Secondary Composite Structures

[verymadmac]

I am working on prepping for certification of a carbon fibre helicopter belly tank (So restricted Cat, secondary structure, FAR27).

The regularity requirements (not located in the US) now include the FAA's Rotorcraft Directorate Policy on Secondary Composite Structure (policy attached). I am trying to get my head around it, looking for a couple of low hanging fruit to pull the test load down a bit, which drives me to a couple of questions.

The environmental accountability factor.
The Wet-Glass transition temperature (Tg) doesn't carry any qualifiers (colour, sun conditions), just a minimum 50 F above max operating temp (which is the same in the small aircraft policy doc for Wet-Glass transition temperature). The EASA CS-VLA AMC 613 uses much the same approach but breaks the knock down factor of 1.5 into 1.25 for temperature & 1.2 for moisture conditioning. Is this separation of the temperature and moisture factors also done in an FAA AC anywhere?

Is there a conservative rapid method to correct the Tg for the fact that the tank has no horizontal surface subject to direct solar radiation (I have chased down the references in CS-VLA, NASA CR-3290 & NASA CP2036 but can't find where they got the data from in them).

Strictly speaking as the environmental accountability factor is to account for differences between as manufactured condition and actual in service condition . When a production limit load test is carried out shouldn't it actually include this factor (doesn't appear to be the intention in the policy doc).

 

[SWComposites]

First of all, a Policy Memo is NOT a regulation. Dont let anyone impose it as such.

Ignore the Tg stuff. That is just a material selection guideline.

If you are certifying by test with a room temp test, you need to increase the ultimate load by an environment factor and a material scatter factor. The env factor can come from coupon tests at your max design temperature + moisture contioned ratioed to RTA data. The scatter factir can also come from coupon data, basically average strength divided by B-basis strength, from data from multiple material batches.

 

[Sparweb]

Air New Zealand just painted their entire fleet of 787's black. Your guess is as good as mine.

There are aircraft manufacturers (Boeing notwithstanding) that forbid their fuselages from being painted colours darker than "X".

I believe you could try modeling the thermal load on a cylindrical vessel, including shading if you wish, based on principles from a good thermodynamics textbook. I have seen an ASHRAE design guideline about air conditioning systems that included calculation (estimation) of heat load on the fuselage given certain ambient conditions (sun, wind, ground condition, etc.) which may give you guesses some analytical rigor.

I have also seen the environmental knock-down factor evaluated in detail. Now that I think of it, I don't believe the individual doing the breakdown was using FAA (or any other) guidance but was relying on "tribal knowledge". Not my tribe, unfortunately. I thought it was in the very policy memo you linked... apparently not... the policy you link is from the Rotorcraft directorate. Try this instead: Transport Standards Branch Lists of Part 25/26 Rules and Their Associated Advisory Circulars and Policy Statements especially under 25.613.

My tribe does, however, include the knock-down factor in load tests. Limit and ultimate tests. I do not recommend certification based on limit tests and analysis to ultimate.

It may just come down to testing some samples with thermocouples on a hot day in Phoenix AZ to determine what temperature they actually rise to.



STF

 

[WKTaylor]

verymadmac...

A belly-mounted [external] FUEL tank has significantly different requirements relative to a belly mounted OIL, WATER or liquid CHEMICAL tank... which is it?

IF this tank is for long-range ferry only, that is one thing... certain risks can be waved...

However if this is a permanently mounted tank intended to extended/continuous use... that is an entirely different beast. Every element should be considered for safety.

CAUTION.
Chemical resistance of the Tank must be robust to accommodate chemistry variations of the primary liquid-load.

Chemical reaction-rates double for every 10F rise in temperature. So... You need to consider tank-material compatibility with Your liquid at mx 'normal' high service temperature... may surprise You how bad the matrix [resin] chemistry degrades. Also flight at extreme low temperature risks 'freezing' the liquid payload... and pipes/tubes/hoses.

Are You aware of the following?

SAE AIR4069 Sealing of Integral Fuel Tanks
SAE AIR5774 Composite Fuel Tanks, Fuel System Design Considerations
MIL-T-18847 TANK, FUEL, AIRCRAFT, AUXILIARY EXTERNAL, DESIGN AND INSTALLATION OF

FAA AC23-10 Auxiliary Fuel Systems for Reciprocating and Turbine Powered Part 23 Airplanes
FAA AC25-8 Auxiliary Fuel Systems Installations
FAA AC120-97A Incorporation of Fuel Tank System Instructions for Continued Airworthiness into Operator Maintenance or Inspection Programs
FAA AC20-53B Protection of Aircraft Fuel Systems Against Fuel Vapor Ignition Caused by Lightning
FAA AC25.981-1C Fuel Tank Ignition Source Prevention Guidelines
FAA AC 25.963-1 Fuel Tank Access Covers
FAA AC 25.981-2A Fuel Tank Flammability
FAA AC120-98A Operator Information for Incorporating Fuel Tank Flammability Reduction Requirements into a Maintenance or Inspection Program
FAA AC25.994-1 Design Considerations to Protect Fuel Systems during a Wheels-Up Landing


NASA CR-3875 Fuel Containment, Lightning Protection, and Damage Tolerance in Large Composite Primary
Aircraft Structures [concepts]

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]

Wil,
My guess is this is a tank filled with water and fire retardants. Carries the load for a few minutes at a time, an hour at most, and drops it all VERY FAST [SMILE]

Mac,
I have some (second-hand) experience with an all-composite firefighting tank that I got to examine in detail and discuss with the designer. If you're willing to share, I can guarantee a keen audience of at least one, and some (hopefully) valuable experiences to pass on. That project came very close to "you can't get there from here" many times.

