HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old May 15, 2012, 03:43 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Build Log
n00b Composite Scratchbuild 32" Slowflyer

Wingitwo: Electric lazy-cruising, small-ish 32" high-wing, target AUW 500 grams (17oz).

I'm planning ahead that I'll screw things up a few times, so I'd like to be mindful of costs (i.e. screw up with cheap stuff). Parts will be done first in fiberglass with EPS cores, until I'm comfortable with the process. Then I'll move on to carbon and/or Aramid with more advanced techniques.

I'm a staunch open-source advocate, so only Linux-friendly software will be used, all plans will be made and shared in the open, and community input received with open arms. Software toolchain is OpenVSP -> Blender -> Skeinforge -> InkScape -> Dead Tree Format.

Phase I:Design & Prototype: Current numeric details and components. Prototype design: Pics and discussion. Version 1.4 plans.

Phase II: Build molds.

Phase III: Make parts (Simple fiberglass + EPS cores at first)

Phase IV: Assemble, fly, & learn.

Phase V: Adjust, rinse and repeat (possibly using more exotic/expensive materials/techniques).
-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 10, 2013 at 10:15 AM. Reason: Updated Progress
Reply With Quote
Sign up now
to remove ads between posts
Old May 15, 2012, 03:53 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Source

Source Files of my design - Blender file is in cm scale units.

I'm not sure what it's called, but the idea behind the downward-curving winglets came from a trainer plane I had. I really liked the way it glided and handled in turns.
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 16, 2012 at 08:32 AM. Reason: Added Content
Reply With Quote
Old May 15, 2012, 03:54 PM
Balsa builder and balsa basher
RC Trainer plane's Avatar
Joined Apr 2012
683 Posts
Cool Chris, botany pics of her yet. Scratch building is my favourite type of building , jut look at my blog.
RC Trainer plane is offline Find More Posts by RC Trainer plane
Reply With Quote
Old May 15, 2012, 05:57 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Once the wing is smooth I'll do some quick renderings. Not sure if there's a better way, but I was thinking of doing the "rib" layout in skeinforge. The mock won't ever fly, but I read its good to have it somewhat light-ish. Connect the ribs with all-thread or steel or aluminum pipe?

-Chris
razortoe is offline Find More Posts by razortoe
Reply With Quote
Old May 16, 2012, 08:38 AM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
For those that don't know, Skeinforge is an app. used to prepare 3d objects for printing (on a 3d printer). One of it's functions is to "slice" an object into layers of arbitrary thickness. Conveniently, the slices export into SVG where I can refine them further in Inkscape. I plan to exploit this toolchain by slicing the thickness of my former/rib material, keeping every 10th layer (more or less?). What should I use as a material? FFF(negates using poly-filler), Hardboard(heavy and cheap), Thick Cardboard(light but flexible), Balsa (Mmmmmm), Something else, or just send them to shapeways for printing ($0.75 - $3 per cu. cm depending on material)?

-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 16, 2012 at 11:51 AM.
Reply With Quote
Old May 16, 2012, 12:07 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Research

In blender, I created a plane 4mm thick and larger than the wing profile. I then duplicated the wing object, and applied a boolean modifier on it, targeting the the plane. I hide the plane, export selected as X3D and upload to shapeways for a quick quote.

Looks like printing one of these in their "sandstone" material would run about $4 or so. I can lower that cost quite a bit by cutting holes in the former, which I'd need for alignment anyway. If I could get it down to about $2 per former, this could be a good time saver as opposed to cutting all the ribs out by hand. It would also have an extremely high degre of precision (error measured in hundreths of a milimeter)

Thoughts?
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 16, 2012 at 12:45 PM.
Reply With Quote
Old May 17, 2012, 10:16 AM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Update: Actually I read the fine-print, and there's a $3 minimum with shapeways, so really this part cost just over $1, so I'd be in roughly $20 to have 20 of these cut. However I still need a method to "string" them together accurately. I found a supplier of 3/8" x 3/8" aluminium square tube, which for a front/back spar would also run about $20. So basically this method is going to run $40 - $60 (estimate), be very accurate to my plans, and I won't have any chemistry problems.

The other option I need to research is the more traditional stick-building method. I'd use skeinforge to slice the plan, print out every 10th slice (more or less), and cut them all by hand. This is by far the more time consuming route, though with laser-cutting could be of comparable accuracy. I could send the slices (in SVG format) for laser-cutting. The ribs are roughly 9cm x 1cm each and I'd need 20-or-so of them. Anyone have any guestimates as to what custom laser-cutting (in some light but rigid material) would run?

-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 30, 2012 at 01:25 PM.
Reply With Quote
Old May 17, 2012, 09:34 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
More progress in blender. I ended up chopping the wing in half because I can always mirror it later. I also chopped off the wingtip, since it's such a complicated shape, I'll probably just get the whole thing 3d printed. The cool thing I did though was cut out 1/4in "tubes" through the wing without any of them spilling outside the exterior form.

