Rapid Prototype of Fairing for End Use Product 1

[Chris843]

I'm basically in the process of designing a junction style fairing that will wrap around some butted tubular sanctions on a modified aircraft. They are somewhat complex fairings and they need to be manufactured in a timely manner. I do not have a lot of experience with composite materials nor rapid prototyping. So anybody have any recommendations on a process/material to make these one-off fairings that will be strong enough to fly on an experimental aircraft as an end product? Fiberglass, Kydex (or some other rapid forming thermoplastic) I'm even open to 3D printing options if there's a industry standard out there for doing it.

 

[rb1957]

external or internal ? (external i suspect)
airspeed ? (sub-sonic ?)
scale of airplane ? ... (large, medium, small, UAV)
permanent (like fiberglass moulded onto the staunction (?)) or removable ?

Quando Omni Flunkus Moritati

 

[Chris843]

The aircraft is a De Havilland DHC-8. The size is a 300 series. So max cruising speed would be around 350-400ish knots. And yes, this is and external fairing application. The fairing will be fastened to the sanctions and will buyt up to a custom pod structure. Hard to explain the geometric application of it because this is for a special mission project. But it will need to be removable for particular reasons.

Not exactly, but something similar to this concept from a Cessna on aircraft spruce's website,

http://www.aircraftspruce.com/catalog/appages/stenefiberglass205c.php?clickkey=6937

 

[rb1957]

don't see a DHC-8-300 doing 400 kts (more like 280) ...

you could make up a two piece fairing like your link ... sheet metal if you don't need to fair into the wing/fuselage (like the cessna one does) ... maybe bend the s/m over a slightly smaller tube than your stanuction (so it'll wrap your stanuction well) and bend tabs to pick up the skin (bending a full flange would be too difficult unless you've got a full shop to work with ... heat treat, form, re-heat treat). make the trailing fairing out of two pieces rather than one with a really tight bend on the T/E, again bend tabs to pick up the skin.

0.032" 2024 would be plenty, maybe 0.025" (much easier to work with).

or make same from fiber-glass ... room temp cure

Quando Omni Flunkus Moritati

 

[KENAT]

There are options for low volume aluminum castings if you want to go that route. If geometry allows they can build a rapid prototype model (including shrink factor) and cast from that.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​

 

[Chris843]

Well...the link of the fairing is something like what I am trying to do but not exactly. I've upload a picture to Teamplatform if you want to take a look. It's just a rough concept, but give's a general perspective of the geometry of the fairing given the manufacturing process capabilities of it. Unfortunately I can't show exactly what it's wrapping around (which is the root cause for the issues of its complexity) but you will get a better idea anyway.

Aluminum casting? Wouldn't that require a ton of machining to make a mold for? Something I'm trying to avoid unless the mold can be made of something non-metal that doesn't require a ton of tooling hours to make. There's only going to be 4 of these.

Here's the collaboration website with the link to the picture.

https://saicbw.teamplatform.com/pages/43339?t=8hcsga

 

[berkshire]

Cris
I am going to give you two options , there are many more.

For this low a volume, find an aircraft sheet metal mechanic that can do hollowed out/ formed shapes i.e. nose bowls for engine cowlings.
Get him/her to pound them out on a sandbag, power hammer, english wheel, or whatever else is to hand. The labor cost is high for this type of work , but you only need 4. They will be a finished part you can put on the aircraft.
If you use 3000 or 6000 series aluminum you can weld the spigots onto the main part of the fairing

Option 2 make plaster/ tool rock molds for part, lay up composite part. looking at your piece you may need a left hand ,and right hand mold.
If you want a smooth surface, make female mold from toolrock part. Then depending on strength desired you can use room temp open lay up, or prepreg from any one of a dozen different materials.
B.E.

 

[GregLocock]

Option 3 wood pattern, light alloy casting

I think a glass fibre splash over a wooden mould is the way to go, but given that the part is reentrant you might want to go with afemale mould and lots of split lines.

Bashing sheetmetal with a hammer and then brazing the parts together is certainly a possibility

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[Chris843]

Thanks guys. We do have an in house machine shop that can do some sheet metal forming and even work working. Unfortunately, there tooling is older than dirt so anything they build would have to be derived from technical drawings. Which by the way, is there any good examples out there I can find fairing drawings? And yes, the mechanics could probably hand construct a fairing on the floor...3D scan it and so on....But I rather keep this in the initial engineering rather than reverse-engineering.

