Category Archives: Maker


3D Printed Bike Badge

This one’s been quite the project in the making, though maybe not for the best of reasons. My co-worker Bob asked me if I could make a Figoli Quinn branded badge for his bike, replacing the stock one there already. Awesome project, only I kept forgetting about it as other things came up. Plus I did actually remember to make it, and then lost it when we moved from Arizona to ORegon, so I had to make it again. Sigh. So what should’ve taken an afternoon to make took all summer.

Once I got to it, the project had an easy start. Bob gave me the existing badge so it was a simple matter of spending a few minutes using my calipers to measure how big the badge needed to be so that it wrapped correctly around the bike tube. Once I did that, I headed into MODO to build out the design.

While I usually try to keep most of my 3d printing work in Shapr3D on my iPad, I really didn’t want to try to redraw the FQ logo. Instead, I just brought an EPS file of it into MODO and extruded it. The entire thing was exported as an STL over to Simplify3D where it was printed.

Once it was printed (a couple different times to figure out the best orientation) it was just a matter of sanding, priming, and painting before handing it off to Bob to glue to his bike.

3D Printing with Wood Filament

3D Printing with Wood PLA

That’s right, 3d printing wood. Is it really wood though? Kind of. It’s a composite made by Hatchbox (https://www.hatchboxfilament.com/products/copy-of-wood-hatchbox-3d-wood-1kg3-00-3d-printer-filament-dimensional-accuracy-0-05mm-3-00-mm-1-kg-spool-wood) that’s basically a mixture of wood particles (think sawdust) and PLA into something that kind of looks / smells like wood, but prints like standard PLA. Depending on how you use it and treat it, it can either look like wood, like brown plastic, or more likely something in between.

Printing

For the most part, you can treat the printing process for this more or less like you would with printing standard PLA. It prints anywhere from 190C – 240C without needing a heated bed, so most 3d printers that don’t require proprietary filament should be able to use it.

Pitfalls

After using this for a few months off and on now, since it is so similar to standard PLA, there’s really only a few pitfalls I’ve consistently encountered, one of which is hopefully fixed by tweaking some print settings. The first thing that’s easy to notice is that while printing… it definitely smells. It’s not really a bad smell, but it definitely has a slight burning wood smell to it. For the first couple minutes, it’s kind of cool. A couple hours into a print, I’m definitely over it.

The second major issue I tend to have is that when it prints, it tends to be a little stringy. Here’s an example of what I’m talking about:

stringy-3d-printing-wood

It becomes more obvious if you’re having to span large gaps. If the entire print is within a pretty tight area though, it’s not too bad. One thing I need to play with though is the retraction setting when printing. I have a feeling that if there’s more retraction while traveling, it could probably solve this problem.

Wood Pattern

I mentioned that with the filament you could get something that really did look like wood, something that looked like brown plastic, or something in between. Well the first thing to help with that involves creating a wood pattern when doing the actual print. Due to this material actually containing some wood particles, you’ll find that the higher the temperature you print it, the darker the print will be (and vice versa). A trick you can use to introduce some color banding to give it a wood pattern look is to vary the temperature of the filament across the layers of the print. I use Simplify3D which makes this process pretty easy — I’m not sure how other slicing / printing packages handle this.

In Simplify3D within the print settings, click on the temperature tab. What I’ve found that works well is roughly every 15-40 layers, change the temperature of the hot end, changing it anywhere from 180C to 245C. Remember, the higher the temp, the darker the color. If you want to have a sudden solid band, do something like transitioning from 180C to 240C, it’ll be really apparent. Just try to randomize it a bit. Here’s an example of what it might look like:

simplify-3d-wood-settings

Finishing tips

Like with most 3d printing, the first step in the finishing process is some sanding. You’ll find that because this does have some wood in it that it sands really well. I tend to start with a sheet of sandpaper in the 100-200 grit range and slowly work my way up to the 400-600 range just to give it a nice smooth surface. When sanding, take it slow. If you try to sand too quickly, it’s easy to heat up the PLA enough that it sort of gets gummy. Start slow, do it while watching a TV show or movie, and slowly step up the grit to get a nice smooth surface. After it’s nice and smooth, wipe it down with a damp cloth to get any excess wood particles off of it.

