about autoland systems with my lead engineer. He said the autoland was too perfect, as the airplanes would touch down at the same place every time. This caused that place in the runway to suffer severe fatigue damage.
- psunavy03 8mo ago
  IIRC, there was a similar problem on aircraft carrier flight decks, where they had to induce some randomized amount of dispersion to keep the tailhook from hitting the same spot over and over again.
  - refibrillator 8mo ago
    Hmm in distributed computer systems similar patterns exist, e.g. adding jitter to avoid thundering herd effects. This feels like an essential pattern of the universe or something…
    - HPsquared 8mo ago
      "Spread spectrum" is used in EE to spread out the frequency ranges used and reduce interference. The extreme version being CDMA.
      - username135 8mo ago
        Intriguing
  - krisoft 8mo ago
    I work at a self-driving car company and we observed a similar problem when we did some off-road testing on dirt tracks. The cars were too precise and they were cutting deep ruts into the soil. We too solved it by adding a pseudo-random offset to the track.
    - bobthepanda 8mo ago
      Before the current wave of automation there was a previous technology to automate buses using optical sensing and lines in the road which had the same issue.
      - namibj 8mo ago
        If you want rails: build rails.
        bdamm 8mo ago
        There are entire subway systems built with tire-on-concrete where the trains ride precisely the same routes down to the millimeter. Montreal’s is a famous one. These systems are not as efficient as rail, but they are quieter and gentler than the typical subway.
        namibj 8mo ago
        That's still rail. It's just not steel-on-steel. Otherwise you'd have to seriously limit what systems you call "monorail".
        bobthepanda 8mo ago
        The problem is that the optical guided bus was built with the intention of reducing cost, since painting lines is a lot cheaper than building rails. The moment you have to build rail-like things you lose most of the cost advantages.
    - mikepurvis 8mo ago
      I believe Google Maps adds a bit of a rng in which route it will recommend when two otherwise similar in distance/time. Obviously the traffic input also affects this, but that's a slower feedback mechanism; better to distribute the cars all leaving the airport for downtown across the 2-3 possible routes upfront rather than dumping them all onto route A until it's a jam and then all onto route B until it's a jam, etc.
  - cucumber3732842 8mo ago
    Citation please. Doesn't pass the sniff test. I suspect the ocean in its various states provides quite a bit of dispersion. Replacing deck plates on a ship is a normal part of maintenance. I find it very hard to believe they'd induce randomness rather than having just that one plate get a different hardness (I know some people will screech about that but trust me, the warship industry is well practiced at such things).
    - Onavo 8mo ago
      Not sure about the ocean industry but carrier landings have full autoland support for a long time (see e.g. magic carpet).
      - psunavy03 8mo ago
        I wouldn't call it a "long time." MAGIC CARPET per se, especially as a default mode of operation, did not become mainstream until the last 10 years, and that's only for the aircraft that have it.
    - Cerium 8mo ago
      I interpreted the effect here not to be on the deck plates but to be the point of impact between the cable and the hook. Sudden bends in cables can wear them fairly quickly in the immediate vicinity. I'm sure you can extend cable life proportionally to the spread of the loading.
      - cucumber3732842 8mo ago
        That makes far more sense.
    - psunavy03 8mo ago
      Citation is my professional network as a former Naval Flight Officer who flew off aircraft carriers, and it's not the deckplates, it's the nonskid coating that goes on top of them. It's expected to wear and be replaced every so often, but when you have to do it more often than that, it becomes a pain logistically and operationally. And the entire point of the system is to counter the dispersion the ocean provides, much like a Naval Aviator is already required to in order to trap successfully. You are literally flying through a very narrow window that is moving in three dimensions.
  - gosub100 8mo ago
    Similar problem with catenary wires on electric trains wearing into the contacts. The wires zigzag to distribute the wear.
pbreit 8mo ago
Isn't it the opposite? Landing stress a sub-section of the runway while departures stress a larger portion? I'd be surprised that a heavier plane on takeoff exerts more force on the runway than a lighter plane landing. And as the departing plane goes faster, doesn't the lift take stress off the runway?
- pc86 8mo ago
  It's the same principle as walking on snow in normal shoes vs. snow shoes. Taking off is normal shoes, a lot of pressure concentrated at the very first part of the runway. Landing is snow shoes because it's distributed across more of the physical surface, and the plane weighs a lot less when it lands anyway.
- rtkwe 8mo ago
  Planes all start their take off from basically the same position and stress the whole runway, slowly lowering as lift increases, but at their highest weight.
  - tadfisher 8mo ago
    And this is because pilots are trained to keep their nose gear on the centerline, and there are relatively few aircraft types in use which receive the "heavy" after their flight number over ATC. So wheels are going to roll over the exact same tracks repeatedly.
    - krisoft 8mo ago
      > pilots are trained to keep their nose gear on the centerline Funily I was learning to fly at a grass strip and we were told to vary our positioning left and right on the runway for exactly this reason. In practice it meant that as we were taxiing to the runway my instructor would tell me “Today we are taking off left/right of center to avoid damaging the grass too much.”
