Henry’s Design – Day 10

1.31.2009 (Saturday) –  Day 10 –

I can’t think of a better reason for building the Science City Labyrinth Project than this. I received an email today with a design for a pneumatic actuated game platform design by Henry. He is 12 years old and was listening as we discussed the project at Thursday’s Cowtown Computer Congress (CCCKC) meeting.

Henrys Design

Henry's Design

John, Henry’s dad, sent me this email:

Hi Vince,

Just wanted to let you know Henry has had a great time thinking about your giant Labyrinth project. On the way home after the CCCKC meeting we had some fun conversation imagining the project and problem-solving different parts of it. Thought I’d share his idea with you — I know you’re well down the design path, I just thought you’d get a kick out it.

He was mostly thinking about how you can get a board that huge to move fast enough to respond to the Wii commands without stripping gears or fouling a servo’s rotation sensors. His solution was pneumatics (he had just seen the virtual waterslide episode of Prototype This! — so he had pneumatics and hydraulics on his mind). And rather than have a motor turn the board (that’s a lot of torque), he thought balancing the board on a single point would allow gravity to do some of the work, so the pneumatics would only have to do up/down — and with a solenoid valve the pneumatics could power on the push up as well as the pull down or hold in position.

Here’s a sketch….  actually, maybe that balance point in the middle wouldn’t even be necessary if the pneumatics were strong enough.

Anyway…. Enjoy the workings of a young hacker’s mind:

Henry and his dad have an important project of their own. They are making plans to build a drum-bot exhibit at Kansas City’s Kaleidoscope. He also noticed we were talking about vacuum forming some parts for our labyrinth game and sent along a link to the vacuum forming machine he built from plans in Make Magazine.

What About Using Pneumatics?

Vex Pneumatics Kit

Vex Pneumatics Kit

Over the last few years the FIRST Robotics Competition (FRC) has included pneumatics in their kit of parts. This kit is available for the Vex Robotics Design System too. My experience with pneumatics is based on this kit of parts for the FRC robots.

Some issues when using this kit are the cylinders don’t give us precision movement. When activated they are either full on or off, making it difficult to go to specific locations along the way. Pneumatics also require an air supply which can be noisy while compressors are running and the constant movement could overpower an air compressor. The amount of time playing the game and moving the platform would need to be considered when determining the size of an air compressor. Our greatest concern for using pneumatics though was its precision control issue. I like Henry’s idea and encourage others to keep those ideas coming.

Oh No, He’s Got The Knack

I learned about this video on YouTube. Seems like this might be an appropriate time to include it here:

Read on, check out: Day[11] = 2.01.2009
Yesterday, check out: Day[9] = 1.30.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows

Advertisements

Closing In On Fasteners – Day 9

1.30.2009 (Friday) –  Day 9 –

TNut Fasteners

T-Nut Fasteners

Finding just the right type of fastener can be a time consuming task. Over the last two or three days, Bob and I have been going back and forth about how we will attach both halves of the game platform without gaining access to the inside when tightening any fasteners. When I suggested using T-Nuts our experience was using the type with spiked pieces that are embedded into wood.

After explaining what I was looking for, one of the Ace Hardware managers thought of these Brad Hole T-Nuts. We’re excited about these fasteners, it looks like the brad holes will help when gluing them in place.

We spent several hours coming to the conclusion that this combination should work. Materials Science played an important role in this selection. Some of the considerations were the material type and if it would cause any problem with corrosion on the aluminum. Another consideration was the hardness of the screw versus the T-Nut and making our selection based on which would fail first. The machine screws in stock at the time were harder so it would cause the T-Nut material to fail first. Fastenal’s engineer suggested another machine screw that would fail first leaving the internal connections intact. After nearly an hour on the phone with the engineer we were ready to place an order.

I went back to Ace Hardware today to get a different size T-Nut with the brad hole for another project I’m trying to get done by March. I am working on a Steampunkt design to incorporate around my webcam.

The adhesive we need to use when assembling the game platform is critical. We are supposed to be meeting with a 3M Aerospace Adhesives expert on Monday to discuss our particular needs.

