projects Archives - Coastal Enterprises
Go to top
Archive
Tag: projects
19 Sep
2013

T-Splines Tip – Panel Line Breaks

I figured I would do a quick little step by step on making panel line breaks with T-Splines.  The first thing is to make sure that you have an isocurve, or isocurve loop that follows the panel line break that you want.  This sometimes can be tricky – in the case of this windshield, I did the topology layout with this in mind.  Here’s the loop of isocurves that I want to form the break on – it’s the junction between the main body and the front windshield: Now just run tsBevel, the options I use for something like this are: Segments = 5, Positioning = Distance, Keep on Face = Yes, Retopo Snap = No.  You’ll need to play around a bit with the bevel distance, in this case I found that 0.09 worked nicely, but you’ll have to adjust it for each model, depending on size.  Here’s what you get: Now select the middle face loop , and set your drag mode to UVN: Now just push that face loop inwards using the normal (blue) arrow.  I like to do it numerically, so that I can make them consistent.  I used a value of -0.05 here, but again you’ll have to play around a bit for each application. Now just run tsMakeUniform.  You should do this after every change in topology.  In smooth mode you get this: It’s quick, it’s easy and it looks good.  Can’t beat that!      

11 May
2012

48″ Convair 240 – Wing Lofting

  Pecking away at the Convair 240 radio control model.  This is being done for a contest on the Ezone forums , and so I really need to get moving on this thing to get it in the air by June 30th contest end date.  The wings are getting there – mostly lofted, just have to add the rear spars and ailerons.  Having designed more than my fair share of RC planes in 2D using AutoCAD, boy, is this a better way to work!  The wings are going to be my favorite style – 1/32″ balsa d-tube leading edges with shear webs.  Super strong, super light, and very “stick and tissue.”  I’m going to try to get the tailfeathers and fuselage formers knocked out this weekend.  Haven’t quite decided if I want to do the nacelles out of balsa or vac formed styrene, but I’ll have to make a call on that pretty soon.

03 May
2012

Using Rare Earth Magnets to Join Composite Structures

You know what I think is a little silly?  Using clecoes to join composite structures.  I mean, sure, they work, but unlike when you are working with sheet metal, at the end of it you have a bunch of useless holes that have to get filled in.  Also, multiple insertions/removals of the cleco tends to open up the hole, reducing the grip of the cleco.  So I started thinking about a better way of doing it, and I think I’ve come up with a workable solution – rare earth magnets.  I’ve used them on radio controlled aircraft in the past, and have always been amazed by their small size and large holding power.  So, I bought some of these 5/16″X3/16″ N42 disc magnets from K&J Magnetics .  On the inside of joggle for the Giles 200 gear leg fairings I put down some electrical tape.  I then placed a disc magnet every 1.5″ along the length of the joggle. The fairing half is on a magnetic building board that I use for building balsa model airplanes, this was very handy because it kept all the magnets in place while I laid them out.  Tucker and Walnut, faithful four legged shop assistants supervise from below: Okay, really, they’re just watching for that cat that walks by the window from time to time.  Anyhow, moving on – I then broke out the hot glue gun, and just surrounded each magnet with a little glue. All we’re going for here is keeping them from falling off the tape once we slide this thing off the table.  Then, I simply clamped the two halves together at the edges.  After that, I just took one magnet at a time, placed it relatively close to where I thoughts a magnet was.  Once I felt it pull, I just let go, and the magnet snapped into place. Well that was easy!  The “hole layout” was less time consuming than if I was using drills/clecos, and more tolerant of edge distance.  Once I bond these together and the resin has cured I’ll simply slide the exterior magnets off one by one, and then pull off the piece of electrical tape that the interior magnets are glued to.  Easy!  Obviously, this will only work in non-blind applications, but it sure seems like a nice method to me.  If your alignment is super critical, I would still recommend using a few holes/clecoes just to make indexing them together fool proof