Once it’s nice and smooth you can either paint it, or if you want to keep a wood look to it (you’re using wood filament, right?), you can use a standard oil or water based wood stain. I found it works well to load up a cheap brush with the stain, brush it on, and use a rag or paper towel to wipe up the excess. When wiping up that excess, wipe it off in a single direction to enhance any sort of banding or wood grain effect. Check the stain for specific directions but I found doing 3 coats with a couple hours between coats worked really well.

wood-stain-3d-printing

The last step is locking in the stain with some polyurethane. It comes in either a liquid you can brush on, or a can you can spray on. I opted for the spray but I’m sure either works fine. Again, follow the instructions on the can. This last step helps lock in the staining you did previously, and gives it a nice sheen depending on whether your polyurethane is matte, gloss, or semi gloss.

3d-printed-groot

And there you have it, not bad right? This Groot model also had a little bit of acrylic paint on him to give that green moss effect, but all in all not a bad wood effect for something that would be pretty tough to make out of real wood, but wasn’t too bad to print. Check back for more tips and tricks with 3D printing.

mbot-demo-makerfaire

Portland Mini Makerfaire

This past weekend I managed to make it to my first ever Makerfaire, or really any sort of maker or DIY event. Even though it was a “mini” one (I have no idea what that actually means), it was still awesome to attend. Going into it, I didn’t really have an idea of what to expect. Maybe something like a trade show with a stage to do some presentations. What it ended up being was much more faire. There were tons of different booths and exhibitors at OSMI, some of which selling things, others promoting their various organizations, but most were just sharing what they do in the maker sphere of things. There was 3D printing, forging, crafts, cooking, astronomy, and even a working Tesla coil.

While the entire event was great, there were a couple key items that stood out in my mind that definitely made it worth the price of admission. First, I found a local shop that can only be described as a junk shop, and the best kind of junk shop. They had a little more store setup, and pictures of the main retail store that was racks and bins of just any and every electronics components and hardware. I’m sure I’ll waste some money there coming up. Another good thing was just the ability to see how I fit into the crowd as a whole. There was a good amount of 3d printing, but even with some nicer printers I can’t say that people were getting any better quality out of their prints than I was. Most of the stuff I saw was also unfinished (I’ll try to have a post here soon about some finishing techniques) and people printing models found off of Thingiverse. I didn’t see much in the way of 3d models that people designed themselves. Finally, I got to talk to the folks at Proto-Pasta about their high end PLA filaments. The quality looked great there and I liked the idea of getting a stronger PLA to work with, so I picked up a small roll to play around with. I’ll try to post some examples when I get around to it.

All in all, the maker faire was well worth the price of admission and something I’ll be doing again next year!

3D Printed Phone Stand

A new site deserves a new post right?

It’s taken far too long to redo my site a bit. I’m no longer in Phoenix, and I wanted to develop my own theme that would better suit more of the 3d printing and making that I’ve been doing lately. In showing off that stuff, I’d love to feature some time-lapse videos because no matter how many times you see it, there’s just something cool about watching a 3d printer build up layer by layer. That is, until you’re several hours in. Then it’s just monotonous.

I wanted to take advantage of the easier to use time-lapse feature on my phone, but didn’t really have a stand to place it on when using it. So why not make one instead of buying one? It’s not like I need to spend more money on this hobby. While I love designing my own prints, I found this AWESOME modular phone stand system over at https://www.thingiverse.com/thing:2194278. How cool is that? While swapping out the printed nuts and bolts with real ones would probably yield a better end result, this definitely worked well enough for my needs. The only modification I made was gluing some felt into it so I wouldn’t scratch my phone.

Not too shabby right? You can tilt and pan it, and the phone fits in nicely and doesn’t seem to knock it off balance. Here’s what it looks like in place, and the time lapse I was able to get from it.

Hopefully I can get more of these going and get some more posts coming soon!

Switching to Simplify 3D

It’s been about 3 months now having switched from using either Cura or Slicr for my 3d printing slicing to using Simplify3D, and it’s been awesome. You couldn’t pay me to go back. If you’ve never heard of it or never looked into it, go check it out. Seriously, head over to https://www.simplify3d.com.

You’ll probably look at the features list and go “Meh, other software has most of that and it’s not $150”. Go read some reviews or watch some youtube videos. This software is 100% worth it.

It’s FAST

It sounds trivial, but especially if you’re working with high poly count objects (pretty much anything out of ZBrush), it sucks to try to import something into your Slicr and have it either crash, or you have to go walk away and make a cup of coffee while it struggles to slice it. Have to do that a couple times? Find a good movie to watch. It just sucks, and coming from an animation background and not a CAD background, I basically stopped trying to print anything I’d done in Zbrush, which limited what I could make.

