Importance of Installing Rivets Within the Interfay Sealant Work Life 3

[nunorodrigues]

Hello,

Currently I am responsible for the sealing process of an integral fuel tank. Due to the several difficulties to rigorously control the temperature and humidity conditions in the assembly line, we are having some troubles with the decrease of the application and work life times of the sealants we are using.

The main problem lies on the interfay sealant we are using to assemble the skins to the spars and stringers. We verify that the squeeze-out sealant that comes out when you install tacking fasteners before the installation of the rivets is dry to touch (tack-free) before the period that is indicated in the sealant's respective TDS. What is the importance of finishing the installation of rivets within the interfay sealant work life? I am asking this because if we can't accomplish the final riveting until the work life time is over, we need to wait the sealant to be cured to finish the riveting process.

I hope I can hear your opinions so I can better understand this issue.

Thank you in advance.

Best Regards,
Nuno Rodrigues

 

[WKTaylor]

nunorodrigues...

WOW, so many factors/permutations to consider. A few important questions, up-front.

Is this for a light/GA acft? Medium weight corporate acft? Heavy weight transport? Military tactical acft?

Is fay sealing absolutely required...?... or would sealant-packing [holes/gaps] with fillet-seals and over-coating be another 'way-to-go'?

Has fay-surface sealant been analyzed/approved for use, considering its negative effects on static strength and fatigue durability.

Is this new structure Assy? New [modified] structure Assy installed on aging structure? Repair of aging structure with locally fabricated details?

What is/are 'tack rivets' as You mentioned above? What temporary Assy fasteners/methods, other than 'tack rivets' are You using, IE: clamps, spring Clecos, wing-nut Clecos, etc.... and/or temporary nominal nuts/bolts, etc.

WHAT sealant(s) are You using? EXACT definition required [IE: AMS-S-8802-C24, PPG P/S 890 Class C, etc]

What finishes are You using on the mating [fay] structural surfaces, EXACTLY: Bare, alodined [chemical conversion coating, CCC] or anodized ONLY, CCC or Anodize + primer [epoxy or polyurethane]; etc. Also, are You using pre-sealing cleaners [solvents] and sealant adhesion promoters?

What fasteners, GENERALLY are You using [IE flush & protruding and shear & tension head: solid aluminum rivets, blind rivets, blind bolts, Hi-Loks [pins/collars], swaged-collar Lock-bolts, etc.

Are You very familiar with typical integral fuel tank design guides and practices, such as...??

SAE AIR4069 Sealing of Integral Fuel Tanks
AFWAL-TR-87-3078 Aircraft Integral Fuel Tank Design Handbook
etc?






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]

 

[nunorodrigues]

Dear WKTaylor,

We are talking about a small commuter aircraft (6 to 9 pax) and the company we are working to demands the sealing process of the fuel tanks to be interfay + fillet + overcoat sealant to be applied. After the MIL-PRF-81733 C12 interfay sealant application we hold the faying surfaces together using spring clecos. The faying surfaces are protected with primer and before sealant application they are cleaned with MEK. For riveting we use solid aluminum rivets. Our process is based on OEM's standards.

I am not familiar with those manuals (SAE AIR4069 Sealing of Integral Fuel Tanks & AFWAL-TR-87-3078 Aircraft Integral Fuel Tank Design Handbook), do you know where I can find them?

 

[WKTaylor]

nunorodrigues... hope this makes sense... A few thoughts.

1. Sealants for Integral Fuel Tank application.

1.1 Sealants qualified to AMS-S-8802* are considered the gold-standard for integral fuel tank sealant. These sealants generally have little/no corrosion inhibitive pigments, hence retain high fuel resistance/adhesion/toughness for decades… at the cost of poor performance in a corrosion-prone [exterior] environment.

NOTE. 8802 sealants are generally over-coated with corrosion inhibited 81733 sealants in fuel-dry environments for increased corrosion resistance.

