Here is the video that we submitted:

Our big idea starts on a small scale: working on building the experience needed to create anything from your imagination, using DIY robots. These will be the toys for the future, where everyone can learn by making, tinkering, soldering, and programming.

Read more about our project on the EYE50 Intel Platform page and the C2 EYE50 page.

More info about the EYE50 competition here:

C2-MTL, in partnership with Intel, is seeking the EYE50, the 50 Emerging Young Entrepreneurs who will transform the world and shape the future. Prove to us your idea is big enough to be the next big thing and get the chance to have an audience for your idea during the C2-MTL conference.

The opportunity to be here and meet other people with interesting ideas, a driven passion, and creativity is great. I really want to learn how people have harnessed their creativity to earn money, and how they evaluated what risks to take to grow further without sacrificing core values.

With the knowledge gained from this week, I hope to put it into action with some of the future robots & projects we are making I’m sure there will be a few moments where someone will say something and it can put things into a different perspective, to look at them differently, and perhaps realize there is something really cool that can happen!

Winning the grand prize would put us one step closer to achieving our goal of making DIY robots as the toys for the future. We would be able to put some money towards more inventory for the RoboBrrd and Buddy 4000 electronics; a highly interactive display for Maker Faire NY; development pieces for other robots; more machines for our rapid prototyping ‘bot farm’; and more development.

Either way, we’ll still be making robots of course! (Stay tuned for a blog post later this week about our Buddy 4000 + BLE iPad App work in progress)

One of the fun things to think about is that DIY robots are really one of several stepping stones to our ultimate goal: making robots become creatures themselves, where they have their own robo-culture, stories that they pass down from generation to generation, and personalities… a real robot society!

A big thank you to everyone who shared and liked our link. It helped our video reach higher up on the page, and we were accepted!

It goes without saying that RoboBrrd wouldn’t be where it is now without the help from others in the Maker community. Read more about RoboBrrd’s history here.

If you are at C2MTL, please come say hi to us (RoboBrrd, Buddy4000, and RobotGrrl)! We’ll be wearing an aqua-blue wearable electronic necklace that lights up white to reveal a lightning bolt design. See ya there!

Buddy 4000 is a classy robot chassis designed to express robot emotions with an artistic flair.

See it in action in this video!

Buddy 4000 is available NOW in the RoboBrrd Store! Special introductory sale, $5 off!

It is everything you are looking for in a robot chassis that looks like a classic robot. Perfect as a chassis for a small project, or just a robotic statue.

Documentation

There is documentation available for building the robot, example electronics, and programming. Check out the main page for Buddy 4000 to see them all.

Open Source Hardware

Yes, Buddy 4000 is OSHW! You can view the .stl’s in our 3D viewer mode, and download an archive of all the files. Hopefully this will let everyone be able to make their own Buddy 4000 (or even a derivative of it). Send pics if you do! Go check it out.

Name Your Price

There is a special ‘name your price’ feature on the Buddy 4000 store page. The extra money is going towards our robotics fund / Maker Faire NY travel fund! We really appreciate your help, and hope to see you at Maker Faire NY!

Inspiration

Buddy 4000 was created to look like a modern version of the old ‘tin toy’ robots. We put much effort into the design of the ‘bot, so hopefully it meets this goal. It took a while to finish this off, especially with all the documentation. Enjoy it! If it makes someone smile, it has been worth it.

—

Thanks to all the twitter people for following this project over the course of developing it. It’s been fun, and the replies were always interesting to read!

Now go get your Buddy 4000 here, or make your own here!

Hooray! The iterations on the ‘Fun Robot Project’ are complete! Since the last post, more work was done on the faceplate/style aspect.

The first idea was to print the style directly onto the piece. However, this made accessing everything inside of the pieces quite difficult. Shown below is the head, with the LEDs and servo in it.

It looks cool from the outside, but it had to be glued together (yikes).

That really didn’t work out. Instead, went with separate pieces for the styles.

In the CAD it was pretty easy to make ‘negative’ extrusions- so they could be printed in a different colour of filament. In order to do this properly, the shapes were scaled down by 0.96. If it was a (what I call) ‘inside-facing-shape’, then it was scaled up by 1.01. This would leave enough space for the kerf, so it could be glued in.

