Saturday, May 7, 2016

FT Tiny Trainer (2015)

After flying the Towel for quite some time, I decided it was time for a slightly more complex RC plane project.  Browsing the Flite Test scratch build library, I found the FT Tiny Trainer, which looked about right for me: it was a 4-channel airplane, construction was not overly complex, it didn't require vast amounts of storage space, and it looked pretty docile.

And so it began...

Layout out the tiled plans.
Cutting out the first parts.  This is much more complex than either the Towel or the F22.
I needed to do some aluminum drilling for my motor mount.  Since my workshop is right in my condo and not in a garage somewhere, I setup my vaccuum to suck in the swarf as soon as it was produced:

No-mess drilling.
This worked quite well.  (Since then, I've acquired a wed/dry shop vaccuum which I could use for this purpose in the future.)

The "power pod" turned out quite nicely:
Assembled power pod.
Testing fit.
Final assembly with nuts.
I proceeded, putting together the body methodically.  It was extremely gratifying to see things come together.  (I was a bit down because of health issues at that point in time and I distinctly recall this weekend as having an incredibly positive impact on my morale.)

Test fit for the tail.
Folding up the fuselage.
Assembled fuselage, with a control rod test.
Putting the tail on the body.
I had to add a bit of extra foam to keep the tail stable because the thickness of mine didn't quite match what the plans were made for.
Attaching the rod to the elevator.
A view of the servos, tucked inside the fuselage.  I only had 9g servos, not 5g as recommended, so things are a bit tight.
It was lots of fun to see the surfaces moving for the first time:

Control surface test.
I carried on, attaching the nose to the body.

Affixing the nose to the body.  In retrospect, that rear mounting point is way too small, it should be bulked up; it gives way at the first crash.

Starting to look like a plane!
Next it was time to put together the wing.  With its curved leading edge, this was a bit more complex than my other constructions so far.
Cutting out the wing and preparing for the big fold.
You can see the curve in the leading edge of the wing.
Putting in the dihedral angle for extra craft stability.
It came out nice and sturdy.
Fishing the servo extensions through the body of the wing.
Once again, testing the control surfaces for the first time was mightily satisfying.

Testing out the wing surfaces.
I made a few thrust tests prior to the maiden flight.
Making sure everything works.
And finally it was time to fly it!

Ready to fly on a beautiful day!
Unfortunately I don't have a video of it in flight, but it was an unmitigated success.  It was a very windy day, but other than that conditions were ideal.  On my first launch, after a few flybys, I looked down at my controller for a fraction of a moment to flick a switch and in the time that took, I crashed into the ground nose first from an altitude of about ten meters.  It was a very hard crash into packed gravel, but the little plane took it like a champ.  One of the servos inside came unglued, but that was quickly fixed.  Other than that, all that was needed was a little piece of tape for one of the points where the nose mounts onto the fuselage.

Connecting everything up.
My friend Greg strapped his Moebius HD recorded to the body on one of my flights up!  You can see a few landmark Montreal structures in the video.  I did a few acrobatics just before the 5:00 mark.

Towel RC Airplane (2013)

After building our first short-lived but fancy F22 RC planes, my friend Pierre-Luc and I decided to go back to basics and build something that would be simpler to build, easier to fly, and more resistant to "adversity".  The build here was much faster than for the F22s, since the part contours were much simpler.  We did use the opportunity to improve our motor mounts using a piece of aluminum angle bracket I found in the trash in my neighbourhood.  Since aluminum is relatively soft, we were able cut it using a hacksaw and drill out some mounting holes.

Positioning the deck on the airframe.  You can see the motor's mounting bracket in the center.
Stress-testing the control surfaces.
Starting to come together.
For fun, we added a home-made ninja "decal" to Pierre-Luc's plane.  (The picture came from here.)  We did this by printing the ninja on a piece of paper.  Then, we added a large layer of packing tape atop the ninja, followed by another slightly smaller layer.  We ran the same piece of paper through the laser printer again, printing the same design, such that the toner would be deposited on the tape.  Then, we carefully added an extra two layers of tape atop that.  By peeling the corner of the second layer of tape we had deposited, the one that's just under the toner, we were able to detach the ninja, cut it to size, and affix it to the plane.