Space for Game Platform Assembly

Bob and I met with Jeff at Science City to look at a possible location to assemble the game platform. We need a place with good heating and a perfectly flat floor. As expected, the room’s tiled floor doesn’t provide a flat enough surface. We’re planning on laying a barrier down then using a plaster and screed to build out a machined surface.

I stopped by Home Depot to see if the extruded styrene foam comes in a 4″ thickness. Their stocked sizes range from 1/2″ up to 2″ thick. They contacted Owens Corning to see if they could special order this in 4″. Our guys at Home Depot were told they couldn’t get it directly since they supply residential products. They did get information about a source in Kansas City that I can contact who can sell this product. I’ll be calling this company on Monday.

Read on, check out: Day[10] = 1.31.2009
Yesterday, check out: Day[8] = 1.29.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows

Labyrinth Simulation – Day 8

1.29.2009 (Thursday) –  Day 8 –

Jestins Simulation Program

Jestin's Simulation Program

Jestin has been developing on a simulation program that takes the Nintendo WiiFit data stream using its Bluetooth communications and is directing the graphics in this program. He was showing it off at the CCCKC meeting tonight. To read more about Jestin’s program check out his Virtual Labyrinth blog post.

Speaking of Nintendo Wii paraphernalia, check out these WiiMote stereoscopic glasses over on Bre Pettis’ blog.

We missed Tom at the CCCKC meeting tonight. He was attending an XNA (XBox) game programming special interest group just getting started. I was wanting to go too but it was at the same time as our meeting this evening. Tom said that they discussed making their group into a CCCKC study group.

Read on, check out: Day[9] = 1.30.2009
Yesterday, check out: Day[7] = 1.28.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows

Logistics – Day 7

1.28.2009 (Wednesday) –  Day 7  –

Kansas City Union Station Web Page

Kansas City Union Station Web Page

Working on a daily blog actually works better if you start blogging from the first day instead of deciding to do it a few days later then catching up. Its a lot of work going through emails and notes from phone calls and I’m still not caught up. When I’m not working on writing and documenting I’m spending time talking to Bob about how we’re constructing the game platform and other behind the scene details with members of the team.

The whole robotic labyrinth needs to be transportable if it gets moved to other locations within Science City or other places around Union Station. I heard somewhere that Union Station gets around 2 million visitors per year.

I think we have an incredible opportunity to help Kansas City’s Union Station, Science City and the Kansas City Robotics Society (KCRS) as we also introduce the Cowtown Computer Congress (CCCKC) to the community. The CCCKC grand opening scheduled for the first week in March is attracting a number of people from around the country as well as rumors of a popular magazine among the Make(r) community.

This is a big project and in many ways it parallels what is going on in high schools around the world today. FIRST (For Inspiration and Recognition of Science and Technology) announces their annual competition the first Saturday in January. From that point on students have six weeks to design, build, and deliver their robot.

To take on this project we have a very similar time frame, of designing, building, and delivering the giant robotic labyrinth to Science City by the end of February. This date was picked to coincide with CCCKC’s upcoming grand opening festivities. Much like these high school students, we have our own homework to do in getting our space designed and built out in time for the grand opening. It is going to take some hard work and long hours for many of us. Like the kids on FIRST teams say, its the hardest fun you’ll ever have.

Things That Keep Me Up At Night

Like so many FIRST Robotics high school students in Kansas City right now, I’m loosing sleep over how we’re going to get this project accomplished. This is a big project with several challenges from technical to logistics and budgets. It would help if our budget was larger than $0. We started this project with a few pledges for materials from KCRS members.

Some of the things keeping me up at night:

  • WiiFit Balance Boards are backordered (anyone know where to get one)?
  • Money for materials and WiiFit
  • Finding space for team members to work
  • 4 Weeks left to meet our deadline
  • All the other details I don’t have listed here

For some reason I’ve been listening to Leon Russell’s Tight Rope a lot recently. I wonder where my old 8 track tape is right now. More on Leon Russell here.