The colours make it look snazzy!

This wasn’t done using dual-extrusion, we’re debating whether it would be worthwhile to upgrade or not. Taking these off the build platform has to be done very carefully, otherwise they jump all over.

Now for the electronics! There is enough space inside of the body of the robot for everything. There are two perfboard breakouts: one for the RGB LEDs, and one for the servos and power distribution. The microcontroller is a 3.3V Arduino Pro Mini, and a LiPower board transforms the 3.7V (or whatever it is, can’t remember) to 5V for the servos and LEDs.

Here is how it all fits inside of the robot. The servo cables and battery are nearest to the front wall:

Everything gets patiently wiggled inside. The servo & power perfboard breakout goes near the bottom.

That breakout then gets folded over, and the servos are plugged in.

Finally, after more poking, it looks like this!

Since it has its battery inside, it will be able to be wireless. Maybe there is room inside somewhere for an XBee as well.

Because of the sockets that are on the arms, head, and feet, it has a few degrees of posability. Combining the poses with the servo movements is going to be fun. Check it out!

Robot is greeting you:

Robot is sad:

Robot is happy:

Robot is cool:

This week we’ll be working on programming it, making a video, getting it on the RoboBrrd Store, and documentation.

But of course before all of this happens- we need to find a name for this robot! We originally made it to look similar to the old ‘tin toy’ robots. Hmm…

Anyway, it is nice to reflect on where this robot has emerged from. Check it out, taking a leisurely swim in the pool of fails!

More robots to come… of course!

ROBOT DANCIN DANCIN TO THE MUSIC OF STEPPERS STEPPIN!

I document & organize the 3D prints very carefully- here you can see the tests in separate bags. Mainly do this in case it’s needed to go back and look at old iterations, but it is also nice to see a real physical trail of the previous design errors

Before we start out on each iteration, here is one of the final-ish assemblies that has some pieces from some of the tests.

Test #9

Here was a test again of the main body assembly. The side pieces (ones with the servo holders) worked out, but the rest of them didn’t really. Also funny: the leg plate was using 5mm sockets for the legs.

Test #10

The design became a little taller! Sockets on leg plate were improved. Beginning the race of iterations for the body top piece (bottom row, in middle).

Test #11

This was a pretty good day of testing! Many more new pieces, also the legs were designed. Initially for the legs, it was going to be a thigh, knee, and foot piece. (A few iterations later, this was discarded)

Also note the quality of this foot print’s first layer. This is one of my favourite pieces from the whole batch

Test #12

Getting the ball joints on the thigh piece was tricky, as they kept popping off. In a later iteration, this was changed to be a socket- with a new two ball piece to join the two sockets. More trouble with the body top piece, also it looks like there are still some nuts in there!

Test #13

The arm of the robot also started out similar to the leg, where it would have a separate upper arm and lower arm piece. The upper arm piece (top right) didn’t work out too well as the sockets sort of disintegrated with the insertion and removal of the ball piece.

Test #14

More tests, and a failed print (on the right).

Test #15

This body top piece was extremely difficult to get right. Many of the iterations were for changing the height of the piece to add more or less room to insert the nuts. In the end, the piece was redesigned from scratch.

Test #16

A new assembly designed- for the head! The head originally did not have the servo in it, but rather an adapter for the socket on the bottom.

This was also the test where the printer stopped working afterwards. The time between test #16 and #17 was actually a few days. Luckily the printer is still working so we could carry on!

Test #17

Switch to yellow filament from VoxelFab. It’s less expensive, and yellow is pretty snazzy. Here was more changes to the head, and of course body top.

Test #18

This was a huge day of testing. The body-side pieces were changed to have the side-mounted screw go through the servo mount. This makes it stronger since now the tab is in the middle of the piece. It also works out well, because the servos hide the screws.

The body-top piece was redesigned to get rid of the nut holders. It works a bit better, but the servo holder was in the way of one of the screws.

Test #19

More tries with body-top, of course. The one on the left in the middle was a weird idea of a design that both didn’t work or look nice. The idea was to have the servo ‘floating’ above the piece, so that it won’t collide with the two arm servos. It didn’t work.

Test #20

This was one of my favourite tests, because it really made the robot ‘come together’ by solving one of the huge issues. The servo was moved from body-top to the head, and now the top has the socket. It worked so much better!