I may expand them a smidgen oversize, but you can see in this pic. what they look like. No matter what prep. method I use (3d or 2d) these will come in handy later for aligning the parts prior to filling. I'll just stick 1/4 OD rigid pipes or rods through them and glue in place.

(This was done by creating cylinders, and using the binary modifier to "drill" them through.)

-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 30, 2012 at 01:25 PM.
Reply With Quote
Old May 20, 2012, 03:09 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Ready for print (almost)

Spent a lot of time in Blender putting on finishing details. I'm really leaning toward 3D printing this wing. However, I can't do the whole thing because it's larger then the print envelope, and it would be crazy expensive. However, by cutting the wing into parts, with holes and hollowed out parts, the cost will be reasonable.

I realized my original plan for holes wouldn't key the wingtip properly so I nuked 'em. I made a new collection, 5 sets of 3 or 4 hole-drillers, all interlocking, and none protruding through the wing surface. After using the boolean operation on them and the wing, it cut produced interior "tubes" of precise alignment and dimension.

Then I created a bunch of boxes around different sections of the the wing, each marking a section I wanted to save. Then again with boolean, I chopped the wing up into sections. I saved all the boxes and original wing in another "layer" for later reference. Anyway, the result is nice neat "formers" with holes that will key and interlock off 1/8" rods (I made the holes slightly larger by about 0.5mm).

My idea here (open to critique / better ideas) is to 3D print all these parts in a material Shapeway calls "sandstone". It's a very rigid, but brittle material with reasonable accuracy, and very low-cost. Then I'll string the parts together on 1/8" rods, referring back to blender for the spacing. Finally, I'll stuff the cavities with polyester fluff, and Bondo filler.

This should keep the model accurate to blender, somewhat light, but also soft enough that I can shape it with a sanding blocks and/or electric sanders.

My main concern with slicing the Blender model up and having some collection of slices laser-cut in balsa is: Balsa isn't as rigid as "sandstone", so I'd need a lot more formers which will bring a lot more measuring and possible headache. Though I'm open to other ideas, what y'all think?

-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 30, 2012 at 01:25 PM.
Reply With Quote
Old May 22, 2012, 09:47 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Ready for print, for real

I knew I was forgetting something, went back to my model, and did a face-palm: It might be nice to have some control surfaces!

Anyway, after some more fiddling in Blender, I got some nice ailerons cut out with bevelled leading edges. Opposite bevelled edges on the mating ribs.

I also dealt with the last item that was daunting me about uploading to Shapeways: Units and Scale. However, I was relieved upon finding a tutorial page answering all my questions. Turned out I needed to stick blender into Metric Units at 0.001 scale. Rotate my models 90* on Z and X. Export in STL format, and import using millimetre scale on Shapeways.

These settings result in blender displaying correct units, i.e. 1cm in blender == 1cm printed out. Grand total is $51.63 delivered to my door in two weeks. As much as I enjoy cutting wood, sure can't beat that price + 40-micron precision. Hmmm, what to do while I wait, oh! I know! I can fly my other planes

-Chris
razortoe is offline Find More Posts by razortoe
Last edited by razortoe; May 30, 2012 at 01:25 PM.
Reply With Quote
Old May 30, 2012, 01:24 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Finally got a nice-weather day to go flying while I wait for my wing parts. I just got an e-mail from Shapeways yesterday saying that printing had begun and I could expect a UPS notice next.

I spent a bit of time refreshing myself on composite construction from this great book. I also caught up on Nodd's new build-log where, where he references two helpful articles on shear loading.

As it relates to the wing of my project, I definitely need to use a bias cloth orientation schedule and come up with a way to use a high/low combination of foam densities for the core. The foam core moulds should be fairly easy to realize later on once I've got a handle on the basics.

I'm also considering (also for later passes) using my silicone supply for expansion moulding (the next-level up from using an inflation bladder system). Basically you do a clam-shell with a solid (or cored) silicone insert. Heating the layup causes the silicone to expand and exert pressures on the laminate, consolidating and pressing out excess resin.

The pressures generated are MUCH higher than what can be achieved with a bladder, and carry none of the leak-risks. Since I already have the silicone, I might as well give it a go. Otherwise, the main downside to this method is the cost, silicone is REALLY expensive. Even this el-cheapo tin-cure RTV was nearly $150 for a gallon.

I've also been working on designing a fuse for this model in OpenVSP and will post some pics as soon as it stops looking like a upside-down guppy.

-Chris
razortoe is offline Find More Posts by razortoe
Reply With Quote
Old May 30, 2012, 02:54 PM
Just fly it!
wyowindworks's Avatar
Cody, WY
Joined Nov 2007
6,915 Posts
Quote:
Originally Posted by razortoe View Post
I'm also considering (also for later passes) using my silicone supply for expansion moulding (the next-level up from using an inflation bladder system). Basically you do a clam-shell with a solid (or cored) silicone insert. Heating the layup causes the silicone to expand and exert pressures on the laminate, consolidating and pressing out excess resin.

