Category Archives: Clockwork Men

First Batch of CNC Parts

Well, I finally got the gumption to drop the cash and buy the specialty parts for my CNC router, and the first batch arrived yesterday, exactly 3 years to the day that I made a post about the crazy treadle lathe.

I made that lathe shaped object in order to make other things, and I plan to make the El Cheapo CNC machine to make things too.  Hopefully El Cheapo doesn’t end up as scrap in a classroom problem solving lab.


Roller chain, drive sprockets, and idler sprockets.  I am not sure how the cat is supposed to fit in the machine.

So it begins.  I’m hoping that within a few months, I’ll have a working machine.


El Cheapo CNC machine

Howdy to all of the spambots and lost souls out there.  I have been wanting to build a CNC router for a long time, but I have finally decided to pull the trigger.  The machine will have a 4’x2′ work area and a roughly 5’x3′ footprint.  The goal of the build is to make a low cost, versatile, medium format CNC router capable of machining hardwoods.

Originally I wanted to make a 4’x8′ machine, but the numbers just didn’t work out.  Also, given that this is my first attempt at designing and building a CNC from scratch, it just made sense to scale back.  Additionally, I’d really like to build a machine that could handle aluminum, but the materials required to achieve that level of rigidity priced me out of the game.

Most of the CNC builds I’ve come across online have pretty poor documentation, and make very bold claims about the budget used to build the machine.  I can’t tell you how many “under $1000” CNC machines I have found where the maker had magical access to an extra $2000 dollars worth of parts.  I will post step by step instructions as well as a complete bill of materials so that others can learn from my comical mistakes design, and hopefully be able to make a good machine for a low price point.


Here’s a model of the machine without drive components.  Stan Lee is included for scale.  He seems a little small.

Davincibots: The instructable

So, it’s been a while, but I finally got around to some dealywhopping.

Hi, I'm a Davincibot.

I put together an instructable for these little robot toys I’ve had kicking around for a while and posted them to thingiverse.  I’ve decided to call them Davincibots, and I have visions of people making them, coming up with all sorts of cool parts to fit on them, and trading them around.  We’ll see what happens.


I’m not sure what came over me, but I even entered the instructable in a contest.

So vote for me!

Wheeling and Dealywhopping

In the past month or so, I’ve found out that we’re moving, had a job interview, which went pretty well.  On top of all that, I’ve managed to get a good bit of dealywhopping done without destroying anything, which is new for me.

First, I’d like to make this public service announcement, since I haven’t seen it explicitly written anywhere.

The MK8 nozzle fits MK7 extruders.

Mr Squiggles. Man of action.

Which is good, because I wasn’t going to wait around for the MK7 nozzle to get back in stock.  With relatively little difficulty, I managed to reassemble my hot-end and get my printer running more or less properly again, but as usual, there was a slight hitch.  I didn’t have any ceramic tape, so I insulated the thermal core with fiberglass cloth instead.  Unfortunately, the fiberglass I chose doesn’t actually insulate very well, so it took several layers to get it to work without tripping my emergency thermostat, but it looks pretty sweet.


I also finally managed to get the braces shaped and glued to my top, which I am super jazzed about.  Many moons ago, I built this insane glue station for just this purpose.  It’s supposed to be a workbench and a Go bar deck.  I built this thing without having ever seen a go bar deck in action, so I didn’t really know what I was doing, not that I ever do.  I also built it without actually having any go bars thinking that certainly I could come up with something.

Shortly, before it came time to glue the top, I decided that I could make go bars out of a bunch of broken tent poles which I so fortunately had in my hoard.  I cut them, tapped them, and then mounted bolts on them to turn them into expanding clamps.  Pretty smart right?


Not really.  The problem with my expando-bolt style go bars, is that when you twist them to increase clamping force, either the top or the bottom changes position, which you don’t want to happen.  Also, since my deck wasn’t very rigid, every time I adjusted the bars, the loosest one would fall out, so I could only put 3 or 4 bars on at time.  I think with a better deck, my bars would have worked better.

Eventually, I got all of the braces glued down and shaped, now I just need to sand them and glue the top.

Looking more like an instrument.

Pretty good for a wild month.

Next time, try to keep your nozzle clean.

In my previous post, I promised to give the Frankinstrument my full attention, but I needed something to do do while the glue was drying and sort of got carried away.

Every spin is a win.

I've got your number.

My brother and I have been working on an open source boardgame so that anybody with a 3d printer could spit one out and play.  I had hoped to get a full set of pieces printed up so that I could send them to him for his birthday, but disaster struck.  Mr. Squiggles’ extruder began failing mid-print.  First, I cleaned the drive gear, and that didn’t work.   So I decided that the hot end was clogged up and set to taking it apart.

