Titanium fastener in tension fitting 1

[A_aerospace]

Hi all,

In one fuselage design of a transport aircraft I saw titanium hi-loks DAN7 in classical channel tension fitting. Fitting connects two parts of a beam and the beam itself is a longitudinal sill over fuselage door cut-out. Primary structure. Usually (to my experience) steel fasteners are used in such locations. However, there are different engineers and different opinions. I listened carefully to their point of view on a subject. Key arguments are:
- DAN7 Hi-lok has shear-tension head type, so it’s designed for tension use.
- Static strength is checked (by calculations). MS is high.
- Fastener fatigue life is calculated. LF is high.

Those bullets are correct. But still I am not fully convinced. Am I too conservative? I tried to find some recommendations in books regarding fastener type/material for tension fittings, but found nothing.
I would like to hear your general opinion as a third side. Is it Ok to use titanium fasteners in primary structure tension fittings? And a reference with recommendations (if exists) would be extremely useful.

 

[rb1957]

if those bullets are correct (and the installation has been properly checked), then yes, you are being too conservative.

It is perfectly fine to use Ti fasteners in primary structure. Ti and Steel Hi-Loks/Hi-Lites are the same strength, Ti actually are cheaper than Steel since OEMs adopted them as their "go to" fastener (being lighter than Steel). Ti gets on better with Al than Steel does, less corrosion >>> less corrosion-protection required.

another day in paradise, or is paradise one day closer ?

 

[A_aerospace]

Thanks for your reply.
Just to be sure that I am clear enough and we are talking about the same thing. My question is only about the tension bolt in that fitting (see picture below).

 

[rb1957]

sure … why should tension bolt only be steel ?

Why do you think Ti is inappropriate ?

If the joint has been checked, why do you think it's not good ?

another day in paradise, or is paradise one day closer ?

 

[A_aerospace]

I don’t have absolutely strong arguments otherwise this post would not exist. Of course the first thing that anybody would look at is static and fatigue analysis. That’s why I mentioned them immediately in the first comment. I am looking for pros and cons beyond those two.

Here is my humble opinion:

- Ti bolts in tension have significantly lower fatigue life. In this way it leads to using bigger fastener diameter (compare to steel) and weight saving declines.

- The meeting surface of L/C channel fitting bends under tension load. The fillet under Ti bolt’s head and the thread are much more sensitive to tension on inclined surfaces than for steel bolts (in terms of static and fatigue). This additional bending of the bolt is difficult to get in big FEM model and usually is neglected. Usually only tension load is taken into account during analysis.

- There is even additional fitting factor (1.15) requirement in FAR25, which sideways shows that fittings are delicate structure and shall be designed with additional attention. Reliability of fitting joint in service is much more important than weight saving. In most cases there are not many of them in the structure (compare to lap/buttstrap joints) so weight saving is not that significant.

- There is such a thing as ‘high tension steel bolt’ (e.g. MS21250) with stiff tension head. They are designed for tension fittings. I have never seen a bolt like this made of Ti. Does it exist?

- Good design of fitting joint requires high level of bolt preload. Steel bolts are better this way.

 

[LiftDivergence]

Initially I was confused by your statement that this was a shear-tension head type fastener. I was thinking, "so which is it actually"? However, looking at the DAN7 spec it does in fact say that it is for "shear- and tension-load".

This is a protruding head hi-lok....comparing the head dimension "H" to other shear and tension head hi-loks, the depth is the same as many other tension heads, and there is no difference between the head dimensions for steel and titanium protruding head hi-loks of the same diameter. For example, for a No. 6 fastener, all shear protruding heads typically have a "depth" of 0.045-0.055". Whereas No. 6 tension heads are 0.064-0.074" regardless of material. So basically, I would just consider the DAN7 a tension head.

A couple things to note which have not been stated so far:

1. "Fastener fatigue life is is calculated. LF is high." Okay?? This is primary structure, assuming it's FCBS (is sill structure). This is not sufficient. What about the rogue flaw LEFM of the mating parts? What about the statistical fatigue life of the mating parts? The fastener type will have an influence on this. Have "they" run this analysis with both materials and compared the results?

