Why I wanted a 3D printer

While having a 3D printer has turned into a hobby in its own right, my true intent in building a printer was to use it to make other projects.

My latest project isn’t anything high-tech at all, but would be almost impossible without a jig that I made with the printer.

Here are some pictures of the jig and the tiki-torches that I’m making.

The jig uses three interlocking parts that are held together by threaded rod. The top part holds a captive nut.

All three parts spaced out.

Interlocking Parts

There is a non-uniform helical path in the side, along with a hole in the smallest section to capture the end of the rod. I #6 copper wire (about 4 millimeters in diameter) around the jig.

Bending a 24″ section of No. 6 copper wire around the jig.

When I bent the first wire, the small end of the jig broke. I filled it with epoxy and let it dry. It worked, but it wasn’t perfect. It still flexed, but didn’t break. I later created another part (not pictured) that went on the outside giving the small diameter better support.

After bending three of these, I assembled them loosely with fine copper wire. I then wrapped each of the top joints with about 4″ of the fine wire to hold them together in their final positions.

Wired together.

Three bent parts

I then took a 1/2″ to 3/8″ copper pipe fitting, filled the 1/2″ end with some aluminum foil, filled the 3/8″ end with epoxy, and then put the bottom of the helical pieces in the epoxy to finish the assembly. After the epoxy was set, I used a needle nose pliers to tear out the aluminum foil.

Bottom epoxied on.

I took a Bawls bottle, wrapped another 1/2″ to 3/8″ copper fitting with teflon tape to make it fit the bottle opening snugly. I fed a tiki torch wick through the fitting, inserted it into the bottle, and then trimmed the length.

I used some hydrochloric acid to remove the labeling from the bottle. After about ten seconds in 30% concentration, I rubbed the labels with some steel wool and they quickly dissolved away. Use some care if you attempt this.

Below if the picture of the mostly finished project. I hadn’t taken the labels off yet. Now I can add any length 1/2″ copper pipe to stand the torch up. I have 8 of them made so far, and plan to make at least a few more.

Almost finished.

I hope you this inspires you to create your own version!

Electronics, Gaming, Hardware, Raspberry Pi, Software

An 8 Bit Detour

I’ve been on hiatus from my printer for a while. While I haven’t done much in the way printer updates, I have been working on a different side project.

This is my NintendoPi:

It’s rather inconspicuous from the outside, and that’s the intent. When the cartridge door is closed, you can’t tell it apart from any other NES.

Most of this project followed this blog, and originally only included the NES emulator and games. I had the original controller ports on the front of the console wired up, and during testing the controllers by challenging my wife in some retro games from our childhood, I decided to go the extra mile and add the SNES portion to it. Back to the lab, where I modeled up some parts and printed them out, and in short order had everything soldered together to include SNES ports where the cartridge slot used to be. This was always in the plan, but I figured it would completely finish the project with just NES capabilities and then add the SNES stuff later.

Under the hood, you can see everything that I’ve done:

I de-soldered the original power supply from the main circuit board and cut the end off a USB to mirco-USB cable and connected the power wires to get 5 volts for the Raspberry Pi, and connected the original power switch inline with these wires to turn it on and off. I disconnected the reset switch from the power connections and used jumper wires to attach it as a button on the GPIO pins of the RPi. I also created a part to accept keystone sockets in the back for ethernet and HDMI connections, and the there is another one under the game cartridge cover for USB connections. This allows me to modify the RPi either remotely through SSH, or locally by plugging in a keyboard, without having to open up the console case. I used the semi-permanent solution of hot melt glue to attach the 3D printed parts the case, as well as the SNES controller ports to the 3D printed parts.

Now I have both NES and SNES controller ports working, but it is much easier to play with the SNES controllers because they have a lighter button pressure which doesn’t fatigue your hands as fast, but the option is there for any of my die-hard friends that feel they must get the full Nintendo experience with the classic controllers.

