The Printrbot has been a blast already, but the adventure has only begun.

This week marked a turning point in my 3D printer project: on Monday I switched from tinkering with getting it assembled, to tinkering with getting the assembly to print! Here's my first test print, to the left of the second:

As with the assembly, where I needed to wait for parts and wait for debugging inspiration, printing requires quite a bit of patience. There are a bazillion settings and tunings for the whole process, but the most mysterious and the most critical part of it all is getting the first layer right. The first pass of hot filament really likes to peel and scatter, and if it gets its way, the print is almost inevitably ruined.

The blue stuff on the glass is just painters masking tape. It provides something of a rough enough surface for the hot filament to stick to, while still being smooth enough to let the part loose when it's finished. It's certainly not foolproof though.

I'm interested to try ABS juice or sugar water at some point, but some Kapton tape arrived yesterday first. It's not particularly cheap, but I found a good enough deal after my first long evening of failing prints to give it a shot. Kapton tape is rumored to provide a significantly better surface than painters tape; I'll find out soon enough!

Meanwhile, even just the painters tape was good enough to get started after some practice this week. The main secret (as I suspect is the case on any surface) is just getting the initial z-axis height right. This means not only getting the bridge ends level and the z-stop adjusted every time the machine warms up, but also being ready to tweak the height by manually turning the threaded rods as the skirt prints. (The skirt is just for this very purpose: a practice round to make sure the filament is flowing smoothly and sticking well.) Don't be afraid to have a very squished/flat first layer if that's what it takes to have it stick. Certainly not ideal, as it requires more clean up on some parts, but better than the alternative.

Next step is getting everything fine tuned, printing at high enough resolution to yield some reasonable quality spare parts for the printer itself. At that point most of the Thingiverse opens up and imagination becomes the primary limiting factor.

I had a few troubles assembling the Printrbot+, but overall it was a very enjoyable process! Imagine a robotic power tool that you get to build yourself out of a lovely wooden Erector Set — in the end the challenges below were all just part of the fun.

Base assembly

It took a couple tries to get the base assembly together "right side up" so that the USB and SD slots on the electronics board lined up with the right slots in the panel. Just as when woodworking you "measure twice, cut once" the solution to this is simply to "test fit twice, carpal tunnel once": not only do you need to make sure the arrows are all pointing to the front, you need to make sure the whole base isn't also inside out by making sure the Printrboard will work underneath before you put in too many screws.

Missing parts

Brook and Co include a handful of extra fasteners so for the most part I was able to make do.

The most major missing parts were the z-axis bearings, which they happily made right. (The original mistake was due to quality issues with a third-party shipper, and they've resolved that too by bringing things back in-house.)

I was also missing a sufficiently long smaller black bolt, and corresponding standoff, for the odd fourth corner hole of the Revision B electronics board. It seemed sturdy enough with only three tie points and so I'm just doing without.

By the time I got to the end, I was running out of the mid-length screw sizes. I had plenty of loooooong ones though and so I was able to hacksaw/file off the excess lengths in each place and finish the job. For the z-axis clamps, even the mid-sized bolts were just slightly too long and would gouge into the wooden bridge as it went past. Rather than trim a few threads off I just added two washers per bolt on the head side of each, so I could tighten the joint properly without the bolt's nether end getting too long.

Broken extruder, beta extruder

The original plastic extruder sent with the earlier kits was badly cracked when I unboxed it and positively broken by the time I'd finished trying to assemble it.

Along with the missing bearings, Printrbot HQ graciously sent me a Wooden Extruder Kit to try. Much much much better! (At this point in the assembly process, I started wondering if I shouldn't have been pining after a laser cutter instead of a plastics printer :-)

I still had two issues with it though: the bolt hob no longer lined up with the filament slot, and furthermore as it spun the nut on the end tended to self-tighten until it had bound the whole mechanism up. I got stuck waiting to hear back from HQ with any ideas, but the fixes turned out to be simple after talking the problem through with a coworking space friend:

The original gear hub had plenty of depth available to get a good start on shifting the bolt hob closer to its target. Even my messy(/dangerous) bit of "hold the part in one hand and the drill in the other" routing got the job done well enough. Once my printing and modeling skills increase I should be able to fix it properly with a custom fitted one!

The self-tightening problem was primarily do to a self-reinforcing nasty feedback loop with the bearings. When tightened down, the inside rolling part would start catching on the wood, which meant the nut couldn't roll along with the bolt as well, which meant it tightened down…. This part would have been easier with a drill press too, but I simply used a knife to carve away a channel for the inner bearing piece. By giving it some empty space, it can continue to roll freely as the static bearing body clamps down.

Finally, switching out the kit's fat metal washers for some sleek "fiber washers" I picked up at Home Depot served two purposes: not only did it stop a lingering self-tightening problem when the bigger washer would bind on the pinch roller part, but it also finished fixing the filament alignment problem.