Types
Sealing compounds covered by this specification are classified as follows:
Type 1 - Dichromate Cured Sealant. Material with a dichromate curing agent.
Type 2 - Manganese Dioxide Cured Sealant. Material with a manganese dioxide curing agent.

Classes
The following classes apply to both Type 1 and Type 2 sealing compounds:
Class A - Suitable for application by brush. Available in the following application times in hours:
A-1/2
A-1
A-2
Class B - Suitable for application by extrusion gun or spatula. Available in the following application times in hours:
B-1/2
B-1
B-2
B-4
Class C - Suitable for extrusion gun, spatula, brush, or roller. Available in the following application times in hours:
Notation: ( ) Assembly time in hours:
C-8(20)
C-8(48)
C-24(80)

1.2 Sealants qualified to MIL-PRF-81733 [any class] are generally considered pressure/environmental sealants with corrosion resistive pigments [fillers]. This spec relates to a family of sealants that at best, is for secondary fuel sealing ONLY, IE: fuel-cell cavities where there is minimal exposure to fuel [except in fuel bladded leak]. The corrosion resistive pigments reduce adhesive strength; and because they ‘leach’ the corrosion protective pigments when exposed to fluids [water, fuel, oil, etc], tend to decay slowly over-time. In an environmental/pressure environment, this can be acceptable… however, this is NOT acceptable for a long-term 100% reliable integral fuel-tank seal.

Now here is where I am concerned.

You stated the sealant You are using is MIL-PRF-81733 C12. Based on specification MIL-PRF-81733, this is meaningless. This sealant is specified in the following order [see MIL-PRF-81733, para 6.2]…

MIL-PRF-81733-{Type}-{Class}-{Grade}-{-Application Time ‘dash number’}, thus…

Types. The types of sealing compound are as follows:
Type I - For brush or dip application
Type II - For extrusion application, gun or spatula
Type III - For spray gun application
Type IV - For faying surface application, gun or spatula

Classes. The classes of sealing compound are as follows:
Class 1 - Polysulfide rubber base material
Class 2 - Polythioether rubber base material

Grades. The grades of sealing compound are as follows:
Grade A - Contains chromate corrosion inhibitors
Grade B - Contains nonchromate corrosion inhibitors

Application time. The minimum application time**, in hours, for each type and class is indicated by a dash number as follows:
Type I, Class 1 - Dash numbers are -1/2 and -2
Type I, Class 2 - Dash numbers are -1/4, -1/2, and -2
Type II, Class 1 - Dash numbers are -1/6, -1/4, -1/2, -2, and -4
Type II, Class 2 - Dash numbers are -1/4, -1/2, -2, and -4
Type III, Class 1 - Dash number is -1
Type III, Class 2 - Dash number is -1
Type IV, Class 1 - Dash numbers are -12, -24, -40, and -48
Type IV, Class 2 - Dash numbers are -4, -12, -24, -40, and -48

>>>>> There is NO designation ‘C-12’ for this sealant.

1.3 NOTES.

1.3.1 Roughly speaking, ‘application time’ is the time span when fully mixed sealant has to be applied to the structure. It will have high-adhesion and good fluidity up-to this point.

1.3.2 Roughly speaking, ‘assembly time’ is the time span when fully mixed sealant that has been applied to structure, can no longer be readily squeezed-out from joints, the sealant [such as fillets] successfully manipulated and/or mating parts be successfully repositioned… without sealant damage… because it has begun the final transition from a tacky polymer liquid to a semi-solid polymer. Assembly time is roughly 1.5-to-2.5 X the application time... unless specifically formulated for an extended assembly period.

1.3.3 Examples.
Type I-1/2 designates a MIL-PRF-81733 brushable material having an application time of 1/2 hour with an assembly time roughly 0.75-to-1.25-Hours.
Type I-2 designates a MIL-PRF-81733 brushable material having an application time of 2 hours with an assembly time roughly 3-to-5-hours.
Application and assembly times are influenced by a number of factors: however, after individual chemistry variations of each batch of sealant, humidity and temperature are the greatest contributors.