The pressures generated are MUCH higher than what can be achieved with a bladder, and carry none of the leak-risks.
The expanding silicone is a cool process but has it's limitations. Model parts rarely are conducive to the process.

The compression force of an expanding silicone is dependent upon the cross-section of the shape. The larger the silicone the higher the potential is for pressure. The inverse is also true. For small/thin cross-sections the pressure exerted can be rather small. With a bladder the pressure is regulated by the air pressure rather than the cross-sectional size and temperature. It's easy to regulate the air pressure of a bladder at varying levels during the elevated temperature cure. With a silicone, of sufficient cross-section, the pressure increases as the temperatures are raised. Many molds can be destroyed by simultaneously using high pressures AND heat, which is exactly what a silicone requires. With a bladder you can use high pressure and keep the mold at mildly elevated temps until the resin sets then reduce/eliminate the pressure and use higher temps. It's also possible for the resin to begin to gel before the silicone reaches a temperature that can generate serious pressures. This reduces the effectiveness of the compaction since the resin can no longer flow. Silicons work best with a CNC cut aluminum mold and prepreg resins that are designed for high temperature use.

Also, expanding silicons only work on structures and processes that allow the silicons to be removed. Bladders can be removed from almost any shape while silicons require large opening/ends for extraction.
wyowindworks is offline Find More Posts by wyowindworks
Last edited by wyowindworks; May 30, 2012 at 03:09 PM.
Reply With Quote
Old May 30, 2012, 08:42 PM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Quote:
Originally Posted by wyowindworks View Post
The expanding silicone is a cool process but has it's limitations. Model parts rarely are conducive to the process.

The compression force of an expanding silicone is dependent upon the cross-section of the shape...With a bladder the pressure is regulated bAlso, expanding silicons only work on structures and processes that allow the silicons to be removed. Bladders can be removed from almost any shape while silicons require large opening/ends for extraction.
Wow, that's fascinating, I never would have guessed it depended on cross-section. Thanks!

Maybe you know too, how come silicone skin-mold + mother mold isn't used for composites? I'm deathy affraid of inadequate release and locking my models into the mold. Seems silicone skin-mold would mitigate both, but I don't read about it anywhere!

Thanks again!
razortoe is offline Find More Posts by razortoe
Reply With Quote
Old May 30, 2012, 10:10 PM
Just fly it!
wyowindworks's Avatar
Cody, WY
Joined Nov 2007
6,915 Posts
Quote:
Originally Posted by razortoe View Post
Wow, that's fascinating, I never would have guessed it depended on cross-section. Thanks!

Maybe you know too, how come silicone skin-mold + mother mold isn't used for composites? I'm deathy affraid of inadequate release and locking my models into the mold. Seems silicone skin-mold would mitigate both, but I don't read about it anywhere!

Thanks again!
A silicon skin is just too soft to get accurate results. Most wing skins utilize a sandwich construction which requires a vacuum bag. The pressure generated from the vacuum will distort the soft silicone mold....even if it has a mother mold.

Even if you don't use a vacuum bag the mold halves need to be pressed together when joining. This will also distort the mold.

You want a firm mold that will hold it's shape if you want accurate parts and nice seams.
wyowindworks is offline Find More Posts by wyowindworks
Last edited by wyowindworks; May 30, 2012 at 10:16 PM.
Reply With Quote
Old May 31, 2012, 06:34 AM
Keep it Simple Stupid
razortoe's Avatar
United States, NC, Raleigh
Joined Jul 2011
244 Posts
Quote:
Originally Posted by wyowindworks View Post
The pressure generated from the vacuum will distort the soft silicone mold....even if it has a mother mold.
Thanks, that makes sense. I suppose even the 'hard' silicone would ultimately distort under pressure. I guess I'll save the silicone for pilot figures and widgets

Wax, carpel-tunnel, and PVA here I come
razortoe is offline Find More Posts by razortoe
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Sold Gaggle of Eflite WW1 Biplane Slowflyer ARF's slogo Aircraft - Electric - Airplanes (FS/W) 4 Dec 24, 2012 12:20 PM
For Sale 32 sets of plans for scratchbuilders $250 101stAB Aircraft - General - Miscellaneous (FS/W) 0 Mar 20, 2012 03:59 PM
Build Log Preelude's Delta Fighter | EDF | Full Fuse | 32" | Scratchbuild preelude Foamies (Scratchbuilt) 10 Jan 13, 2012 02:53 PM
Sold GWS slowflyer props lot and more.$29.99 flyarmy58 Non R/C Items (FS/W) 2 Sep 04, 2011 05:22 PM
For Sale slowflyer 32 epp w/2 wings jim e Aircraft - Electric - Airplanes (FS/W) 6 Apr 02, 2009 06:01 PM