Miles Aerovan 36" span scratch build
 

This build will use keel and former construction, using a 3-view for the plan. As usual, the necessary formers and keels have been derived from the 3-view drawing without CAD assistance. For whatever reason, there are few color photos of the plane, even though it was built after WWII. One interesting color scheme that would certainly be original, would be the IAF camo scheme, which I have in mind. There is a multi-view color rendering of it, which seems to agree with B/W photos of the plane.

http://www.britsim.com/components/co...mg_1601_02.jpg

http://warbirds-online.org/images/Pi...sAerovan01.jpg

No one will ever accuse you of building ordinary airplanes.

Looking forward to this one.
Paul

Seconded :D

Barry

Yep this one's unusual. I thought I had an unmodeled subject, until I came across the Nijhuis build after deciding to build the model. There's a lot of options on this one, as far as locating the gear is concerned. With the tail heavy tendency, gear will be located forward. The simplest method for tail control would be to mount the servos at the base of the boom, but in the interest of forward weight, I now have a tail assembly with long pushrods hanging out of it. The servos will probably mount in front of the wing, and slightly below. Since the tri-rudder and split elevators are a bit unusual, I figured on starting with the stab fabrication. I now have a stab frame with a "Y" pushrod configuration for the elevator halves, and a bellcrank setup for the rudders. The pushrods travel about 5/16" before binding, which provides ample elevator and rudder throws with proper control horn linkage geometry.

The rudder setup is similar to what I used on the Dornier Gs build, although this plane has a center rudder adding to the complexity, which will be driven from a bellcrank mounted pin that inserts into the rudder bottom. The bellcrank center aligns with the rudder hingeline. The outer rudders are driven by a pushrod mounted inside the stab, which exits the stab bottom at the hingeline, roughly 2 inches from the rudders. The distance from the pushrod exit point to the rudder horns is necessary to provide ample distance from the exit point to the rudder horns, allowing for some lateral movement when in motion to prevent binding. A slight downward pushrod angle from the exit points to the rudder horns will prevent interference with the elevator when moving downward. The stab will be sheeted with 1/32" balsa.

As Barry said before, one advantage of twins is that the needed motor power and size is reduced. Looking at the nacelles, some cheat will be needed to house outrunners, slightly lowering the props from scale location. It shouldn't be much however, as 200 class outrunners should power the plane, with reasonable weight. I have small Axon outrunners that are barely 200 class size flying a 19oz Guillows DC3, probably with less than ideally efficient props.

Another unusual but oddly handsome airplane.

Interesting way to do the rudders. You'll have to show a photo of the rudder pushrods when the elevators are attached. Is there a "Z" bend downward in the pushrods to clear the elevators at the hinge line?

I've got a DH-91 Albatross at the top of the build queue for this winter and it has twin rudders; always looking for better way to do them.

I'll have to show the other side also. There is a groove channeled in the bottom of the hinge frame, where the pushrod exits from. A small downward angle is all that's needed, as the pushrods will remain close to the hingeline such that the elevator would require reasonable down deflection before hitting the rudder pushrods. The pushrod ends may exit just a slight distance from the actual hingeline, so as to not contact the elevator at the initial exit point. The rudder horns will mount slightly below the hingeline, and reasonably close to the rudder hingeline, although not too close so as create slop/softness, such as 10mm distance. Not much down elevator travel should be required either, unless I were to attempt to fly this thing inverted. :Q

One of my favorite types. I've done quite a bit of research over the years. You'll probably enjoy this color video. The Aerovan fly-by is brief, but enticing. http://www.rafjever.org/stationpic1039.htm

There is little footage of this plane, given that it's post WWII. Even the photos are almost all B/W. I'll have to find what plug-in is required to view the video clip, as I've become leery about downloading media software from a site that says you need to get it from them to watch a clip. I also noticed that there is a sim version of the plane. I would imagine that some of the top sims have become advanced enough to simulate flight based on the overall shape, weight, power, airfoil sections, etc. It would be interesting to see what the full scale flew like.