1.3.4 CAUTION. I always recommend that the first a bead of each mixed sealant batch be applied to piece of primed sheet metal. Flatten the bead using a squeegee to ~0.020—0.040 thick [notch-out the squeegee blade to attain this flat profile. The sealant tech should note the mixing start time adjacent to each sealant bead [sharpie pen]; then check the consistency of that sealant frequently, to ensure it is still within the assembly time… experienced mechanics will ‘know’ when this point is crossed.

1.3.5 NOTE. I thought that ‘C-12’ might have come from the manufacturer’s ID on the label… however I had no luck confirming this possibility.

2. You seem to need more than a briefing on this topic. Here are useful references [plus where to find on web]…

NOTE.
I tried to find a web-link for the AFWAL-TR-87-3078 Aircraft Integral Fuel Tank Design Handbook which is an extremely detailed design document. HOWEVER... even though it is almost 30-YO... I think, it was recently reclassified [with a lot of other/similar documents] based on original cover-page markings to ensure DoD controlled/restricted access. Perhaps a corporate, university or national library has a copy You can review/use. IF there is a US defense aspect to this work, then DTIC

http://www.dtic.mil/dtic/

might release a copy to Your company.

MIL, MS, etc specs…

http://quicksearch.dla.mil/

Search Document ID MIL-PRF-81833 and MIL-S-8802 [predecessor to AMS-S-8802], etc…

SAE documents…

http://www.sae.org/aerospace/

AIR4069 Sealing of Integral Fuel Tanks [must have]

http://standards.sae.org/air4069B/

AIR786 Elastomer Compatibility Considerations Relative to Elastomeric Sealant Selection [good-to-have]

http://standards.sae.org/air786B/

AIR3270 Aircraft Sealant Removal Techniques [good-to-have]

http://standards.sae.org/air3270A/

AGARD-R-771 Fuel Tank Technology [must have, poor quality scan]

http://www.dtic.mil/dtic/tr/fulltext/u2/a217832.pdf

PPG Aerospace Sealants
Various PPG products qualified to meet sealant (and related) specification requirements [see individual data sheets]

http://www.ppgaerospace.com/Products/Sealants/

USAF Integral tank sealing, and related, T.O.s (general technical manuals for practical applications)

http://www.robins.af.mil/About-Us/Technical-Orders

1-1-3 INSPECTION AND REPAIR OF AIRCRAFT INTEGRAL TANKS AND FUEL CELLS [must have]
1-1-8 APPLICATION AND REMOVAL OF ORGANIC COATINGS, AEROSPACE AND NON-AEROSPACE EQUIPMENT [good-to-have]
1-1-691 CLEANING AND CORROSION PREVENTION AND CONTROL, AEROSPACE AND NON-AEROSPACE EQUIPMENT [good-to-have]


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]

As usual, Wil, when you hit a home run, it goes way over the park fence.

STF

 

[rb1957]

question for the OP ... if you're following OEM practice, how come you're having trouble ? Clearly you're going to have trouble if you "need to wait the sealant to be cured to finish the riveting process" ... is this "sealant" the fay surface sealant or the tank sealant ?

I'd've thought (quite possibly incorrectly) that you'd apply fay sealant, reassemble with clecos, install rivets, apply tank sealant.

Are you applying tank sealant over the rivet tails ? This sounds like old practice. Fillet sealing the spar flanges to the skins/webs should be sufficient. If you're rebuilding an old design the default is to follow OEM procedures, but if we've learnt something over the years, I'd've thought you could apply this as a product improvement.

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

 

[WKTaylor]

Guys... second-phase to my answer: general assembly and fastening practices for structural integrity and durability.

Again, I hope this makes sense...

