Great, I'm happy to answer all. ( I just read this back after writing it all. It's a bit ramble-y, but I think all the info is there. Feel free to do some heavy handed editing, hah)
PatchBOTs (Patrick's Robots) really grew out of my day job as a 3D illustrator. I've been a freelance 3D artist for about 7 years since quitting my full time job in the same field. Working for yourself has it's ups and downs but one of the best perks is being able to serve a pretty wide range of clients. Corporate work is kind of my bread and butter (store design renderings, merchandising, product renderings), but I get a lot of cool jobs that push me out of my comfort zone and give me some different types of challenges. For example, Air Hogs got in touch with me last year about doing the packaging art for their Star Wars toy line. This was like a dream come true for me, but creating models of spaceships and speederbikes and TIE fighters is something I had never done before, let alone doing it for one of the largest audiences I can imagine. I did a good job with it and they ended up coming back to me for subsequent runs. All in all I think I've done about 30 toys for them.
All this variety of work has really kept my 3D modelling on point, so when desktop 3D printers became a feasible thing I dived right on it. The idea of being able to model something and then print a physical copy was really appealing. However, when I started getting into it I quickly found printing little plastic knick knacks got boring really fast, and functional prints to fix things around the house (which was how I sold the purchase to my wife) were few and far between. After 6 months the novelty wore off and the printer kind of just sat there collecting dust, which always made me feel bad. There was so much potential in this little machine, but I never found a purpose that would be both fun and worthwhile.
At some point before the new Star Wars movie came out I started looking into the R2D2 builder's club. Here's a group of guys who have been building droids since the 70s before the idea of 3D printing was even spoken. I wanted to see if I could take on a project that big using this fancy schmancy new tech, and to be honest, the answer I found was was "no." WAY TOO BIG! In hindsight, the project wasn't too big, just the amount of information was too overwhelming. As I start my new Chopper build I find it much easier to digest with a little context.
Anyway, with the new movie came a new droid, BB8. When I first saw the teaser trailer my though twas "oh, CGI droid." But then then there was Star Wars Celebration 2015, where they rolled out a WORKING BB8! I was blown away. This got me super interested in how that was done. I started doing some research and found the BB-8 builder's club. The club exists both on Facebook and in it's own forums at bb8builders.club At the time I joined it had about 800 members but now has over 15 thousand. Most are just casual observers, but there is a good number of builders working towards a finsihed droid. The same deal as the R2 club, but focusing on the new droid. It was this big community of people just like. People who wanted to create this thing, but had no idea how. That collective "huh?" led to a huge community of sharing. Everyone sharing their work, everyone sharing their ideas, everyone sharing their parts. The club is free, the information is free and everyone contributes. It's really awesome. I was able to get peoples blueprints, compare them to other blueprints, compare all of them to different scenes in the movie and make changes. Everyone saw something different, so everyone's droid was different. Eventually the builder's club released the "club" files." Which were all the external pieces one would need to build a droid based on all these initial blueprints and sketches. A lot of my parts are modified "club parts" and some I found to be even better then mine, so I kept them. If you join the club now you can download all the parts you'd need to build a bb-8 prop for free!
I bought a new printer which was a little larger, a Lulzbot Taz5. It's a little pricey, but the printer is incredible. A lot of the typical 3D Printer problems just seem to go away with a well built printer, and I benefited from having the larger build space with a project like BB8.
Making it drive was a whole other beast. In the beginning, there were two main ideas for how to make him drive. "Hamster wheel" style, which is basically a robot inside a ball moving the ball. And "single axle" style which is like a wheel or a pendulum inside a ball. I went with the latter. There is an amazing builder on youtube named James Bruton. He really pioneered the Single Axis style droid and made a whole video series about it. This was like my build bible. I watched all those videos a million times and ultimately followed his design for my internal structures. I built and modified all the parts to fit my the hardware/motors I bought, but the main idea came from James Bruton. After the internals were done, I didn't really like the way James approached the rest of the droid (the external parts and the head), so I kind of went off and did my own things there. I really can't emphasize enough the sense of sharing within the builder community. It's a pretty rare thing these days. And I was happy to give back with my 3D parts and ideas.
