DrawBot – The Breakdown, Part II

Since my prior cost breakdown post, I’ve invested a little more in this project.  Thankfully, this hasn’t been a project where I’ve been throwing money at the end goal.  Overall, it’s been incredibly cheap with the cost of failure at the most delicate points being quite low.this help in choosing the steppers and power adapter.  Here’s what I bought and from where:

Right now the robot is probably close to its final form and I don’t anticipate any further expenditures.  Someone who has a box of electronic stuff could almost certainly build this robot for next to nothing.  Many other people have spare Arduinos, steppers, stepper controllers, and power supplies.  If I just purchased the bare minimum for this project and used scrap stuff from around the house, I probably could have built this robot even cheaper.

Default Series Title
  1. I couldn’t think of anything else I couldn’t live without!! []

DrawBot – The Assembly, Part VIII

Well, my newly designed motor mounts, while cool, just aren’t working.  If I were committed to using that long pine box, I could probably make a go of them.  I’ve tried several variations on this robot, and nothing has really worked so far.  Here’s what I’ve tried:

  1. In a box, mounted to the wall
    1. This just didn’t work.  The spot I found on the wall was in our robot room, behind my monstrous card catalog.  It was difficult to get to, the laptop I had it hooked up to wasn’t really talking to the robot, and I didn’t even get it fully hooked up in that location before I decided it just wasn’t going to work.
  2. In a box, under an easel
    1. In a fit of inspiration I realized that I had a painting easel, a rather nice one at that.  I took an old new canvas1 , put it on the easel, put the easel’s feet inside the long pine box, ran the monofilament line from the motor, through holes in the pine box, up behind the canvas, through an eyescrew on top of the canvas, and down over the front.  This was terrible.  It creaked, shuddered, and the pen sputtered around the front.
  3. In a box, on an easel
    1. Not ready to give up on Project Easel, I tried to put the pine box on top of the canvas and hold them together in the easel.  This really didn’t work.  Because of the location of the motors in the pine box, the canvas was either going to block the holes for the monofilament line or I’d have to run the line through the top of the box, which would have resulted in the same creaky, shuddering, sputtering problem of using the line through a hole in the box as if it were a pulley.  Either way, the entire assembly just wasn’t stable on the easel which did not want to hold it together.
  4. Bolted to a chunk of wood
    1. I actually have some spare plywood, but it’s been in the garage for a while and has a weathered look to it.  It also has a large 1.5 foot radius quarter circle cut out of it.  I figure if I’m going to keep this inside, I might as well spring for a new piece of wood.  A bit over $14 at the hardware store and I was in business.  Now that I’ve given up on my nifty motor mounts, I designed and printed a dead simple motor mount.  I could have accomplished much the same result by just drilling some holes in a piece of scrap wood. 2  I’ve also abandoned a gondola of my own design. 3
    2. The final result is essentially as follows – a 2’x4’x0.5″ piece of plywood, wrapped in butcher paper, two steppers bolted to mounts and screwed to the left and right sides at the end of the plywood.  The steppers are wired to the stepper motor in the center, and the Arduino and stepper motor controller are in an ill-fitting plastic holder which is duct taped to the front top of the plywood.  The plywood is then propped up against our built-in on which my most useless laptop is sitting, ready to run the robot.
Default Series Title
  1. Old because I’ve had it for about 10 years; new because it’s still in the shrink wrap.  Now, to my credit, I have actually painted on my other canvases – this is just the last one and I completely forgot about it. []
  2. These will probably have been the very very last thing I’ll have ever printed with my black ABS. []
  3. While cool, it just did not work well at all – I made the pen holder so large it would only hold the largest sharpie made by man, it didn’t do a good job of holding the pen against the paper, and it held the weights rather awkwardly. []

DrawBot – The Assembly, Part VII

Yesterday I accomplished a little more with my DrawBot.  Here’s what I did:

  1. Drilled holes in my project box1 for running mounting wires through the box.
  2. Drilled holes in my project box for the power cord and USB cable.
  3. I added several yards of monofilament line to each spool.  The way I have the motors mounted into the box I had to add some filament line to it on the spool.  Obviously, I had no way of knowing how much line I would need, so I just added a few feet.  Today I cut off several additional yards, tied the new section to the shorter section already attached to the spool2 , cut off the little loose ends, and rolled the spool with a finger.
  4. Found studs in the wall, put two huge screws into two studs3 for holding the box on the wall.
  5. Ran some sturdy steel wire through the box for mounting on the wall.
  6. Cut a huge sheet of cardboard out of a box we had in the garage.  I wrapped three edges of it in packing tape, so that it would remain flat and not shed little bits of cardboard paper.  The fourth edge I covered with duct tape for a sturdier hold.  This was then mounted on the wall underneath the pine box using more huge screws going through the side with the duct tape.  The purpose of this is to give the DrawBot a flat surface to draw on, rather than the textured wall.
  7. I then plugged the power cable and USB cables in and tried to fire up the Polargraph software.  Since this is all being done in a different room, I’m connecting all of this to a totally different laptop using different software.  I have an older laptop running Windows Vista that is connected to my Thing-O-Matic and Cupcake.  Now it is attached to my DrawBot as well.
  8. At this point the only physical components of the DrawBot missing are the gondola which I haven’t installed and blank paper.
  9. Unfortunately, that laptop I connected the DrawBot to is having some kind of problem communicating with the Arduino.  But, that’s another problem for another day.
Default Series Title
  1. A long pine box []
  2. Square knot []
  3. True story:  My wife heard the new electronic stud finder going all wacky from the family room.  She called back, “If it’s making that beeping noise, you’re doing something wrong.”  I replied, “Yeah, I’m just running it over my belly just to see what it will do.”  My daughter ran over in disbelief and then called back to my wife, “Mom!  Daddy’s rubbing the finder on his tummy!”  Say what you will, that damn thing is accurate. []

