Poking holes 2

[RoarkS]

So round two... I'm clear of the first group, getting ready to hire another set of DER's but want to guide where this is going.

part 23 turboprop. aft of pressure bulkhead. I have the OEM FEA analysis results for the area.

-previously mentioned there is an existing hole done by the OEM near where I want to put my hole(s) that isn't addressed by the FEA I was provided. Their solution is a lot like the "porthole" I talk about below...
-The bays have a high aspect ratio... long stringers very narrow space, and wide frame spacing.
-The hole I need to poke needs to basically remove the skin from one of the bays entirely. I find this to be annoying because now every bit of guidance I have in front of me says cruciform patch, but the high aspect ratio of the bays makes it a BIG patch... which results in a shim, and the actual doubler. Lots of rivets, lots of paint. That said looking at some of the other BIG holes the OEM did, yeah there is a big doubler built under the frame's stringers, inside of the skin... so looks like pretty obvious guidance that's a path... but at that point I'm saying not economical for this mod.
-Looking at some slightly larger commuter cat aircraft on similar situations for environmental holes... reminds me of a naval ship porthole... basically a stringer formed into a hoop... but machined. What I'm thinking I want to do is cut one of the stringers, hole saw the skin, then rivet in this machined porthole, and splice it into the stringer. Could even machine louvers that could "carry the stringer through". ... I just need some basis for doing something like that. Would make install super easy and basically remove any external visual modification... except for the hole itself. Any suggestions for references? The predicted stress distribution is well within what the porthole can handle per FEA... just trying to find a method that isn't novel/I can throw some hand calcs at. //I have Niu on my desk now. Flabel and Bruhn in my bookcase at home...

 

[rb1957]

1) get your DERs under contract first (there is a cart and a horse, and here the design is the cart !).

2) there are several design approaches depending on what the DER likes ...
a) put the hole between the stiffeners, taking out the web; or
b) put the hole inline with a stiffener and include a bunch of stiffeners to offset the lost stiffener.

I put a lot of "faith" in "equivalent strength" ... cut a hole in the web, add a doubler so the net area thru the hole is higher than without the hole.
I would not patch the entire bay, even though that may be "good practice" I think you have an example where this practice is onerous. Big doubler giving plenty of space (and fasteners) for the load to transfer.
0.063" and BJ4s should be plenty, machined rings if you like that kind of stuff !?
I'd make the doubler between 3D and 5D square (where D is the diameter of your cut-out).
DTA may not be trivial, but I'd aim for a visual inspection of the bulkhead web (so place the doubler accordingly, fwd face or aft face of the bulkhead so you can easily insepct the bulkhead. This may make your installation harder, but it will help the maintenance.
You might want to make a larger hole in the bulkhead web, so that the connector sits on (and is clamped against) your doubler.

If you're interested in the DTA, I'd recommend Broek's Practical Damage Tolerance Analysis (or something like).

GL !

 

[Sparweb]

Niu and Flabel will tell you what you don't want to hear, then.
Typical whole-bay cut-out will get a reinforcement patch that covers the "other 8" bays of the 9-bay grid surrounding the cut-out bay. Like a big game of X's and O's. Pretty typical, and par for the course for lots of structural mods. Sometimes you can justify cutting a diagonal around the 4 extreme corners.

If you want more nuance, you'll have to look to Safarian and other "practical" guides for information to help you make your doubler simpler to make. Not knowing the make of the plane, I can't guess as to how much the OEM made your job easy, such as protruding-head rivets in the original skins or CSK.

Aft of the pressure bulkhead, and presumably on the belly, you're dealing with loads that alternate in direction in tension/compression pretty much completely and in shear in both directions. Meaning you have a fatigue stress ratio of R=1 in both shear and axial load conditions. At least you don't have pressure hoop stress to contend with.

Use the OEM's FEA as your guide and anchor for equivalent stress, but Don't try to reproduce it. There are 3 generations of published materials to help you use conventional analytical hand-calcs which will be faster and more transparent to the DER's who look at your work. If they have any gray hairs, they may prefer that anyway.

 

[RoarkS]

rb1957, sorry to mislead totally understand... I'm aft of the pressure bulkhead, in the aft equipment bay. wouldn't touch the actual pressure bulkhead with a 10ft pole no JAL123 for me please. The 6.8" circle I need is going to blow out at least one stringer... and honestly that hand calc is still a little over my head. a round slot (flat oval) will fit, but I have to justify that, thankfully the OEM data looks like I can easily justify equivalent stress... especially with that machined ring all polished up made out of titanium or something lol!

Sparweb, I think the OEM did make the job easy honestly... just trying to wrap my brain around the whole job and figure out what's best. Just today I finally got the area of the cut out and it's awkward, but still a good bit smaller than the bay itself. I have data data on all of the bays on the side for all the load conditions. There isn't much room on the belly...lots of stuff going on so I'm looking at one of the bays on the side. We were looking real close at putting it on the crown and venting into the dorsal fairing but I don't have that set of data from them and don't think I will get it. I think it was an act of god we got what we did.

I will check out Safarian a little more. See if I can take his class next year.

 

[rb1957]

if you really want to (I think it is "overkill") add some lateral stiffeners to break up the long panels ... this clearly increases buckling allowables. Overkill of overkill would have 4 lateral stiffeners so you make the "3x3" pattern that is referenced, with the middle panel having your cut-out and the surrounding 8 have the doubler replacing the lost area.

Re previous post ... doubler need to be on the unstiffened side of the bulkhead web (obviously), and these lateral stiffeners on top of the doubler. This then sets up how to inspect the bulkhead web (under the doubler) ... visual or LFEC depending on which side it is (or which side you want to access).

Assuming a flat bulkhead ... then stress state is awful to predict ... but you have th eOEM FEA as a guide. Bruhn has a nice method for this type of cut-out (redistributing internal loads to the adjacent panels), no doubt Niu and Flabel too. I think adding the doubler is sufficient to take the loads from the lost cut-out around the cut-out ... but 1st get your DER signed up so you can proceed in a direction he likes. A good DER (in my mind) will work with you on the conceptual design (but you do the detail design and analysis) ... they will probably charge for this. A bad DER (IMHO) says "I will assess the design and analysis you give me and tell you if it is compliant, and if I'm really nice I might tell you where it is non-compliant and maybe suggest ways to make it compliant." Having a good relationship with your DER is priceless.