Not much progress to show, other than building the rudders, sheeting the stab, and cutting some wing formers. The basic wing framing is now drawn out also. Once again I'll still try to push the idea of scoring a plan with a dull exacto, which leaves a print on the wood below for the cut line. It's a fast and sufficiently accurate way to make parts.

I understand how the rudder linkage works now. From the earlier photo it looked as though the rudder linkage might have exited at the hinge line instead of in front of it.

Shouldn't be noticeable at all unless you flip the plane over. Easily accessible for adjustment too.

I've used micro E-Z links and also the small thumbwheel Flyzone links for adjustment, but I'll probably just set the surfaces straight without adjusters. I've done this before with dual rudders and it works well. The idea is to use thin ply control horns, making the 90 degree bends on the pushrod ends as close as possible to where they should be, to fit into the control horn holes with the rudders parallel. The control horns are pre-fitted into slots cut into the rudders, with an interference fit tight enough that the horns won't move if you look at them the wrong way. Next, fine adjust the control horn positions in the rudders such that the rudders are parallel to each other. Finally apply a drop of thin CA to the horns to fix them in the rudders. Once set properly, they shouldn't lose adjustment unless damaged from abuse, and the horns can always be cut out and reset. All 3 rudders would still be adjustable as a unit, at the servo.

I've changed my mind and decided to build a tugboat. :Q (see first pic below)

The steerable nosegear mount has been added, using a modified nylon wing hold-down mount for the steering mount/pivot assembly. Using methods from past builds, the gear is removable. A length of rubber tubing captured inside aluminum tubing provides a friction fit for the main strut wire. Control horns are pressed onto both ends of the aluminum tubing shaft that turns inside the nylon mount. A key pin, made using a small machine screw mounted into the lower horn, indexes with the nosegear for positive steering. Searching through my servo horn collection, I found horns that could be pressed onto the tubing with a vise. I could not manage to cause them to slip with my hand strength after installing, so they should be robust enough to hold up under use.

The lower wheel fork was bent from a single piece of wire, where the 90 degree bent ends insert into the aluminum tube wheel axle. The fork is flexed open to install the wheel. The bent wire ends are cut such that they insert into the axle tubing only a few mm on each end, which is done so that the fork is not permanently distorted, when it is flexed open to install the wheel. Before installing the wheel, the fork is covered with plastic tubing, which is heated to create the bends at the top of the fork. These sleeves are then slit lengthwise and inserted over the fork wire. The slit seam is finally glued with thin CA. To attach the lower fork to the nosegear main strut wire, the strut wire is nicked numerous times and slightly bent at the end, to ensure it does not break free or slip inside the glue joint which will attach the two parts. A CF horsehair wrap with CA applied is used to permanently attach the two parts, after tack gluing together with thick CA. Finally the aluminum tube main strut sleeve is slid over the nosegear assembly and glued, after pressing the servo horn onto the top end, which will index with the steering locking pin. A slot is filed into the tubing end, to ensure that the servo horn does not slip.

Great work Bill.:D
I always enjoy seeing how you solve those engineering issues that arise in the course of your builds.
It may look a little like a tugboat at the moment but I'll bet you're pleased with all that internal space for installing the works.

Barry

Thanks Barry. It is pretty roomy inside. The only areas that will be cramped, are ones of my own doing :D such as the nose steering servo. I was barely able to feed the pushrod s-bend into place, while installing the servo with it's mounting plate in the small area. The only concern I have with this one is keeping the gear/battery forward, to avoid the need for ballast. I've been on a roll for quite some time in not needed to add any nose weight to a build, by careful gear and battery placement. The nose servo was crammed into the nose area, so that the battery can mount directly behind it, likely installed from the bottom. With the tail servos installed directly behind the cocpit area, balance shouldn't be too far off.

Nice

Thanks Paul. I've been keeping an eye on your Astro Hog build also, which turned out really well.