- duskwuff 8mo ago
  > And as the departing plane goes faster, doesn't the lift take stress off the runway? Only for a short period between rotation and liftoff. Most of the takeoff roll is spent building up horizontal speed; the pilot doesn't command the aircraft to pitch up before it's ready to lift off.
cmurf 8mo ago
Yeah, the higher departure stress due to greater fuel weight at takeoff was mentioned in this video. I'm now curious about the engineering of the displaced threshold. This is a portion of the runway that aircraft can taxi onto and use for takeoff but not for landing. I thought (assumed) that the landing was harder on the runway surface than takeoffs, hence the displaced threshold wasn't designed for that force. The displaced threshold could also be used to ensure obstacle and terrain clearance on landing - simply disallow that portion from being used in order to create an offset from the obstacle. But I don't know whether this is a very common reason for displaced threshold usage. -- Video also mentions https://skybrary.aero/ https://skybrary.aero/ which I'd not heard of previously. Looks neat. I'll have to check it out.
metadat 8mo ago
Do you know how they keep the concrete from cracking? All the pads in general are in way better shape than my driveway, and the driveway has decent support underneath and is subjected much, much less load. Maybe they use plentiful jagged interlocking sharp granite as the base l? (like railroad track foundation) Next time you're at SFO, SJC, or any other major hub sitting in the plane before it backs out of the gate take a second to gaze upon and admire how pristine all the concrete pads are, it's really impressive.
- quickthrowman 8mo ago
  Mostly the extensive subgrade work, as I understand. Similar to a road, there’s a bed of sand and aggregate under the concrete surface. The concrete they use is very precisely mixed to a specification and then it’s tested for adherence to the spec. A runway is also going to be 3-4x thicker than a 4-6” thick driveway slab. Probably they also use fiberglass or PVC coated rebar instead of plain steel rebar. Definitely not an expert here but I can read a civil site plan and hire civil site work subcontractors frequently. Here’s some information on concrete testing: https://www.concrete.org/frequentlyaskedquestions.aspx?faqid=1 https://www.concrete.org/frequentlyaskedquestions.aspx?faqid... Also, Grady is one of the best creators on YouTube, I can’t help but watch his full videos whenever they pop up. I always learn something, even if I’m familiar with the subject. Edit: Granite is not one of the listed materials in Part 4 - Base Courses of the FAA runway construction guide, here’s the entire thing for reference - https://www.faa.gov/airports/engineering/construction_standards https://www.faa.gov/airports/engineering/construction_standa...
- blauditore 8mo ago
  Not sure if that's a serious question, but your driveway might lack a proper foundation, so the surface is moving and cracks. Also, it's likely not concrete, but tarmac (which is much softer).
  - Arainach 8mo ago
    A significant number of American driveways are concrete. I'm not going to look up numbers, but I would have to believe that more are concrete than asphalt/tarmac. Unpaved driveways could outnumber both, who knows, but most people with paved driveways have concrete.
gerdesj 8mo ago
"stress" (in engineering terms) has a particular meaning and is not a generic term. It is not really a synonym for "forces" or "what makes other stuff break"! Let's look at just the downward forces: I need some quick figures 1 - an early Boeing 747: 330 tonnes (metric) fully loaded and 160 tonnes empty. A tonne is 1000 Kg. According to 2: 240 feet per minute vertical is a hard landing which about 1.2m/s. 60 - 180 is considered ideal, so let's go for about 150fpm which is about 0.7m/s. We have to estimate the maximum downward force on take off. At the point of just before lift off, the plane has rotated to say, let's say 45 degrees, and its engines are delivering enough force and its wings are delivering enough force to push it into the air. Surely at take off, that vertical force is simply the weight of the aircraft, which has remained the same all the time. It doesn't suddenly push down harder than its weight, that's just what it feels like for a passenger. So let's allow our jet to be empty on landing and also let the acceleration due to gravity be 10m/s/s So what is the instantaneous downward force of a mass of 160 tonnes dropping at 0.7 m/s compared to a dead weight load of 330 tonnes. Both are in a gravitational field of 10 m/s/s (or m^s-2). Now this is where I get a bit lost because force = mass x acceleration and the landing plane is descending at a constant velocity of 0.7 m/s. Mind you, the ascending plane is also ... ascending, or will do but it does not have an instantaneous upward velocity so at wheels off it has a vertical acceleration of zero. Help! 1 https://measuringly.com/how-much-does-boeing-747-weigh/ https://measuringly.com/how-much-does-boeing-747-weigh/ 2 https://aviation.stackexchange.com/questions/47422/what-is-the-typical-touchdown-vertical-speed-of-a-large-airliner https://aviation.stackexchange.com/questions/47422/what-is-t...
- laurencerowe 8mo ago
  Isn’t the acceleration here just the difference in velocity between the plane and the runway so 0.7 m/s/s?
  - petschge 8mo ago
    IF the velocity would change from 0.7m/s to 0m/s over one second, the acceleration would be 0.7m/s/s. But if the time span over which that velocity change is (much) shorter, the acceleration would be (much) higher.
roelschroeven 8mo ago
According to the video the much larger weight is the main or even only cause of takeoff exerting more load on the runway than landings.

