Project: Carbon Quad Finished

Finally all the parts needed have arrived and it's time to complete the build. (link to previous project log part).

Motors and ESCs lined up. They need some preparing before connecting everything up.

Motors are Turnigy NTM Propdrive 900 kVs. I had decided that after my earlier experience with my hexacopter's dt700s and their constant eating of the motor bearings, I was going invest a little into more "premium" motors. I also decided to try out the Afro series ESCs as they were getting praises for the improved handling when used in multirotor setups and I wasn't interested in going through the re-flashing of the hobbyking's ESCs, because I wasn't sure which microcontrollers they got these days.

Counter pieces for the motor mounts. The motor mounts are a accessory pack for the motors and do not come with the motors as standard.

The material is 3 mm thick milk white plexiglas plastic, which I have used before in various projects as an enclosure building material. I would have used thin plywood, but I didn't have any in the right thickness.

Looks pretty neat and is mechanically super simple and incase of a crash it will allow the motor to twist a little bit on the boom. The nuts and bolts are M3 sized.

This is my simple anti-vibration solution to get rid some of the motor vibrations. It's meant to be used for furniture legs to prevent scratches to floors, but I find it much more useful for this kind of use. Why didn't I use the solid rubber pads? Well the rubber pads that I found in my local hardware store was so hard and solid I wasn't very convinced it would absorb the vibrations as much, then just pass it right through it to the boom and into the frame.

Of course to make sure everything looks nice, neat and somewhat professional, the white motor mount counter pieces were painted blaaaack (not with the marker).

All the motor mounts test mounted.

The foam pads applied to the mounts. I had to change to longer bolts (20 mm), as the original ones (16 mm) weren't long enough.

Ok, time to start doing some wiring. I desoldered the 3,5 mm plugs from the motor leads and soldered some 18 awg wire from them to the ESCs, which had their original wires removed, with plugs.

When testing the motors and ESCs, a good habit is to mark which direction they turn so you don't end up with 4 clockwise rotating motors in the end. Although that won't be such a hassle if you use the motor wire plugs.

Boom ends look nice and neat.

The afro ESCs had their heatshrink wrapping removed for the duration of the soldering and got some new sleek black ones after the procedure.

Test weighing, 618 grams with all frame's heavy parts and looks very promising. We'll have to see what it will be AUW (All Up Weight = ready to fly).

Use zoom

Offtopic. I was working on some stuff very late in the summer in my room and had my window open and desk light on, which attracted a couple of friends... I could hear an audible buzzing sound in the room and happened to gaze up into one the upper corners in my room and saw this... I did not sleep in my room that night. Mosquitoes and other nightly insects, this picture doesn't do justice how many there were.

ESC battery wire soldering.

I decided to mount the ESCs so that they were poking out a little from the sides, this prevented them from using valuable real estate on the carbon fiber plate, where I need to mount all the electronics. I also added a little servo lead straight from the main battery lead, in which I would connect my FPV gear.

Everything squished and hidden away between the bottom and top CF plates. The FPV power lead peaking through the middle hole on the topside and I decided to pull the ESCs' servo leads from the front and rear side so they wouldn't have to go so close to the noisy ESCs and power leads, they were also twisted to a great extent to provide more noise shielding.

AUW weighing with no FPV gear, 1066 grams, great!

FPV gear mounted and all the essentials on the table to go flying. You can read more on the groundstation here. 

1,3 GHz cloverlead antenna and transmitter. You can see under the KKboard 2.0, a servo connector peeking out, which is connected to the battery and supplies the FPV gear with power.

I'm very pleased how it turned out and it turns out the key to this success was pre-planning and visualizing the whole setup. Now I have very sleek and neat looking multirotor, which I can take with me to video with. I am getting a GoPro to increase the visual quality of my coming videos. Now I just need to improve my FPV gear with better antennas and I should be pretty set.

Update: Now that I have flown the quad quite a lot I can give some new opinions. This is couple of months later after finishing the build and having time to fine tune things. These are in no particular order.

1. As a flyer the thing is super stable, especially when there is no wind in the evening it just flies so well. Unfortunately during straight vertical descent it tends to have quite bad wobble, but that is the flight controller trying to do it's job, but ends up doing it badly. Unfortunately due it being so light, it is a little sensitive to gusts and wind in general, but ultimately it is the best flying multirotor I have built so far!

2. One of the motors has bad bearings and is creating absolutely terrible vibrations in the frame and at least triples the noise of the quadrotor when in flight. I could have tried to change the bearings, but I didn't want to ground the entire copter, so I ordered a whole new motor. Then after I receive the new motor I can try to replace the bad bearings in the old motor without worrying breaking it and grounding my copter.

3. I noticed very early that there was quite a lot of noise in the power leads from the battery, which created noisy lines in my FPV feed, so I built a LC low-pass filter to filter some of the noise of the power leads and put some servo connectors to it aswell so I can just connect it to the FPV circuitry plug-n-play style. I wound the coil myself using motor winding wire. It works and I get no more lines when I'm at hover power and little over, but at full throttle the filter is too small to handle the greater power spikes and the lines appear again, but because I'm a very relaxed flyer I very rarely go to high throttle

4. I also need to get better antennas for my FPV system and I am going to make them myself and test them at the tech lab at my university to make sure they're good and up to the task.

5. The rectangular carbon fiber booms are terrible in crash landing durability, because they are extruded carbon fiber and not woven, but woven carbon fiber in rectangular profile is still a little pricey so I guess I should just avoid crashes *shrugs and rolls eyes. Also they are very hard to drill through cleanly. Round booms would be supreme in every other way, but they're a little hard to mount without adapters.

The current look of the copter as of the moment of posting this article. GoPro on the front giving live video feed to the VTX (video transmitter). I'm going to 3d print in the future a 1 axis gimbal for the GoPro and use a servo to be able to turn it up and down.