One of these days I'll have to build a vacuum molding machine, as the hand pull method produces less than perfect parts, when making larger parts. For this one, it wouldn't have worked however, as the mold is a sheeted foam core, which would melt. Since I didn't want to wait for a trip to the LHS to buy solid balsa block, I sheeted a sculpted foam mold with 1/32" balsa. The outer perimeter of the non sheeted surfaces of the foam core have balsa frames epoxied to them, which are used to glue the sheeting to, using CA. The sheeting provides enough heat protection to prevent the foam core from melting, with the hand pulled, heat gun method. The mold turned out well, although the hand pulled parts seem to always have some imperfections in one area or another. Version 2 worked out well enough to deem useable. The canopy will probably be mounted on a frame, so that it can be removable for batt/ gear access. I can always make another, if done that way. Lessons learned from the past have shown that it's good to make a large canopy removable, especially true on planes with nose glass such as my AR234 that's on canopy #5. :eek: It can be easily replaced if damaged, versus having to cut it off and replace, re-trim, etc. At lest this plane has a solid nose under the glass, reducing the chance of damage.

Since the lousy sheet plastic costs a fortune at the LHS, I may have to collect a few clear 2-liter bottles from neighboring recycle bins. I've seen a method where the bottle is slipped over the mold, and then pulled downward with a stick slipped inside the bottle, while heating the bottle with a heat gun. The bottle should be reasonably close in shape to begin with, so it may work well.

The build is coming along nicely.

A warn of caution on using bottles for canopies and cowls; they can impart significantly more pressure (and temperature) on the form than vacuum forming. I usually make my molds from Pine or Basswood.

Weird and wonderful!Just my type of aeroplane[popcorn]
The only thing that worries me a bit is the short coupling.It could be a little tricky.The cg could also be a potential mousetrap.
But those are the challenges of doing your own thing.
Good luck with the rest of the build.:tc:

Bill, I think the post regarding encouraging the bottle neck to form around the front of the mould by using a leaver was one of mine.
I was moulding the canopy for the Saetta which had a very sharp angle on the front screen.
I found that it helped to mount the mould onto a wooden block with a horizontal groove in the front face of the block. The idea was that the bottle plastic could be tacked to the rear of the block as it extended out of the rear of the bottle. heat could then be applied to the bottle working from the rear going forwards shrinking the plastic as you go and taking up the slack. The blade of a large screw driver was placed through the neck of the bottle into the groove and then as heat was applied around the neck of the bottle at the front of the mould the driver can be levered downwards to prevent the neck from riding up on the mould and at the same time forcing the plastic down over the sharp angle on the windshield.
You need a good supply of bottles as I've found that it's very hit and miss and there are so many variables that each attempted can turn out very differently. I hope this helps in some way.:D:D

Barry

DHC I was having the same thoughts about the relative short coupling/tail moment. Probably a good plane to not start with an aft CG. :D Too far forward would not be good either. I did enlarge the stab a slight amount for added stability, although it is reasonably sized at scale. It's interesting that you can never really tell until you fly them however. I recently flew a Guillows P51 with a short tail moment, and thought I would have a bit on my hands. Turned out to be one of the best dang flying airplanes I've ever flown.

The foam mold does have it's limits Pat, in how much force it can take, although I was pretty hard on it. If I do any more molding with it, the plastic will have to be heated enough to pull reasonably easily, while not heating for long enough to overheat the mold. Probably barely enough time to do the job. As you were saying Barry, there is definitely some technique for working the material across the mold, with parts like this one. The bottle idea seems to have possibilities, as I could probably keep the material taut, much easier than I could with my two hands, as I really needed 2 more. If the plastic was pulled so that the canopy top was well stretched, then the sides would tend to bunch up. With Barry's method, I could keep the heat applied while pulling at the same time. I likely will just keep this canopy however, as I made a frame for it, that will allow it to be removable. The removable frame allows me to trim the perimeter a bit further, and the outer edge will be painted to cover the frame. Trimming off that extra material will pretty much remove the areas around the edge that have a bit of ripple/warping.

Main gear
Fabricated the main gear legs using aluminum tubing mounted across the inside of the fuse, for the gear legs to plug into. I had thought about making spring shocks, but decided it was more effort than I cared to go to, for a 36" span model. The idea could be easily done however, if the wire struts were installed in the swing arm location (see photo below) versus the shock location that they are currently installed in. Stop collars could be mounted on the swing arms inside the fuse to retain them, while they could individually rotate inside the aluminum tubing. The aluminum tubing joiner also serves the purpose of supporting the gear wire, versus stressing the fuse sides due to side loading.

they way you built the fuse give me good ideas i can use on floating hull planes. ive been wanting to make a few but just never figured out how to make fat curved fuselages. i think i can use your setup and try a few now. thanks!

