Just Another RepRap

So I’m without the RepStrap until Thursday or Friday, and have time to post to the blog while waiting for pizza. This blog post will mainly be pics and video.

First, Saturday was successful in that I printed several parts for my Mendel RepRap. Some pictures of success:

z-leadscrew-base-bar-clamp_2off

z-axis-opto-spring_1off. I need to try printing this on its side.

z-opto-bracket_1off

z-bar-top-clamp_4off

There were also less-than-successful prints. Here was one I was particularly sad about, because it was looking so good before it died (and broke the entire RepStrap, yet again). I’d rotated it in Blender first to try printing it on its back:

pinch-wheel-bracket-NEMA17_604-bearing_1off, rotated to print on its back

Here’s a pic of all of the parts I currently have printed for Mendel (printed Friday/Saturday):

Mendel Parts So Far

And another view of that “spring” piece, which didn’t print too well but will probably work:

Video of those pieces, and of some of the recent failures:

Now, a suggestion about unattended printing.. Don’t rely on a Rovio that’s on the floor to see if your 3D printer is merely still moving. Once it’s done, you might see something like this:

And finally, you know you’ve been thinking about RepRap too much when you go to a football game, see this, and think about a RepRap extruder:

Not a RepRap extruder

It’s Printing Day! Currently I’m looking at 5 completed frame-vertex pieces (which each take 1.5 hours to print on the MakerBot I’m using), and I’m currently printing the 6th and last required instance of that piece. I’m hoping to print out enough of the pieces from the first steps of assembly that I’ll actually be able to start assembling part of the frame before long – but I don’t want to get my hopes up.

Here’s a pic of the 5 completes pieces:

Five completed vertex-frame pieces. The one on the left was my first Mendel part. The one on the right hasn't had the raft cleaned off yet; I should be doing that now, but instead I'm blogging.

One thing that I always I’m always fascinated by and enjoy is the list of strange “side-game” routines that you discover when doing a project. When I was designed and built the Lego Halo 3 Foundry Forge Kit (a large Lego model of the video game Halo, that lets you design maps), I purchased pieces directly from LEGO. But when I built my next large kit (and decided to buy my bricks elsewhere), the unexpected side game I discovered was finding where to purchase what parts I needed (412 of the 4×4 blue plates from here, 1240 grey 1×1 bricks from here, etc). I had completely overlooked the detail that would go into acquiring parts. I describe the feeling as if I were an old woman in a supermarket shopping for good cuts of meat, looking for a good bargain, weighing pricing options, etc. This store has these bricks that I need but not those, this other store has more things but charges and arm and a leg for shipping, etc. That same side-game existed for my Mendel parts purchasing as well, but that’s not the new side-game I discovered today.

No, today, I discovered another unexpected side-game in building myself a Mendel: scheduling which parts to print, and when. By my count there are around 106 different parts I need to print out, some taking 2.5 hours to print, others reportedly taking 11 minutes. I’ve calculated that it should take sixty solid hours of printing time to print out these pieces (don’t quote me on the accuracy of that estimate – it’s based on other people’s documents, me throwing them into a simple spreadsheet, etc).

Which to do first? Print out the most time-expensive pieces first and get them out of the way? Well hold on.. Certain pieces make sense to print at certain times of the day, it would seem (at least for me – I’m married and have four kids). If I have a reliable piece that always prints well that takes an hour and a half long to print, maybe I should save that for when it’s time to watch a movie with the family. When I finally get some time that everyone else is either out of the house or asleep, maybe that’s a good time to do a bunch of the 11-minute pieces.

..but then there’s the order you want them in.. I’d love to actually start assembling the frame, if only to give me a reason to bother cutting down the threaded rod to size, etc. But that requires certain pieces. So I’d want to print those first, right?

That’s the game going through my head – coming up with the best strategy I can. Add to that the fact that I’m away tomorrow, so I’m handing the RepStrap (MakerBot) to Chris so he can have it for a week, and I now have a time crunch for today! Given a set amount of time today to print, which parts make the most sense, etc?

I decided to print the 6th frame-vertex piece now (even though I’m using up one of those reliably-prints-over-1.5-hour pieces that would print easily during a movie) so that I can use the 1.5 hours to install more software and learn how to rotate some pieces around before printing, etc (except that I’m using that time to blog now). Oh, and I have David here (who’s been very kind to sleep through most of the past piece).

More pics as the day goes on.

Woohoo!!! After days of anguish over failed prints of various Mendel pieces (and broken extruders, and broken thermistors), I’ve finally printed my first successful piece of my Mendel RepRap (printed on the MakerBot RepStrap, which I just fixed). This piece is “frame-vertex_6off”. I completed it sometime after 4am on December 11, 2009 (my brother Jon’s birthday.. Happy Birthday!). So happy. Awesome.