DrawBot – Printed Parts

So far I’ve got three types of printed parts:

  1. Spools
    1. These spools hold the monofilament and are friction fit onto the motor shaft.  You can check out the designs on Thingiverse.
  2. Motor Mounts
    1. The Thingiverse page actually has a lot of information about the motor mounts.  They’re designed in OpenSCAD and are mostly parametric.  Since I’m mounting these motors inside a box, the mounts are designed to go into the corners of the box.
  3. John Abella’s Gondola
    1. I haven’t hooked everything up yet, so I don’t know how well this will work.  I can’t wait to find out!
  4. Arduino Mount
    1. I’m kicking around some ideas for how this would work.  Ideally, I’ll end up designing a bracket that the Arduino and motor shield can just snap into.
Default Series Title

DrawBot – The Assembly, Part VI

Okay!  It was a good day for drawing robots in the MakerBlock household!  Here’s what we got accomplished:

  • I designed some motor mounts, printed one… and had to redesign it.  Gotta love having a 3D printer!
  • I selected a long shallow pine box to hold the motors and electronics.
  • I tried a variety of sound insulating materials.  Bubble wrap was easily the most effective, but least practical.  I settled on some corrugated cardboard.  I placed some cardboard between the motor and the motor mount and between the motor mount and the wood.
  • I soldered some longer wires to the motor leads.
  • I drilled holes in the wooden box, attached some monofilament line to the spools, put the spools on the motors, mounted the motors, ran the leads to the Arduino, powered it up… and it moved around and stuff!

As this point the next step is to mount it on the wall and actually attempt to draw stuff.  Yay!

Default Series Title

DrawBot – The Assembly, Part V

I just finished soldering the Adafruit Motor Shield … this time with all the bits in the correct places.  The Adafruit directions are incredibly detailed, I just soldered a few parts in wrong like an idiot last time. This time it went much smoother and much quicker.

I just wired up the two steppers, plugged in the power adapter, plugged in the USB cable, uploaded the Polargraph firmware, fired up the Polargraph software, and asked it to try to draw something.  And it started to twitch away in what appeared to be a meaningful pattern!!!

Huzzah!

Now… for sleep!

Default Series Title

DrawBot – The Face Palm

Okay.  Confession time.

First, if you have been following along at home, I’m sorry – I have been leading you on a fool’s quest.  I did warn you at the outset I had no idea what the hell I was doing.  Carry on, don’t worry about me, I’ll go down with this ship.

Second, here’s the actual confession.  Despite amazingly detailed directions from Adafruit’s website, I soldered the Adafruit Motor/Stepper/Servo Shield for Arduino Kit v1.0 totally totally wrong.  I soldered one of the L293D chips into the location for the 74HC595N integrated circuit chip.  And I wondered why it didn’t work.

This would completely explain why the shield made a pop noise, why there was the smell of burning, and why one of the L293D chips (which was not an L293D chip AT ALL) was getting super hot while the other was just fine.  I just tried to de-solder the L293D chip, but that’s not going to work.  I’ve de-soldered a part before, but surface mount parts are MUCH easier to de-solder. 1

At this point my choices are, from wisest to most foolish:

  1. Abandon all hope and prayer of building a DrawBot.  Power down my laptop, unplug my router, take the battery out of my phone and stop inflicting my delirium on the world.
  2. Order another Adafruit Motor/Stepper/Servo Shield for Arduino Kit v1.0, and start again. 2
  3. Keep trying to de-solder that damned L293D chip, hope and pray I didn’t destroy both the L293D and the 74HC595N, then try to put them in their proper spots.
  4. Damn the torpedoes!  Full speed ahead!  Who cares that one sad little chip refuses to act like another sad little chip?!  More power and brute force will carry the day.  If I cannot make this work, I will ride this project straight into HELL!  Mu-ah-ha-ha!!!3

I’ve taken a stab at desoldering the L293D, but it just isn’t working.  The solder isn’t really coming out.  Even if it does, I’m going to have one hellishly ugly board by the end of the day – and I’ve probably already burned out the L293D and/or the 74HC595N.

At this point, I think I’ll opt for #2 above.

Default Series Title
  1. Score one for SMT, eh? []
  2. It’s a nice day to start again. []
  3. The world is ours, let’s use it up. []