Builds on hold

I have placed my X8 octorotor build on hold. Building and working on a multi-rotor is a lot of work in that there is a lot of planning, measuring, and soldering  that needs to happen. When I do a build I tend to take over the only working space that I have; our kitchen. With granite counters, lots of outlets, and a big dining table this is the perfect spot to assemble electronics. I cannot stress enough the importance for a well organized work space.

My daughter is due to be born on November 25th. I can honestly say that the amount of planning, space, and assembly of things like a crib and changing table dwarf any multi-rotor build I have ever done. The preparation for a child have spilled into the kitchen, living room, garage, storage shed, and our bedroom. We have been going 24 hours a day for the last couple of weeks trying to get ready for the birth.

DJI Ground Station and Apple iOS 7 problems!

I picked up the DJI Ground Station about a week ago. The ground station give a pilot the ability to set up missions with way points and even navigate beyond visual range. I finally started messing around with it today and figured out that DJI's Bluetooth module flat out does not work with iOS 7. Needless to say I am annoyed.

Contra Rotating Brushless Outrunner motors

Contra rotating motors are nothing new. They are essentially two motors that are connected through a bracket ass to ass style and share a common shaft. The two motors still need separate ESCs. With the X6 and X8 modes it is possible to put this type of motor on the same side of the UAV instead of having one on top and one on the bottom. It might be something to look into with a future design.

FAA regulations

It looks like UAVs are about to get seriously regulated. I have not read the entire document but so far it looks like little drones such as the Phantom are about to require some sort of certification to fly. 

http://www.faa.gov/about/initiatives/uas/media/UAS_Roadmap_2013.pdf

More flight time

I initially built the F550 hexarotor with the desire for more flight time. The F550 is more stable in the air, has some redundancy in the motors, and it can lift more but it does not have better flight time.

I found out about a design that my give me what I want. It is called an X8. Basically it is a quad copter with contra rotating propellers on the top and bottom. I have most of the parts for what I need. I just need the time to throw it together.

B.A.L.A.N.C.E, Balance!

A perfectly tuned and maintained Phantom will be rock solid in the air. It will do what you tell it to do and will perform that you tell it to perform. All UAVs, Drones, or whatever you want to call them need to be maintained and tuned in order to fly properly. At least two hours of maintenance is needed for every hour of flight. I have been working and tweaking my Phantom for the last week. I had a chance to fly and it soared like a bird!

Lipos

To follow up on my Lipo issue. Hobby King gave me a store credit. All I have to do now is dispose of the bad Lipo.

Agile vs. Stable

The DJI Phantom is based on an F330 frame. This means that it is small and agile and can do and as a flying machine is very versatile. It can zip in and out of tight spaces with no issues. The F550 frame has an emphasis on stability. It is for getting from A to B and then filming.

To give a sports car analogy the Phantom is a sports car, The F550 is an SUV.

Back to the drawing board

I needed to do some wiring work on another project and all I could find at the time was 16 gauge in red only. I decided to use what I had remaining on the F550 build. The design itself does not call for this but I used this wire so that I could have removable connectors for the ESCs. This would make maintenance easier and would be easier to fix in the even of a crash. 

My original design called for 20 gauge wire, using 16 gauge wire was a mistake. It is not flexible enough for where I am placing it and has started to fray. The only way to fix this is to rewire the entire frame with a thinner wire. 

A shorted wire is a disaster waiting to happen. 

Here goes the balancing act

Building a multirotor is a balancing act between available power and weight. I added a set of landing gear and I plan on adding a gimbal from the Phantom.

Build, refine, tinker, adjust.

I was having an issue where the F550 was gaining about 3 meters in height and then dropping by 3 meters. It was essentially acting like a yoyo. I fiddled with the gains to no effect. What I needed to do was a second calibration of the Naza flight control unit. I now have a stable platform.

Never charge LIPOs unattended!

These are two fairly small LIPOs that I got for my Futaba transmitter. I charged them one at a time. Notice the battery on the right is puffed up. When I checked the charge it was at 99%. I am not sure if the balance charger would have stopped charging or if there would have been a fire. 

LIPO fires are serious and you can burn your house down if you are not careful. 