There are at least 3 ways to solve any problem, the right way, the wrong way, and my way.  Typically, there is some overlap between the last two options.  This time, the overlap approached 100 percent.  The first thing i did wrong was ignoring my instinct to put anti-seize on the nozzle when I installed it.  The second mistake I made was assuming that the brass nozzle would be stronger than the plastic, which had invaded the threads.

It turns out that you should heat up your hot end before taking it apart.  I never found anything which explicitly states that, so I will consider this a public service announcement.

You should heat up your hot end before taking it apart.

I probably should have gotten the hint when I broke out the vice grips and started cranking, but I guess I was in the zone or something.  I gave it couple of heave-ho’s, and the nozzle twisted off, leaving the threaded portion firmly lodged in the thermal core.

After cursing my own name and insulting my intelligence for several minutes, I inspected the hot end.  It was definitely clogged, so hopefully that’s my problem.  Otherwise, I guess I’ll have to get a new stepper for my extruder.  Thankfully, the MK7 nozzle is a relatively cheap part, when it’s in stock, which apparently, is never.  I’m proud to say that I resisted the urge to turn this repair into a full on dealywhopper, even though I think I came up with a way to build a new hot end.  Instead, I bought a MK8 nozzle.  Hopefully, it is compatible with my extruder.  We will find out soon.

In any case, I noticed that quite a bit of plastic had pushed its way through the threads of the thermal core.  The way this part is designed, there is a gap between the nozzle and the barrel which provides a place for pressure to build and push molten plastic up through the threads.  I think a better setup would have the nozzle attached directly to the barrel.  When I reinstall the nozzle, I’m either going to put a little anti-seize on it, or possibly some teflon sealing tape.  I’ll mull it over while the part is in the mail.

In the meantime, I will have little excuse for not working on Frankinstrument, which now has mounting brackets and a full set of braces.  My new genius plan for that block seems to have worked well, hopefully as well as my first genius plan, which I forgot.

Hey! Shut down that bracket!

I'm sorry. These are not covered by your dental plan.

Work continues, and distractions self-destruct, next time on Dealywhoppers.

Meet Dr. Scratchy: Clockwork man #2

After I put together Mr. Squiggles, I started examining the economics of 3d printing.  Owning a 3d printer gives you the ability to manufacture things on a small to medium scale, but there are certain things the technology is better suited for.  It’s not great for cranking out thousands of something, and the build area of most printers pretty much rules out large items.  What the 3d printer really excels at is making small batch or one off, specialized parts.  It’s perfect for prototypes, and it’s even more perfect for robots.  So, I decided to start building some robots.  The first robot I decided to have a go at was a polargraph.

An English cat named Sandy Noble has spent a huge amount of time and effort on this drawbot.  He’s got some excellent tutorials on building it, and has put together some impressive software to control it.  The mechanism is simple.  There are two stepper motors attached to some ball chain and a pen holding apparatus called a gondola.  The steppers change the lengths of the chains attached to the gondola and in doing so, slowly drag the pen along the desired path to create a picture.  The process is totally mesmerizing.

This picture looked better in my head.

It was easy to find most of the parts.  The only hitch is that the steppers Noble uses aren’t readily available in the states.  Being new to the world of electronics, and a headstrong idiot, I got some that pulled way too many amps, and would have fried my motorshield.  Thankfully, I had the sense to check the specs, and eventually located a set that would work.  I ended up buying these.  They work, but I think there are probably steppers out there that would be better suited to the task.

You'd best be steppin'.

I printed up some ball chain sprockets using this design from thingiverse.  I took the default sprocket to the hardware store.  They didn’t have a ball chain with the correct pitch and ball size, so I bought the smallest stuff they had and printed up new sprockets that fit.

Oddly enough, I didn’t like the gondola design that actually introduced me to the polargraph, so I designed and printed my own.  It’s a pretty simple affair, actually inspired by the gondola in this video.  I like it because I can quickly and easily pull the pen out of the gondola.  It also holds pens and markers of varying thickness.

Go! Go! Gondola!

Once I had all of the parts, I followed Noble’s instructable and stuck my polargraph to a scrap of MDF I had.  I guess I was feeling greedy, because I was not content with the drawing area of my little test polargraph.  Bigger is better right?

It's small but fierce.

