Category Archives: troubleshooting

Plutarch’s Debut is Nigh


Plutarch 1.0 is scheduled to make his party debut two days hence at J’s Halloween 2016 bash. The active duration of the party is roughly four hours. Here are my goals for Plutarch 1.0 during that time:

  • he must not fall off my shoulder
  • his head must remain attached to his body

Spontaneous decapitation has been a serious issue for this bird. His good-enough-for-prototyping attachment system might not be ready for primetime. We’ll see.

Whether his electronics work for the frightful fiesta’s duration is almost secondary.

While I celebrate the success of Plutarch 1.0’s ahead of schedule completion, I’m excited to get his successor out of the prototyping stage.

Here’s what Plutarch 2.0 looks like as of Halloween 2016. Keep in mind his deadline is 368 days in the future:


He’s a discombobulated mess. But he’s much improved over his predecessor:

  • got 3D printed bevel gears working
  • added avoicebox with categorized, individually addressable sound calls
  • wrote a Maya-to-Arduino animation translator for the jaw movement
  • wrote a file processor that handles the WTV020SD-16P’s wonky-ass file format
  • amplified Plutarch’s speaker so he can be heard over party din
  • made a 3D printed chassis that’s modular and easy to assemble/disassemble
  • added multiple microphone inputs so that he can turn toward the loudest sound in the room
  • replaced soldered joints with connectors

Plutarch’s jaw opens and closes in sync with his sound calls. He’s got reliable, if shaky, 3-axis head movement. All of these systems mostly work, at least in isolation.

I just have to put everything together into a single, functional organism, at which point I can start working on the final challenge: his animation decision trees. Plutarch 2.0’s almost there and I’ve learned a tremendous amount about electronics and robotics over the last few months.

Unfortunately, I’ve just run into a couple of walls.

too many wires

Too many wires. Not enough space inside the body cavity for Plutarch’s y-axis servo to rotate freely. Voltage mismatch between the Pro Trinket and the voicebox and no room to cram a regulator in there.

Working in this cramped space is difficult (especially with a splinted pinky), and the microcontroller’s going to have to move to a more accessible spot. The 90° connections between Plutarch’s control wires and the microcontroller are eating up way too much servo rotation room.

Complexity could be conserved by tying a couple of grounds together. Connections could be used with higher-gauge wire. Options exist, but we’re gonna need a hogshead of brain juice over the next few weeks to figure all this out. And cash. Cold, hard cash.

It pains me to buy a $13 part to fix a problem I might have forseen but when one considers that $13 is an average morning at Starbucks for some folks, the bad feels wither away. Also I take comfort in the fact that I really have no idea what I’m doing so expectations really shouldn’t be that high in the first place.

A year remains before Plutarch 2.0’s unveiling. Still much to learn.

Make: 2015 Digital Fabrication Shootout

I’ve just returned from Make: Magazine’s 2015 Digital Fabrication Shootout. Huge, huge, HUGE thanks to Make: for granting me the opportunity to participate, and a thousand quatloos to generalissimo Matt Stultz for keeping the cat herd on point during the tests.


We testers have been asked not to reveal or discuss which machines we tested until after Make publishes in November, so I’ll do my best to provide a description of the testing process without letting the cat out of the bag or the horse out of the barn or whatever mammal-preposition-structure metaphor you choose. Onward we go.

Full disclosure: I traveled to the Digital Fabrication Shootout in San Francisco at Make:’s expense. I did not receive any financial compensation from Make for the time I spent at the shootout. Fortunately you don’t have to declare having your mind blown to the IRS, because that totally happened. At least twice.

Most of us left for SF on Thursday the 30th. A weather snarl on the east coast of the US caused many of the New Yorkers/Bostonians/Rhode Islanders among our crew to arrive late, but the midwestern contingent, myself included, got in right on schedule. We agreed to meet up at 10AM to let the late arrivals sleep in a bit. This would be the last time any of us got a full night’s sleep for the rest of the weekend.

By some combination of Ubers and hoofing it, all members of our party arrived at the Maker Media lab, tucked inside the Innovation Hangar next to the Palace of Fine Arts.

Several times I witnessed the death throes of the taxi industry as six nerds waited RIGHT NEXT TO AN IDLING TAXI for our Uber to arrive. I’m convinced the killer app for Uber is the convenience of expense reports; reciepts get emailed to you, you forward them on to your corporate overlords, and bam. Done. Also, no tipping, which is darn handy.

The People

Ehmahgerd. The people. An assortment of keen minds at the tops of their games. Some of these folks I know from Maker Faires past, and the rest are now new pals of mine. Once we went around the circle once doing introductions I realized I was hanging out with the Justice League of 3D printing. That’s me, down at the bottom left.


Those of us with prior CNC/laser cutter experience broke off to work on machines of that ilk– with only 3 days to test over a dozen machines there’s not a lot of time to climb a steep learning curve. I’m a 3D printer guy, so I grabbed the nearest printer, moved it over to an empty bench and started working through the testing procedure. Everything is branded at the Maker Media Lab, even the workbenches. Note to self: brand my workbench at home.