by jld 8mo ago

WalterBright 8mo ago

When I worked at Boeing, I talked about autoland systems with my lead engineer. He said the autoland was too perfect, as the airplanes would touch down at the same place every time. This caused that place in the runway to suffer severe fatigue damage.

psunavy03 8mo ago
IIRC, there was a similar problem on aircraft carrier flight decks, where they had to induce some randomized amount of dispersion to keep the tailhook from hitting the same spot over and over again.
- refibrillator 8mo ago
  Hmm in distributed computer systems similar patterns exist, e.g. adding jitter to avoid thundering herd effects. This feels like an essential pattern of the universe or something…
  - HPsquared 8mo ago
    "Spread spectrum" is used in EE to spread out the frequency ranges used and reduce interference. The extreme version being CDMA.
    - username135 8mo ago
      Intriguing
- krisoft 8mo ago
  I work at a self-driving car company and we observed a similar problem when we did some off-road testing on dirt tracks. The cars were too precise and they were cutting deep ruts into the soil. We too solved it by adding a pseudo-random offset to the track.
  - bobthepanda 8mo ago
    Before the current wave of automation there was a previous technology to automate buses using optical sensing and lines in the road which had the same issue.
    - namibj 8mo ago
      If you want rails: build rails.
      - bdamm 8mo ago
        There are entire subway systems built with tire-on-concrete where the trains ride precisely the same routes down to the millimeter. Montreal’s is a famous one. These systems are not as efficient as rail, but they are quieter and gentler than the typical subway.
        namibj 8mo ago
        That's still rail. It's just not steel-on-steel. Otherwise you'd have to seriously limit what systems you call "monorail".
        bobthepanda 8mo ago
        The problem is that the optical guided bus was built with the intention of reducing cost, since painting lines is a lot cheaper than building rails. The moment you have to build rail-like things you lose most of the cost advantages.
  - mikepurvis 8mo ago
    I believe Google Maps adds a bit of a rng in which route it will recommend when two otherwise similar in distance/time. Obviously the traffic input also affects this, but that's a slower feedback mechanism; better to distribute the cars all leaving the airport for downtown across the 2-3 possible routes upfront rather than dumping them all onto route A until it's a jam and then all onto route B until it's a jam, etc.
- cucumber3732842 8mo ago
  Citation please. Doesn't pass the sniff test. I suspect the ocean in its various states provides quite a bit of dispersion. Replacing deck plates on a ship is a normal part of maintenance. I find it very hard to believe they'd induce randomness rather than having just that one plate get a different hardness (I know some people will screech about that but trust me, the warship industry is well practiced at such things).
  - Onavo 8mo ago
    Not sure about the ocean industry but carrier landings have full autoland support for a long time (see e.g. magic carpet).
    - psunavy03 8mo ago
      I wouldn't call it a "long time." MAGIC CARPET per se, especially as a default mode of operation, did not become mainstream until the last 10 years, and that's only for the aircraft that have it.
  - Cerium 8mo ago
    I interpreted the effect here not to be on the deck plates but to be the point of impact between the cable and the hook. Sudden bends in cables can wear them fairly quickly in the immediate vicinity. I'm sure you can extend cable life proportionally to the spread of the loading.
    - cucumber3732842 8mo ago
      That makes far more sense.
  - psunavy03 8mo ago
    Citation is my professional network as a former Naval Flight Officer who flew off aircraft carriers, and it's not the deckplates, it's the nonskid coating that goes on top of them. It's expected to wear and be replaced every so often, but when you have to do it more often than that, it becomes a pain logistically and operationally. And the entire point of the system is to counter the dispersion the ocean provides, much like a Naval Aviator is already required to in order to trap successfully. You are literally flying through a very narrow window that is moving in three dimensions.
- gosub100 8mo ago
  Similar problem with catenary wires on electric trains wearing into the contacts. The wires zigzag to distribute the wear.