27 Apr
2012

The Method Part II – Demolding the Parts

Following up on my previous post on my method of making high quality, re-usable tooling with CNC machined foam and Stretchelon bagging film , here’s how things tend to go on the demold side.  First, turn off the pumps, and then remove all the secondary fabrics/films that are bagging the part: I’ve left the peel ply on here, because the parts are still pretty darn soft, but everything else has gone into the trash.  Now, simply get some air underneath the layer of Stretchelon covering the mold, and then pull the parts up, along with the film: Now, just simply peel the Stretchelon off the parts – if you’ve used mold release, it should come off very easily more or less in one piece: Presto!  And the molds are totally unharmed – same as they were when we started.  Here’s a closeup of the surface texture of the finished part: Yes, a bit of texture, but like I said before, a few swipes with some 80 grit to prepare for primer and the texture just vanishes.  Or, if you really want a smoother texture, just get more dense tooling foam.

26 Apr
2012

Creating High Quality, Reusable Molds with CNC Machined Foam and Stretchelon Bagging Film – aka “The Method”

This post is really the culmination of a few years of research, research that I’ve decided to set loose into the world.  We call this “The Method,” or sometimes “The Stretchelon Trick” and it’s my primary method of creating tooling for composite parts these days.  As  an example, I’m using the Giles 200 gear leg fairings that I describe in depth in this post here.    You might read that post and think “well that’s all well and good, but what the heck are you going to make those molds out of that’s cost effective?”  The answer is medium density polyurethane tooling foam, normally 15 lb. density foam.  For these gear legs, the depth of the mold is about 1.5″, so I used 2″ thick foam.  A 2″X48″X96″ block of tooling foam is about $500, and this particular mold is about 23″X40″, so the material cost is a fairly small fraction of that.  Here’s the machined foam of the gear leg fairings, cut yesterday: Okay, granted, it’s a bit hard with the light here to really see what’s going on, but if you look at the post I link to above, you’ll see this is just the CNC machined version of the gear leg tooling.  Since one of these molds is two part, the next step is to join the two parts together: We usually join them together with some fast setting epoxy and then drive some dowels into them for good measure.  Now at this point you’re probably thinking “well that’s nice and all, but now you’ve gotta spray some sort of primer/sealer onto that thing, and hand finish it.”  But, you would be wrong.  You see, I hate sanding.  Actually, that’s not really true, I like doing body work, but I hate sanding when there’s no point to it.  And now let me show you why there’s no point to it: This, my friends, is the heart of “The Method.”  The green stuff over the mold is an elastic vacuum bagging material called Stretchelon .  Stretchelon is a high elongation vacuum bagging film.  I hate dealing with bagging tape, and so I’ve routered channels into my work table here, coated them with shelf paper and then used hardware store screen door spline to create the seal.  This method works quite well, but you can just as easily make a tube out of the Stretchelon and put the mold in the tube.  I pull my vacuum for the mold from the bottom of the table.  The mold is sitting on a piece of breather so that the air gets evacuated from all around the mold.  The slight porosity of the foam does wonders to allow the air to travel to the edge of the foam.  Closeup, it’s looks like this: The Stretchelon pulls down right onto the foam, sealing it for use.  Pretty cool huh?  Yes, there is a bit of a texture to the mold.  If you want a smoother mold, just buy higher density foam.  But what I’ve found is that since the first thing I do when I go to prime a part is to scuff it up, that texture simply vanishes with a few strokes of a sanding block.   The final thing to do is put some mold release on it – the film is polyurethane and so epoxy will stick to it.  I’ve used Frekote in the past, but I like to use another Airtech product called Safelease 20L .   It can be dispensed from one of those hand squeeze spray bottles you get at the hardware store, so I like the fact I’m not using a bunch of aerosol cans.  And it works, so that’s nice.  Anyhow, once you spray your mold down with some release and then wipe off the excess, this mold is pretty much like any other – yes, it’s slightly more ding prone than standard hard tooling.  If you want to make lots of parts (let’s say, more than 10) then I would simply go with a more dense foam, say 25 lbs.  If you get any dings, you can simply patch them with spackle.  When I first starting doing this, I was really concerned about mold wear, but frankly it just hasn’t been a real problem.  Some of my molds have been used 10-12 times now, and for the most part they look just like they did the first time.  So now, it’s just a matter of laying up the laminate, in this case 3 layers of 5.7 oz carbon: And then, you vacuum bag another layer of Stretchelon (or regular bagging film) over on top of the part: Your standard peel ply/perf/breather fabrics are on top of the carbon.  The vacuum for the top bag comes from a standard bag tap.   So we’ve got two separate pumps running here – one for the mold, and one for the part.  They’re both around 25-26″ of mercury.  You’ll get a few more inches of mercury with a bag tube made with bagging tape, but the cost and labor savings of this whole screen spline method are nice

