Wednesday, August 31, 2016

Part 107 and more

Well, it's official, today I passed the 107 test. I did pretty well on it, although I was aiming for over 90 and fell just short. All in all, I thought it was an excellent test that will ensure that yahoos who buy a Phantom on Amazon can't go out and sell their services while endangering the public, or at least not legally. It isn't easy, but that is a good thing as it will absolutely advance the industry and start to build the credibility we deserve.

If you're looking for some insider info, I'll provide a few pointers:

1. Know. Your. Charts. Charts are a HUGE part of the test. My estimate is that they comprised about half the test. I thought I was pretty squared away on sectionals but I still missed 3 or so questions on them. You will need to be able to interpret charts to answer multi-level questions. Further, you will be expected to understand the 107 regulations and determine what is and is not allowed in complex airspace delineated on charts.

2. Know the more obscure parts of 107. Of course you should know that you're limited to daylight operations (or civil twilight with anti-collision lights visible for 3 SM), 87 kts, 400' AGL, 3 SM visibility, 500' vertical and 2000' horizontal cloud separation, etc. But you will also be asked about the non-operational aspects of the rule, too. Be sure you know accident reporting requirements, registration rules, rules about your airman certificate (once you have it), etc. You will have multiple questions on these non-operational aspects of 107.

3. Understand performance. This material isn't inherently more difficult than the rest, but it is easy to get confused, especially if you aren't already familiar with the physics behind how an airplane flies. Know how CG affects performance. Know how loading, including dynamic loads (i.e. turns) affect performance. And know the difference between pressure and density altitude as well as their effects of performance.

Well, that is it for my 107 advice. Now for a couple quick updates on the Mugin. The biggest thing is that I've been working on getting wiring installed and mounting up all the electronics. The wiring isn't looking quite as clean as I was hoping but it isn't bad. Once it is all said and done I think I can make it look pretty nice.

White 3D-printed blocks are epoxied to the fuselage structure and small carbon plates are permanently mounted to those blocks. Vibration isolators then attach the avionics stack. It is hard to see here but the Pixhawk is mounted bottom center, the Piksi is bottom left, Dragonlink bottom right, and Powerbox is on top. Microhard radio will be mounted beside the Powerbox once I get a functional case printed.
The avionics stack will hold everything you see above plus the Microhard radio. The radio amplifier, PMU, and backup battery will be mounted just behind the nose gear. I'm planning to rail mount them so they're adjustable for CG.

Also, I fully assembled the plane to show off and got some pics of it with v1.

The fleet looks nice, doesn't it?

Well, thats it for now. Look for another update in a few more days once some more electronics are in (and hopefully servo linkages for the flaps are here).

Saturday, August 27, 2016

Mugin Updates

Here is the latest on the Mugin and a few other updates. First of all, I got the fancy fuel-injected motor mounted to the bench, ran it, and then got it mounted on the plane. That is one big step toward a maiden. Also, it is insanely powerful, as compared to the Saito FG-57, so this plane will have much better climb performance, payload capacity, and ability to fly at high altitude.

Servos are all in and wired up. I still need to make the wiring harness for the fuselage but I'm taking it slow until I get all the electronics mounted in their final locations. And, speaking of, I just finished the custom vibration-damped avionics stack. Hopefully it does its job and doesn't end up amplifying vibration...




All the parts are in except for the alternator and radio amplifiers. Hopfully those will be in soon so we have everything we need. In the meantime, I'm going to finish up getting the electronics we have mounted and wired in. Also, I got around to powering up our Micasense RedEdge today. It is an impressive little camera and it took me no time to get it triggering off the Pixhawk. It is really configurable, which is nice. It means that on the Drak we can simply pull out the QX-1, unplug the trigger from the servo wire, plug the RedEdge into said servo wire, and we're ready to roll.

The other major thing I'm working on is an accuracy comparison of DroneDeploy, Photoscan, and Pix4D. I'm using the campus imagery from 2 years ago, which isn't the best dataset, but it is at least a real-world example. I have a CAD file for the entire campus so I'll be measuring points in it and comparing to their position in the imagery and elevation in the DEM.

There's bound to be some kind of mischief I can get into having all this information...

Well, that's it for now, I'll update in a week or so, hopefully with a nearly-ready-to-fly plane.

Monday, August 15, 2016

The latest from the Skunk Works

Well, it has been quite a while since the last time I posted on here, huh? It looks like the last time I was posting about endurance quads. Well, I still hate quads but we do have some updates since then.

First of all,  the more mundane stuff. We have a fully functional Ritewing Drak. I built one earlier this year for a client and since then the lab has picked one up. It is a really excellent platform that carries more payload, faster, longer than a Z3. It is basically better in every way. The only downside is that it is significantly bigger (1.5 vs 1.2 m wingspan). Here is a picture:

This isn't actually ours, this is the one I built for a client and currently resides in Canada. But you get the idea.

Ok, now on to the really exciting stuff. We are gearing up for another campaign in Peru but, this time, it won't be with any anemic little electric plane. We are building V2.0 of the 3m Mugin and it is shaping up to be pretty incredible. A lot is different from the first, including the airframe. The newest version of the Mugin airframe has a lot of improvements including, but not limited to:

  • Carbon fiber fuselage frame (formerly wood)
  • Thicker fiberglass fuselage skin
  • Fiberglass wing and tail skin (formerly wood)
  • Improved fuel tank (so we might not be throwing this one away immediately)
  • Beefier main landing gear (unfortunately still the same crappy nose gear)
  • Flaps! (split flaps, at that)
  • H-tail providing redundant elevator and rudder
And then there are the improvements I'm making to the guts. Again, including but not limited to:
  • DA70 motor modified by HFE International for fuel injection
  • Sullivan UAV 300W alternator and power module
  • Microhard pDDL broadband data links (for telemetry and video)
  • TriadRF RF amplifiers (to boost that pDDL and get us 100km range with the help of...)
  • Actively tracked 23 dB parabolic grid antenna on the ground
  • All around better avionics mounting, layout, wiring, etc and a variety of custom 3D-printed mounts for different bits and pieces
So, it is shaping up to be pretty exciting. I've put some picture below to show off the goods and I'll post again sometime soon(er than last time).

Thats one helluva antenna

V2, with V1's fuselage in the background

Mmmm, redundancy

Love me some EFI