Leon Russells Tight Rope

Leon Russell's Tight Rope

Read on, check out: Day[8] = 1.29.2009
Yesterday, check out: Day[6] = 1.27.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows


Robotic Labyrinth Build – Day 6

1.27.2009 (Tuesday) –  Day 6  –

Top 40 Arduino Projects

Top 40 Arduino Projects

I found another web site linking to the wooden robotic labyrinth project. They seem to like our project. Ranked at 14 in their top 40 projects and only a week after the YouTube Video was uploaded.

Source Code Progress

Jestin was called out of town this week for a meeting at his company but reported in with his progress,

Thought I’d write a quick update from my end.

Since I got called out of town, I haven’t been able to put a lot of time into the project since Thursday. I have updated the code that runs the labywiinth so that it now can use a regular wii remote instead of the balance board. It doesn’t seem as fun, but it’s something we talked about and it didn’t take much effort. For both wii devices, I have added the ability to zero it by pressing the A button (only button on the balance board). This will set whatever the device currently reads to turn the servos to 90/90. The next step will be to create some main runtime loop (probably in shell scripting) that will call this program, or an auto-run program, or a remote/network control program depending on whether or not someone hits the “Play” button on the display. Ideas are still needed here.

I have not yet decided what device all this code will run on, but I have received numerous suggestions from the CCCKC mailing list. People there also have experience with the devices they are suggesting, so I’ll probably be able to go to them for help if I get stuck. If anyone has any preferences, let me know.

On a different front, Will Holtorf who was at the CCCKC meeting Thursday has told me he wants to help the group. I have added him in the CC section of this email. I’m not sure what exactly he brings to the table, but I think we can use all the help we can get if we want to hit our deadline. Perhaps he can tell us all what his areas of interest/expertise are.

Though we don’t have much to show yet, I think we are getting off to a good start.

Jestin added more information to his blog. Check out his LabyWiinth Update here

Labyrinth Game Platform Update

We have an interesting variety of technological challenges. Materials science is something we’re dealing with on our game platform surface. We’re aware of the need to be able to gain proper adhesion between similar and dissimilar materials. Bob and I are working on getting a meeting set up to discuss this with a rep from 3M Aerospace Adhesives to identify the best adhesives to use in our case using readily available building materials.

Platform Mechatronics

We talked about using four stepper motors on Saturday. One of my concerns is about about the binding. The last concept we discussed was using three motors offset in a triangular pattern from the center. Tom has expressed some concern about using a universal joint as a gimbal in the center also causing some problems with binding.

I like the two motor approach and it was my first thought but I don’t have experience at sizing the motor and didn’t know how much your motor controllers would be able to drive.

I think the triangular approach could be doable but would make the software side more complicated. I like the two motors along an X and Y axis but wonder if the motors we have will be big enough.

I need to get some new estimates on the weight of the game platform.

Terry’s Perspective:

I was thinking about the motors last night. If we use four motors and assume each pair is in lockstep. If something happens and the steppers get out of sync (maybe by manually moving the table or dirt in the mechanism) there will be a torsion set up between the pair of motors. Depending on the mechanism you chose this torsion could be enough to freeze movement.

The problem is eliminated if we use two motors instead of four.

Jestin had to miss the meeting on Saturday and wasn’t aware of this discussion until today (pretty good reason to keep the blog up to date).

I didn’t even know that more than 2 motors was even an option. So long as the motors are powerful enough, I say 2 axis = 2 motors. It would also look nice if the game actually looks like a regular labyrinth. The only reason I can see adding more motors is if we are unable to move the maze quickly enough (or at all). If we want to complete the project on time, I think simplicity is key. The less we have to worry about keeping motors synced in lockstep, the better.

Jestin added this comment later today:

Truthfully, it doesn’t matter that much to me how many motors we use. We have discussed a protocol, and that’s what protocols are good for; not knowing or caring how the other parts work. I like to think that my lack of knowledge about the design is more attributed to good planning rather than not poor communication.