To sum up the basic drive mechanism: Within the ball exists a 20" plastic wheel. This is where the main front to back motion of the droid comes from. That 20" wheel is driven by a Windshield Wiper motor from a 1997 Dodge Ram. You wouldn't think it, but windshield wipers have incredible torque-y motors. Being able to move all that snow off your windshield from such a far leverage point takes a lot of power, and I harnessed that power to move a 70lb ball dressed up like BB-8. Yes, you read that correctly. 70lbs. A lot of that weight is legit LEAD. I have 8 pounds of lead shot in little compartments at the base of the ball as stationary mass, and an additional 15lbs of lead shot as swinging mass. When I swing that big 15 pound lead weight side to side I can tilt the droid which allows it to turn. The stationary mass in the base of the ball counters the weight of the head to keep it happy up on top. In addition to swinging the weight side to side, I have it attached to a lazy susan bearing which can spin at really high speeds and allow the entire droid to spin on the spot. Here is a longer video explanation I made:
The rest of the weight breakdown goes like this:
20" wheel inside the ball ~ 8lbs
Stationary Mass ~8lbs
Swinging Mass ~15lbs
Body Skins ~13lbs
Motors, batteries, structural parts ~25lbs
Head 2lbs 14oz
The head needs to be as light as possible. People on the forums always ask "how heavy should the head be?" and I always answer n - 1 oz. Where "n" is equal to the current weight of your head. It really can't be too light. And yes, to answer your question, the head is by far the trickiest part of the whole thing. It's held on with magnets. Really strong, finger crushing magnets. However, strength of magnets isn't the only factor at play. You have to find this perfect balance of "strong enough to hold the head on, weak enough to allow it move smoothly without pulling down on it too hard." This is the make or break for so many droids out there. I was smart enough to build a test head, and thank god I did, because my head hit the ground about 8 dozen times while testing. Finding that balance is tricky.
One of the other hard parts/tricky parts of the build is the fact that the final droid rolls and moves on the finished parts. If you were to build an R2-D2 there would be wheels in his feet which would carry the load across the ground. On BB-8 it's your freshly painted and meticulously labored over body panels that are being rolled over on the dirty floor. It's a bit heart wrenching to see little dings and cracks appear on the panels, but constant aesthetic maintenance is a inconvenient reality with BB-8.
Future of PatchBOTs: I have a few goals with this youtube channel. One is to create projects like BB-8 and Chopper. I consider them "long builds" which I plan to document as well as possible and link all my resources and share all parts in hopes that someone can follow along and build the droid exactly as I did. I plan to keep everything free of charge as far as my files go, and make money for the channel through Amazon Associate links and maybe a future Patreon campaign. I will never charge for my 3D parts.
A second goal of my channel is fill a noticeable void in the robotics world. If you want to start doing robotics you will find 1 million beginner tutorials. And if you want to do some really advanced stuff you can find people talking about it. However, there is a huge gap of education at the intermediate level. For example, My BB-8 droid is controlled by remote control via Bluetooth. Finding a tutorial on how to pair two blue tooth modules was a huge task. I had to combine the steps of about 4 different videos to get the end result. I plan to release a series of "functional intermediate tutorial videos." The first, which I hope to make tomorrow will be, you guessed it, pairing two bluetooth modules.
I love sci-fi, I love droids. I also love pop culture. I plan to do projects within different sub-groups to attract a larger audience. I think the BB-8 and Chopper builds will give me some decent exposure in the Star Wars world, I have plans for projects from the cartoon Rick and Morty, a project from South Park, and a few other of these "cult" like followings in hopes of appealing to as wide a group of people as possible.
I hope I answered all of your questions! Let me know if there are any points I didn't expand on enough.