Converting a DJI Phantom to an F450 or F550

People want to know if they can just convert a Phantom to an F450 or an F550 frame. The answer is yes but if you got your first computer in the 1990s and it was a Dell or an HP and you wanted to add a bunch of drives you had to change the case and a lot of other things. You were better off just buying what you needed and selling the original computer.  Here is my reply from a forum post:

The price difference is not about DJI making money. The Phantom is fully integrated with the shell serving as the electronics protection, that is why it is cheap when compared to the stand alone Naza. The Phantom has a Naza 2 as the brains but everything else is much bare electronics and can only be laid out in one way. The stand alone Naza that you see for $439 is everything that is inside the Phantom flight controller wise but has been broken out into protected component parts for customizable integration into open air frames like the F450 and F550.

The Phantom's ESCs are bare electronics with no protection. An F450/F550 has a wrapping that protects them from moisture and shock. When you look at the Phantom, the compass on the leg is linked via wire to the GPS in the top of the shell. The stand alone Naza has a GPS puck with an integrated compass with a can bus cable. It would be a challenge to engineer an F450/F550 to integrate the Phantom's GPS and compass. The LED in the Phantom is just a light with an LED port that is connected via a servo cable to the Naza. The stand alone Naza LED actually has a shell around it and connects to the Naza; again the connection is via a plastic coated can bus cable.

It is possible to do a conversion of a Phantom to an F450/F550 frame but you are much better off buying an ARF and selling the complete Phantom.

Dumping the Phantom?

I am going to remove the Phantom's gimbal and eventually place it on the F550. The F550 is a much better platform for aerial photography. The F330 frame that the Phantom is based on is very cool in that it can do some amazing things but it is not much more than a toy when it comes down to it. Maybe toy is too harsh of a word. The Phantom is more training wheels for the larger multirotors. It is great for FPV, learning how to fly and for experimentation with a soldering iron but I can no longer take it seriously as a stable photography platform.

Mulitrotors are a balancing act; there can only be so much weight piled onto one before it reaches the limits. The Phantom is a perfect example of what the limits are when it comes to weight and performance degradation. The F330 frame is small and cannot be pushed much beyond it's design limits. Anything over 1100 grams is going to dramatically shorten flight times. Adding anything after market can easily push the Phantom into and over the 1100 grams range. Most people do not take weight into account when they buy their shiny new set of landing gear or a gimbal.

Even thought the Phantom is a lot of fun to fly, I may end up selling the it and flying the F550 exclusively. The F550 is a much easier platform to modify and maintain. It also has a lot more redundancy in that if a motor fails the whole thing will not fall out of the sky.

4S Lipo batteries

I plugged my only 4S lipo battery into the F550 today and it took off like a rocket ship. I had a good 20 minutes of flight time. Wow what a ride.

F550 is flight worthy

The F550 is flight worthy but not everything went according to plan. My first power up resulted in one of the prop screws flying off. Even though I had tightened them it wasn't enough. Unfortunately I am on my very last replacement. If I lose any more I will not be flying for a while. 

I did have a few successful test flights using small 3S 2200 ma batteries. The second I put the larger 5000 ma battery on things got a little unstable. The Futaba transmitter does not have a self centering throttle like my Phantom does.  It takes about a second or two for my command to be carried out. The Phantom reacts almost instantly so minor delays can magnify into bigger issues. I did have a minor crash and snapped a prop. Tri-Blade props are very cheap and I have about 30 of them so it was not a big deal. 

I need to tune the gains so that the F550 is not so wobbly in the air, install the IOSD, transmitter, gimbal and then finally navigation lights. This is at least 24 hours of work. It is going to have to happen sometime in December. The baby will be here within the next week and I have a ton of school work to catch up on. 

Plug n play or lack of

Nothing about the Phantom is plug and play. I have had to make a connector for nearly addition and modification. When I needed a gimbal I needed to splice some wires and run them to the flight controller. The most challenging was the video hookup because it required some trial and error. The video hookup is for the DJI 5.8 gHz video transmitter. I had to take a 3.5 mm video jack and split off a video and ground wire. The result is a small connector that I would have paid for before I knew how to solder.

When I bought the parts to make the video connector there were two in the bag. I decided to sell one to see if I can make any money off parts. Hopefully this will save somebody some time.

Here is the Ebay listing:

http://www.ebay.com/itm/161139036026?ssPageName=STRK:MESELX:IT&_trksid=p3984.m1555.l2649

Post crash flight test

The post crash flight test went well. Actually it went very well, my DJI Phantom was rock solid in the air and it stayed up for about 10 minutes before I decided to bring it down. Everything looks like it is working the way it is supposed to.