Game Base Update

Tom is keeping us up to date on his progress with the game’s base portion:

Ok, another update from me on the project base. I talked with Jay Lind today, he’s is the coordinator for the metalsmith union training shop in Raytown. He has agreed to take my design (our design…), set it up in Autocad, verify it w/ us & use the training shop to fab / weld up the design for us. He has given permission to let Vince come along & take pictures / video of the whole process as well. He is going to get the metal cost estimates for use prior to the build, but he believes that the cost will be minimal. They have a cad driven plasma table to use to fabing the plates we need, so no water jet time for us. Since part of there job is community service, all the time from them will be ‘Free’. I can put up $500.00 for the metal & parts for the table base.

We need some kind of contact to get this thing powder coated as well, it will make for a nice pro finish to the product as well. So in the next day we will be passing around the dxf files, along w/ scanned print outs.

I also talked to the bearing guy today & what we want is $400.00. So they sugessted a large ball joint from a 70 series chrysler / dodge truck. You can still get these new. I will be trying to look into that tomorrow & get some pictures posted.

I also need some input on power supply / distribution for the table as well. I have designed in a lot of mounting points on the table base, but need help with what and how much we need. So any help with that please.

Looking into the I2C stuff for communications on the micro network we talked about, thats how I remembered power? what power??

Signing off. (I have a Ham call sign some where…)

Team Member Focus

We have a number of volunteers helping with the project. I received this email from a new member to our team from the CCCKC group: Will Holtorf

I have a strong science background, mostly physics and chemistry,currently I support Navy and Marine Corps VIPs on the NMCI network for all remote sites in the continental US, I am pretty good at manual labor, I can lift heavy things, I can help build, solder, I would not mind learning to weld, whatever you got.

I cannot bring the level of expertise that you guys can on specific areas, but I can ask pretty intelligent questions. Also, I want to stop dragging my butt and add content to my science website, and I would not mind documenting the construction with photos and whatnot… Oh yes, I am a photographer, most of my paid work has been for real estate, but I would love to shoot the work you guys do.

Read on, check out: Day[7] = 1.27.2009
Yesterday, check out: Day[5] = 1.26.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows

Soccer Ball Return – Day 5

1.26.2009 (Monday) –  Day 5  –

I went shopping at fabric stores looking for some kind of netting to catch the soccer balls as they fall through the maze and return them to a loading area. I was looking for something with a mesh about the size of a tennis court net. I didn’t find anything that big but found a couple of candidates.

Rug non-slip mat

Laundry bag material

The receipt calls this fabric Mesh White. It looks like the material of some laundry bags we’ve bought. I’m a little afraid that this material won’t hold up to daily use with soccer balls.

Rug Liner

Rug Liner

This fabric is called Rug Liner, it is used underneath rugs as a non-slip material. It is rubberized and feels stronger than the other material. There is a sale on this material until the end of the month (five days from now).

If anyone has suggestions about where to get a suitable mesh please let me know.

Game Platform Components

I spent a few hours today looking for parts to use with the game platform. We need to create holes through the platform for the soccer ball to fall through. A reinforcement is needed at the edge of the hole which supports the top and bottom aluminum skin. Our first thoughts were to use the Sonotube cardboard concrete form tubes. Home Depot didn’t carry that brand but they did have a similar product. The only sizes they carried were 8″ and 12″ tubes, too small and too large. I asked but they don’t carry a 10″ size.

Since Lowes is across the highway I decided to check there too and the 10″ size was in stock so I bought it here instead.

There is a discussion among our group about making the game maze removable. I checked with my plastics supplier, Laird Plastics, about possibly using Sintra on the surface. I can get it in 4′ x 10′ sheets in a 3mm thickness. Each sheet weighs about 14.78 lbs. with a cost of around $28-$30 per sheet depending on market prices.

Game Base Progress

Tom sent this update,

Ok, an update on my end of things. I talked to the metalsmith union guy on Sunday and it looks like I can get 1 or 2 apprentices from the union shop to help out. It looks like part of the gig is to do community service as well as general fabrication work.