The gold standard for structural assembly is fitting parts together 'perfectly' [and with perfect fitting metal or non-metal shims if required] to virtually eliminate gaps; then match-drilling and fastening with assistance of tooling that tightly clamps parts together to attain minimal/no gaps. In this case, solid-driven rivets, blind rivets, blind bolts, lock-bolts/Hi-Loks [pins/collars], bolts/nuts/washers, etc attain their best mechanical fit/tightness with the parts tightly fitted-together, IE: the assembly develops its highest potential strength/stiffness and fatigue durability. This assembly practices is defined as 'dry' assembly: no coatings/sealant between matting parts and in fastener holes. OBVIOUSLY coating systems (alodine [CCC], anodize, passivation, primer, etc) for corrosion protection of the base-metals, and/or Isolation between mating dissimilar metals, is mandatory and can't be avoided. However, in this case these coatings must be held to a bare minimum thicknesses needed to protect parts and minimize separation of the structural metals when structure is pulled/crushed together with fasteners.

When structure is assembled 'wet with [fay/fastener] sealant', then there is added physical isolation [including fluid-tight barrier sealing] between structural members and the fasteners resulting in a huge increase in environmental durability and the possibility for pressure/fluid-tight [integral fuel tank] structure. HOWEVER, even with the best-'wet-sealant'-assembly practices, there should be at-least a ~10% knockdown in static-strength/stiffness and fatigue durability when fay surface sealant is placed between structural members. This means that the resulting cured-sealant thickness added between members, and at/around fasteners [shanks/heads/tails], should be reduced to the thickness of a primer-coating [~0.0020-inch maximum dry-film thickness of sealant, after squeeze-out/cure] then this is the best assembly practice possible... and there is still this real reduction in mechanical efficiency that MUST be accounted for, as a practical matter, due to the sealant-rubber filling the gap between structural parts.

NOTE.
During ‘dry-assembly’, practical matters such as making and installing shims/fillers to minimize/eliminate gaps between parts can be readily accomplished. During ‘wet-sealant assembly’ this process of installing fillers/shims is more difficult and prone to errors… allowing excessive gaps between parts… which become permanent after the sealant is cured too hard rubber. In one of the worst cases of excessive sealant fill [in-lieu-of proper shimming], I ever ever witnessed, the cockpit longerons of an F-4E were installed with a 0.002--to-0.160" [YES>>> 0.160, NOT 0.016] thick sealant fill from the longeron flange to the thick fuselage side-skins, over an distance of ~50-inches. needless to say moisture intrusion, corrosion, elongated holes with small cracks, etc were rampant.

IF done absolutely properly ‘wet-sealant assembled structure’ MUST be made ready for sealing and fastening assembly, as-if the sealant was not going to be applied, IE: structure is assembled/drilled/reamed/csk'ed/disassembled/deburred/etc... ‘dry’ up-to this point. This allows tightest fit between parts when reassembled ‘wet with a thin fay-coat of sealant’ and assists/forces the settling-together of parts with clamps and temporary fasters closely interspersed within the fastener pattern. Permanent fasteners are then installed in the ‘open’ holes using a rough cross-pattern (crisscross-pattern) sequence. Once these ‘open’ holes have been properly filled with permanent fasteners, then the remaining temporary fasteners are removed one-or-a-few-at-a-time, until all of the permanent fasteners are installed [also working in a cross-pattern sequence] in the structural working area.

NOTE. Automated assembly practices rarely ever apply/use sealant for a multitude of critical reasons, not the least of which is these dry-assemblies maximum strength/durability/sealing built-in... and sealant will always 'gum-up' the automated assembly equipment [important practical reason].

WARNING.
IF structure is assembled 'wet with [fay/fastener] sealant' that is poorly applied, IE: applied too thick for adequate settling/squeeze-out; or too viscous for efficient squeeze-out; or the sealant begins to transition to a semi-solid before full squeeze-out can occur; etc... and there are built-in gaps exceeding 0.002-inch 'shimmed with rubber'... then the degradation in static-strength/stiffness and fatigue durability can be dramatic! Some researchers have claimed up-to 90% knockdown is possible [~10% residual strength/stiffness and/or fatigue durability]. This can also occur when fasteners holes are not mate-drilled when the structure is ‘dry’… the drilling/reaming process being ‘too intense’ to integrate into the assembly process with ‘wet-curing-sealant’.