Note that I also had to take out some of the screws and turn them in from the opposite side instead (versus what the store's pictures of the assembled extruder show), so that they wouldn't interfere while the thin washer lets the gear ride so much closer to the assembly.

Cable management

I still don't really have this solved, but it works I guess.

Seeing all the wooden parts work together so brilliantly, I was a bit surprised when I got to wiring everything up to the board. Cable management seems like a total afterthought — most cables are way too long, some are a bit too short, and there's really not a whole lot provided in the way of guidance holes or tie points.

My general strategy was to ziptie cables together where they would naturally "bundle" For example, there's three separate wires from the extruder that all need to end up at the board. So I lined them up, tugging gently on the ends as I ran my hand towards the extruder, tightening them together until I got to where they needed to go their separate ways. That spot got a zip tie, and then back down towards the board every so often to turn it into one thick cable. Similarly for the bed wires, the x-axis cables sets, etc. Basically, you want one nice bundle in place of wherever there were two or three sets of two or four wires each twisting all over.

Under the bot, the other trick that kind of worked was to make a loop of the "slack part" of extra cabling, so that at the bottom of the loop the rest of the line was a nice slug length. I then ziptied that line at the bottom of the loop, leaving two parts: the "main" cable run in place, plus a somewhat more manageable single loop I could sort of sweep under the rug with all the others.

I still ended up with kind of a mess, especially after adding the power supply which comes with its own nest of cables. I'll probably get the Power Tower to help with that (as well as the other Printrbot nesting material: the filament tangle!) at some point. It's messy to see cables pushing out underneath the stand, but so long as they don't catch on the underside of the bed as the y-axis moves, it's not a particularly urgent problem either.

Gathering the heated bed

It's possible to print directly (well, I assume there'd still be tape on top) onto the included heating plate. However, I was getting pretty proud of what I'd built (well, at least assembled) and wanted to finish it right; according to the discussions online, that meant getting a plate of glass and insulating the flat heated surface from the wooden bed.

I wanted to keep it simple, too, so as not to deviate too far from the Printrbot LC "birch and bolts" aesthetics. I started from this post as well as another encouraging using simple binder clips on the glass, stopped by an Ace Hardware just up the road from Room to Think, and here's what I came up with:

Basically just two needed springs chopped in half, holding the four corners against the screw head above and the nut beneath. I didn't bond the nuts underneath like the forum post — screws not glue™ — a socket wrench will work just fine if I need to adjust more later.

This was actually one of the funnest parts. Doing a bit of research, and then deciding how I wanted to set up my particular printer, and gathering the parts needed to make it happen— it's one feeling to have something "just work" when it's fresh out of a shiny box, it's quite another to realize that picking up some springs and struts at a hardware store will never be out of warranty. With that being said,

Missing instructions

…there are a lot of good resources for getting this thing built — not the least of which is engaging the creative "DIY" side of one's brain! — but going to http://printrbot.com/instructions/ as your starting place is (currently) not one of them. The PDF there is great for getting started after assembly, but (especially for the Printrbot+ specifically) I'd recommend starting with Craig the Fabricator's better-paced assembly videos, moving on to the wiki instructions when those run out, then falling back to your inner maker spirit and the forums as you tie up all the loose ends. At least, that's what seems to have worked for me ;-)

[via]

Locking yourself in a garage until you emerge with a made-from-scratch homebrew CNC exruder a month or two later has been the rite of passage in the original RepRap community. So having to find some better step-by-step assembly videos ALL BY MYSELF was kind of a pansy problem. But I still hope that Brook and Co continue polish up the experience, making it almost (though perhaps intentionally not quite) completely "plug-and-play". The hardware itself is 99% of the way there, basically just plus/minus a few fasteners and a bit better base plate labeling. All it's missing is a booklet that goes from step 1 to step 97 with instructions specific to the pieces in the box, reviewed and user tested to find any lingering pitfalls, and I really think their "your first 3d printer" could be assembled by just about anyone in just a few evenings!

Of course… assembly's just the first part…

I was excited by this morning's announcement of Amazon Glacier.

I've been interested in automatic offsite backup of my photo archive but found the cost for "cloud storage" prohibitive so far. (I'm assuming that attempts to actually take advantage of e.g. Flickr Pro's cheap "unlimited" storage will not end well. Also, any sort of "we'll happily take all your stuff" offer also tends to imply insecure storage of data.) A penny per gigabyte each month is a game changer — an order of magnitude cheaper than anything else I've found so far.

There's a slight catch in that Amazon charges an extra fee for retrieving more than 5% of the total amount you have stored each month. What happens if I'm using it as a backup for all my photos and vidoes, and would like to quickly recover it all after a data–tastrophe? It gets complicated but here's my best estimate for the real-world situation I'm interested in. If I've crunched the numbers right, I like what I'm seeing!