DJI Phantom Crash post mortem

I took the Phantom apart today to figure out what exactly happened. I have ruled out the 4.02 Naza update as the cause. I noticed some sand on the inside and wiped it out with a q-tip and a can of air.

There was some minor damage to the frame. One of the mounts is cracked. A little Super Glue should fix it up. 

 I took off the protective cover that was over the GPS and checked the connections. This was the first time I had seen this particular piece of the Phantom. Note the tiny battery connected to the GPS. I am assuming that this battery records the last position of the GPS making future satellite acquisition faster. 

The back of the GPS is the top of the Phantom. The GPS has a ceramic antenna. Some crashes have been caused by cracked antennas. There did not seem to be anything wrong with mine. 

I ran a motor test and everything looked fine. I started to put the top of the Phantom back on and one of the screws near the motor arm would not seat properly. The DJI Phantom is a precision machine. Even though there was some damage to the frame the top should seat properly and all of the screws should go back into place. When I checked for blockage I found one of the motor cables had a short. I am pretty sure this is the cause of the crash. This wire must have been pinched the last time I put the Phantom together. 

I repaired the cable by cutting the short and then re-soldering. I then ran some motor tests. Everything seems to be working but I need to do a flight test. To prevent this from occurring again I taped down the wires and insulated the ESCs. I made sure not to use too much tape so that it would not cut off air flow from the outside thus preventing cooling. 

My next step will be to replace the crushed landing skid and a flight test.

More LED Lights

It took me a long time to solder together 6 sets of LED lights, probably the better part of a day. I did perfect the design and started using better materials. I am not sure if $20 per set is enough. I might have to raise the rate.

Crash recovery

I powered the hooked the Phantom up to my computer, powered the motors and pushed the direction stick forward. The Naza said that I was doing something but the motors did not respond. I am going to need to tear the entire thing apart to diagnose the issue. I am starting to think that the Phantom is more trouble than it is worth. It might be better to convert the whole thing to an F450 frame. I would still keep most of the hardware but it would be far easier to deal with these sorts of issues.

DJI Naza software 4.02 problems.

Crashing is scary. Realizing that your bird is out of control and there is nothing you can do about it is the worst feeling ever. Hoping that it does not cause damage to anyone or anything is the primary concern and hoping that the damage that is going to be sustained will be minor is always secondary.

Multirotors run a small computer that that runs their own brand of software. My birds fly with DJI Naza brains. I updated both the F550 and the DJI Phantom to Naza version 4.02 and have had nothing but issue. The sticks reversed on my F550. Up is down and so on. The Phantom flies for a bit and then just fell out of the sky. It was a horrific crash and it is going to take a lot of time to repair the crushed landing gear, prop guard, and possible shell damage.

I will need to run a post mortem later this week.

Calibration and testing

Flying a multicopter is not something that you just go out and do; it is something that needs to be planned. The F550 is 98% complete. DJI came out with version 4.02 of it's Naza software today. I installed it and then could not get the F550's motors to start. It turns out that the transmitter was set up in reverse. I needed to figure that one out but at least everything is working properly.

I need to mount some of the lose electronics like the Naza's PMU and the GPS. The largest hurdle is still the Futaba transmitter. There are about 100 settings and each of them need to be specific to the F550's characteristics.

I also lost a prop nut when I did a motor test with propellers attached. It is a good thing that I have spare motors but I need to get a couple of 4mm prop nuts...from China. Ugh.

Motor quality

Quality is everything when you are talking about RC motors. My Phantom has Tiger Motors at $50 per motor they are considered very high end. I bought six lower end AX-2810Q-750KV motors from Hobby King. These motors ran me $20 apiece.  I noticed that one of the motors had it's E-clip jammed into the motor. It took me a good 10 minutes of prying to get it free. I have yet to fly. I am hoping that this was a fluke and that the motors perform to an acceptable level. 

If I had to do it over again I would have spent $60-$70 per motor and got myself more Tiger Motors. Unfortunately it takes at least 2-6 weeks to get anything from China. I waited for about a month for the motors that arrived yesterday. I wonder if I should just order a set now and save myself the trouble. 