In comes Simplify3D. No more waiting (I mean a little, but seriously it’s almost nothing). Check out the image below, a screenshot from Simplify3D. It’s a model exported from Zbrush, about 960k polygons of Maui’s magic hook from Moana.

In Simplify3D, it took about 10 seconds to slice it on a 2010 iMac. It’s a slow machine, and it did just fine. I did the same thing on my 2015 iMac using Cura (two machines in the same room) and it took several minutes. Doesn’t sound huge, but in the span of a minute I could try out several different settings for slicing until I found one that worked great. To do the same thing in Cura — I’d be there a good half an hour.

It just prints better

It sounds weird, or it did to me originally, but it just flat out prints better. I have fewer issues in printing (most now due to me not leveling my print bed correctly), fewer drips when traveling, and just smoother prints. Yes you can probably get similar settings out of others, but look at how simple this is.

There’s a ton of settings here if you really want to dive into it. But honestly, 90% of my time is spent choosing between ‘Fast’, ‘Medium’, and ‘High’ using the presets, and then toggling the checkbox next to Generate Supports. That’s it, and it’s way better than anything I was getting before.

Let’s talk about supports. If you’re doing stuff with supports, you need to be using this. The generated supports are better placed, you can manually edit them, and they print soooo nice. They’re easy to break off afterwards, and when they do, it’s like a little accordion. Here’s an example of an elephant themed sunglasses stand I made for my wife, a before and after removing the supports on it.

It totally came off in one clean piece. How cool is that?

Of course, there’s a bit of a trade off with these awesome supports. First off, it’s not the fastest thing in the world printing these supports. Second, if you’re not careful, they’ll use up a ton of material. But personally, I’d rather waste a little extra material on some supports over having to reprint several times because a support failed. That’s all I got. I know it’s expensive, but if you’re into 3d printing and your printer will allow it, I definitely recommend giving it a try.

Animatronic Door for Dr. Seuss Week

This past week was Dr. Seuss week at the school my wife teaches at. One of the ways they were celebrating and getting kids engaged in Dr. Seuss was having a door decorating contest. Each classroom would decorate their door in a Dr. Seuss theme, complete with some visuals and a quote. My wife asked for some help, and the video below shows what we came up with.

Pretty cool, right? She didn’t end up winning, but I thought I’d do a post on how we went ahead and did this. The entire project was completed over 2 evenings.

The basic design

My wife teaches elementary music, so we thought we’d stick with a music theme. Problem is, there’s oddly not a ton in the Seuss world that screams music. In fact, we could really only come up with the scene from How the Grinch Stole Christmas where the Grinch is describing all the musical instruments they use, and how much he hates the noise. A fun irony for a music teacher, right? So we decided to run with it. We started with the simple sketch below, figuring we’d need a grinch head, with some thought bubbles around him showing the different images.

Pretty fun…. but what if we could get it to move like he did in the movie?

Construction

Since I had an Arduino and some parts that were lying around not doing anything, I suggested maybe we try to make this thing move, or light up, or do something. We settled on making the eyes rotate and the mallets move back and forth if someone stepped in front of the door, similar to how the animated movie worked.

With that idea and our sketch, we knew we’d need 4 servos to move all of that, a way to determine if someone was standing in front, and make sure the items were all separate so they could move. We started with creating the assets for the door. The thought bubbles were printing and mounted onto foam core, though we decided not to have them interactive at all. Since the Grinch would be interactive, we needed his face, each eye, and each mallet to be complete separate pieces. The eyes also had to fit behind the face so when rotated, the pupils would turn correctly in the non-circular sockets. Most of this was created by handing using foam, and then mounted onto foam core with some marker coloring. The Grinch face was traced using a marker and projector (with a frame grab from the movie) to accurately draw and cut his face.

Assembling the electronics

The Arduino piece of this was actually pretty straight forward. The movement would all be driven by 4 180 degree servos. Each servo has 3 wires for power, ground, and data, and is wired similar to below.

So all we had to do was use a small breadboard (didn’t want to solder any of this) so tied together all the related ground / power wires, and then the servos ended up using 4 separate data points on the Arduino. For the distance sensing, I decided to use an Ultrasonic Ping sensor. Maybe not as useful in this case as a motion sensor, but we already had one on hand. With this sensor, the wiring is still incredibly basic, consisting of a power, ground, and then 2 separate data wires (triggering the ping and receiving the response).