I’d love to hear your comments or suggestions about this project! Until next time, happy modding!

Uncategorized

What I’ve been up to lately

Edit- Guess I left this as a draft for the past few months… oops!

I don’t have too much to show you as far as progress, but I haven’t been sitting idle. As I write this I have a cross slide vise being shipped to my house, which gets me one step closer to being able to machine some of the parts I need. I have also thought about ways to refine the hot end design, and will try to create models of this to publish, and hopefully someone will be willing to try it out before I get around to it.

I moved my “lab” from the basement of my house to an extra bedroom on the first floor, as it’s getting colder here, and nothing keeps you warmer than sitting still at a computer in a 60 degree F basement.

My current plans are to upgrade the printer with the details I have shown (and a few things I haven’t shown yet), get it nice and reliable, and then start working on some other projects. I’ve read some MIT thesis papers lately that have gotten me brain storming. Couple this with my frustration of trying to get parts machined, and a lot of my thoughts have been directed towards designing a small 3 axis CNC mill, and probably a small CNC lathe as well.

From my quick searches online, I haven’t found any great open source designs for these, but there is definitely more information about building CNC mills than there are about lathes. I think either of these types of items would be a great challenge and learning experience, which is what I’m after in the end.

3D Printing, Hardware

Version 2.1 Part 1

As promised, here is an overview of the changes being planned for version 2.1. This is an upgrade to the X and Z axis of the existing build, with the Y axis being kept the same.

This post jumps around a bit, but I figure it was better to put it out faster than to spend more time organizing it all better. I may rearrange it later (or probably not).

Overview of the changes. New parts are in green.

First up, changing the orientation of the Z axis rods. The new motor mounts are still under construction, but will follow along the same philosophy as what I have in the picture below. The motors still mount in the same location, with the big change being the addition of an angular contact bearing below the coupling. I may look at changing over to acme rods to decrease the runout of the drive rods, but I haven’t looked at the cost of this yet. It may just wait until a future date as it would be a minor change.

Z axis concept with angular contact bearing in blue

The angular contact bearing (blue in the cross-section above) is designed to withstand pressure both axially and laterally. I will be including a second one at the bottom of the drive rod to support pressure in the opposite direction. I don’t have this part designed yet, so it isn’t shown in the overview at the top.

The second change is rotating the X axis smooth rods to be aligned vertically. This allows me to get the second bar out of the way and run the belt between them to decrease the skewing and the increase in friction that is experienced from pulling off center. This picture shows the new carriage.

New liquid cooled X carriage design

My major goal in starting this redesign was to gain the ability to print two materials at once, with the intent of running PVA plastic as support material that can be dissolved away easily after printing. Along with this change, I plan to move away from the PTFE bushings and use recirculating linear ball bearings. To retain these I am using screws as shown in the cross-section below of the new X carriage. I dislike the use of cable ties to retain bushings and bearings. It seems no matter how tight you think they are, they always either work their way loose, or stretch slightly. I was after a way that wouldn’t distort the diameter of plastic bushings before I decided to go with linear ball bearings.

Bearing (dark grey) being retained by an M3 screw.

Water channels that will be drilled into a 12.7 x 50.4 x 100mm aluminum block. All of the fittings for the water channels are 10-32 thread with Buna-N seals.

The design will use two PTFE tubes to deliver the filament from two new drives that will be mounted on the back of the machine. These are also a new design I’ve been working on, in conjunction with the axis changes.

Dual filament from remote extruders

The slot in the center of the cooling block is for routing the wires to the heater blocks on the nozzle ends. I plan to run the wires up through the middle, then bundle the PTFE filament tubes, water delivery and return tubes, and the wires together for a neat appearance.

I’ve been making some on the fly changes as I don’t have access to a lathe to make any circular parts and sending anything out to be machined is out of my budget currently. I’ve been redesigning any parts that would have needed a lathe to something I can buy, (nozzles and peek insulators) or to parts that can be made with a drill press and a simple two axis clamp.