I got a hold of an old cart made to display those pull down maps that you see in schools.  It was on its way to the dumpster, and thankfully I was there to save it.  The cart is a little over 6 feet tall, and about 4 feet wide.  I decided to mount a 4 foot by 4 foot board on it as my drawing surface.  I couldn’t fit a board that big in my bitchin’ grand am, so I joined together two boards to make the the same dimensions.  I did it the same way I joined the Frankenstrument’s top, except instead of tape, I used nylon straps.  That worked surprisingly well.  I glued some scraps perpendicular to the joint on the back to brace it.  I stuck a 2×4 shelf at the bottom on the front, so that the bot could double as an easel, in case I ever learn how to paint.

Pay no attention to the stuff behind the cart.

I'll never join you!

I hung the board on the cart with hose clamps and strapped some shelves to it the same way.  This also worked surprisingly well.

Just add electronics.

The crossbars on the cart are perfect for holding a 3 foot roll of paper, I couldn’t find paper that size locally, so I bought some 2 foot paper and rolled it over the top.  I slapped on some nylon straps to hold the paper taught, and I was ready to draw.

Gaze and be amazed at the monolithic Dr. Scratchy.  I doubt if he’s the biggest polargraph out there, but he’s certainly no wimp.

Hiii everybody! I'm Dr. Scratchy!

I get a huge kick out of running this machine.  It’s not really built for speed or precision drawing, though with some tweaking it can probably be done.  Operating it is really more of a zen exercise, sort of like watching a plant grow.  I really like to drop the speed way down and run it at night.  It makes soothing scratching noises as the pen slowly drags along, and I feel content knowing that it’s busy drawing incredible digital patterns as I sleep.

Dr. Scratchy enjoys mediterranian ferryboat rides.

And appreciates ancient architecture.

Meet Mr. Squiggles: Clockwork Man #1

As usual, I got sidetracked from the Mythical Musical Instrument project, this time by the world of 3d printing.

Ask any kid if they want a robot, and I guarantee they’ll say yes.  Even in ancient Greece, the god Hephaestus had a troop of clockwork men, which helped him out.  Last October I decided it was about damn time that I had a robot.  I had been reading about 3d printers for a while and found out about the reprap project, but eventually decided to cough up the dough and buy a Makerbot Thing-o-matic, rather than source all of those parts myself.

Makerbot has excellent build tutorials, so that’s not what this post is going to be.  I guess it’s somewhat of a review, but mostly just ramblings about how rad it is to have a robot that makes stuff for you.

Sticker shock.

After a few weeks of anticipation, my Makerbot actually arrived about two weeks early, which was exciting.  They require that you send it somewhere it can be signed for, so, I carried it around all day at work like an idiot.  I was actually afraid to open it before I got home and could begin assembly.  When I actually did open it, I suddenly realized what I was in for. Prior to this, I had virtually no experience with electronics.  I had to buy a soldering iron.  Luckily, there’s not actually that much in terms of electrical assembly.



I only have two main criticisms about the Thing-o-matic.  While I think that the electronics(though virtually inaccessible), and most of the mechanics are good, the actual structure is way more complicated than it needs to be.  There are scads of bolts on the blasted thing.  Hardware is expensive, so when I make something I’m always looking for ways to eliminate nuts, bolts, and screws when reasonable.  The design is great if the only tool you own is a laser-cutter, but I think it could be drastically simplified.  Which is one reason I like the looks of the Printrbot, though I am skeptical of how stable it would be.

My other criticism is a little more serious, and was a much greater disappointment for me.  One of the reasons I went with the Thing-o-matic was the automated build platform.  I had visions of my makerbot happily printing thing after thing all day while I was at work.  It turns out that the ABP is not reliable enough to install, but there are upgrades out there which I haven’t had time to fool with.

I’m still glad I went with the makerbot.  They’ve got a good user network that will help you fix and prevent problems that crop up.

Anyway, once you’ve got a 3d printer up and running, the first thing you should print is a filament holding apparatus to automatically feed plastic.  Ironically, my printer failed while making its spool holder because I neglected to feed it plastic.  Thankfully, enough of it printed that it is still functional.

The second thing you should print are x and y belt tensioners.  These will drastically improve your print quality, and make it so that you don’t have to take apart half of the machine to tweak your belt tension.

I can’t get over how awesome it is to have a machine that builds stuff for you.  You can literally build things while you sleep.  You can even build things while you build other things.  Going from idea to object happens in hours instead of days, and best of all, the machine can easily make copies… while you do something else.

Introducing Mr. Squiggles

As I explored Thingiverse, looking for interesting stuff to print, I ran across some parts for a curious machine called a polargraph.  Naturally, I had to build one.