Anyway, I'm still really keen on sending out a model for 3D printing. Anybody have any experience with this similar to the application I'm trying to do? Since this is not a structural part and just needs to take on the aero forces I feel like 3D printing has come a long way to suffice for this application. What are opinions on the best type for me to research? FDM is my best conclusion so far. Or maybe still leaning toward some kind of thermoplastic. Any website links would be helpful for me.

 

[MikeHalloran]

The first thing that came to mind is the water- activated glass fiber reinforced tape that's sometimes used for making casts for broken bones. Rapid is the operative word, because the stuff sets up in like 15 minutes.

You could put a release layer of Saran on the pod and the stanchions, bulk up the fairing in situ with modeling clay, or styrofoam, or balsa, or whatever, apply the tape, then optionally a layer of Saran and some balloons and duct tape to smooth out the surface texture and densify the part, working _rapidly_ of course. ... or use a slower curing prepreg and work carefully.



RP machines in general, especially the 'affordable' ones, have a pretty limited build envelope. The biggest RP part of which I am personally aware is a custom developmental intake manifold for a Big Block Chevy, built in four parts with mating dovetails. The producer (also incidentally a RP machine retailer) asserted that epoxy parts lasted long enough for testing; the heat riser channels charred but didn't burn through right away. The RP parts were only used long enough to get dyno and field test data; the good ones were replicated in aluminum for production.



Mike Halloran
Pembroke Pines, FL, USA

 

[rb1957]

"We do have an in house machine shop ..." good
"anything they build would have to be derived from technical drawings" ... only if your guys are thicker than planks ... good shop guys should be able to work from a sketch on a "fag packet" (which isn't, of course, a gender preference slur).

why the focus on "rapid prototyping" ? is this a one off or a production run ? fairing something on a DHC-8 is pretty much cosmetic ... any sort of fairing will do, maybe something with a 20deg angle. i wouldn't bland into the mating surface ... i'd use a P-seal to interface (to protect the mating surface).

and once you have your prototype, you'll have to make an engineering part out of it. mike's suggestion of fiber glass (as in arm casts) is certainly rapid but is it certifiable ? the problem being prove it doesn't "depart the airplane in an unsafe manner (should it choose to do so, ie depart (= fall off) the airplane)".

if you reaaly mean rapid prototype ... styrene foam and "speed tape" would work to give you a good shape (and easy to adjust) to productionise.

Quando Omni Flunkus Moritati

 

[Chris843]

Fag packet....I do work with an English guy on occasion. Guess we call it napkin sketches sometimes 'round here. But this is a DoD project and don't think AED would accept that type documentation!

This is a onefour-off production run. And the fairing isn't part of the DHC-8 OML. It is a "trussed" style structural frame that runs up into ttached pods on the aircraft and suspends a antenna ground plane on the underside of the aircraft. The structural frame is fair'ed and this is where the sanctions all come together on the pod. Very hard to explain I know.

I refer to rapid prototype that will also be used as an end product too, or engineering part per say. The 3D file is used for the design activity and is QA'ed with. I'm really trying to explore these options for a few reasons. We (my company) does the design/engineering. Another company does the integration of our designs. So the way the contract is set up my company would have to contract another company to come in and do the molding work and then we document to build for another one or capture what was done. It's hard to run down the logistics of the contract, but doing reverse-engineering is costly and time consuming on our side. My interest in 3D printing is that there is no costly tooling to make for an one off composite part and lead times are short. Problem is there's several different techniques to this and material properties are shady from my research. Even using something like Kydex is an option. But I just don't know how labor intensive making a mold is for shrinking the shape to is would be. Bottom line is that I have the shapes. Who (links if possible) can I send 3D files to and can make an end product.

 

[rb1957]

"napkin" ... can mean something else too !

seriously, why fair the struts ? is flight in icing a concern ? is the pod the junction of the truss, or do multiple legs pick up on the same pod fttgs ? the latter is harder to fair ... fiber glass is the only way there ... make a dummy fairing from polystyrene and use this to create the fiber glass fairing. the struts can be faired either with fiber glass or sheet metal ... i'd mount these pieces off the legs and use them to secure the fiber glass fairings.

personally, the most i'd do (being a -8 guy) would be to "squeeze" the round tube into streamline shapes (and keep the round tube at the ends for easy of joining.

Quando Omni Flunkus Moritati

 

[Chris843]

Take a look for yourself...

https://saicbw.teamplatform.com/pages/43437?t=q4nuq

. Gotta love that website. But the 3D is stripped down dramatically and cut down not to compromise the design. But I think you will understand the area that needs to be fair'ed in. Anything above the struts/sanctions would be inside a pod and will not be subjected to the airstream. The struts/sanctions go up into the pod and are secured to a hard point. So there is nothing really being transferred in relation. Even though the struts/sanctions and pod are not tied into together as a junction point they do "junction" around each other in close proximity.