NOTES.
Assume for a moment that a thin layer of epoxy adhesive [thermosetting plastic] is between metal parts, instead of sealant [rubber]. The flexural/shear strength and stiffness of epoxy adhesives is MUCH higher than cured rubber. In this case there will be a notable load-sharing between the highly stiff metal fasteners and the thin structural adhesives. This load sharing is usually on the order of 70--85%-Fasteners and 30--15%-fay adhesives. This is significant load sharing/transfer. HOWEVER, due to the inherent LOW flexural/shear stiffness [and strength] of cured sealant ['rubber'], there is no possibility that the sealant can 'pick-up' any significant loading in a mechanically fastened joint, IE: it is 99%-fasteners, 1%-rubber [or worse]. When the reality of thick sealant [rubber] gaps is factored in, then the structure flexure increases dramatically due to lowered-mechanical stiffness between parts... and the sealant rubber will be stretched/squashed over-time and will degrade/crumble in service... further aggravating the ability of the metal structural members and fasteners to withstand deflections and load reversals [fatigue]. NOT only will fluid sealing become impossible [becomes a ‘leaker’]; but fatigue crack initiation/extension… and fastener-hole deformation accelerates as a vicious self-reinforcing cycle.

Each type of permanent fastener installed… solid driven rivets, blind rivets, blind bolts, lock-bolts/Hi-Loks [pins/collars], etc… have unique installation aspects that can be negatively affected by ‘wet-sealant’… fay-surface or wet-fastener installation… that must be accounted for.

Also, I’ll try to summarize my personal story related to integral wing-fuel-tank design/fabrication in an all-metal homebuilt aircraft in ~1972. At that time my dad was modifying his aircraft for record flights [Thorp T-18, N455DT], so he needed a LOT more fuel [=range] than the original 30-gallon main tank and a 30-gallon seat tank could provide. At that time I was amazed, and somewhat baffled-by, John Thorp's proposal for making the outer-wing panels and inboard wing-panel-leading-edges into ‘fuel-wet-wings’. As far as I know, his ‘wet-wing design’ for this aircraft proved-to-be one-of-the-first ever built/installed on a small single-engine homebuilt/GA aircraft. To make this more weird/confusing, there was a scathing EAA Sport Aviation article describing how crazy-bad this assembly concept was. However as decades have gone-by, I now understand the intimate details of the what/why/how the designer actually did to overcome these potential devastating structural issues brought-up in the article. This was especially true considering the ‘amateurs’ building/sealing the new wing-set: Dad and my ‘small-armed baby-sister’ who was the only one who reach into narrow tank spaces... thru tiny access doors built into each open bay... with a sealant-brush [and remembers/hates the sulfur smell of curing sealant to-this-day].

I will have to discuss these aspects in my next posting. Enough for now… I’m pooped.


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]

 

[tbuelna]

Well done WKTaylor. I hope nunorodrigues appreciates the huge effort you put into your posts above. I don't know anywhere else where one can get such detailed aerospace-quality M&P guidance just for asking.

 

[Sparweb]

Wil,
Eagerly awaiting your reassurance that your sister was not sealed inside the wet wing of a Thorp!


...thank you very much for the detailed thoughts and the story that brings it all to reality.


STF

 

[WKTaylor]

Guys... busy with my day job [including an-house presentation] and honey-dos. In-process of drafting fastener discussion. Will try to post by Thursday.

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]

No rush there, Wil. Just don't forget to reach in through the lightening holes and hand your sister a snickers bar or two.

STF

 

[WKTaylor]

Guys... haven't forgotten about next reply... just too busy right now to finish the draft I've started.

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]