Introduction: Guided Hand
Hey everyone! I'm Yeliz Karadayi and I'm going to be showing you how to easily augment your haptic feedback device to integrate a 3D printing pen, as seen in my thesis project Guided Hand at www.yelizk.com/projects/guidedhand.html. This project was finished before I thought to do an instructables of it... hopefully the documentation I had done will suffice!
The reason I'm putting this instructable out there is because I strongly believe in open-source and collaborative effort. I started with this idea and I know that there are a million different things that could be done with a system like this. If you think about this beyond 3D printing, if instead of a 3D pen you used a paintbrush, for example, imagine the ways you could use haptics in your painting process.
The software that I made was developed purely for the purpose of exemplifying some of my favorite ideas but there is a lot of room for growth and completely new and novel implementation and applications. If you are interested in the software I developed or need any help at all, please email me at firstname.lastname@example.org.
Step 1: Take Apart the Haptic Device
So the first step, and it's probably the most intimidating one, is to take apart the haptic pen. If you choose to cut the wires as I have, this DOES void the warranty. I do recommend doing this on an older or used device.
You can pull off the back end of the pen to keep it out of the way. You also have to pull off the metal pen nib, which requires even more force, and a tool for gripping it. I actually used the nail-remover on a hammer to pry it out with one hand while holding the pen still with another. Both these pieces of the pen will probably require more force than you'd be comfortable with.
Notice the Y shape holding the device pen. On one side of the Y you will see three plastic circles capping some screws that are holding it together. On the other you won't see caps. There are wires that feed from the pen into one of the Y prongs. If you are looking at the side without the caps, then the wiring is on the left side, as you can see in the photos.
What you will have to do is pry apart these prongs so that the pen pops out on the side that does NOT have the wires, so that you can slip it out. I used a wooden handle to get between the prongs and pull them apart. It needs a lot of force. This will leave the pen dangling from some wires on the left side along with a pivot bearing on each side, which you can pop out from the prongs. There are six wires connecting the pen to the device. Three of them are connected to a rotating sensor in the pen. The rotating sensor isn't necessary to my project, as it simply detects the pen's rotation around its length. We don't need to track this rotation, the location of the pen tip is what's important.
The other three wires, which are colored red, green, and blue, will feed into the pen in order to detect the two buttons on the pen. Blue is ground, green and red are power. At this point you have a couple of options. Those who are more weary of breaking the device, it is possible to simply pull the remains of the pen including the swivel bearing off to the side and tape it there out of the way if you do not want to void your warranty. If you choose this option then you can skip step 2.
If you do want to connect your new 3D printing pen to the haptic device's button to allow for in-hand control of the haptics, you will want to mark the wires to keep track of them and then cut them as close as you can to the pen as to maintain maximum length. I soldered them to 3" wire, just to be safe about the length. At this point you'll want to touch the red and blue wires together while running Geomagic's diagnostic tool to see that the buttons still activate as they should.
After that, just push the three wires for the rotation sensor back into where it was feeding out of, to keep them safely out of the way. I only implemented one button for my software, so I only needed the red and blue wires. I pushed the green one back in along with the rotation sensor's wires. If you're into coding you can make your own control options with the second button. At this point you'll begin modifying your pen to communicate to the device.
Step 2: Optional: Mod Your 3D Pen to Incorporate Push Button Controls
If you're going to mod your pen, you'll need a small screwdriver, wire-cutters, wire, a soldering iron, an Exacto knife, and two push buttons. Acrylic and a laser-cutter is a bonus but really any hard plastic in the shape of a button will do the trick at this point. Be sure to take note of which buttons do what, and then use the screwdriver to take apart the device and remove the plastic button covers. Remove the circuit board and pen cap.
In the second photo you can see I've highlighted in red where I cut out a hole through the plastic case. Because it was difficult to reach this area with a drill, I did it in a creative way using a soldering iron to melt a hole large enough to feed two [or three] of the haptic device's button's wires through and then used the Exacto to cut out smooth edges.
At this time, you can solder two wires onto the button. I chose colors to match the red and blue wires from the device, as they will be soldered together later on.
As you can see in the photos, I did NOT choose to re-route the 3D pen's existing buttons to the haptic device. This is because I did not want to risk shorting anything, knowing both the haptic device and the 3D pen have their own power source. It's probably possible but I don't know enough about electrical circuits with two power sources of different voltage and amplitude to deal with it.
Instead what I did is locate the button that you push to extrude plastic. You'll see that the button in the Samto 3D pen is aligned to the edge of the circuit board. Then, super glue down a push button, onto the side of the circuit board such that it aligns perfectly with the extruder button. This way, when I place a plastic cutout that spans over both buttons, when I push it, both buttons will be activated. This was my quick hack to enable my software to know when I am printing, and thus to record my movement.
If you are adding the second button, I would recommend adding a button in its own location and cutting out an entirely new opening to enable independent functionality unlike the first button which is meant to be synchronous to printing.
After all of this you can try closing up the pen and lining up your button cover to fit over both buttons in such a way that you can comfortably push on it and feel a satisfying double-click. Do NOT glue the button cover on to the buttons yet.