pbreit 8mo ago

Isn't it the opposite? Landing stress a sub-section of the runway while departures stress a larger portion? I'd be surprised that a heavier plane on takeoff exerts more force on the runway than a lighter plane landing. And as the departing plane goes faster, doesn't the lift take stress off the runway?

pc86 8mo ago
It's the same principle as walking on snow in normal shoes vs. snow shoes. Taking off is normal shoes, a lot of pressure concentrated at the very first part of the runway. Landing is snow shoes because it's distributed across more of the physical surface, and the plane weighs a lot less when it lands anyway.
rtkwe 8mo ago
Planes all start their take off from basically the same position and stress the whole runway, slowly lowering as lift increases, but at their highest weight.
- tadfisher 8mo ago
  And this is because pilots are trained to keep their nose gear on the centerline, and there are relatively few aircraft types in use which receive the "heavy" after their flight number over ATC. So wheels are going to roll over the exact same tracks repeatedly.
  - krisoft 8mo ago
    > pilots are trained to keep their nose gear on the centerline Funily I was learning to fly at a grass strip and we were told to vary our positioning left and right on the runway for exactly this reason. In practice it meant that as we were taxiing to the runway my instructor would tell me “Today we are taking off left/right of center to avoid damaging the grass too much.”
duskwuff 8mo ago
> And as the departing plane goes faster, doesn't the lift take stress off the runway? Only for a short period between rotation and liftoff. Most of the takeoff roll is spent building up horizontal speed; the pilot doesn't command the aircraft to pitch up before it's ready to lift off.

cmurf 8mo ago

Yeah, the higher departure stress due to greater fuel weight at takeoff was mentioned in this video. I'm now curious about the engineering of the displaced threshold. This is a portion of the runway that aircraft can taxi onto and use for takeoff but not for landing. I thought (assumed) that the landing was harder on the runway surface than takeoffs, hence the displaced threshold wasn't designed for that force. The displaced threshold could also be used to ensure obstacle and terrain clearance on landing - simply disallow that portion from being used in order to create an offset from the obstacle. But I don't know whether this is a very common reason for displaced threshold usage. -- Video also mentions https://skybrary.aero/ https://skybrary.aero/ which I'd not heard of previously. Looks neat. I'll have to check it out.