With everything wired up, I’d used a total of 5 data ports on the Arduino. The more pressing issue was trying to hide as much of this as possible, so it wasn’t a door covered in wires. We decided the only place we could really do this was right behind the Grinch’s face. We also wanted to hold a battery pack so we could run this off of AAs (this didn’t end up working, but it was built in). So to hold most of this, I went ahead in Shapr3D construction a simple custom box that would hold the Arduino and a battery pack, while having holes that would allow wires to pass out the back, which would be covered in paper, hiding the wires. This was then 3d printed, and the wired, with the Grinch face taped to the face of it.

As you can see, due to time, it ended up turning into a bit of a messy rat’s nest of wires and tape. But hey, it totally worked. The tape helped keep all the wires within the box (and going out the back under the blue paper) and keep the Grinch face secured to the box. Each of the eyes and mallet were secured to the servo using small screws. The foam core allowed us to use pretty cheap lightweight servos that helped keep the weight and cost down on the project.

The Code

With the entire thing build, wired, and mounted, all that was left was getting the code to run. Like much of everything else on the project, the code is sloppy and could be refactored, but it worked. I’ve included the code below, but the main thing it needed to do was sense the distance in front of it, and if something got closer than a specified threshold (roughly 8 feet), it would trigger the entire animation, which in turn drove each of the 4 servos.

The only tricky part here was I wanted to make sure we could run each of the 4 servos completely independent of each other, which meant I couldn’t use the typical Arduino delay() function, because it would halt all processes, not just the specific animation or servo. So, without a proper threading library on the Arduino, each process I wanted to run had to keep track of the last time it was run, and an interval in order to know if it was time to run again. All of this allowed it to have the servos run at different intervals and speeds, and allow the animation to start, run for 30 seconds, and then kick off until it was triggered again.

#include <Servo.h>
#include <NewPing.h>

/*************************************** 
 * Distance
 ***************************************/

#define TRIGGER_PIN 9 // Arduino pin tied to trigger pin on the ultrasonic sensor.
#define ECHO_PIN 8 // Arduino pin tied to echo pin on the ultrasonic sensor.
#define MAX_DISTANCE 160 // Maximum distance we want to ping for (in centimeters). Maximum sensor distance is rated at 400-500cm.
NewPing sonar(TRIGGER_PIN, ECHO_PIN, MAX_DISTANCE);
unsigned long distancePreviousMillis = 0;
int distanceInterval = 50;
int distance = 0;
int previousDistance;



unsigned long animationPreviousMillis = 0;
int animationInterval = 30;
bool animationShouldRun = false;

/*************************************** 
 * Arms
 ***************************************/
int armMaxRange = 115;
int armMinRange = 65;

int armInterval = 15;
int armSpeed = 1.5;

// Left
Servo leftArmServo;
int leftArmPos = 90;
bool leftArmDirection = false;
unsigned long leftArmPreviousMillis = 0;

// Right
Servo rightArmServo;
int rightArmPos = 90;
bool rightArmDirection = true;
unsigned long rightArmPreviousMillis = 0;



/*************************************** 
 * Eyes
 ***************************************/

int eyeMaxRange = 179;
int eyeMinRange = 1;
int eyeInterval = 15;
int eyeSpeed = 5;

// Left
Servo leftEyeServo;
int leftEyePos = 90;
bool leftEyeDirection = false;
unsigned long leftEyePreviousMillis = 0;

// Right
Servo rightEyeServo;
int rightEyePos = 90;
bool rightEyeDirection = true;
unsigned long rightEyePreviousMillis = 0;



void setup() {
 // Attach the servos
 leftArmServo.attach(13);
 leftArmServo.write(leftArmPos);

rightArmServo.attach(12);
 rightArmServo.write(rightArmPos);

leftEyeServo.attach(11);
 leftEyeServo.write(leftEyePos);

rightEyeServo.attach(10);
 rightEyeServo.write(rightEyePos);

Serial.begin(9600);
}

void loop() {

if ((unsigned long)(millis() - distancePreviousMillis) >= distanceInterval) {
 distancePreviousMillis = millis();
 previousDistance = distance;
 distance = getDistance();
 int delta = abs(distance - previousDistance);
 Serial.println(distance);

if (distance <= MAX_DISTANCE && distance > 2 && delta < 6) {
 animationShouldRun = true;
 }
 else {
 if ((unsigned long)(millis() - animationPreviousMillis) >= animationInterval * 1000) {
 animationPreviousMillis = millis();
 animationShouldRun = false;
 }
 }
 }

if (animationShouldRun) {
 runAnimation();
 }
}

int getDistance() { 
 return sonar.ping_cm();
}

void runAnimation() {
 // Animate the arm
 animateLArm();
 animateRArm();

