View Full Version : [Contest Entry] Beetlejuice

06-21-2009, 07:37 PM
Abstract: Beetlejuice is a beetle-shaped hybrid vehicle. It will be able to travel by either wheel or foot and navigate by either sonar or remote control.
Inspired by the Chiba Institute of Technology’s robot, the Halluc 2, I am building a roller/walker hexapod called Beetlejuice. The wheels, powered by continuous servos, flip up to bring rubber-tipped feet down on the floor. The legs will be powered by 333oz/in and 168 oz/in servos.
Beeltejuice will initially use an Atom Pro 28 controller for initial testing but will then be upgraded to an Axon microcontroller. It will include four sonar sensors for autonomous navigation but can also be controlled remotely via a PS2 controller. Beetlejuice will also include a series of multi-colored LED’s for a psychedelic effect.
I will begin fabrication this week, by laser cutting the plastic plates, and CNCing the shell for the back of the robot. I will be cutting the back from high density foam, which I will then lightly coat in epoxy resin. I will then take the epoxy painted foam shape and vacuum form over it with a plastic sheet. I am planning on a rapid fabrication time for this robot and will hopefully start programming a finished product within three weeks. Too optimistic? We shall see…
YouTube - Beetle Robot

06-21-2009, 08:15 PM
That looks awesome! I'll be anticipating some updates :)

06-21-2009, 10:49 PM
Wow. that thing looks amazing. will be a great achievement when complete.

06-22-2009, 03:26 PM
I hereby declare this project as awesome! :P

06-24-2009, 11:23 AM
Well, I have begun the CNC process for creating the vacuum plug. Thorough research has found that CNCing the positive mold (aka plug, in vacuum forming parlance) would take considerable time and money were I too use wood or plastic. Even balsa can only be milled at a rate of 30 inches per minute or less while also costing considerable sums of cash.
So I have found a high density Styrofoam that is relatively cheap (about 30$ for twice what I need) and can be milled at the breakneck speed of 3 inches a minute. Unfortunately, it is likely to melt when vacuum formed, so I have taken the positive mold and painted a thin layer of tooling epoxy over the surface in the hope of providing heat dissipation and structural integrity to the mold should it begin to melt. Below are some before and after pictures of the mold covered in epoxy.
In addition to the foam plug, I have successfully CNCed the negative mold for the rubber wheel out of a small piece of sugar pine.
Tomorrow I will begin the process of vacuum forming onto this mold and report back the results. By the way, thanks for all the nice comments :). It is nice hear such positive feedback.

06-24-2009, 01:45 PM
Well the epoxy covered Styrofoam failed to vacuum form for multiple reasons; One, the vacuum on the vacuum former was broken and, two, the high-density Styrofoam melted far worse than anticipated. I will try again with CNCing SignFoam3 next. SignFoam is supposed to be a temperature resistant, low-cost, and high-density foam which should provide far better results.

06-24-2009, 02:13 PM
Awesome, awesome, awesome, awesome, awesome. Did I mention this is awesome? Definitely keep us updated on this project.

What about covering the plug with fiberglass?

06-24-2009, 04:19 PM
Or milling a negative plug, casting a plaster mold (cheap, heat resistant) and then vacforming over it?

Might be harder to get a good finish on a negative mold, but you can easily sand out imperfections on the plaster cast

06-24-2009, 06:12 PM
Fiberglass is a valid option I am considering. As per plaster, I am told by the plastics experts down at Tap Plastics that plaster doesn't vacuum form well. On the other hand, I have a barrel of plaster left over from a family member's project that I was given permission to use, so I may try that if all else fails.

06-25-2009, 03:41 PM
Wow, awesome project!
Keep the updates and pictures coming.

06-26-2009, 11:06 PM
Well I purchased the sign foam yesterday and machined it today.

The sign-foam was a real steal. Thanks to some barganing and a real nice sales-lady I was able to get a 200$ peice of the material for only 50$.

The foam CNCed beautifully and I will post pictures in early tommorrow, as soon as I have finished the vacuum forming. But now, I must go eat :)

06-27-2009, 10:21 PM

So Yesterday i started out with the above piece of Sign Foam. It can be machined by a router at breakneck speeds (up to 350 inches per minute) because of its very low friction, temperature immunity, and completely inert chemical nature. It really is an amazing material and I recommend it to anyone who can pay the price. The following are some pictures of the foam being CNCed:

http://forums.trossenrobotics.com/gallery/files/2/9/7/4/a_depiction_of_the_difference_between_the_roughing _pass_and_finishing_pass.jpghttp://forums.trossenrobotics.com/gallery/files/2/9/7/4/the_cnc_machine.jpg

The above CNC machine is a Shopbot PRT Alpha which operates at 18000 rpm. The next picture is of the finished beetle shell (with a little plaster patching on the side).