2. You say you are considering steel steel fasteners instead. What type? A286?

3. What type of fasteners are used elsewhere in this joint? You need to be careful about using fasteners of different types in the same joint. This exacerbates areas of different stiffness.

4. Steel fasteners are MUCH stiffer than titanium. Steel is on the order of E = 29*10^ psi. 6Al-4V on the other hand is only 16.9*10^6 psi or so. Steel fasteners will be worse for the hole joint severity factor. This is why you need to check the joint Statistical fatigue life, not just the fastener itself.

5. WHY is the joint designed with a spotface from production??? This is a poor Damage Tolerance detail and should be avoided if possible.

6. How thick are the flanges of those angle fittings and what material are they? If you are really concerned about pin bending like shown in your second figure the diameter of the fastener is of more concern than the material.

I would wager the fastener "fatigue life" you mention is not going to be critical relative to the damage tolerance characteristics of the other parts in the joint.

Keep em' Flying
//Fight Corrosion!

 

[rb1957]

I'm surprised that a Hi-Lok (Steel or Ti) is being used in a tension situation, as the preload is limited by the collar … unless you use the MS nut.

As for weight … I think it's a wash, but in any case the gains are minimal.

I think corrosion issues may be the dominant design driver … that or the Design (or the Manufacturing) Director who says "thou shalt use Ti tension bolts" if for no other reason than to differentiate his reign from the previous junta.

another day in paradise, or is paradise one day closer ?

 

[A_aerospace]

Thank you guys for comprehensive comments.

“fatigue life of the mating parts”- It was checked. Results are Ok.
“Have they run this analysis with both materials and compared the results” – No.
“You say you are considering steel fasteners instead. What type?” - If it were for me, I would consider MS21250 or NAS6603 from 4140 alloy.
“designed with a spotface” – there is no spotface in my case. The picture in the first post is not the original design but from commonly known book (Niu). I cannot post original design.

“What type of fasteners are used elsewhere in this joint?” – I’ve just notice (after your question) that the picture in the first post has one significant difference from my case. There are two load passes in the picture.

In my case there is a single load pass. Only through tension bolt. So there are no fasteners elsewhere in this joint. Only tension bolts in channel fittings. Sorry for mislead.

“flanges of those angle fittings“ - They are from Al 7XXX alloy. Mating flange thickness equals bolt diameter.

“about pin bending like shown in your second figure the diameter of the fastener is of more concern” – The pin itself (cylindrical part of the fastener) can handle the bending. My concern is about radius under bolt head. There is high stress concentration on the radius due to load eccentricity (red on the picture below). It’s like a notch. And Ti does not like it from static and fatigue point of view. The same story on the other side with the nut and thread.

“I'm surprised that a Hi-Lok (Steel or Ti) is being used” – Totally agree.

 

[moon161]

I think your FBD of the joint is incomplete and the deformed sketch is inaccurate because the flanges of the fittings are clamping the web of the beam shown in section. The joint preload should keep them all clamped parallel. A proper preload would would keep the faces from separating at all design loads including margins wouldn't it?

 

[A_aerospace]

Good point. Bolt preload definitely helps.
Probably it would look more like this (exaggerated of course):

 

[rb1957]

and this is why tension fittings have thick bases,

and this is why we have prying loads on the bolt.

If the tension bolt can't develop preload ('cause of a Hi-Lite collar) then the joint is designed for a vey low tension load or gapping is acceptable.

another day in paradise, or is paradise one day closer ?

 

[WKTaylor]

AA...

Use of Hi-Lok tension head pin and mating/compatible tension collar or nut... are not typical for SINGLE pin-tension joints... but are usually suitable for multiple-pin [frame] tension joints. Otherwise deep-tension head hex or 2X-hex or internal socket-head bolts... with generous/rolled-fillet radii... and high reliability tension nuts are principle used.

