Reinforcement to Thin skin - DTA Concerns 3

[edmeister]

Lately projects keep getting more difficult with a new twist each time.
I am required to install an antenna Reinforcement to a .032 Thick crown skin. Customer insists on having no external Doubler.
Since aircraft is CAR 23.573b regulated (DTA Analysis requirement)- the following options are available.
1/ Dimple NAS1097AD4 rivets in .032 skin & CSK into Doubler.
2/ Use greater quantity of NAS1097AD3 rivets
3/ elect to reduce CSK head depth of NAS1097AD4 rivets
4/ Metal to metal Bonding.
I would prefer (1) but require fatigue data for Dimpled fastener locations.
(2) would be the simplest to certify - but unorthodox & have to consider bending effect thru approx .080 matl.
(3) reduce the shear strength of the fastener & possible add more fasteners. also need data.
(4) this would require coupon testing -
Are there any other suggestions out there or relative DTA data available?

 

[rb1957]

so TC approve your limitations, published in the MMS ?

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[LiftDivergence]

Hey, at the end of the day there's always different ways to solve the same problem, sounds like you have a plan and guidance... but as a follow-up, my "two cents":

I'm not really a fan of the modification intercostal design idea. I have seen this done before when I was working on major STCs but it's usually reserved for very large antennae with radomes like for 2KU wifi. Your antenna is small enough that it would generally be installed with a simple skin reinforcement. I still don't really see the issue with having a small number of rivet heads protruding on the OML with an internal doubler. I doubt it would even be noticeable. Having said that, your design would alleviate any concern over out-of-plane skin bending. BUT... there are other drawbacks.

1. More complexity. Instead of a simple skin doubler you now need in internal intercostal and whatever clips/angles etc. you need to interface with the fuselage frames. More material, more manufacturing, more complexity.
2. I'm guessing the crown frames are also fatigue critical structure, just like the crown skin (which you are still also affecting). So now you're touching 2 or 3 primary structural elements (PSEs) instead of 1 or 2 (depending on status of the stringers). This means more analysis because now you need to assess the frames. Ring analysis is more challenging than skin analysis. Yes the loads on the frames will be low, but beware that you are also going to be hard-pointing them. Your DTA will now need to analyze the skin and the frames, and you'll wind up needing inspections for multiple details.

The nice thing about having an internal skin doubler is that your ICA can be an external MFEC and you wouldn't even need to gain access to the interior for inspection. With your design you're going to be tearing down the ceiling every inspection interval.

3. An unreinforced hole in the skin is...unorthodox. I think RB said it's only 0.7" (not sure). But still, that's a bad continued airworthiness detail even if we think this aircraft won't see a ton of cycles. At the very least if you go this path I would strongly recommend cold working that pass-through hole.

You mentioned you want to bond a doubler for reassurance. Be wary... bond strength and load transfer would need characterization. There are always attendant drawbacks to reinforcement. Skin patches tend to draw load based on stiffness and you can easily create a bad hard-point detail (think 757 crown skin issues). I think you need to either commit to installing a reinforcement and fully analyze it's effects, or just don't install it.

I've worked on a lot of small and large antenna STCs and the golden rule, like anything else... keep it simple, if you can.

Keep em' Flying
//Fight Corrosion!

 

[edmeister]

.. I am listening to the potential issues .. better to be prepared ahead of time.
Small antenna .. even with abuse loads - load transfer into the frames should be minor.
Should this become a repeat job - will attempt to provide a test coupon for the bonding next time so i could get some credit.
- agree .. Ext. Doubler is my first option as well / but given location is "nose-level" when entering the aircraft - Owners qualm at anything that distracts from the smooth aesthetic outline.
.. Not installing the Bonded Doubler? .. Im just uneasy about an open hole (without reinforcement) in a pressurized fuse. Is there a difference between a bonded Doubler or a riveted Doubler?
One could also argue that the bonded Doubler is a crack-check retarding the crack growth. Never had to deal with additional stiffness WRT riveted doubler. - but then again is was almost universally a Strg-Strg / Frame-Frame span.
again .. appreciate input.

 

[rb1957]

yes, there is a massive difference between a bonded dblr and a riveted one.

A riveted one has an assured load transfer, albeit somewhat ineffective as it is discrete and remote from the stress concentration. But the process (rivetting) is well understood and well controlled.

RT metal-metal bonding is IMHO "awful", it is completely dependent on the installation process. Ok, LD and me have different opinions, based on our experiences ... NP.

Yes, a ring dblr will be very effective on retarding crack growth from the hole. Having it and not accounting for the benefit ... ok, I can see this (being conservative).

You say this at "nose level" ... so near the entry door ? That is an important detail. It means the hoop stress has been funneled away from the skin, into the door frames (as the door cut-outs would be ineffective for hoop stress). So assuming hoop stress is "wildly" conservative.

I get that you're nervous about an unreinforced hole, but if the stress level is low enough have confidence in your calcs. You should also see that the crack growth is slow, even from a "obvious" 1" crack (heck, conservative 2" visually detectable), if you insist a 1/4" HFEC detectable.

If you want to make a project out of this (maybe a uni project ??) do flight testing with a s/g ... I think this'll convince you of the issue you have.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[MNLiaison]

Ed,

Document, DOT/FAA/AR-98/33 - Uniaxial and Biaxial Tests on Riveted Fuselage Lap Joint Specimens, contains fatigue data on riveted dimpled, csk, and csk/bonded joints. A good read.

Regards