That brings us to today when I started vacuum forming the mold to create the plastic shell. I tried using clear acrylic, white styrene and black abs. The results were mixed; of the six sheets i tried vacuum forming only 4 came out satisfactorily. I will post pictures of them later tomorrow once my camera is at hand, but the following are a picture of the vacuum former and of one of the acrylic sheets cooling on the mold (hence it is covered in water I sprayed at it).


Tomorrow I begin experimenting with a small fiber glassing kit i purchased and I will post some images of the completed and cut plastic sheets.

06-28-2009, 11:36 AM
Well I have finished cutting out the Styrene and ABS Shells. The cuts were a bit rough but I think the end results turned out nicely.


Unfortunately, a few test cuts of the acrylic sheets showed that the hand dremel would not penetrate the acrylic easily nor cleanly. I need to find another method of cutting the acrylic. Would any of you dear readers have an idea?

Meanwhile, I will be trying out the fiberglass today.

06-28-2009, 11:46 AM
That looks really good. I especially appreciate how your taking us through the process, and showing us how you learn from your mistakes.

Can't say I know anything about cutting acrylic...maybe a fine toothed band saw blade?

If you have any extra, make some test cuts.


06-28-2009, 11:50 AM
Actually - fine-tooth is what you don't want, in my experience.

6t/in for bandsaw and scroll saw - if the blade's fine, in melts not cuts.

I cut it with plywood ripping blades on band/scroll/skill/circular, or with a router - dremel router works ok too.

06-28-2009, 11:52 AM
I agree with DB, this thing is going to look awesome and it's great how you show every step of the way.

'fraid I can't help with the cutting part either though

06-28-2009, 07:24 PM
Well I managed to get some mixed results cutting the acrylic using a small metal disk attachement to the dremmel. I picked it up this morning at the hardware store. It provided a clean and quick cut compared to the sanding disk. Unfortunately, of my two clear acrylic shells that I cut, one of them cracked beneath the dremmel. At least I got one good cut out of the batch :)

As per the fiberglassing, I have spent the day testing it on small samples of sign foam to see how it turns out. I tried applying it to the raw foam surface, to a foam surface treated with mold releaser, and to a surface of foam treated with car wax. My hypothesis is that the raw foam surface will rip and tear when I remove the fiberglass, while the waxed surface will release easily. I give mold releaser a fift fift chance of succes based of m observations of it.

I will post tomorrow with some more pictures and the results of the fiberglass tests. I will also start the laser cutting process.

06-28-2009, 09:59 PM
The RC guys have been fiberglassing Foam for quite a while. Here is one such thread:

The terms to search for is usually "glassing foam".

06-29-2009, 09:26 AM
Cutting acrylic can certainly be a b#$ch. When I was in high school (back in 95), we had one of the top ranked plastics program in the country. We had a special blade in a band saw that we used to cut acrylic, although I can't remember off hand what the name/type of it was off hand (I'll do a quick google and see what I can find). I remember that aside from the chipping factor, another reason why we used a special blade is because blades not designed to cut acrylic will dull much faster. I do know if you put masking tape down on both sides of the sheet, it'll help prevent the chipping.

EDIT: US Plastics has some nice tools for cutting acrylic: http://www.usplastic.com/catalog/search.asp?search=cut+acrylic&category=59

06-29-2009, 01:14 PM
Well the results are in for fiber-glassing onto sign foam.

The untreated surface tore horribly when the fiber glass was removed. The wax treated surface came of very nicely but the wax left behind a terrible residue that did not entire wash away under soap and water. In short, it was very messy and ugly. Finally, the mold releaser treated surface released well only where I applied a double coat of the releaser. Unlike the wax, it did not leave residue behind and look very nice when finished. When my camera is recharged, I will post some new pictures of the tests.

I believe that, from this experience, I can get a good fiber glass shell from the sign foam, however the risk to the mold is high; I am very likely to damage if not destroy the sign foam mold in the process. Therefore I propose I make a couple more batches of vacuum formed shells using the sign foam mold and then when I have more shells than I could possibly need, I will create a fiber glass shell.

Thank you very much for all the advice about fiber glassing and acrylic cutting, dear readers. It has been most illuminating and even more so, it has been quite helpful.

06-29-2009, 01:31 PM
Did you see Jes' suggestion? http://forums.trossenrobotics.com/showpost.php?p=31874&postcount=18

06-29-2009, 01:59 PM
yes I did; Even though sign foam is a very difference kind of foam (its more like a soft sandstone than foam), the tutorials were very helpful in figuring out the 'art' of applying fiberglass properly.

