Fuselage Expansion due to Pressurization 4

[Stezza]

I am about to set up a model the fuselage cross section of a 767 to determine the expansion between the passenger floor and the crown of the aircraft under a 9.1 psi pressure.

Does anyone know roughly what sort of magnitude of expansion I should get so I can be confident that my model is correct?

Our problem is that we have an interior panel spanning from the passenger floor to crown (but not attached on the sides) that has to react some vertical loads in the crash condition, but because the radial expansion joints (to prevent loading due to pressurization) at the ceiling of the aircraft the vertical loads can only be transferred to the floor.

This configuration is showing that the panel will buckle and we are considering limiting the travel of the crown expansion joint so that in the crash case when the panel buckles (and has large displacement) it will then have a load path in tension to the crown. Note this load path is only required in the crash conditions.

We are looking at a few other options but I would be interested to know if this sort of feature has been used by other people in the past.

 

[GBor]

If you are in the fuselage, the floor would be pressurized on both sides and shouldn't see any vertical deflection due to pressure. The skin deflects like a pressure vessel. If you want to be conservative, you ignore the frames. Deflections should be small, but they are tied to the radius of the aircraft, so it is hard to say precisely what to expect without knowing the type of aircraft or at least the dimensions. If you are using FEA with stringers and frames, the deflection pattern should look something like a waffle, but the magnitude should be small.

 

[RPstress]

Just do a 2 minute check treating it very simply: If 60 thou skins E 10.5e6, 767 fuse height 17' 9" = 213", width 16' 6" = 198", 9.1 psi => 900.9 lb/in => 15 ksi => 0.0014 strain => total height expansion 0.3" =>~ 0.15" floor to ceiling. Smear the frame area out into an equivalent thickness to roughly allow for them.

 

[rb1957]

yeah, the problem with the prediction is the frames ... it's easy to calculate the expansion of a cyclinder under pressure (hoop stress, hoop strain, change in length of circumference, change in radius). this'll be the upper limit of what would happen in the real world. in the model world, fine white elephants do funny things that you won't see in the real world.

i'd done similar models ... get a cyclinder working, then mess it up with stringers, frames, dblrs, doors, longerons, etc !

 

[Andries]

Stezza,

It is always good to do a hand calc to get a feel for your FE results. There is a good chapter on stress and deflection analysis of pressurised fuselages in the book "Introdution to the Theory of Aircraft Structures" by D. Williams. This is an old book (1960), long out of print, but good for classical methods. There is also a NACA report by Flugge on this topic.

Regards

Andries

 

[SAITAETGrad]

Do you know enough about the existing floor panel to do a direct comparison rather than going to the applied loads on the aircraft? Think globally but calculate locally?

GBor...what about decompression?

I agree with Andries...hand calculation first. Further, I'm not sure FEA is really necessary for this problem.

 

[rb1957]

why have it attached (reacting vertical loads) at both ends ? why not have it "nailed" into the floor and floating at the ceiling ?

 

[GBor]

SAITAETGrad
9.1 psi is an overpressure relief valve lift for the 767...not decompression.

 

[Stezza]

Thanks for the replies.

The FEM gave me a differential deflection between the crown and the passenger floor of 0.17 inches.

There was some deflection of the passenger floor downwards (appears to be due to the post being pulled downwards due to the lower fuselage expansion and the SOB attachment to the floor essentially remaining in the same vertical position) which pushed the total deflection above the 0.15 inches calculated by RPstress (nice work - you got a star from me).

Thanks

 

[Stezza]

Oh also thanks for the NACA TN 2612, STRESS PROBLEMS IN PRESSURIZED CABINS pointer Andries.