The build took 1 hour and 28 minutes to complete. Two days ago I’d tightened all of the belt tensioners and turned the Y pot up 1/4 turn, but never got to test it out because of the heater barrel problem. That night I built a new PTFE insulator, screwed it on the heater barrel, and in the process damaged the thermistor, causing the temperature to constantly rise when the machine was turned on. Tonight I replaced the thermistor, taped up the extruder again, and then tried printing a standard test part with no success (the part bent up, causing the nozzle to hit it and knock everything all over the place). So I took off all of the double-sided scotch tape that I’d put on the acrylic bed and replaced it with new double sided tape. That held everything down perfectly. Once the raft was laid out without flaw, I knew I had a chance.

I actually fell asleep a bit of the way through the build, then woke to see it had completed the teardrop holes on the side.

Here are pictures I took throughout the process, showing it progress. Too tired for captions now, maybe later:

The one thing I was a bit surprised by was that after the entirely successful build, the nozzle then plunged into the piece, which left the tiny brown mark you can see in one of the photos. Was this somehow deliberate, like “tie off the last bit of plastic so it doesn’t unravel?”. I can’t remember if I stopped it or if it stopped on its own. But anyway, the piece is great!

Oh, and just one more time… Happy Birthday, Brother Jonathan!

Something’s wrong here, and I’m not sure what it is. There are a bunch of things it could be, and I’m frustrated.

Tonight, again, I watched the MakerBot plastruder push the heater barrel right out of another PTFE insulator.. First it had extruded a small amount, then the entire barrel pushed out. I should create a gallery of PTFE insulators..

I was just ready to print out the test piece that I’d read about for months. I had just tightened up all of the belts, and I was convinced everything was going to work great.

Unfortunately, the barrel is now caked in this seemingly unremovable glaze that’s practically filled the grooves to the point where it seems smooth, which is why it’s not getting a grip anymore. Actually, I don’t know that. I suspect the thermistor is giving inaccurate results, but I’m not sure which way. When I saw it push out, I’d told it to be at 220C. If it was actually colder than that, the plastic would have been tougher to push through the nozzle so the motor would have pushed the heater barrel out (I DID see it extrude some through the barrel, so at least some of it melted). Or it could be way hotter than that, and have affected the PTFE. I don’t know.

I’m burnt for a while. This is extremely frustrating. The extruder needs to be strong as steel, and there needs to be some ability to detect when we’re pushing waaay harder than we should need to, so firmware can stop the motor before it tears the extruder apart. I’m tempted to use some of my Mendel supply of brass rod and PTFE rod to create another extruder from scratch (and use kaptan tape instead of fire cement, since that’s what the MakerBot’s extruder uses), but then I also have to use more nicrome wire, which I don’t have too much of. Agh.

Frustrated. Hulk mad. RepStrap no work.

[Update: Before giving up for the night, I clamped the existing heater barrel down in a vice, and successfully ran an M6 die from my new metric tap & die kit along the barrel, completely cutting away all of that awful glaze. Later I plan to cut off another piece of PTFE, drill it on both sides, use the M6 tap, and rebuild the insulator again. I do not plan to turn anything on though until I can find our meat thermometer, borrow one from a friend, or buy one at the store, so I can see what temperature the barrel actually is when I tell it to go to 220C.]

After printing the K block (see my previous post) with the RepStrap, I immediately went ahead attempting to print my first Mendel part. I ambitiously chose the pinch-wheel-bracket-NEMA17_604-bearing_1off.stl file, which is the green part of the following picture:

The green part is what I need to print

I’ve had varying amounts of success and failure. I have several aborted attempts. Since these were aborted, most of these are just the bottom part of the above piece:

You leave the thing alone for one minute...

Multiple levels of failure

Almost good enough!

This was a very fast print, but it almost wrecked the extruder

The bottom of one of the failures before clearing away the raft

The one on the right is closest. I took a spare ruined PTFE rod and stuck it in the piece on the left (and then fed a piece of ABS filament through it) for illustration purposes.

So, the above one on the right is the closest I’ve come to a functional piece. Worst case, I’d drill through the top, sand off the rough parts, and make it work, but it’d feel better to print it correctly. Clearly I need to do some kind of configuration for this device that I haven’t done (I thought we’d calibrated it but I’ll try again).

Video:

About two weeks ago I almost posted a blog entry titled “%*@$-ing surreal!!”, whose first sentence was “Ok, my head’s about to explode”. I decided to hold off posting it just a bit – here’s a bit of back story behind that, that post, then more..