// Animate the eye
 animateLEye();
 animateREye();
}

// Animate the arm
void animateLArm() {
 
 if ((unsigned long)(millis() - leftArmPreviousMillis) >= armInterval) {
 leftArmPreviousMillis = millis();

// One direction
 if (leftArmDirection) {
 if (leftArmPos < armMaxRange) {
 leftArmServo.write(leftArmPos);
 leftArmPos += armSpeed;
 }
 else {
 leftArmDirection = !leftArmDirection;
 }
 }

// Other direction
 else {
 if (leftArmPos > armMinRange) {
 leftArmServo.write(leftArmPos);
 leftArmPos -= armSpeed;
 }
 else {
 leftArmDirection = !leftArmDirection;
 }
 }
 
 }
}

void animateRArm() {

if ((unsigned long)(millis() - rightArmPreviousMillis) >= armInterval) {
 rightArmPreviousMillis = millis();

// One direction
 if (rightArmDirection) {
 if (rightArmPos < armMaxRange) {
 rightArmServo.write(rightArmPos);
 rightArmPos += armSpeed;
 }
 else {
 rightArmDirection = !rightArmDirection;
 }
 }

// Other direction
 else {
 if (rightArmPos > armMinRange) {
 rightArmServo.write(rightArmPos);
 rightArmPos -= armSpeed;
 }
 else {
 rightArmDirection = !rightArmDirection;
 }
 } 
 }
}

void animateLEye() {
 if ((unsigned long)(millis() - leftEyePreviousMillis) >= eyeInterval) {
 leftEyePreviousMillis = millis();

// One direction
 if (leftEyeDirection) {
 if (leftEyePos < eyeMaxRange) {
 leftEyeServo.write(leftEyePos);
 leftEyePos += eyeSpeed;
 }
 else {
 leftEyeDirection = !leftEyeDirection;
 }
 }

// Other direction
 else {
 if (leftEyePos > eyeMinRange) {
 leftEyeServo.write(leftEyePos);
 leftEyePos -= eyeSpeed;
 }
 else {
 leftEyeDirection = !leftEyeDirection;
 }
 }
 }
}

void animateREye() {
 if ((unsigned long)(millis() - rightEyePreviousMillis) >= eyeInterval) {
 rightEyePreviousMillis = millis();

// One direction
 if (rightEyeDirection) {
 if (rightEyePos < eyeMaxRange) {
 rightEyeServo.write(rightEyePos);
 rightEyePos += eyeSpeed;
 }
 else {
 rightEyeDirection = !rightEyeDirection;
 }
 }

// Other direction
 else {
 if (rightEyePos > eyeMinRange) {
 rightEyeServo.write(rightEyePos);
 rightEyePos -= eyeSpeed;
 }
 else {
 rightEyeDirection = !rightEyeDirection;
 }
 } 
 }
}

Like I said, messy, but it worked. It was also great to be able to change the timing and how far things moved after it was all assembled via software updates.

From Flat to Physical: 3D Printing a Logo

I recently had the fun opportunity to do something I’d never done, take a 2d logo and turn it into a 3d physical object. I was looking to create something for a friend of mine using my 3d printer for their Christmas present. They were the one that bought the kit for me, so I figured it was only fitting to give them an example as a thank you. I struggled for a while thinking of what to make, but ultimately decided it’d be fun to create the logo for his consulting business and make it into a sign for his home office. A business isn’t official until it has a sign.

Now the logo already existed, and was something we’d created for him at Figoli Quinn & Associates, so I had access to the Illustrator file for the logo. But how to get that to the 3D printer? The 3d printer needs an STL file, not an AI file. Also, how do I visualize the following into a 3d object?

I decided to keep it pretty simple. The entire logo would have a flat backing that could be wall mounted, with the outline of the fish, and then the inside of the fish and the letters all stacking on top of this is 3 or so layers. Basically, a simple three dimensional extrusion of the flat logo. This meant working in a 3d program, but first, I needed to get the curves from the logo into my 3d program which in this case was MODO. While I could remodel the entire thing in 3d, why go through all that trouble when the curves were already built?