The keel and former method works well for flying boat hull construction. It's pretty much the same method that most Guillows designs use, where all you need is a decent 3-view. I've built a number of flying boat hulls using this method, as well as a Dornier Gs build thread posted here recently. It's actually easier than built-up construction for several reasons. The keel and former method lends well to tab and slot design, making it easy to build a frame that self aligns fairly well, where the parts are made directly from the drawing which creates an accurate profile. With built-up construction, the curved areas have to be fully sculpted, with tri-stock corner fillets inside. To get an exact shape, you would have to use templates to check the curves while sculpting, as well as having to flex sheet balsa to create overall fuse curves. With keel and former construction, the curves are already cut into the keel and former profiles. Inset planking glued between the stringers on curved areas can easily be sculpted to match the profile of the formers, which is the exact profile that you want. This has been done along the lower fuse corners. Inset planking the curved areas provides a bit of added latitude for final shaping after sheeting as well as reinforcement, and allows the fuse to be sheeted with thin sheeting for light weight. This plane will be sheeted with 1/32" sheeting, to keep the weight low.

Outrunners are now installed in the nacelles, bought from Heads Up RC.
http://www.headsuprc.com/servlet/the...0-Sport/Detail The Power Up Sport 250 outrunner claims 10oz thrust on 2s lipo with a GWS 6030 prop, which is the largest prop size that will fit on the plane, without relocating the nacelles off-scale. Next to build the wing, where the wing parts are notched for tab and slot assembly. The ailerons will be constructed as part of the wing assembly, and finally cut away.

The wing frame is nearly complete and going together well. The full scale plane uses a NACA 23018 inner section and 2412 tip section. To keep the wing reasonable scale looking, I used a 23012 changing to a 2412 at the ailerons. There is one airfoil former section that is a bit modified during the transition. Sometimes I wonder if I should have just used a single section like an E197 though. :eek: The two sections used in this wing are quite different, and it becomes a bit difficult to set washout without some flexing of the LE. An overlay of the two sections shows the difference in entry point locations, as well as other features. Irregardless it's a good bit of work to properly sheet a light frame without stringers to begin with, so it makes little difference in the end. I may add a few stringers, mostly for the purpose of making slightly more rigid frame. It would then be a bit easier to sheet and avoid having any warps or mismatches across the panels.

It's starting to look like an Aerovan now.

That is one cute airplane! Good work Bill! As usual, you talented (and patient) guys make me green with envy...

Thanks Paul
I have no patience! :Q Seriously, once you build a few of these using 3-views, it really takes little time. In actual hours, I don't have much time in the build, at this point. Covering :eek: is actually the worst part of the ordeal, by far.

To start, you have to pencil in clean lines and curves, since the enlarged drawing is grainy. It actually takes little time. The keels are all derived from perimeters, using the drawing views. For a fuse like this with flat sides, you simply take two dimensions for each former, after laying out the formers, keels, and spacing the stringers. The rounds are drawn using a circle template, and keel/stringer notch locations are taken from the drawing. There's probably no more than a few hours drawing and a few hours part cutting for the entire build. With a simple wing layout, the nacelles probably took as much time as the wing to draw and construct. I rarely cut part templates, simply using a semi-dull knife to score a mark on balsa placed beneath the drawing, which doesn't even cut through the drawing. It's proven to be as accurate as using cut out templates, as you score right on the line, and get a reasonably clean mark on the balsa sheet below. Going through the motions of cutting paper templates and using them as a cutting template easily introduces as much error as the "tracing" method I use. It's a good speed trick, that allows you to have a part set in short order.

Since the plane will have functional flaps, I figured why not add lights. I'll probably skip the tail light mounted on the center rudder, adding the wing lights and nose mounted landing light. The plane should be ready to sheet soon.