(Quick glossary for friends reading this that don’t know the terms: a RepRap can print most of its own parts, and is made from parts printed by another RepRap. A RepStrap is a device you make out of whatever you have lying around, that works just well enough to bootstrap you into the system by letting you print out the necessary pieces for a RepRap (so a RepStrap might be made out of wood, or might be a modified existing machine you have, or you could theoretically make one out of Lego bricks, etc)). Darwin is the name of the first version of RepRap, and Mendel is the name of the 2nd version of RepRap.

First of all I’ve been acquiring parts for and building my Mendel RepRap since October 29th, when I threw my hat over the wall and ordered most of the 3rd generation electronics. I really wanted to make a Mendel at this point, not a Darwin, and since there were no Mendel-like RepStraps in the world, I was initially torn – I didn’t want to build a Darwin and then immediately go ahead and build a Mendel (cost and time).

So, I decided I’d throw caution to the wind and just plunge ahead acquiring parts for a Mendel, doing all of the assembly that I could before requiring RepRapped parts, and eventually I’d figure something out. The closest thing I had to an exit strategy was posting a picture of all of the purchased and assembled parts I had, and begging someone who wanted to grow our community to print me parts (with the true promise that I have a bunch of friends lined up behind me waiting for me to print them parts).

While building and not having any real access to RP parts, a rationalization started to build in my mind. Rather than feeling like I should have gone down the longer path of building a RepStrap of my own design (which would be destined to fail, by the way – I’m married, with four children, and it’s amazing I have any time at all for any of this.. my building experience is ok, but not great..), I’d convinced myself that it’d be even better to not build a RepStrap, my rationale being that the quicker I get on the actual RepRap platform, the sooner I’ll be able to apply my creativity and design to that platform, to make things better for everyone (rather than just designing, building, becoming attached to, and then enhancing my own tangent RepStrap design, which I’d only really needed to get on the RepRap platform in the first place). Some people have done amazing innovative work on their RepStrap platforms, don’t get me wrong, but I really wanted to get to a RepRap soon.

On Sunday, November 22nd, I was explaining my new rationalization to my dad and my friend Kevin on the way to the football game we were attending (Go Pats!). After describing my desire/justification to bypass the RepStrap process, I said:

“Don’t get me wrong, if someone dropped a MakerBot on my desk tomorrow, I wouldn’t say no, it’s just that I want to spend my time/energy on getting a Mendel.”.

Now, here’s the post I was going to post on November 24th, two days later:

Title: %*@$-ing surreal!!

Ok, my head’s about to explode.

Today after lunch, two coworkers returned from some talk and said…

…that there are MakerBots in the building, that we can go ask for. And take.

(For anyone following this blog that doesn’t know, a MakerBot is a RepStrap build by a company called MakerBot Industries in New York. It’s like a RepRap kit. It can’t print another one of itself, but it can print a RepRap, which could then print another RepRap).

The parts are sitting on my desk right now. They bought three, for groups to build. It’s part of a technology library they have here, to foster creativity/imagination/etc.

Chris and I picked it up, opened the box, and plan to build as much of it as we can on Saturday.

I still can’t quite believe it. It went from “yeah, no I think you can just walk upstairs and ask for one”, to “no, umm, really?”, to a bunch of discussion, to “why not, let’s go ask”, and then bam – parts all over the desk.

What’s really killing me is three (fully assembled) Stepper Motor Driver v2.3 packages, sitting right here, next to a window on my screen that’s checking the website every 30 minutes to make sure they’re still out of stock on Stepper Motor Driver v2.3 boards. >:(

I fully expect to wake up and wonder how I believed any of this was real.

Two days after I told Kev that I wouldn’t say no if someone dropped a MakerBot on my desk, and someone literally dropped a MakerBot on my desk.

That Saturday, at Chris’s house, from 11:30am until 4:30am, we built the MakerBot that will someday print out parts for many Mendels.

(BruceW, if you’re reading this, keep printing. Larry and potentially I will still need them, as will others in the new england group. I’m not so good at reliably printing parts yet (after many, many attempts). And theoretically we don’t get to keep this MakerBot forever).

This whole thing has been a big (but useful!) change in direction for me. Before I was building. Now I’m printing, with a functional 3D printer. It feels a bit like reading the last chapter of a book before reading the rest. I’m now stuck figuring out what temperature to print what at, how to get something to print without it being destroyed by the printer (or destroying the printer), etc. But I’ll get back to that fire cement soon enough.

So, many of the blog posts after this will reflect my use of a functional RepStrap, and the problems involved in printing. Hopefully it won’t take long before I get enough parts printed (either by myself or from others) that I complete my Mendel and I’m talking about things I printed with my own instance of the Mendel RepRap.

p.s. After a few failed attempts to print whistles and minimugs, here was the first successful print. (I’m now going to save the minimug tradition for when I get my Mendel printing):