07-03-2009, 08:27 PM
Well I have finally perfected the vacuum forming process leading to beautiful shells like this one:

Unfortunately that is more than I can say for fiberglassing; The fiberglass process seems to rip away parts of the signfoam samples I apply it to no matter the wax/releaser coating. Since I have excellent vacuum form molds already and am pressed for time, I have decided to shelve the fiberglassing to some future time when I am bored.

In the meantime I have started fabricating the other parts using the laser cutter (shown below). The Epilog laser cutter is an easy to use and though slow (compared to cnc) fabricating device. One uses it exactly the same way one does a normal printer, except that one has to focus it first and watch it carefully in case its starts a fire. Below is a picture of the laser cutter in action.

So far I have completed a set of wheels and the antennae (shown below). I also have started ordering the necessary parts online and should start the construction process sometime next week. The Axon microcontroller has already arrived and I have begun trying to get it to receive commands from the PS2 controller. I am using a Lynxmotion PS2 controller which is causing some trouble since it is different enough that the Axon's PS2 tutorial software refuses to cooperate with it. I am going to start writing my own version of the ps2 code from scratch tonight but if anyone has some code or advice, let me know. :) Thanks.


07-03-2009, 10:49 PM
Man it looks gorgeous so far!

07-04-2009, 10:15 AM
I am using a Lynxmotion PS2 controller which is causing some trouble since it is different enough that the Axon's PS2 tutorial software refuses to cooperate with it. I am going to start writing my own version of the ps2 code from scratch tonight but if anyone has some code or advice, let me know.
Its because they use a different communication protocol. You just need to find out how they are different, and you can simply make the adjustments in the existing code. Please post up code after you get it working!

07-04-2009, 11:25 AM
Yes i have surmised that they have different protocol and am deconstructing the basic code for reading the Lynxmotion controler as we speak. However, there are a number of code snippets in the PS2 tutorial I do not understand since this is my first time working with an Atmega controller. I will start a seperate post for my questions shortly.

Rest assured I will probably write up a tutorial or something for getting the Axon and Lynxmotion controller talking once I am done and it will include the code too :).

07-04-2009, 11:39 AM
I was just wondering, why don't you make an extra vacuum formed shell and use that to make you fiberglass part? I remember a tutorial done by a fiber glass supply company that showed how to make a fiberglass part that was based on a vacuum formed part. That way you wouldn't have worry the foam mold getting messed up.

07-04-2009, 11:40 AM
oh btw, that shell does look amazing. You should consider making a tutorial describing how you did it.

07-04-2009, 12:03 PM
The trouble with fiberglassing a vaccum formed part is that the the parts are precisely sized to snap fit to the top of the robot's upper plate. If I fiberglass one of the v.f. shells, it would be slighty too big or too small :(. However, I could probably hold it in place with some velcro though so maybe it would work afterall. I will consider this.

07-04-2009, 12:26 PM
Could you machine a new slug that is slightly smaller than you want accounting for the fiberglassing? Actually...I wonder if there is a way to tell the CNC to just cut it at 99% instead of 100?


07-04-2009, 11:43 PM
I might not be thinking about it the same way you are. What I meant was fiberglass over the shell, then pop the vf part out. Then you fiberglass on the inside of the first fiberglass part (the new mold), that should give the correct size. I don't really know how it is all supposed to go together though so this may not really work at all.

07-05-2009, 03:01 PM
Update time! I have begun the coding phase of the project this weekend. Thanks to the sage assistance of you forum goers, I have succesfully gotten the Axon microcontroller to control my SSC32 board and have gotten it talking to the Bot Board II as well. I treid getting the PS2 wireless controller talking directly to the Axon but in the process something went wrong and now the wireless module refuses to work anymore :(.

So rather than risk a frying a new PS2 controller on the Axon, I have set up the Bot Board II to receive the PS2 data and then report it back to the Axon. That code is complete and tested as far as I can without a working controller.

As per the fiberglassing, the cnced mold was sized perfectly so that a shell formed around it would snap fit to the size nessecary. That menas the vacuum formed shells should fit snug as a bug. Ideally the same would go for any fiberglass built over the mold as well.

07-07-2009, 04:15 PM
So now the ABS body plates are complete (pictured below). They still have that toxic ABS smell on them. On the programming side, I have completed the C code for servo control and have begun the Inverse Kinematics. Unfortunately I now have the flu and it is slowing me down quite a bit this week, but I promise not to let it stop me from working on robot. Expect more updates as further parts arrive this week :).



07-08-2009, 07:48 AM
Simply a masterpiece...where you get facility for machining?

07-08-2009, 08:55 AM
Brilliant work!
Hope you get to feeling better soon.
Keep the updates coming!!!