Test #21

Socket improvements on body-top.

Test #22

For some reason, body-top decided to be mean once more. The sockets just wouldn’t work with the ball-servo piece. So it was iterated on a few times, finally finding one that works.

The ball-servo piece began as something that would clasp onto the servo nub using friction. It would sometimes work, but eventually come loose in the end. The winning design was something that can fit a small servo horn onto the piece, and be attached with hot glue.

More to iterate

There will still be some more tests to do, especially for the panel pieces that attach on to the frame.

I’m really excited to finish off this robot! It’s a tribute to all of those old tin-robot toys.

It is surprising how long it has taken to get to this point of the robot, I was working on it for the entire month of March. Kind of discouraging, I guess I have much more to learn.

What you see now is just the beginning- there will be a few different themes for these robots. Back to printing & designing!

The inner diameter of the ball bearing was wrong, and of course this is the weakest point of the design, so added on some heat shrink tubing and some glue.

Pressing down on the arm to make the spring compress before placing the screw in the top left:

Here’s what it looks like with the final faceplate on and filament through:

For disassembling the extruder, I didn’t end up taking any photos (it was too difficult). Here is one of the stepper motor and all the filabarf (filament dust).

Even more around… pretty crazy.

This was the first attempt at mounting the new filament pusher onto the stepper. It didn’t fit very well, for some puzzling reason.

After assembling/disassembling a few times, that reason was because of this piece of support material (i guess?) on the back of the assembly:

Now it fits much better:

Now that it was fully assembled, it was more fighting with it to make it work. It ended up not working (after all that time and effort- arrrrgh!!). The reason seems to be that the spring was too weak. I couldn’t find many places to buy springs in real life (and paying $15 shipping for a spring from online is just lame), so this isn’t really going to work out.

The delrin plunger didn’t have any grooves in it (just a slight marking of where the filament was), so I put a few more grommets and washers on it. The next day when trying it, it worked like a charm!

Now I can print reliable prints at 80mm/s, 220degC, using normal (non-Makerbot) ABS filament. WIN!

Not sure if I’ll ever revisit this experiment in the future, or just order some plunger-replacement kits. Pretty happy with it right now!

They have to be cleaned up, time to dive in. On this side is the belt that moves the encoder wheel and gearbox (which is on the other side). The encoder wheel is made out of metal.

There are two light sensors to detect the encoder stripes:

After cleaning all this up and screwing it back together, it’s time to flip to the other side. This is the part that is protecting the gearbox:

After taking it off- tada! Gearbox! This was pretty dirty before cleaning it.

Did the same cleaning for both wheels. This should make them run better!

Here’s what we’re going to be driving them with! The Pololu Dual VNH5019. This motor controller is pretty sweet, it even has current feedback. Special thanks to Solarbotics!

Here’s what the setup looks like. I wasn’t sure about using it as a shield and the pin connections, so it is wired up with the breadboard for now. The battery is 11.1V (same one that MANOI uses).

It works GREAT! The motors spin and the wheels move. Sensing the current works as well. One of the wheels, however, had a tendency to slip when pressure was applied. I found an elastic roughly the same size and installed that instead of the black o-ring. It slipped a little less this way.

There is a little problem with the Roomba wheels. How on earth to mount them to a robot? Well, luckily with the 3D printer and some calipers, we can rapidly prototype! Here is test #1-4 (left to right). It started as a C, and once determining the dimensions, some boundary extrudes were added on top.

Here is test #5-8 (left to right). #8 ended up being the final version, but had to be reprinted with a smaller number of shells. In test #6, the ‘circle’ part in the middle now has an empty boundary around it. Without it, the piece cannot flex together and clamp onto the motor. The ‘clip’ areas were greatly thickened in test #7. This is important because these pieces will need to be quite strong!

Now, one would imagine that printing the final version would be quite simple. For some reason, this was definitely not the case, and the prints kept failing one after the other! 6 fails in total! Arrgh!

This one is my favourite fail.

ANYWAY, finally it finished. Here they are attached to the motors, with a 1.5 inch 6-32 screw holding it all together!