CAUTION.

Your structural examples/concerns are 2D and simplistic.

Your 'textbook image' example is of a 'corner fitting' with spot-faces... not a simplistic 2-flange-angle-fitting ['cartoon-images'] You show with edge-loaded 'heads' and 'nuts' and shank bending.

None of the examples use washers or fillers under the head and nut. IF there is a slope where these elements meet the structure, then 90-degree spot-faces, self-aligning nuts and/or washers or nested fillers [with built-in counter-slope and radii or chamfer to nest into fillets] can be applied to relive prying loads.

ALSO...

Low alloy steel bolts made for tension loading do very well in that case... however there is a primary concern for inevitable potential for in-service 'rusting' [rusted = busted]… so careful painting/sealing for moisture mitigation is absolutely necessary. And, as previously mentioned, LA steel is being replaced by fasteners that [essentially] are impervious to environmental deterioration [normal structural temperatures and chemical exposure], such as A286, Inconel 718, MP-35 or Ti-6Al-4V titanium are far better options... with appropriate finishes for isolation to the mating structure... such as aluminum [very dissimilar-to/reactive-to ferrous, nickel and chrome, etc... less-so titanium alloys]. Typical finishes: cadmium+chromate, IVD-Aluminum, aluminum-pigmented coating, SFL or anodize [Ti], etc...

NOTE.

The 'long-time standard' LA steel HT160 bolts for high reliability tension joints are typically MS20004-to-MS20024. A new standard for equivalent titanium and A286 CRES bolts is NAS9921 BOLT, SHEAR, INTERNAL WRENCHING, SHORT THREAD, SELF-LOCKING AND NON-LOCKING, 6AL-4V TITANIUM AND A286 CRES, 160 KSI Ftu [0.1120-up to-0.875-Dia, only]. CAUTION: mating/compatible nuts and washers are essential elements of the installation.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]

 

[Sparweb]

I don't get how the splices, which are 4 rows of double-shear fasteners, are supposedly beefed up with a single skinny bolt neck in tension. To get through all that fat build-up, the bolt shank is at least 1/4" long, may be 1/2", while right beside it are numerous other rows of much stiffer joints. The angle fittings are also splice joints in themselves: look how they overlap on the vertical flanges to the T-splice there, too. You could take that tension bolt out and hardly notice! I doubt that Hi-Lock gets much tension at all, except close to ultimate load. So, yeah, the fatigue should be perfectly fine since in the linear range the load is very low.

http://www.sparweb.ca

 

[A_aerospace]

WKTaylor, NAS9921 is shear type bolt? It says “SHEAR… SHORT THREAD”. Is it suitable for high tension load? Or in other words: what Ti bolt works well in tension?

 

[WKTaylor]

Aa...

Suggest You compare tensile and shear [specification] allowables between MS20004-to-MS20024 and NAS9921, side-by-side for comparable diameters.

BTW the title for MS20004-to-MS20024 is not as definitive regarding 'tension' [use] as You'd expect.

Also NOTE that NAS9921... with 'short threads'... will comfortably/fully accept an NAS1805 [A286 HT180] tension-fatigue rated NUT and a mating washer. TaDa!

BTW... I wish I could discuss a current problem of severely corroded [scaling-rust] MS20005, MS20006 and MS20007 bolts in [aluminum] structure in Class-3 [very-loose-fit] holes. OH yeah... the holes with these corroded bolts are themselves also corroding, badly, as a direct result of exposure to ferrous rust by-products with constant 'wet/moist' conditions [electrolyte] and corrosion prone high strength aluminum alloys... perfect storm.

Regards, Wil Taylor

o Trust - But Verify!
o We believe to be true what we prefer to be true. [Unknown]
o For those who believe, no proof is required; for those who cannot believe, no proof is possible. [variation,Stuart Chase]
o Unfortunately, in science what You 'believe' is irrelevant. ["Orion", Homebuiltairplanes.com forum]