MODO actually makes this a pretty easy process. First in Illustrator, you need to expand everything within Illustrator as curves. This includes text, groups, compound curves, etc. Everything needs to be broken down into just basic curves. Once you’d done that, export the file as an EPS file, but make sure it’s a version 3 EPS. I honestly can’t tell you if that’s actually something that’s vital, but I remember hearing before it was, and you know what, it’s worked for me.

With the Illustrator work exported, that EPS file can be imported straight into MODO via the File > Import command. When it’s brought it, the curves will look more or less like they did in Illustrator. You just need to go into item mode, select the curve you’d like to fill, and go to Geometry > Freeze. Make sure that it’s set to fill curves with faces, and if you have holes you need to preserve, check the box to make holes and select the appropriate axis for the holes to be created in. You can do that for all your necessary curves. Once that’s done, you can use the Thicken tool to start thickening these flat faces. For the sign here, I needed to make sure the outline of the fish was one layer, and the inside of the fish and text were different layers that stacked. With those correct, I added a simple beveled box in the back to act as a backing for the pieces. I then added some simple holes in the back corners to act as mounting holes. With that done, here’s what I ended up with:

With the model complete, I then cleaned this up just a little more to make it easier to print. The plan was to print each of these individual pieces separately and assemble them back together afterwards as I wanted to print this larger than my printer could handle. So with that in mind, I made sure each piece was it’s own mesh item in MODO , and then I split the back piece in two because I knew it’d be too big to print in a single piece. With all of those items separated, I just selected them all, right clicked, and exported the layers. Make sure it’s set to export each layer as a separate file, and as an STL file.

That’s all for MODO. With MODO done, the next step was to send it over to the 3d printer software for slicing. Since I didn’t model in MODO at any specific scale, this was my first chance to see how big it’d actually print out. Much too small, but that’s OK. Since everything from MODO was to scale in comparison to itself and the other pieces, I just needed to scale it all up in the slicing software before sending it to the printer. After that, it was just printing each piece individual (well, I could print several at a time).

It’s off the printer, all done, right? Not quite. The first step was coating the “top” side (the side that would be visible when done) of each of the small pieces in a little bit of liquid putty filler. This would help fill in any lines left by the printing process and give it at smoother finish. The back took a little more work since it wasn’t just filling in some lines, but joining two separate large pieces into a single item. First, I ran some Krazy glue down the edge that would be joined and glued the two pieces together. It wasn’t very strong, and there was now a big line down the middle.

Rather than simple model filler, this was solved using all purpose Bondo. Just slather on a side and use a spatula to spread it to the thin finish. Once one side dried, I went ahead and did the same to the other side. While it make things pretty smooth, there were still some rough patches here and there, so it was onto sanding the Bondo down. I started with 80 grit sand paper and eventually worked up to 150 grit until the surface was flat and smooth end to end. I went ahead and did the same to the smaller pieces from before, giving them a nice smooth finish. After that, I coated each of them in some XTC-3D epoxy which helps smooth out their coat and give them a bit of a finish, and once that was dry they all got some fine grain sanding, giving everything a nice smooth matte surface.

Next, painting. Everything except the inside pieces of the fish was either black or white, so they got a quick coat in spray paint. The fish, to give it a nice gradation effect, was airbrushed using the airbrush system I had for my Copic Markers. If I had an acrylic airbrush system I would’ve used that instead, but this was all I had. With all of the pieces painted, it was time to glue them together, and then give them a final coating in varnish to help seal the paint in. Looking back, I should have either used a spray varnish (I painted mine) or just not used the Copic Markers as the ink ran quite a bit during the varnish process. Oh well, you live and learn.

And here’s what we ended up with. All the way from a flat 2d image on the computer to a 3d object ready to hang on the wall using some 3d printing, and a good amount of hand crafting. I hope this helps you see how you can do something similar whether it’s doing a prototype for something that will be make into a larger sign, or just a fun project at home.

DIY Christmas Decorations with a 3D Printer

It’s the holidays again. It’s getting cold (even here in Arizona), Christmas music is playing, and I’ve found it’s a pretty good excuse to use the 3d printer for some DIY christmas decorations.

While we were at IKEA, my wife found a geometric deer model she really liked, but wanted a couple smaller ones to go with it. Sounded like a pretty good way to use the printer, and to save some money on buying more Christmas decorations.

First step, as with a lot of my 3d printing designs, started over in Shapr3D on the iPad. I didn’t do any measurements of the original deer, but just eyeballed it so it looked close. Since they were going to be scaled down anyways, getting an exact match wasn’t super important. For this, the basic plan was to sketch out the basic outline of the deer, and use those curves to extrude the pieces needed.