metadat 8mo ago

Do you know how they keep the concrete from cracking? All the pads in general are in way better shape than my driveway, and the driveway has decent support underneath and is subjected much, much less load. Maybe they use plentiful jagged interlocking sharp granite as the base l? (like railroad track foundation) Next time you're at SFO, SJC, or any other major hub sitting in the plane before it backs out of the gate take a second to gaze upon and admire how pristine all the concrete pads are, it's really impressive.

quickthrowman 8mo ago
Mostly the extensive subgrade work, as I understand. Similar to a road, there’s a bed of sand and aggregate under the concrete surface. The concrete they use is very precisely mixed to a specification and then it’s tested for adherence to the spec. A runway is also going to be 3-4x thicker than a 4-6” thick driveway slab. Probably they also use fiberglass or PVC coated rebar instead of plain steel rebar. Definitely not an expert here but I can read a civil site plan and hire civil site work subcontractors frequently. Here’s some information on concrete testing: https://www.concrete.org/frequentlyaskedquestions.aspx?faqid=1 https://www.concrete.org/frequentlyaskedquestions.aspx?faqid... Also, Grady is one of the best creators on YouTube, I can’t help but watch his full videos whenever they pop up. I always learn something, even if I’m familiar with the subject. Edit: Granite is not one of the listed materials in Part 4 - Base Courses of the FAA runway construction guide, here’s the entire thing for reference - https://www.faa.gov/airports/engineering/construction_standards https://www.faa.gov/airports/engineering/construction_standa...
blauditore 8mo ago
Not sure if that's a serious question, but your driveway might lack a proper foundation, so the surface is moving and cracks. Also, it's likely not concrete, but tarmac (which is much softer).
- Arainach 8mo ago
  A significant number of American driveways are concrete. I'm not going to look up numbers, but I would have to believe that more are concrete than asphalt/tarmac. Unpaved driveways could outnumber both, who knows, but most people with paved driveways have concrete.

gerdesj 8mo ago

"stress" (in engineering terms) has a particular meaning and is not a generic term. It is not really a synonym for "forces" or "what makes other stuff break"! Let's look at just the downward forces: I need some quick figures 1 - an early Boeing 747: 330 tonnes (metric) fully loaded and 160 tonnes empty. A tonne is 1000 Kg. According to 2: 240 feet per minute vertical is a hard landing which about 1.2m/s. 60 - 180 is considered ideal, so let's go for about 150fpm which is about 0.7m/s. We have to estimate the maximum downward force on take off. At the point of just before lift off, the plane has rotated to say, let's say 45 degrees, and its engines are delivering enough force and its wings are delivering enough force to push it into the air. Surely at take off, that vertical force is simply the weight of the aircraft, which has remained the same all the time. It doesn't suddenly push down harder than its weight, that's just what it feels like for a passenger. So let's allow our jet to be empty on landing and also let the acceleration due to gravity be 10m/s/s So what is the instantaneous downward force of a mass of 160 tonnes dropping at 0.7 m/s compared to a dead weight load of 330 tonnes. Both are in a gravitational field of 10 m/s/s (or m^s-2). Now this is where I get a bit lost because force = mass x acceleration and the landing plane is descending at a constant velocity of 0.7 m/s. Mind you, the ascending plane is also ... ascending, or will do but it does not have an instantaneous upward velocity so at wheels off it has a vertical acceleration of zero. Help! 1 https://measuringly.com/how-much-does-boeing-747-weigh/ https://measuringly.com/how-much-does-boeing-747-weigh/ 2 https://aviation.stackexchange.com/questions/47422/what-is-the-typical-touchdown-vertical-speed-of-a-large-airliner https://aviation.stackexchange.com/questions/47422/what-is-t...

laurencerowe 8mo ago
Isn’t the acceleration here just the difference in velocity between the plane and the runway so 0.7 m/s/s?
- petschge 8mo ago
  IF the velocity would change from 0.7m/s to 0m/s over one second, the acceleration would be 0.7m/s/s. But if the time span over which that velocity change is (much) shorter, the acceleration would be (much) higher.

roelschroeven 8mo ago

According to the video the much larger weight is the main or even only cause of takeoff exerting more load on the runway than landings.