F550 build

I did 90% if an F550 build today. There are a couple of things that I need to get to complete the project. The motors I got did not come with mounting screws. Fortunately I found a source on Ebay that can have them to me by Saturday.

I made the build slightly more complicated than it needed to be when I added XT60 connectors to make the ESCs easier to remove. I did this because I much longer arms on the way. Adding connectors added more wire to the build. I might need to revisit this decision at a later date.

The GPS/Compass is not calibrating. It was probably exposed to a magnet. I will need to degauss it when I get a chance.

The biggest hurdle is the receiver/transmitter.  It is the one area that I have zero experience. I might have it flying by Saturday.

Cold

The air is turning cold and this is bad for battery life and flight times. Moisture in the air reduces transmitter range.

Wire clean up

I cleaned up some dangling wires on the Phantom. I noticed that some of my video from the Gopro has interference lines. This means that I am getting EM interference from something on the Phantom. I am going to shorten some of the external power lines this week and see if the issue goes away.

Dirt, dust, and sand are the enemy!

I took off from Ocean Beach yesterday and realized that there is a fine coating of sand on the Phantom. It is not too much trouble to wipe off but if I had not already lubed the motors the other day I would be doing so again.

The GoPro took a face plant into some fine sand when I  It is okay and there are no scratches but I need to get a protective lens cover. Just a little sand in the wrong place can cause a really bad day.

More then the Phantom than just a toy

I have made some money off soldering some lights together. Not bad considering the investment in time and resources was very small. I noticed that there are some postings for Phantom pilots to do film shoots. These are paying gigs too! We are not talking about hey can you do some photography for the exposure, I got asked up front what my rate was and if I could travel. Flying a Phantom around is an in-demand job!

Now here is the rub; fly a drone in the US for money and the FAA will come down on you like a ton of bricks.

Do you want to meet your neighbors?

The best way to meet your neighbors is to have something so outside of peoples experiences they cannot help but to ask questions. I have been flying the DJI Phantom for about a month now and just about every time I step out of the house and start flying around somebody comes out of their house and asks questions about flying. I have met no less than 8 of my neighbors in the last month and had positive interactions with all of them. My neighbors are amazed at the Phantom and the ability to take pictures from 200+ meters up. They are wowed by the technology and how easy it is to fly a multirotor. The Phantom has the coolness factor that an iPhone would have in the 1990s.

Wait until I roll out the F550 kit that I am working on.

Apply what you know in an area that nobody occupies

Owning a DJI Phantom is like building your own computer in the 1990s. Things could always be tweaked and modified to meet a higher level of performance. There are always additions that could be added on to make life with a multirotor better. People are innovating with all sorts of extra additions such as lights, landing gear extensions and so on. Some of it is useful and some of it is just bling. I started soldering together my own design of navigation lights, they are useful and they have been selling.

Pre-Flight

The importance of a pre-flight check cannot be stressed enough. I moved the Phantom from my desk to the floor today and noticed that one of the prop guards had come loose. Upon checking I noticed that I was missing a screw. This screw goes directly to the motor mount. I can fly with out it but there will be some vibration. That one screw will need to be replaced. One loose screw always means there are more.

Enhanced navigation

I flew out to 100 meters today and never lost sight or orientation. Red for forward. Green for aft.

Night flying and dogs

Dogs are dumb, I mean really dumb. As soon as the Phantom flies over head dogs lose all sense of self control and need to chase it. This particular dog had a blinking LED light on so his owners could see him. What we had was an LED lit Phantom with an LED lit dog chasing it. I am amused.

The cost of inaction

The Phantom is working perfectly after I greased the motors with white lithium grease. Yesterday's interaction with the sand took up far too much of my time. No flights today. I am far too busy with studying.

Flying over Ocean Beach

I flew the Phantom along Ocean Beach in San Francisco today. The sky was amazing and it was the first time that I had the chance to not have to avoid power lines. The major problem with going anywhere near Ocean Beach is that sand will kill the Phantom.

All of my landings were impossible and had to be done on some grass. Of course the Phantom fell over on my first landing and sand went everywhere. I needed to take apart the Phantom, clean and then re-lubricate the motors.

I am still getting jello in the GoPro. Time to balance the props or switch to carbon fiber.

New Phantom

I have been flying a DJI Phantom around for the last week. Honestly, I am hooked but I noticed that it is taking up too much of my time and I need to focus on school.