This is what it looks like with just the curves. There’s a lot of intersections, but it allows for the extrusion of a lot of different pieces out of this same plane.

And here’s what those pieces look like. They’ll be glued after being printed — it’ll make the printing process really easy this way, and it’s easy to glue.

After that, it was exporting and sending over to the printer, Shapr3D handled the entire creation process just fine. I printed these at 0.1mm without any supports or rafts, and after some issues with leveling the bed, they didn’t come out too bad, and were pretty quick to print.

Next, “post processing”… that is, glueing and finishing. The glueing is really straight forwarded. These PLA pieces do fine with basic crafting Krazy glue, and look like this when assembled. One trick — these glue together a lot better if you lightly sand the contact points first.

After that, it’s just finishing the surface. Depending on what the model is, that could mean a number of different things. For this, it’s really just some light sanding followed by painting, and then a light coat of varnish to help hold the paint on. All that’s needed her is just a rough sanding sponge to rough it up a bit and flatten out some of the uneven edges in the printing process. Sand. Inspect. Repeat.

After that comes painting. If I’m doing small work I’ll use a brush but for large uniform surfaces, I find the spray paint works great. My wife wanted it metallic, so this is just using some basic metallic spray paint that was lying around. A couple coats later and after drying, here’s what it looks like (the bigger model this is modeled after is in the background).

Not too shabby. Finally, just a light coat of varnish to hold this together and give it a slight finish. The big model has a pretty dull finish, so I’m using a Satin varnish I picked up on black friday. Dull, but just a small bit of gloss. Honestly, with a model like this it’s pretty subtle as to the varnish effect, but it does help give the paint a little extra protection.

And there you have it, completed and added to the other 3d printed items on the table (I only printed those, didn’t design them). Custom printed decorations that took a little time and filament but were otherwise free to make.

If you’d like to print this, I’ve gone and posted to model here: http://www.thingiverse.com/thing:1930156

Get the Most Out of Your Printer

It’s been a couple weeks now that I’ve atad a fully functioning 3d printer, and other than the weekend I was away on business, I’ve been printing at least every other day. Some of my prints worked well, and some were abysmal failures. Good thing is, they seem to be getting better as I go. I thought I’d put together a quick list on some of the tips I’ve learned over the last couple weeks.

1. Make sure the extruder is the right height off the bed

While there’s probably a good measurement for this, I honestly haven’t found it, and I’ve found an almost intuitive feel about it anyways. Here’s what I do, and it goes hand in hand with leveling the bed anyways. I start by lowering the extruder to 0 on the Z axis, or just basically setting it to it’s home position in one corner of the bed. When doing this, keep an eye on it. If it looks like it’s going to crash into the bed, you should probably stop it.

Next, slide a sheet of paper under the extruder. If it doesn’t fit, you need to move the stop for the z-axis higher off the bed. If it does fit, I’ve found it needs to be so that you can slide the paper around and feel just a small amount of resistance. Not a ton, but there should still be a little there. I know, right? Not super helpful, but it should be pretty easy to move that paper around. Adjust the z-axis stop to make that happen.

2. Level the bed

This step basically makes sure when the extruder is at the 0 z-axis position, it’s actually the same distance from the bed no matter where on the x/y plane it is. I do this very similar to how I described settings the height of the extruder. If you’ve already done that, move the extruder to another corner of the bed and do the same thing, but this time, instead of moving the z-axis stop, adjust the bed leveling for that corner. You’ll probably have a screw or something for adjusting that. If you do, you don’t need to move the screen very much, it’s fine adjustments. Go ahead and do that for each corner of the bed, and you’ll probably find that as you do one corner, another will go slightly out of alignment. Keep going back and forth until they’re all about the same resistance with that paper.

3. Ensure the filament is sticking to the bed

The first step in having a good print is laying a good foundation. If that first layer sucks, you’re going to have a bad print. Now setting the extruder height and leveling the bed are great starts to this. Next, it’ll depend on what sort of 3d printer you have. If you’re lucky enough to have a printer with a heated bed (like I do), consider yourself lucky, they’re awesome. Basically the filament needs something to help it stick to the initial surface (it sticks to itself really well though), and what the heated bed is doing is warming the print surface, which makes it stick better. The specific temperature depends on the material but somewhere in the 55C ballpark is a good place to start. Once the print is done though, you’ll notice it continues to stick rather well, probably too well to the surface. When your print is done, turn off the heated bed and let it cool down. When the glass plate cools down, the print will pop off pretty easily.