STF

 

[verymadmac]

Its a spray tank, for which the loads are covered by AC27-1 MG5, its much the same as a firefighting tank except no controlled drop function (dump is all or nothing) & more complicated plumbing. Typical cycle time probably about 10 minutes. The max specific gravity fluid to be used in this one is about 1.1, the previous version run to a max SG of 2 (they spray suspension fertiliser using really big nozzles), although this mainly impacts the spray booms.

The previous engineer involved with this had suggested limit testing each tank as a method to reduce the material / manufacturing variability factor, hence the question. I would rather take tensile test coupons from the hatch trim outs also provides a realistic data population for future tank projects. All the composites on this one will be demonstrated by test to ultimate.

I haven't had a look yet at ASHRAE design guideline about air conditioning systems but it looks interesting.

We use to have a joke about rotorcraft spray systems, when asked to certify minor changes to a spray system, it usually meant that it looked exactly almost the same but was completely different.

The regulator here won't allow belly mounted ferry fuel tanks on helicopters for obvious reasons, hasn't stopped at least one helicopter operator here having nice pics of their belly mounted auxiliary tank on their website (I can guess how much paper work went with that one).

I guess AirNZ has spent some time finding a paint that while black, has low absorbancy in the IR spectrum.

 

[WKTaylor]

VMC, SW...

My military-mind just pointed me at fuel tankage... what other Tanks are there???... maybe water-alcohol for engine cooling/performance and napalm???

OH YEAH.. fire-fighting and crop-dusting for starters. Dohhhh

Comments... for what they may be worth... now that my left foot is wet with saliva...

What are the mixed water-retardant properties like? Almost 'water-milk-like' [low viscosity]... or thin mud [high viscosity]... etc? Does the mixture outgas [fizz] or retain entrapped air bubbles.

Ensure adequate baffling to minimize 'slosh' inertia effects.

I'm sure You've considered over-pressure [blow-out] loads... what about under-pressure [suck-in] loads... just in-case the tank vent becomes blocked [if that's even possible] or during rapid altitude changes, etc [delta-P].

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]

Mac,
I have recently perused the NCAMP files without delving too deeply. Have the layups/processes/allowables for your tank structure been substantiated already, or is that also on your plate?

That ASHRAE suggestion was just a fun idea thrown out from the top of memory... probably not a valuable avenue if you already have a thermo textbook with a heat transfer chapter (convection+radiation). You have much bigger fish to fry.

-it looked exactly almost the same but was completely different.

That sounds like the tea on the starship Heart of Gold....
...And all of the external mods I've done on helicopters, too, for that matter.

I would rather take tensile test coupons from the hatch trim outs also provides a realistic data population for future tank projects. All the composites on this one will be demonstrated by test to ultimate.

That's probably a good way to succeed. Will you have enough samples to do some environmental conditioning on, say, half of them? It's still a bit risky (from a regulatory point of view) to get into sample testing because making overtures to testing all environmental conditions, but testing too few samples, leaves the margins realy wide, and you still get stuck with an awful knock-down factor.

I need to ask a stupid question now: Is there any (ahem) sandwich core in this shell?

STF

 

[SWComposites]

Limit testing each tank does NOTHING for a material variability factor, as there are no failure data to use.

Tensile strength is typically not critical for composites and typically not environmentally sensitive. Compression testing would be more appropriate.

 

[Sparweb]

SWC,
Good point, we were talking about ultimate tests, but I wasn't thinking specifically about compression loads, thanks.

STF

 

[verymadmac]

No honey comb in this tank, just laminate.

Over pressurisation is prevented by having two filler ports (one each side, not capped).

Heat transfer model is simple except that the tank is subject to secondary IR radiation sources i.e tarmac etc which doesn't make for a nice clean comparison (not a lot of assumptions in solar radiation levels), although thinking about it, the thermal mass of tank contents will probably win but not quite so simple for the spray booms.

Layups etc were spec'ed by a composites engineer, demonstration of compliance is my problem. Unfortunately testing to failure of the tank and other large structures isn't on the cards.

Not sure about the number of samples per tank.

For compression testing, what standard is most practical for this.
I see ASTM D3410 while it appears the most common, is noted as been sensitive to loading procedure & fixture setup, which for something that is batched produced by a small company, one would need to the samples out for testing to get repeatable data.