The Process:

The testing process was fairly straightforward; every tester put his or her printer through its paces by running specially designed test prints, one at a time. The test models are designed to isolate one specific aspect of the printer’s capabilities; speed, resolution, etc.

In the interest of impartiality, testers were prevented from using any specialized knowledge/3d printing voodoo to ensure successful prints; we used manufacturer-recommended default settings for and followed manufacturer instructions to the letter.

I can’t show you any of the prints but I can show you these shiny and attractive stickers:


These stickers help to eliminate bias from the scoring process, and here’s how: each print begins with a logging a sticker’s number in a spreadsheet. Other data: the printer, slicer, filament, tester name, temperature, etc wind up associated with that sticker number. One print, one sticker.

Finished prints got labeled with their sticker and dropped into a ziploc bag.

At the end of the day, each tester’s bag is brought anonymously to the scoring room, where it was stored until the prints could be evaluated. The scorer has no idea which tester or machine made the print; as far as they’re concerned each print is just a number.

We had one print labeled 24601 that escaped to the countryside. No idea what’ll happen to it, but I’ll bet it’s miserable now.

All during this process we were ensconced behind a small barrier that kept muggles from getting into a space where safety and liabiity might be an issue. But you could still have come by and thrown peanuts at us.


My Saturday highlight was talking to a gaggle of tourists from Beijing who wandered up to the rope line. Always a treat to see the mental shift in a native speaker’s face when they realize I’m trying to speak broken Mandarin at them.

The testing period was intense; Make: thoughtfully brought in catered eats which were consumed while hunched over laptops and printers, working, working, working. I think I saw the sun maybe twice the whole weekend– day one went until just before midnight and day two went even longer.

This is not to say there wasn’t the usual horseplay and shenanigans you’d expect from twelve geeks in a lab full of technology; far from it! There’s some downtime while print tests are running, so we used that time to document our tests, update social media (check #digfabshootout on Twitter, lotsa good pictures there), and, most importantly, chat with each other.

In between the prints and the coffee and the prints and the Red Bull and the prints there was a lot– I dare say a metric butt-ton of– 3dprinterati cross-pollination. Meeting up with this crew and sharing ideas has my mind going in all kinds of new and different directions now. Also, I gotta up my 3d printing game.

#staytuned, my friends. New stuff on the way soon, once I rest up a bit.

Clearing a Filament Jam on a Type A Series 1

single bolt

Neighbors take out the trash at 3AM, gregarious dogs hump your leg, and 3D printers jam. These are unfortunate but unavoidable facts of life.

I’ve invested in nighttime earplugs and my leg’s been celibate lately, but the other day the first ever maintenance issue with my shiny new Series 1 cropped up. Filament jam. Booooo.

The support section of the Type A Machines website is unfortunately silent on the subject of clearing a clogged nozzle. Scouring forum threads is seldom productive for me, so I grabbed a hex wrench and dove right in on the off chance that I might learn something without breaking something expensive.

The first thing I’ll do is check for obvious problems. Is there a blob of plastic blocking the nozzle? No. We’re good. Just to be on the safe side I’ll insert a pin into the nozzle’s opening and wiggle it around a little.

What about the hobbed gear? There is some powdered filament on there that can reduce the gear’s ability to grab the filament and push it into the extruder. I keep a small, stiff paintbrush around the 3D printers for just such an eventuality. Type A has made the gear very accessible, so it’s easy to clean. Ten points to Ravenclaw.


Neither of these easy fixes got me up and running again, so the next step is to take the extruder apart and see what’s jamming things up inside.

The extruder comes apart with a few turns of a single hex bolt in the middle of the extruder; ten more points to Ravenclaw for keeping things straightforward.

extruder open

Things are so simple here that any blockage should be obvious. The gray schmutz is probably thermal paste but definitely not melted plastic, and it’s not messing with the printer’s feed tube anyways.

This indicates that the clog is further down in the needle assembly. There’s really only one way to handle this short of replacing the needle and that’s to manually force whatever’s stuck in there through.

The extruder, unaware that it’s been vivisected, will happily heat up as normal with a few clicks from Octoprint. Set the target temperature to 300° and then use a long piece of thick wire to push the molten goop through the nozzle.

You’ll want a pair of pliers to hold that extruder block while you do this. A prehensile tail will be handy if you’d like to photograph the process and put it on your blog.

(My tail’s just a crusty little degenerate twin so no photos for you.)

unforeseen problem Big old warning: if you push too hard, you can accidentally force the needle through the extruder block when you do this. Turns out the needles are designed to be removed, but I didn’t know this when I began the process.

So, gentle pressure it is, with the extruder block held in a pair of pliers. You’ll get some zit-popping levels of satisfaction when the hot plastic plug finally bursts free through the nozzle.

Just to make sure there’s a free flow of filament, keep the needle at 300° and use a second pair of pliers to push a length of narrow gauge wire all the way through the needle and then work it back and forth. You’d be surprised what kind of crud you can floss out of one of these.


