32
$\begingroup$

Pretty much what the title says. Surely a circular window would allow more window area while not compromising on strength?

Some planes(Gulfstream jets) do have circular windows. Is there a reason for this? Does it have something to do with fitting windows without in between structural components?

$\endgroup$
6
  • $\begingroup$ i think, honestly, it's just an "how else would it be?" issue. Note that in a house, 1000% of windows are horizontal, not vertical or square! One could equally ask "why are the doors vertical and not square". $\endgroup$
    – Fattie
    Jul 29, 2020 at 13:32
  • 3
    $\begingroup$ @Fattie I'm not sure where you've seen that all house windows are horizontal and neither vertical nor square. All the windows on the building right in front of me are clearly vertical, and there are plenty like that around here. Similarly, the very common vertical sliding window is... vertical. $\endgroup$
    – jcaron
    Jul 29, 2020 at 16:28
  • $\begingroup$ @jcaron so sorry, I made a typo for "vertical" of course! can't edit now. $\endgroup$
    – Fattie
    Jul 29, 2020 at 16:36
  • 2
    $\begingroup$ @Fattie Conversely, there are also LOTS of horizontal windows, like the one I'm sitting in front of :-) There are really all sorts of shapes for house windows... $\endgroup$
    – jcaron
    Jul 29, 2020 at 17:14
  • 5
    $\begingroup$ @Fattie in that case this will blow your mind $\endgroup$
    – Chris H
    Jul 30, 2020 at 7:48

3 Answers 3

48
$\begingroup$

It's done mainly for the passengers' viewing benefit in making structural trade offs over total opening area. For a given area, a round window has less "meat" of fuselage structure between them. Aesthetically, a round window looks too much like a ship's porthole unless you make it really large, and then it looks like a fish bowl.

So for a given area, a window that is elongated vertically allows for more variation in eye height than a circular one where the viewing area is equally distributed vertically and laterally, and provides more fuselage structure between the windows (important as an inflated tube in circular tension, or under "hoop stress").

The 787 went with very large passenger windows to take advantage of the carbon structure and bring as much light inside as possible. But they still stayed with the round-corner vertical rectangular shape so that short and tall people both get a good view.

The CRJ200, as a corporate jet derivative, suffered from very small windows that were set low, and if you're ridden in one and are tall, you have to bend over to see out. When the follow-on model, the 700, was in development, a lot of work went into moving the windows up and enlarging them so that most people got a decent view (actually it was solved by moving the floor down a bit).

And finally, actually the Gulfstream's windows aren't round, but ovalized horizontally. This is likely to accommodate lateral seating variations as a priority over vertical variation because of the corporate seating arrangement and is disadvantageous structurally in terms of the fuselage tube's hoop strength.

$\endgroup$
6
  • 4
    $\begingroup$ If they cared that much about the passengers' viewing benefit, you'd think they'd line up the windows with the seats so you're not craning your neck forward or backward to be able to look out. (Of course this is because most planes were originally designed for fewer rows of seats, and they just keep cramming more in there to increase revenue at the expense of passenger comfort. Just wish they'd take that into account when building the hulls on newer planes...) $\endgroup$ Jul 29, 2020 at 13:59
  • 10
    $\begingroup$ @DarrelHoffman Each airline can put seats inside the body however they see fit, so it's not really possible for the manufacturer to build one plane that aligns windows with every airline's seat configuration - airlines would have to agree to a fixed seat pitch across the industry for this to be possible. Blame the airlines, not the manufacturer. $\endgroup$ Jul 29, 2020 at 19:58
  • $\begingroup$ Even more than @NuclearWang says, considering the typical widebody with multiple seat classes (and therefore seat pitches), the window spacing wouldn't be constant within a single body. That would make structural aspects of the design even more complex. And you'll never make an integer ratio between all of economy, premium economy, business, and first class seat pitches $\endgroup$
    – Chris H
    Jul 30, 2020 at 7:51
  • 2
    $\begingroup$ It's also worth noting that the rectangular windows caused extra localized hull stressing and caused some flight to fail on the De Havilland Comet. en.wikipedia.org/wiki/… $\endgroup$ Jul 30, 2020 at 17:15
  • $\begingroup$ It wasn't necessarily the square window's fault. You can make the window corners square if you accommodate that in the structure with lots of meat where the stress concentration is. The effects of pressurization wasn't a total mystery, with the designers unknowingly blundering along. That is a bit of a myth. The root problem was they simply messed up the calculations. They thought the square corners were strong enough, but under calculated. $\endgroup$
    – John K
    Jul 30, 2020 at 17:26
26
$\begingroup$

Speaking in ship terms (from which airplane structure has evolved), modern passenger airliners use longitudinal framing, which relies on a lot of small tightly spaced longitudinal members (stringers) placed across heavier, widely spaced frames. This design improves overall lengthwise strength.

Any cuts in the framing require local reinforcement, adding weight.