Say I've got 500 GB backed up to cold storage, something goes horribly wrong, and I need to locally restore a complete copy from the cloud. My ISP at home can "burst" around 10 GB/hr but hates my guts if I download more than 250 GB/month. So, the quickest I can expect to have it all home is 2 months, ten times faster than Amazon's "preferred" 20 month free retrieval period. This means I'll be "charged a retrieval fee starting at $0.01 per gigabyte".

Let's work through Amazon's pricing example with my round numbers. For 500GB, my peak daily allowance would be 0.833GB. Assuming I steadily exceed that, downloading an even 8.33GB/day, my peak hourly retrieval rate for the month will be 0.347 GB/hr. This results in a billable peak rate of 0.312GB/hr after subtracting 1/24 of their daily allowance. So my excess retrieval fee would be only $2.25/mo (0.312 GB/hr * 720 hr/mo * $0.01 GB) — less than 10% of an increase on top of the corresponding $24.90 of standard data transfer cost each month!

The numbers get a bit better for 1TB of data (which I'm heading towards). My ISP is the limiting factor; doubling the data at rest actually decreases my retrieval overage fee to an even $2.00 each month. So as my photo archive grows into that ballpark I'd be looking at:

• $12.50 for 250K UPLOAD requests over four months (or around maybe $100 to speed it up using their sneakernet option)
• $122.88 a year for storage
• $122.50 for a complete data recovery ($12.50 for 250K RETRIEVAL requests, $8 in overages and $102 for AWS transfer of 1TB), taking 4 months

Four months seems like a long time, but it's way longer than I've been hobbying upon ShutterStem — which, of course would be a perfect fit for Amazon Glacier! In fact, since all my main ShutterStem apps rely only on the "always present" 512-pixel photo thumbnails, having such cheap storage means that my hard disk archive is almost just a local cache, mitigating Glacier's 4 hour storage latency. Theoretically I could toss all my big drives, run my medium-resolution ShutterStem library off SSD, and just process any full-resolution exporting for prints/uploads overnight. Of course, why would I do this, and what would happen if AWS were to lose/corrupt that copy of my data, but it's fun to consider…

Overall, Glacier is looking like an excellent fit for ShutterStem's design and intended audience. When used as a true just-in-case offsite backup, anticipating catastrophic recovery happening over average US broadband speeds, the pricing is very competitive. The crazy numbers come only if you've got a gigabit connection at home or suddenly feel like slurping your entire Glacier archive into an S3 bucket por el pronto. The fact is that cloud backup is "slow" for mere mortals, and slow is what Glacier wants. I expect to see a number of backup apps spring up around this new offering, and am already pondering what the interface for ShutterStem should look like.

Last week a friend gave me the Printrbot Plus he backed on Kickstarter! I suspect he gave up finding the time for it; the box label says it originally shipped on June 14 :-)

I'm super excited to have a 3D printer. Toby is too; he's eager for me to get done "fixing" it and start printing toys and parts we've found on Thingiverse. [He recently broke his femur and will be in a hip spica cast for a while. He's been doing a lot of Duplo at the table lately. Being able to extend his toy sets in interesting ways could be fun.]

Opening the box and pulling out myriad parts and pieces and fasteners, I was a bit overwhelmed myself. The official instruction page don't provide a very smooth start. (There are three different Printrbot models to sort through, none of the parts diagrams include big-picture context or step-by-step instructions, and the "Building the Printrbot LC and PLUS" videos I tried to start with were sort of confusing and lite on any actual assembly.) It seemed like the official stuff was all aimed at people who had already built other printer kits or who had been closely following the project from day one; I just had a huge box of parts and desperately wanted someone to please say which two pieces I should start with!

What really helped was an (in progress) series of unofficial Printrbot Plus–specific assembly videos by "Craig the Fabricator". The ones that have been uploaded so far got me off from a great start to a point where the official videos became easier to follow. I found a wiki with reasonably good step-by-step assembly instructions that has been indispensable as well.

I'm currently stuck waiting for Printrbot HQ to send some missing bearings for the z-axis as well as (hopefully) some replacement plastic parts for the extruder head that got cracked. When it's done, I'm hoping to get competent enough with it to have it be something of a community resource for Room to Think members and friends. In the meantime, I've got a replacement balloon to keep flying and plenty of work to keep up with.

Last week I launched an early preview of the data that's been rolling out over at Argyle Tiles. There's a good portion of the global coverage already in place, as well as a single-city sample of what I'll be able to provide in most US urban areas. You can check it out via the signup page.

I have been prioritizing client work over this speculative venture, but I've also been outsourcing some of the tiles work to make sure the ball keeps rolling even while I'm meeting my other commitments. The list currently looks something like this:

1. finish generating and clean up some global coverage issues (Finland spot, coastline resolution)
2. start adding basic coverage for the whole Continental US
3. beta launch: provide API keys and documentation for early customers
4. begin processing more popular high-resolution urban areas

I'd love to hear what you think, and would also be grateful for any more leads — if you know developers who could benefit from the service in its beta, please send them over.