The decal in place.
An experiment with big z-bends in the control surfaces.  This make the plane floppy and very hard to control; the control surfaces had no authority.  We tried this out because we did not have proper control rods and/or linkage stoppers.
This time, since it was still very much winter, we even bothered to make our planes water-resistant.  We followed some directions from Flite Test.  Though this discolored the plane a bit, the time we spent on the treatment definitely paid off.

Out comes the Minwax...
...and onto the plane it goes.
About to step out and fly!
Out again into the icy, windy winter night we stepped.  But this time we met with success of a sort!  My plane flew a bit but control was impossible due to the give and play in my control rods, and Pierre-Luc had slightly better luck.  However we had difficulty getting our planes to gain altitude properly.

Back home, we ran a few experiments to figure out which propeller gave the most thrust for our battery and motor combination.  We did this by suspending the plane upside down on a luggage scale:

Measuring thrust.
By measuring the increase in weight seen by the scale when the plane was at full thrust using various propellers, we were able to ascertain that in our case, the 10-inch props were the best to use.  They provided about 650g of thrust for a 530g plane, which is a little over a 1:1 thrust-to-weight ratio.

To further test the thrust and make sure we would be able to climb, we built an extremely dubious launch ramp in my living room out of political signage.

Testing thrust-to-weight ratio indoors.
The next time we went out with our planes, it was a little warmer (but still below freezing) and there was no wind; with all the tests we had run, we were feeling pretty good about our chances.

The planes did climb!  In fact on the first launch, Pierre-Luc's plane's surfaces had so much authority he did a very tight loop close to the ground and immediately crashed.  A few tweaks and adjustments later, we were flying!  It's hard to see anything in videos but here is one.  We fly a bit in the second half:

Testing the waterproof coating.
Here are some videos of my other friend Greg flying my Towel on a nice day while I film:

I took the Towel out on many other occasions, even travelling with it a bit.  I've changed the control rods for wooden dowels, and that did wonders for control surface authority.  It has given me an incredible amount of fun over the years.  As of May 2016 I still have it in my workshop, though it hasn't flown in a while and is due for an airframe replacement.

F-22 RC Plane (2013)

In automn 2013, inspired by an article in MAKE Magazine about the Brooklyn Aerodrome's Towel project, my friend Pierre-Luc and I decided to get into the RC plane hobby without knowing anything about it.  We spent a few evenings reading up on the technology behind electric planes and wound up ordering a few hundred dollar's worth of batteries, receivers, transmitters, servos, and various other bits and bobs from HobbyKing.

As it turns out, it takes some time for these things to arrive from the other side of the planet.  In the meantime, we spent more time reading about the hobby, and we stumbled upon a design for an F-22 ("based on a design by Kosh, adapted for EPP by Dekan, with plans by Firetrappe") that piqued our interest.  Now don't get me wrong, the Towel is a fine airplane - but who wants to build a flying triangle when you can build an F22??  We figured we could simply use the parts we ordered for the Towel in the F22 and call it a day.

Little did we know..!  (At the time of this writing, I've built and successfully flown two other planes, so I get quite a few chuckles out of our inexperience at the time.)

Waiting for the parts, we decided to start cutting the plans out from foamcore:
Pierre-Luc pinning down sheets and tracing outlines.
Cutting out shapes from the foamcore.
More cutting.
Cutting auto-awesome!
And a first part is out!  You can see this bodes well, the front right portion of the nose was cut improperly.
 My cat insisted on helping out:

Finally, the parts arrived! :D  We spent an evening soldering connectors, plugging stuff together, testing out the motors and servos:

Some of our first connectors.  XT60 is not completely trivial to solder; there is a lot of thermal mass, and you don't want to deform the plastic.
Trying out the servos.
Firing up the motor for the first time.
Feeling flush.
We continued with the assembly, testing parts for fit.  We know the motor, propeller and battery we had were way larger than recommended, but we figured "meh, the plane will just be more powerful".  Hahaha...  Ha.