Reassemble the extruder (one hex bolt. ONE!) and you’re back in business.

back in business

Curiasser and curiasser

Faire Play backer rewards are all I’m thinking about these days. The Athena Makeover Kits are done and should be winging their way towards my midrange backers shortly, and now I’m printing a suit of field plate armor for one of my top backers. My printer is working on a lovely pair of sabatons in the background as I type this.

Nothing’s ever completely straightforward at this level of 3D printing, and of course now I’ve got a strange little gremlin that didn’t surface the last time I printed a cuirass: I’m getting a small line up the side of the print where my printer’s print head stops, lifts, and begins a new layer. Take a look at the cuirass on the left.

cuirass lines

I’m doing my best to make this suit as nice as possible and I’m aesthetically offended by artifacts of the printing process like this one.

What’s wrong here? The geometry hasn’t changed since the first print, the filament’s exactly the same, and as far as I know I’m using the same slicing algorithm. I have replaced a worn-out print nozzle, but I can’t imagine that’d be the source of the issue. I’ve changed every setting I can think of and I’m still getting this weird little printing track.

Truth be told, it’s a fairly minor aesthetic problem in the grand scheme of things and now that I look at the original armor I do see some traces of this artifact there, too.

Even if it can’t be completely eliminated, it can at least be put on the back side where it’s less visible. The best way to do this is to just rotate the model 180° around the Z axis before printing. Easy peasy: take a gander at the cuirass on the right.

Troubleshooting a Shapeways Print

April Fools’ Day (愚人节) has come and gone and, to the Internet’s credit, most folks weren’t taken in by my last post, Printing With Steel on the PrintrBot Simple. Good on you, clever people.

But now the blog post just lurks there, context-free and malignant, a coldly crystalizing piece of unexploded Internet ordinance, waiting to disembowel the unwary traveler who pays no heed to timestamps.

Also! Faire Play’s funding period is winding down and only a scant five days or so remain before pre-production on Barbie’s parade armor is set to begin.

Shapeways managed to deliver the first metal proof of one of Faire Play’s backer rewards ahead of schedule! I was expecting this steel Aegis Pendant to arrive after the funding period, but we got a little lucky this week.

Here it is, printed in polished gray. I’m impressed with how well Shapeways was able to reproduce the meandros motif on the shield’s face.


I’m a little less enthused about the stepping patterns on the gorgon’s face, but I think I know how to fix it. First, a simplified explanation of how 3D printers do their thing.

Hobby 3D printers and Shapeways printers’ work on basically similar principles. First, 3D models are digitally sliced up into multiple layers from bottom to top. A print head traces a pattern on an XY plane for each layer, either putting down a thin bead of hot plastic or sintering tiny particles of steel powder into a solid form.

Once the XY pattern for the first layer is completed, the printer moves up one layer and repeats the process.

unforeseen problem The moiré patterns on the gorgon’s face are an artifact of the printing process. If you look carefully you’ll see the divisions between layers, like elevation lines on a topographic map.


Rotating the model 90° on its X axis might mitigate these artifacts. I’m fairly confident that the face of the pendant will be smoother if Shapeways’ printer prints the pendant as if it were balanced on its edge rather than laying on its back. This kind of edge-on printing isn’t possible with a hobbyist 3D printer, but shouldn’t be any problem when the developing print is suspended in Shapeways’ matrix of steel particles.

I’ll check with the gurus at Shapeways and see if this is possible, or if they’ve got some kind of auto-orientation function in their printing software that I can’t do anything about. #staytuned

Plunger Fail

Lately there’s been a proliferation of 3d printing failure blog posts percolating around the twitterverse. I’ve got plenty of these on the blog already, so I’m dropping another one in the spirit of mutual commiseration.

Plunger Fail

I’ve been printing dozens of bloxen in preparation for Maker Faire NYC lately, and finally managed to wear out one of my delrin plungers. I had two or three days of intermittently failing prints before I figured out the problem.

I printed a new one, but printing a replacement plunger while using a faulty plunger leads to interesting results.

Ultimately I held my filament in place manually while the replacement plunger printed. It’s a quick print, about five minutes, so it was a good time to put a little dent in my Netflix queue.

What a difference a year makes.

I’ve been printing with my Replicator1 for just about a year and a half now. Here’s what my Penny Catapult prints looked like when I started out. (This print’s a veteran of many Seej battles.)

And here’s what my prints look like after countless hours of learning and frustration and failure and learning.

That cross-grain seam in the catapult’s side arm is probably caused by two pieces of blue painters’ tape butting up against one another. I do all my PLA printing on painters’ tape whether I’m using my Rep1 or Printrbot Simple.

One major difference between these two prints is that the top is ABS and the bottom is PLA– after a year of working with both I’d have to say I definitely prefer PLA. It smells better and heats up more quickly, which saves precious minutes of printer warming when repeatedly iterating through a design.

You can get a mirror-finish base with ABS, but apart from that I can’t see a reason to bother with it. Most of my stuff doesn’t wind up in high-stress situations so the added strength isn’t much of a draw for me.