Though, here is something peculiar. The two motor mounts are exactly the same, except mirror images of each other. The mirrored one printed fabulously the first try. This one seems to keep not having some of the layers get printed. It is slightly worrisome, as the layers could break apart and cause a tremendous robot breakdown.

Notice the difference? Quite weird.

The motor mount was modeled in Autodesk Inventor. Here’s what it looks like:

Stress simulations are fun. This is a simple one with two 10N forces on the clips, and the model is fixed at the ‘circle’ part. You can see where the point of failure will probably be, as indicated in red.

The clips actually don’t move this far when attaching to the motor, and hopefully the motor will also provide some sort of support at the stress points.

It was fun to be able to make something that wasn’t created before, so that these great wheels can be used. Though, I didn’t expect to take so much time on a minuscule task. I guess there is still lots to learn about 3D printing.

Up next, I have to create the shell of the robot. The two roomba wheels will be the only drive wheels, meaning two other wheels will probably be 3D printed as well. RoboBrrd will definitely be able to sit in/on the robot!

Here is a two part video recapping the adventure!

In part 1, it is a summary of all of the tests so far. This includes the ball & sockets, sliders, sliders with fillets, cross section of a ball & socket extruded, and with sectioning.

In part 2, we discuss the mechanical pieces and tabs, capturing the hex nut, and more about 3d printing. Also show the models in Inventor and how they were made. Show the failed prints, which were most often caused by the extruder stopping. Plus, a look at the log book of all the test prints from #1-10! Wooh!

Hopefully these videos help others on their journey of attaching pieces together. It’s worth trying out your own ideas too, or even possibly refining some of the tests! Good luck to everyone who tries, leave a comment if you find a way that also works- we’d be interested in learning about it.

It was fun to make progress on these pieces, as well as progress on the quality of prints.

Here’s how it started out. Using a raft, evident warping, quality not that great:

Somehow this one managed to split in half. Kind of like a waffle:

Getting better prints, this looks interesting (although it did not work)

After more testing and levelling (it took a while), it started to get good enough to be able to print more quickly!

The gcode viewed in Pleasant3D often has stunning results with its vibrant multi-colours!

And so here we are, with the test #11 partial assembly

The purple filament has a decent dent in it from all of the test parts and the RoboBrrd prints (still have a few more long prints to do), but there is still plenty of orange filament.

Currently the printer is printing some higher quality pieces that didn’t need modifying from test #11. It’s starting to get really exciting, just some more designing and printing to do… then adding the electronics! I can’t wait!

These are all the 3D printing files for creating your own RoboBrrd! You can download the .stl files and get printing right away and also view the dimensional drawings. Check out the webpage!

The dream of having RoboBrrd as a widespread toy all around the world is what is really motivating us. OSHW combined with availability of 3D printers, it can be possible!

One day, there will be more RoboBrrds in the world than Furbys, and everyone will make it themselves.

It seems odd that we go to a store to purchase a toy when we can create one ourselves. So let’s do it, and make a positive impact! Manufacture our own toys, customize them, learn about them, build them into robots, make them better.

We put a lot of effort into the webpage by making the pics look fun & groovy. With the 3D Viewer Mode, it’s great to be able to spin, zoom, rotate 3D files. Many people dismiss RoboBrrd as goofy/stupid/useless, so it is also nice to show the thought that went in to RoboBrrd.

On the technical side, creating the pieces to work with both the laser cut pieces and 3d pieces was interesting. The two variables were the extrusion width, and the kerf. Figuring out the extrusion width was straight forward- we measured the mdf with calipers, and they said 3.05mm. As for the kerf, it was exactly +0.2mm different than the laser cut ones. Funnily enough, the laser cut kerf adjustment was -0.1mm, which meant all that needed to be changed was the – sign to a + sign.

There is a donation button on the webpage, which will help us to create even more robots. Each donation will come with a ‘RoboBrrd Name Card’ that you can fill out and give to your newly created RoboBrrd. If you do donate, thanks a bunch!

Well, what are you waiting for? LET’S GO PRINT A ROBOBRRD!

They are waiting, just glued together

Some of them have been painted now

Here they are drying their nice coat of gloss. Look at all of the beaks!