Testing the propeller for fit.  That's a nine inch propeller in a plane designed for a 6- or 7-inch one.  What could go wrong?
First airframe almost complete!
Working on a custom motor mounting bracket, made from tie-wraps and coroplast.
Putting things together.
Getting there!
At this point I was really thinking about cool additions we could make to the plane.  A few hundred grams more or less, what's the difference?  I mocked up a light system that could be driven by a microcontroller.  I spent some time reading the FAA's rules about lighting systems on planes, watched some planespotting videos, and came up with this:

Remote-controlled plane lighting prototype.
We also noticed that once mounted in the slot, the propeller was quite loud.  We did some research online and found some very interesting information about slot shape and noise, so for the second plane we decided to use the information to alter the shape of the propeller slot.

Bigger slot for less noise.  There are a few centimeters of foamcore left on either side, surely that's enough??
For the record, this did significantly reduce the amount of noise the plane made, though I still would have considered it loud by any standard.

We made some durability enhancements to the leading edges of the wings.
Taping a dowel rod in place.
Part-way through taping.

Finished result.
I really wanted the plane to look nice, so I printed out some decal outlines, "laminated" them with packing tape, cut out stencils, and made some tests on foamcore.

Home-made stencil.
About to test out the stencil with dollar-store paint.
After a few tests, things looked decent!  The trick was to lightly dapple the paint on with a cynlindrical-ish brush.
We finished the planes by adding the servos and control surfaces:
Testing the servos.
Never mind that the rods are under the surfaces, and thus that to climb, the plane needs to compress the control rods (unless you have stiff rods or good channels, it should be the opposite); never mind that we are using soft steel wire that flexes and bows under compression.  All in all these would be but minor factors in the aircraft's overall performance, as you will see below.

More tests.
Finally, we completed one plane!!  Completely giddy with excitement at the prospect of flying it, we strapped our 180g, 2200mAh battery onto our tiny plane and headed out.  You have to understand - by this time, it is the dead of winter in MontrĂ©al, it is around -20C (-4F or so), there is a lot of wind, it is night time, we are headed for a small-ish park with neighbouring trees, and we have never flown a plane outside a simulator before.  And then after a 20-minute walk in the cold, smiling and almost skipping, this happens:

Fail.  You can't see anything but you hear our exclamations at the end.
When the moment came to launch the plane, we realized we had wired the motor backwards and the propeller was spinning the wrong way.  And we had not brought any tools!!  What a pair of newbies.  We headed home with our tails between our legs and waited for our next opportunity to fly.

You can tell that I was looking to make some nifty bombing additions to the plane before we maidened them:

Looking for a place for a bomb drop mechanism. 
When next week rolled around, both planes were ready to rock!!  We took a few pictures of our completed masterpieces - hard to say how much time we put into them, probably on the order of a couple dozen hours total to make both - and then we headed out into the winter night again, completely overflowing with anticipation.


Ready for takeoff!
This time out it's even worse, it's raining, everything is frozen with ice sheets, we are freezing our hands off holding the controllers and we have to warm our hands down our pants every few minutes.  The planes are not waterproofed and the electronics are wet before we even throw the first plane.  While plugging something in, Pierre-Luc rips something off his plane - can't quite recall what and it's hard to see in the video.  Then we realize his transmitter is not correctly programmed, and overwhelmed with excitement, we decide to launch the plane anyway and "make up for it".

The videos are terrible - you can't see anything in the night - but I will post them here for posterity.  French speakers might get a chuckle.

(warning: some foul language, we French borrow from English but the words lose their punch)

Long story short, the planes "flew" a combined total of approximately five seconds each, which is just barely more than the ballistic time predicted by Newton's laws of motion.  The planes were way too heavy, the control rods were way too flexible, the control surfaces too small, the airframe too flexible, the prop too big, and generally nothing worked.  We trekked home with our "lightly used" planes:

And... that was that.  We tore off the electronics, which were not damaged, chucked the airframes into the garbage, and got down to business working on our next planes.  In retrospect, it's amazing we managed to cause so much damage to the frames in so little time.