Step 3: Prepare the Pen Cap
I only took photos of this piece after finishing with it, but it's not too complicated. The goal here is to make the pen cap resemble the original haptic pen as closely as possible. What you will need this time around is a laser cutter and acrylic, super glue or a strong plastic solvent [I used both], and those pivot bearings from before.
To start, I slid the pen cap into where I wanted the pen to be positioned to find an optimal position to connect the pen to the haptic device. Because the cursor of the haptic feedback device is located at the pivot point of the pen to the arm, I wanted to place the tip of the pen as close to that pivot as possible while ensuring that the Y shape doesn't block the pen's reach or my view. As you can see, this ended up being about mid-way up the cap.
Remember that the wires are feeding through the left side of the pen, so that side needs to have a hollow connection. If you did not mod your pen for buttons, then you don't need to worry about the hollow connection.
The next step is to make the rods that will stick out from the pen cap to fit, as the haptic pen did, into the arm. I chose 1/16" acrylic because there are strong solvents to bond plastics, ensuring a sturdier build, and the rod needs to be a thickness of 3/16". It took a bit of playing around with the circumference so that it didn't fit too tight; for me, a circle of 4.9 mm in diameter to fit perfectly snug into the bearing. The inner diameter was about 1.5mm less than the outer, allowing enough room for the wires to feed through it comfortably while maintaining strength.
The next step is to put a hole on the left side of the cap if you're doing the hollow connection. Again, I used an old soldering iron and Exacto to make an opening that matched the hollow rod to be glued to that side. Slip the bearing into position so that you can test the fit of the cap in place. If you place in the rods into the bearings on each side and hold the cap in place, you'll notice that the pen cap is at a rather steep angle to the rods in the bearings. After cutting out the hole, scrape off the plastic where the rod will meet the cap so that the angle of the surface that the rod will be attached to is perfectly aligned with the rod when properly positioned. This was trial and error, just finding a good comfortable fit. Once this is done you can remove the bearings from the arm and the arm from the rod again and glue and bond the hollow rod to the cap, but not the right rod. I repeated the glue/bond step a couple times to be safe. It was messy, but once dried it is sufficiently sturdy.
You might have noticed on the original pen, on the right side, instead of the hollow opening, there is a protrusion, which is actually a key that locks into a rotational sensor on the right side of the Y. I simply laser cut a replica of this protrusion and bonded it on. It was surprisingly rigid with the plastic bond, but I was still worried about this piece.
Note: Tracking this rotation axis is not extremely important as it detects the orientation of the pen relative to its tip, when all we really care about is tip location for recording prints. However, the more information you can collect, the more you are capable of in the long run. I wanted this rotation tracked because I wanted to see if I could get an ABB robot to completely mimic my printing, which could involve me maneuvering around my own print to avoid collision. I didn't get that far. Maybe someone else can.
At this point you should have something very similar to what I have shown, excluding the bearings and the right-side rod. Unlike what I did, I'm having you wait to glue the rod on the right because doing so will mean you have to pry apart the Y shape again to let the piece through, and it was really stressful to worry about.
Step 4: Attach the Pen
The pen is ready to be attached to the arm. To do so, you will need to feed the wires through the hollow rod and slip the pen cap into place, with the left side safely in place and the wires fed through.
, which is the hard part. On the right side, if you added the key, you will have to rotate the pen until it aligns with the rotation sensor key hole, snapping into place, and that's a lot of force being placed on that tiny piece, so it could break if you aren't extremely careful. Instead you hopefully did not glue the right rod on yet, and instead can easily slip it into place, then, holding the pen cap in position such that the left rod is completely in place with wires fed through, you can glue the right rod down.
The pen extruder and insulation can be placed into the cap at this time as well, but make sure the wires are coming through. When attaching the pen case don't forget to feed the wires through the small hole you had created there, and re-attach the pen.
In the second photo you can see I simply taped the wires from the haptic device to the wires on the button to test out whether the buttons would actually activate the haptic device. Use the diagnostic tool to confirm button functionality. If it's working, go ahead and solder it up. If it's not working, try taking the red and blue wires and touching them together to see if the buttons activate at all. If they still don't, there is something wrong with your wires. If they do work, the button itself could be flawed and need replacing.
After ensuring the buttons work with the taped wires, remove tape and solder. At this time you can optimize the length of the wires as well. I left extra to be safe. Run one last test to make sure everything went smoothly. The pen is ready to be reassembled and closed up, and last of all, glue the button covers to comfortably cover both buttons so that pressing it pushes both down.
Step 5: Try It Out!
Awesome! You should at this point have a fully attached, fully functioning pen. The real fun begins now.
The way I started testing the device is by using some example applications provided by Geomagic Touch. They have applications that demo vibration, path following, surface snapping, and object confinement. These are the simple examples that don't perfectly lend themselves to the printing application but are a great way to start.
I imagine this being used by people with coding skills who want to test out new types of control, and for that you should use Geomagic's API which they have documented pretty well.
I also have my own sketch software that I wrote for the application, if you want it just email me at email@example.com and I can send it very quickly to get people up and running with a basic starting point. There is a lot of work left to be done and the more people play around with it, the better it can be.