This was bad. We were trying to get the top face on, but it must have been stuck by some excess glue, gloss, or paint. Nothing we were trying was working, and the slots were already carved out with a knife. Standing on it seemed like a good idea, and worked for the previous Brrds. Except for this one. We were really upset about this

Here are all of the Brrds so far

Rivets installed, wing servos added

Red RoboBrrd takes a break to enjoy looking at a multimeter

Time to solder up all of these RGB LEDs and their resistors. This part is so mind-numbing and boring.

Soldered up, labeled, and with heat shrink

It took us ages to make all of these… phew!

Soldering up the LDR, resistor, and wires

Installing front face onto Blue RoboBrrd

Wiring up Black RoboBrrd

More LEDs to wire! AAAH!

The CTV RoboBrrd was sitting around doing nothing… painted it PINK!

Time for the TMP36 and Piezo speakers!

Remade Orange RoboBrrd

Black RoboBrrd is set up, calibrated, and working!

That is it for the progress so far. Much of the time gets chewed up in the little things. Here is a list of fine details that no one will know about, except the custom RoboBrrd, myself, and now you the reader!

• Excess glue was carefully taken away
• Lighter colours were two coats of paint
• Gloss was added everywhere on the Brrd (except the Skotch-Yoke pieces)
• Gloss left to dry for 24h
• Slots & holes ‘careved’ out from excess gloss & paint with knife
• Actually got a blister on my finger from all the carving
• Test fitting and more carving is done
• Adding heat shrink and labels to all the wires
• Twist wires by hand for neatness, and add additional electrical tape

It takes time to create a RoboBrrd. Hopefully you will appreciate the time & effort spent into the them. We really care about the Brrds, and it’s the same amount of effort that we put into our own as well. It’s pretty tough trying to stay focused on them, especially when there are quite a few to create! We’re aiming for the end of this weekend to finish them, but we’ll see how it goes.

Back to building the Brrds! ^_^ CHIRP!

Interested in making a robot (perhaps) with ball and socket joints. The balls were in increasing size to 5mm. The sockets had an increasing overhang. Here was the result of the first test.

For some reason early in the morning 5mm seemed like it would be big enough. Clearly it isn’t, so here was the next test (based on some of the results from the first test). The scale of the balls is 10mm.

The result is the socket with the 1.5mm overhang works best with the 9.7mm ball. The 9.8mm ball can also fit, but it is tougher to move.

If there is a coating on the ABS that makes the plastic slippery (not sure if there is or not), then this could be bad if it is stripped off when moving the ball & socket around too much. Krafter suggested to use some acetone to make the ball smooth as it eats away at ABS, this is on the ‘to experiment’ list.

Next up is a creature created in 123D Creature! This is software from Autodesk that runs on your iPad. It’s pretty sweet, really easy to create imaginative creatures.

Here is my first creature, the Ancient Exploring Mlipto!

Mliptos are tiny creatures that go on huge expeditions. They reside in wetland areas. You know when one is nearby by its distinct high pitched buzzing sound.

Printing it was fun:

Definitely needed lots of supports, it almost looks like it’s buried in plastic!

Here was the result:

From the behind:

The details are nice:

It’s still a work in progress. One of its feet fell off because its legs are thin. A few more modifications are needed before releasing it into the RoboBrrd Shop and Thingiverse.

This software is quite enjoyable. Here is the next one that will be created, we’re really excited to see how this one will turn out!

Next up are improvements to The RED^GREEN Machine! There were a few days that were really dry here, and we accidentally zapped our Message Pump Through Hole. This meant having to monitor the bot temperature from Serial Monitor… and that’s no fun. SpikenzieLabs was nice to send another one along, now it’s back up and running! THANKS GUYS!

Actually, we have even modified this further to display a bar of ‘:’ characters along the bottom row of the LCD to display how the temperature is (scaled value from 16 to 38 to 1 to 16). When it hits 27 degrees C, the Plasma RoboGlyph will start to blink.

SpikenzieLabs also sent A LOT of stickers, with some cutting it became nice wallpaper for the bot!

The ‘hinge’ on the front panel was slipping off. Electrical tape is not that sticky. We bought some duct tape (by the way, that tape is really expensive!), and there is a new improved hinge now. It works much better than the electrical tape.

That sums up the fun for now! The white filament is depleting quickly now, soon the prints will be in PURPLE!