So to continue the story of the Horus Harpy, we have to go to the IDRA finals and think back to my previous chapter, where I reference the dominance of the 'X' frame.
Sitting in the pits at the race, I became incensed with the 'X' frame adoration. As I mentioned, I see it as a stagnation, a step backwards to a simpler design. While I admire simplicity, I think we have too much to explore before we hit that point of reduction.
I became determined to design a VTail to compete in any event next year. A straight-forward design. Devoid of strength robbing cut-outs, using the lightest and strongest carbon I could get my hands on. I did the design and printed my first prototype in 12 hours.
I focused on a couple of changes I wanted to make versus the Kestrel. The Kestrel is optimal with it's own mixer, I wanted this to work with any Cleanflight and Betaflight mixer, to do this I shorted the tail, thereby bringing the rear and the front closer. Then I designed the arms and tail to press against one-another, optimizing their individual strength. Next, I reduced the thickness of the arms down to 14mm, just wide enough to fit the new smallest ESC's and provide only the slightest airflow occlusion possible. For this I'm leaning heavily on the 4mm 6K twill dry carbon to provide me with the impact strength I hope to get.
The changes didn't stop there. When it came to the main plates, I reduced their width to just barely cover the flight controller, 26mm FPV flight camera, Lumeneir 200mw VTx and FrSky X4R. This was to reduce the surface area exposed to the air. I also designed the standoffs to bolt on to the bottom plate vs. through the arms and tail. This was for three purposes: the first, to prevent the need for different sized standoffs for those attached to the arms vs. those attached to the base-plate. Second, it is a built-in lower-profile design, it reduces the standoff height by 4mms, so a 30mm standoff becomes in-effect a 26mm, a 40mm becomes a 36mm and the later provides ample room for the 26mm flight cam tilt assembly. Thirdly, it forces the flight controller a little higher, in to an optimal position for the CG.
Lastly, I redesigned my VBlock. This was critical to reducing weight and improving the CG. I began by hollowing it out and using the section of circles placed at the outer edges to provide strength, finishing by providing positive triangle shapes to lock the tail 'feathers' in place. Once complete, I 3D printed all the components to test fit them and once I knew they fit, I shot the files to my carbon fiber partners and they cut me a few frames for testing.
Once I received the parts I began building the frame. For this particular airframe, I used the SP Racing F3 Acro controller, which despite all the hoopla, I enjoy. A Matek PDB is distributing the power, which I'm not a huge fan of, but this version has a lost-alarm built in to the PDB, which I want on my prototypes but never remember to bring to the testing grounds. As for ESCs the decision came down to the UBAD 30A's or the Hobbywing 20A. Looking at all the testing the Hobbywings are a little better, but require a bit more work to wire-up and update. I chose the UBAD's because their performance is nearly the same but the 30A peak will come in handy for testing. Speaking of that, I strapped on RCX 2206 2633kv motors, which I'll be running with a range of 5in props from APC for testing. APC's Robert has been kind enough to give me a bunch of 5in prototypes for testing: bullnose 5x4.3, 5x4.4, 5x4.5 and 5x4.6. The APC props have a superior airfoil design and with the bullnose, I'll be drawing more amps than HQ or DAL, so the UBADs seemed like a no brainer. While the frame is a true 250, because of the minimization of all the surfaces, it can actually accommodate 6in props, so that will be explored as well. FPV Camera is a PZ0420M using a 3.6mm lens. I've used the 2.8mm in the past but never gave the 3.6 a shot, so I'm testing it out and worst case, I can just swap the 2.8mm in without worrying about changing the wiring. VTx is the Lumeneir 200mw (based on 2016 IDRA rules) and FrSky X4R using S.Bus.
As you can see, the arms are just wide enough to fit the ESCs... heck the whole frame is just wide enough for everything to fit. Unfortunately I didn't have any standoffs bigger than 30mm so my camera is locked in position for the time being, but the angle is suitable, so I'm not complaining. It also forces the VTx to be mounted on the outside for the time being until it's done being tuned, which is intended to go within the body for protection, but I don't want to remove it every time I have to plug in my USB cable.
For the wiring I did one little wierd thing, I flipped the ESC pins upside-down and forward to keep everything tight and while I was going to directly wire the ESCs to the FC, I thought I ought not to in case I want to or need to replace ESCs down the road.
Once it was all wired up I threw together a quick body design which I printed in my new carbon fiber reinforced polymer. Its a crazy material because its printing so well at a low temperature (230°F), but hardens in the nozzle much harder than any other filament I've used before.
I'm not especially fond of the body design yet, but I am happy with the space/room allotment and the antennae openings. I feel like there is a lot left to be desired and I'll continue refining it as time goes on.
Now nothing I do is ever perfect nor would I want people to think that. There are some clear errors, to me, which will need to be addressed before production. As you can see the front motor mounts are tilted in such a way that the wires don't align with the arms. Yeah, stupid mistake as I normally address this from the start, but just didn't for whatever in this reason. Their position in the tail is actually optimal for wiring the ESCs on the arms as I have, which for now, seems optimal for cooling. Issue number two, is the FPV cam mount hole. It isn't perfectly centered. Its a fraction of a millimeter off, but it's noticeable to my eye and it won't stand on when its such a simple fix. Third, I'm concerned with the thickness of the landing gear. While I've designed the cnc files to include two or three sets of spare landing gear, that'll get annoying quick if the break in the slightest tumble.
Now all that is left is to update the firmwares of EVERYTHING and maiden it. As it turns out my Taranis, X4Rs and ESCs will need updates then we'll be able to maiden it. From my experience, I'll need a big C battery, so I'll be using Tattu's, Dinogy or Rebel's, 75C minimum for racing. Keep tuned for the next chapter and I'll have more insights from the maiden and test flights.