25 Apr
2012

VariEze N388DT – Fuselage Refinishing

I’m back from vacation!  Had a very nice time, but I’m happy to be back in the States, and working away on some cool projects.  James has been working on scraping all the old paint and fill off of the VariEze.  The paint is Imron, which has been holding up quite nicely, but the fill used by the builder was Featherfill, which in some spots is practically exploding off the airframe.  It’s only further cementing my opinion that polyester based products have no place whatsoever on aircraft.  It’s not a matter of if they break down, only when.  Here’s some pics: Do you know what tomorrow is?  It’s the one year anniversary of the launch of this blog!  And I’m cooking up a very special post to celebrate the occasion!  Stay tuned…..

01 Apr
2012

Fun with MAXshot/EXAscan – Digitizing “The Racer” Fuselage

Last week we were up in Reno, doing among other things, a digitization of a plane that is simply called “The Racer.”  The Racer is an as of yet unfinished Unlimited Reno racer.  It’s pretty much the smallest plane you can build around a Wright R-3350 , which is still a pretty big plane.  The Racer is the brain child of Dave Cornell, the current crew chief of Rare Bear.  It’s been a 23 year, on again, off again labor of love, and now it most definitely seems to be on again.  We’ll be doing a whole range of projects on The Racer, starting with a new canopy plug.  The current canopy doesn’t come down far enough, and so a new plug needs to be made.  Rather than do it by hand, we’re going to take the scanned data, create a perfected T-Spline surface and then CNC machine a new plug.  After that we’ll be re-lofting the aft bottom fuselage, and making all new form blocks.  Then maybe some cowling ducts.  And some canopy skirts and frame.   And drag strut braces.  You get the idea – there’s a whole host of projects we’ll be working on, which is great, cause this thing is really, truly cool. The fuselage digitizing was our first real chance to use the new MAXshot photogrammetry setup.  I’m happy to report that the MAXshot worked flawlessly.  The through process is very much the same to using the EXAscan, but covers a much larger area more quickly, and more accurately.  Here’s some pics from the scanning process: Here’s a screen shot of the capture session for the right side of the fuselage: For the straight lofted panels we just scanned the perimeter – no need to scan every square inch when you can get all the data you need from the edges.  James will be up in Reno again this coming week to scan the wings.  We’re really excited to be working on such a cool project!

21 Mar
2012

T-Splines Tip – Panel Line Breaks

I figured I would do a quick little step by step on making panel line breaks with T-Splines.  The first thing is to make sure that you have an isocurve, or isocurve loop that follows the panel line break that you want.  This sometimes can be tricky – in the case of this windshield, I did the topology layout with this in mind.  Here’s the loop of isocurves that I want to form the break on – it’s the junction between the main body and the front windshield: Now just run tsBevel, the options I use for something like this are: Segments = 5, Positioning = Distance, Keep on Face = Yes, Retopo Snap = No.  You’ll need to play around a bit with the bevel distance, in this case I found that 0.09 worked nicely, but you’ll have to adjust it for each model, depending on size.  Here’s what you get: Now select the middle face loop , and set your drag mode to UVN: Now just push that face loop inwards using the normal (blue) arrow.  I like to do it numerically, so that I can make them consistent.  I used a value of -0.05 here, but again you’ll have to play around a bit for each application. Now just run tsMakeUniform.  You should do this after every change in topology.  In smooth mode you get this: It’s quick, it’s easy and it looks good.  Can’t beat that!