Cutting across the stringers to make the windows is relatively cheap, as there's so many of them. Additionally, the ones that contribute the most to structural integrity are at the top and the bottom, acting as the flanges of a beam, carrying the tail and the nose and supported by the wings.

Cutting across the frames is not cheap, requiring very significant reinforcement. This is especially worse because the section with windows takes up 70% of the fuselage's length, but less than 10% of its perimeter. So percentage-wise the loss in frames is much greater.

Passenger airliners generally cut every other frame to fit the windows. This is the sole factor determining their window spacing. Cutting more stringers for a vertically-extended oval (or, for anything modern, rounded-rectangle) window is lighter than cutting, say, 2 out of every 3 frames.

Bizjets, simply put, can afford to spend more weight on passenger comfort and pleasure, as that's a more important part of a bizjet's value proposition than fuel economy.

$\endgroup$
2
  • $\begingroup$ The SE 210 Caravelle had teardrop windows. $\endgroup$
    – Léa Gris
    Jul 29, 2020 at 18:30
  • 1
    $\begingroup$ Hadn't thought about the fact that removing material for the window ends up in a net increase in weight, due to the reinforcement. $\endgroup$
    – Cireo
    Jul 29, 2020 at 20:01
6
$\begingroup$

I see two good answers, but would like to add a bit to that from another perspective. I don't know if it's the reason, but it definitely is taken into account.

Stress

Stress is the force per area. It's used for analysis of structures. The most important thing is that too much stress can lead to (fatigue) failure.

Aircraft cabins are pressurised. The result of this is that the entire outer shell of the cabin is under stress. You can split this into an axial and a hoop stress. The equation for these are:

hoop stress

axial stress

It's clear the axial stress is significantly lower. Keep this in mind.

Stress concentrations

First, a simple sample calculation:

Let's say you have a plate of 1 x 1 m with a thickness of 1 mm. You can place a uniform load of 1 kN/m on one side. This means the plate is supporting a load of 1 kN. The stress is the load (1 kN) divided by the area (1m x 1mm) = 1 MPa.

When you have a plate with a (small) circular hole in it you might think this only has a very small effect, but the effect is quite large. Just at the side of the hole, the stress is raised to 3 MPa*.

Adding the hole increased the stress by a factor 3 under the same load. Effectively the whole reduced the maximum loading capacity by a factor 3.

This effect is due to stress concentrations. A stress concentration factor is a multiplier for the maximum stress in a plate due to holes or other geometric properties. The shape of the hole has a large effect on the stress concentration factor. For an oval shaped hole, the stress concentration factor is:

stress concentration factor

2A is the width of the hole, 2B is the length of the hole. It should be clear that increasing the height of the hole along the axis with the most stress reduces the maximum stress, making the plate stronger!

Removing area

The other effect of adding a hole is that you remove area over which the load is spread out. If your hole has a diameter of 0.9 m, you can imagine that the maximum stress would increase much more than if it were 0.00001 m.

For more information on these interactions, you can check out

Putting it all together

The stress in the hoop direction is much higher than that in the axial direction. By elongating the holes (windows) along this axis you can reduce the stress concentrations. This decreases the maximum stresses, resulting in a stronger aircraft.

You can also more easily increase this length as you can more easily afford to lose material along this axis.

Additional factors

The fuselage is also subjected to bending forces due to the fuselage tube being supported only at the wings. However, these forces only cause large stresses at the top and bottom of the cabin and much less so at the position of the windows which are located near the centerline.

Furthermore, according to @Therac most of the load is caried by the frame members, which also reduces the importance of the stress concentrations around the windows. I personally am not certain about that as apparently windows add about 200 kg for an aircraft seating 150 passengers. There are also suggestions for window-less passenger aircraft with wraparound screens.

*slightly higher due to the plate also losing surface area, but I will ignore that effect at this point

$\endgroup$
4
  • $\begingroup$ When considering the stresses, don't forget the longitudinal stress resulting from the very long fuselage tube being supported solely by the wings. Of course, the peak stress from that is far away from the window section, but overall it's the principal load the fuselage is designed for. Hoop stresses take second place. That said, the point about stress concentrations is valid, although it's not quite as extreme due to most of the load being on the frame members. $\endgroup$
    – Therac
    Jul 29, 2020 at 20:43
  • $\begingroup$ @Therac thanks for your addition. I didn't know that but it does make sense. Should I update the answer to reflect your comment (I'm not very active on this forum)? $\endgroup$
    – Nathan
    Jul 29, 2020 at 21:14
  • $\begingroup$ Sure, the answers here are editable, and it's common and expected to update them to account for comments. $\endgroup$
    – Therac
    Jul 29, 2020 at 22:19
  • $\begingroup$ @Therac do you feel this is a correct addition? $\endgroup$
    – Nathan
    Jul 30, 2020 at 8:30

You must log in to answer this question.

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