07-08-2009, 11:20 AM
The majority of the machining for this project is done at the Techshop (http://techshop.ws/). For a price of 75$ (for students, 125$ for everyone else), I have unlimited access for a month to vacuum formers, cnc machines, laser cutters, milling devices etc. They even do 3D printing at 15$ per cubic inch (which is hideosly cheap). If you are interested they have franchises up and down the west coast and will soon be expanding to Boston.

07-08-2009, 11:36 AM
Cool. You did you drawings on solidworks?

07-08-2009, 11:46 AM
Yep I got a student version of solid works from work.

07-08-2009, 11:57 AM
A student version from work? sounds a little odd :p

07-08-2009, 12:04 PM
Well the work was academic research

07-10-2009, 01:27 PM
Well the robot is coming along nicely. Here are some updated pictures of my progress:


http://forums.trossenrobotics.com/gallery/files/2/9/7/4/a_leg.jpg http://forums.trossenrobotics.com/gallery/files/2/9/7/4/partial_body.jpg

07-10-2009, 01:45 PM
Wow! Yes, yes it is coming along nicely! Kudos! And a highly sought-after monster +Rep! ;)

07-10-2009, 02:40 PM
I'm really impressed with this project...I can't wait to see it all come together. Your inspiring me...I almost want to make a beetlejuice of my own. (ok I really want to make a beetlejuice of my own...if only I had more money and time!)


07-10-2009, 03:34 PM
+Rep from me as well.

Awesomeness³. looks like a great platform for expanding upon (electronically) as well

07-12-2009, 11:30 AM
Wow, thanks for all the nice complements. I really enjoy reading them. For today's update, I have completed the tibia and feet of the Beetlejuice, shown below:

I also have completed construction and programming of the sensor systems. Below is a picture of the sonar antennae, as well as a video demonstration of the sensor system in action:

YouTube - Beetlejuice's Sensor System

07-12-2009, 04:36 PM
wow. nice work. this is a great project, keep it up. im intently waiting the next update

07-13-2009, 10:17 AM
Very impressive craftsmanship!
Keep up the good work, and keep sharing!
One more +rep cause this is awesome stuff.

07-16-2009, 10:38 AM
Wow this is looking great!!! Don't know if anyone said this yet, but it looks like a giant Hexbug!

07-17-2009, 07:33 AM

Excellent work and very nice pictures and video!
Very fun to see other roboteers that also uses ABS plastic ;)

I loved you very nicely shape top-plate.

Keep up the great work.


07-18-2009, 12:06 AM
Oh no! My robot has gone belly up!

(PS Thanks for the great compliments)

07-19-2009, 03:04 PM
Well the robot is now physcially complete; it only remains to finish the calibration and programming aspects (the later may take a while :( ). Here are some completed photos of Beeteljuice. Expect a video (hopefully) later this week showing Beetlejuice rolling about on its wheels.



http://forums.trossenrobotics.com/gallery/files/2/9/7/4/complete_beetle.jpg http://forums.trossenrobotics.com/gallery/files/2/9/7/4/complete_beetle_2.jpg

07-19-2009, 03:23 PM
Man that turned out awesome!

The high gloss on the body sure takes the eye.

Good luck on the programming!

07-19-2009, 03:57 PM
Ok...here is something that I find upsetting. That bot start to finish came out absolutely awesome. Why am I upset you ask? Because I freaking want one!

Oh well...can't have everything!

Great Job...Can't wait to see it move!


07-19-2009, 04:08 PM

Thats awesome! Looking good and LARGE !
What are the outher dimensions and the total weight?

Looking forward to see the first video.


07-19-2009, 07:00 PM
It has a very trilobyte appearance.

01-16-2010, 10:12 PM
Now that I have finished the axon based hexapod control code, I have moved on to working on my older project, the Beetlejuice. This weekend, I completed the wheeled portion of the Beetlejuice code then added some LEDs and a sound module to the project. In order to test out my new movie-making software, I decided to make a vignette of the Beetlejuice robot in action and share with a few thousand strangers on the internet :happy:.

I hope you enjoy it:
YouTube- Beetle Cameo

01-16-2010, 10:27 PM

01-16-2010, 11:53 PM
Wow! Very nice work! I'm looking forward to seeing more.

01-17-2010, 01:11 AM
Fantastic. +Rep

01-17-2010, 12:35 PM
Thanks very much for the kind comments; I think I will be posting a tutorial soon on adaptive wheel systems.

01-17-2010, 10:00 PM
Thanks very much for the kind comments; I think I will be posting a tutorial soon on adaptive wheel systems.
Reminds me a lot of my 4 wheel system tutorial :veryhappy:

01-23-2010, 09:53 PM
Awesome! One up

02-23-2010, 01:39 PM
Will you still add the walking mode too?
Not that this isnt awesome already, just wondering

02-24-2011, 12:58 PM
So where is this project at now? Always been anticipating an awesome update!