This constraint got me brainstorming about designing a small mill next. I’ve just started looking into it and coming up with some concepts. This project would be a long way off, but it could be a very interesting learning experience. That’s what it’s all about in the end.

If you have any ideas or comments, I would love to hear them!

Uncategorized

From the ground up

I’ve been working on two new printer designs lately. One will be an upgrade to the current design with simplified geometry, and the other is an entirely new design from the ground up. I’ve been concentrating on the upgraded version, and now I’m waiting to get parts machined for the new hot end designs. Once I get my current printer upgraded, I’ll be putting more effort on the new design.

I’ll be posting some pictures of the upgrade designs soon.

Electronics, Hardware

And then there was light

And God said, “Let there be light,” and there was light.

It may have been that easy for God to create light, but it took slightly more work to create these LED lights. This has been on the project list for a while, but I actually started it on a whim while contemplating what to do with my afternoon. A trip to Radio Shack and I was underway. This is a pretty simple project, and the only setback I faced is that I didn’t really approach it with a real plan. This caused some extra work and I plan to redo them eventually, but they work well enough that I’ll be leaving them alone for a while.

I created an array of series of 3 LED’s and tested it to make sure my math was correct when picking out a resistor. If I was wrong I’d rather know before I solder parts together. After the first series worked, I added the other three in parallel and then tested the entire array.

LED’s on a breadboard

Testing the entire array

Then I moved all the parts to some perfboard and soldered them together, testing each series after soldering to make sure I didn’t overheat the LED’s.

Testing LED’s

After soldering on lead wires and cutting the light sections apart I used foam based VHB tape to mount them to the top of the printer cabinet. This is where a plan would have helped. There were two things that could have been done better:

1. I cut each series into its own section and gave them their own leads. I wasn’t sure where I was going to mount each light originally, but I ended up mounting three of the series on the roof right next to each other, so with a little planning I could have just left them all on a single board. Also because I cut them apart, I gave each series its own set of lead wires. If I had left them together I could have just used two wires instead of using six. I also had to find a place that I could tie those six wires together because of this. I also left the leads as long as I could, but this means I had to find places to coil the extra wire until I come back and clean up all the wiring.

2. I cut the perfboard sections too small which didn’t leave me very many options for mounting them. This turned out fine as the VHB tape I used has 1/8″ foam in the middle so it was able to adhere fairly well to the uneven surface on the back of the soldered assemblies. I had to use a large piece of tape for each series to make sure the soldered contacts were covered so they couldn’t touch the cabinet frame they were mounting on. The solder of the fourth LED series was not as flush with the board and so the tape didn’t stick to it as well, so I left it off for now. I plan to have my mounting solution figured out for the set of lights in the future.

LED mounting

This is the result shown with the lights on and off.

Lights off

Lights on

I added a switch for the lights, and while I was doing that I added a master power switch for the entire printer. Now I don’t have to keep unplugging it from the wall.

3D Printing, Hardware

Back in Business

After the previous fixes, I leveled the bed today and started some calibration prints.

Printing the first calibration part

After completing the first print, I measured the part against the theoretical dimensions of X = 100.0, Y = 100.0 and Z = 10.0. X and Y were pretty close, Z was off by quite a bit.

Z was off so far because the I had not adjusted the software from the previous version, and it was set for 5/16-18 threaded rod, but V2.0 used M8 x 1.25 threaded rod. I adjusted the settings in the firmware, and these are the results of the second calibration print. (Sorry for the poor pictures, I don’t have very good artificial lighting in this room.)

X and Y are pretty much set, but I’ll have to give Z some more adjustments. So far I’m pretty satisfied with the progress, as getting to this point with V1.0 took weeks instead of a day.

Automata Intus

Exploring robotics, artificial intelligence, and everything else.