So, tomorrow I am going to meet with them and see about getting a cad file for the base that we discussed. We will then pass that around for approval and start our cost estimates as well.

After getting hold of a drive shaft and U joints, I have come to the conclusion that it is not going to give use the type of independent movement that we are going to need for tilting the table the way we need to. I have calls into the bearing person that we know and am looking into a table spring connector.

You can think of it as a square plate (2 top and bottom)  that has four stiff springs sanwiched inside of them. I will be going over this with the metalfab guys as well for any suggestions.

Stepper Motor Controller Boards

Terry has developed his own electronics to control stepper motors.

Since Saturday, I’ve decided that the motors Rich wants to use are too large for my small stepper motor board.  I’ve built a large stepper board (the same one that drives my CNC) and I’m testing the board for soldering errors.  I’ve also started coding the commands that Jestin and I talked about where the commands are N(orth) S(outh) E(ast) and W(est) followed by an angle 0 to 30.  I’ve reserved the command Z for zero and eventually that command will read the accelerometers and level the board.  The large stepper board also has the ability to handle 8 switches (either input or output) as well as 5 A/D (voltage) sensors.

Other Progress

I ordered three stepper motor controller boards from the RepRap Research Foundation today. These probably won’t be used in this project because they only support up to 2 amp motors but I may end up using a couple on my Steampunkt WebCam project.

I’m looking for a location where we can assemble the game platform. Bob wants a perfectly flat floor that doesn’t have to be level but it must be planar. I’m trying to get in to see some space that Science City has to see if it might be suitable.

Read on, check out: Day[6] = 1.27.2009
Yesterday, check out: Day[4] = 1.25.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows

Game Platform Planning – Day 4

1.25.2009 (Sunday) –  Day 4  –

Vacuum Bag Testing

Vacuum Bag Testing (c) Bob Thurn

I met up with Bob to work on some designs for the game platform. We did a quick vacuum bag test to get an idea of how much pressure can be created to squeeze materials together.

Small Homemade Vacuum

Small Homemade Vacuum (c) Bob Thurn

Photo of the small homemade vacuum rig used for testing. The readout on the scale is misleading. I bought the scale at a thrift store for about $1.00 with the intention of taking it apart. The scale runs about 18 pounds heavy which was verified when compared with another scale.

Vacuum Pulled At Corner

Vacuum Pulled At Corner (c) Bob Thurn

This pulled the plastic fairly tight at the corners. The  plastic covering is 4 mil thick

Vacuum bagging could play a big role in squeezing together laminated pieces of the game platform. This was also an interesting test as we consider vacuum forming bumper rails on the maze.

Game Platform Design Concepts

A few hours were spent considering the connection detail from yesterday’s thoughts. Looking for advantages and disadvantages in the design.  It seems like the game platform’s design is looking a lot like an airplane wing but with enough holes to look like swiss cheese. We are already thinking about using a strong adhesive to bond the styrene foam to the aluminum if we can find something that doesn’t destroy the foam.

We began thinking about using two panels instead of three. If we can make a strong enough joint then we could cut the middle panels to be 2′ x 10′ and assemble this together forming a 6′ x 10′ section. We began looking at alternatives for connecting the two halves and came up with a new sketch of that joint.

Game Platform Center Connection

Game Platform Center Connection

Game Bumpers and Blinky LED’s

Something mentioned at last Wednesday’s Science City meeting was about putting LED’s on the game surface. The question came up today as to how many LED’s and where they should be placed.

It is important to get the game layout determined in the next few days. So many decisions in the game platform structure could depend on this for bolt layout patterns and the game hole positions so we can start building.

The original game has 24 holes – that doesn’t mean we have to include 24 holes on ours. The number of holes may be a decided from a structural issue we need to look at too.

What works best, to have flashy LED’s at each hole or try establishing regions where several holes will exist within a region and flash together? Maybe this could be related to how many sections are in Science City’ s map.

Read on, check out: Day[5] = 1.26.2009
Yesterday, check out: Day[3] = 1.24.2009
Or start from the beginning: Day[0] = 1.21.2009

About SomeoneKnows