If you don’t have a heated bed, there’s a little bit you need to do to help it along. You’ll probably find that your printer needs a little help getting the filament to stick to the bed. While I have a heated bed, one of my cats actually pulled off one of the wires to mine, and I had to be without one for a little while until the soldering iron I ordered showed up. While there’s a couple different methods, the one I found that worked best was using blue painters tape. Before printing, lay down several strips of blue painters tape to cover the print bed (I used the extra wide rolls). Make sure you don’t have edges handing off that’ll get caught on something if your printer’s bed moves. This blue tape surface will help it immensely with laying down those first layers, and when it’s all done, the tape comes cleanly off the bed. However, I did find that it was much tougher getting the print off the surface that it was with the heated bed, and I had a couple prints (large base area) that stuck to the blue tape and I struggled to get the blue off. I ended up painting the model anyways so it wasn’t a big deal, but it was kind of a pain.

4. Use good filament

I severely underestimated how varied the quality of 3d filament could be. When I got the printer, I had some filament that was coming in the mail, but it was taking a while and I was impatient so I went to Fry’s Electronics and picked up a roll of whatever they had on the shelf, and it’s been a struggle. I originally though it was mostly my print settings, and configuration, and spent a lot of time trying to improve those, which probably helped me in the long run anyways.

When my new Hatchbox roll showed up in the mail, it was night and day. Where previously I was having problems with the material extruding at a constant rate and sticking to the bed, this roll from Hatchbox was butter. It went down smooth, seemed consistent, and my prints instantly went from “meh” to “wow that looks pretty nice”. I haven’t experimented too much with different brands, but I will definitely continue to buy all of my filament from Hatchbox. You can find them on Amazon or at http://hatchbox3d.com. Literally one of the best things you can do for your prints.

Repeat after me: If you buy cheap filament, you’re gonna have a bad time!

5. Orient your parts so they print well

This is definitely a try and see what happens, but you’ll pick it up after a while. Basically because your objects exist in 3d, you can print them in a number of different orientations, on it’s back, face, left, etc, and sometimes this really matters. For once, you’ll find that things like circles often print better in the X/Y plane vs using the Z-Axis, so orienting it so circles print that way will result in better prints. A big part deals with printing supports as well. Your slicing software may build in supports to your model for printing, which are basically little pillars or other shapes of filament that’s lightly added on and supposed to be broken off when it’s done. They’re usually added for overhangs, or spots where there needs to be a layer of filament where there isn’t anything to really attach to. The more you can limit supports, the better your print will turn out and the less of a chance you’ll have of supports that don’t cleanly break off. If you can rotate your model to avoid them, definitely do.

Like I said before, this list is far from exhaustive. I also haven’t talked about what happens when things go wrong, like an extruder gets clogged, because I thankfully haven’t had to deal with anything major like that. When I do, I’ll let you know. If you have any questions or any additional tips, I’d love to hear about them in the comments.

Custom Mute Holder Version 2

After using the last version of my custom 3d printed mute holder, it was obvious it was be improved on, mainly by:

  1. Making it just less bulky
  2. Ability to collapse when storing
  3. Print in multiple pieces to allow better printing orientation
  4. The hole was a little too big for my trumpet straight mute to actually rest within it

I liked the overall design, so it was really just tweaking that wasn’t working and leaving alone the parts that were broken. Overall I made the pieces just thinner to lighten it up so that it was lighter and used less than half the material the previous one did. Issues 2 and 3 were both tackled by adding two hinges that allowed the top and bottom of the item to fold in.

After that, it was onto printing and finishing. Splitting it up into multiple pieces allowed me to better position the printing of these pieces, which definitely helped with the quality, and the hinges worked pretty well. In fact, I really only had two problems with this. The first problem was that in the finishing process, I left the pieces outside for a little bit to dry, and they must have gotten some direct sunlight because they ended up warping. Nothing that stopped it form working, but it was definitely noticeable. The other problem was the pins that went into the hinges to keep it together were too big to fit. I’ll go back and make them smaller, but for now, the problem was solved by just straightening paper clips and using those instead.

The paint was hardly dry and I was using it at a jazz band rehearsal. Best part was that no one seemed to really notice and just assumed it was an accessory I bought. Maybe there will be a version 3.0, but I think I’ll leave it as is for the time being.

If you like this project and want to print it out yourself, you can find the files over at http://www.thingiverse.com/thing:1885650.