Also, there are cables that will run from the inside of this pod down to the ground plane via the strut fairings. Even if the CFD data was good those cables can't be exposed environmental factors and need to be shrouded.

Anyway, take a look at the collaboration site link. Any input on the shape is accepted as well. I'm not an aero guy by far. Just keep in mind my original comment about sourcing this to a third party manufacturer to build.

 

[berkshire]

Chris,
I suggested "Tool rock" thinking that you wouls be fabricating parts from a drawing in a tool shop setting.
From what you are saying you have the brackets for your pod already mounted on the aircraft together with the struts.

Now if you want to get down and dirty, you can take a plastic bag full of two part self expanding foam, wrap it around the fairing junction, wait for the foam to expand and to harden, then whittle away anything that does not look like a fairing. Carefully cut the result into two or three pieces to remove it then putty and smooth the outside. you can tape that on and use it for a test flight article, modify it as needed then use the final result as a basis for a mold or a final drawing. You cannot do production and QA until you have a final result that is satisfactory to all of the participants.
B.E.

 

[Chris843]

berkshire,


Yes, the pod and struts are mounted on the aircraft now. The task is to fair in the traffic jam of struts and the flat bottom part of the pod.

I do not want to get down and dirty sorta speak. Meaning I have already have my shapes in 3D CAD now and that is my mock-up....3D printing will be the physical prototype. The original part of my question is has anyone used this type of application or other rapid prototype methods derived from 3D models to be used as and end product as well. The process of getting there isn't primary concerning to me -unless- it alters the manufacturer-ability of my overall design. I'm merely only going to accept the final product that derives from my intent. Whether that's a rapid prototype method or me sending drawings out for a third party vendor to make.

 

[MikeHalloran]

The logical next step is to contact a bunch of RP contract houses and ask them what their maximum build envelope is. That will tell you how many pieces you need to cut your solid models into, in the (likely, I think) event that you have to make the fairings from more than one piece.

Also ask what file format they prefer, so you can generate the appropriate files from your solids. You shouldn't need to send drawings except for reference purposes.

I suggest adding some self-aligning/ self-fixturing features (e.g. dovetails) to help in assembling the parts.

This would also be a good time to ask about what materials are available (lots) and what mechanical properties they will guarantee (probably none). I'd expect the produced articles to be anisotropic, e.g. relatively weak in the Z-axis direction. Ask for any quantitative data they might have.

Note that the surface finish in the Z-axis direction is also usually pretty bad, because striving for good finish in that direction, i.e., thinner layers, increases build time, and price. Anyone doing the assembly of sub-parts will need sandpaper and Bondo or similar.



Mike Halloran
Pembroke Pines, FL, USA

 

[berkshire]

You might want to contact 3D systems inc, they make these machines. As you already know they range in price from a few thousand to hundreds of thousands of dollars depending on what you want to do with them. Ask them if they can give you names of their customers who have bought production machines who would be interested in taking on contract work.

http://www.3dsystems.com/

I am not affiliated with these people.

B.E.

 

[rb1957]

if you've got a cad model of a fairing, why not go the 3D printing route ?

cover the shape with "speed tape" or glass fiber. i'd suggest the former for flight test, get satifsactory results then productionise it ...
you'll need to think about joining segments together (shouldn't be too difficult ... use sections of the 3D printer output, carve away a little to allow for an overlap,
you'll need to think about how to attach to the struts, probaby little stand-off brkts,
and seals to protect the mating structure

Quando Omni Flunkus Moritati

 

[KENAT]

These guys do rubber plaster mold aluminum casting which is fairly well suited to relatively low volume production.

http://www.aljcast.com/index.html

Depending on geometry etc then hey may be able to make the mold from a rapid prototype part as I mentioned before so no it may not require lots of machining of the mold. Other companies offer similar low volume casting processes. For some low volume parts we had getting them made by this method was cheaper than rapid prototype selective laser sintering.

From the Rapid Prototype plastic parts I've played around with I'm not sure they'd be my first choice for in airflow flight components but I may be missing a trick. I've done some FDM parts made from ABS and they were good enough for production on in lab scientific instruments. However, some other parts that 'clipped on' proved a bit too fragile and couldn't take many cycles - if nothing else I suspect the layering caused some stress raisers.

From the links I couldn't size information. Large rapid prototype parts get expensive quick.

Posting guidelines faq731-376

http://eng-tips.com/market.cfm?

(probably not aimed specifically at you)
What is Engineering anyway: faq1088-1484​