RepStrap.dk

It moves - in 3D

2011-10-13T12:27:00.000-07:00

I now have movement in all three directions at the same time, after some futzing with the Z axis. Here is proof:

The noise in the background is the lasercutter. You can faintly hear the motor noise if you know what to listen for.

TODO:
Find, test, fix and mount opto endstops.
Talk to Dieter about right screw for extruder.
Or go to RepRapSource.de/RepRap-Fab.org and buy a full extruder for ~€75 (without motor).
Mount extruder and go! :)

A new life in a new land

2011-10-03T13:59:00.000-07:00

When last we heard from our intrepid robot, all three stages were almost built. Only the right way to keep the Z-stage from falling was missing, but a move had interrupted further building. Since then, another move has occurred, but things are now calmed down enough that I have been able to join FabLab München, the local maker group.

They have taken me and my robot in and kindly lent me the bits and pieces I hadn't brought over yet. The drivers have been re-tested and now all work. The threaded rods have been oiled (this helped the X-stage quite a bit), and a solution for the dropping Z-stage has been found: Place a ball bearing on a block underneath, so it is supported without getting extra resistance. This has been implemented (with several layers of cardboard:), and the last few kinks of the Z-stage are being worked out, mainly bolts that are too long.

I have decided to not go the purist way and rather have the extruder head printed on one of the other machines (if they work by then). Especially since there's hardly any designs for non-printed extruder heads out there. Maybe, once all the other problems are worked out, and I know a little more, I will go back and make it pure. Maybe.

In the meanwhile, it has so happened that I have inherited a broken MakerBot. Some of the electronics smoked, and we're not sure which. Since the owner has moved on to other robots, he's happy to let me try to fix it and get it running. Sweet!

Update: Nothing has happened

2011-07-02T12:04:00.000-07:00

Since my last post over a year ago (gasp!), nothing much has happened on my RepStrap. I got stuck at trying to mount the Z-axis threaded rod in such a way that it wouldn't just fall down, and then suddenly I got hired by Google and moved to München. Since I haven't had the opportunity to get all my stuff down, no further development has happened.

However, München does (now) have a maker group, FabLab München, of which I am planning to become a member once the last moving woes are over.

Motor testing (with several digressions)

2010-05-24T11:30:00.002-07:00

I'm starting to collect motors and getting more ideas for what to do with them. Most of them I have no idea what the characteristics are, so I'll have to experiment with them. This page and this page gives some useful information on how to test unknown motors. Testing methods include turning the motors at known RPMs and measuring current/voltage and running the motors at known voltage and current while measuring RPMs. To measure RPMs, I would have to build a tachometer. There's an Arduino-based test at Instructables. Preferably one that gives a nice output, e.g. with 7-segment LCDs. I just happen to have a bunch of those lying around, so why not test them all?

This is how my projects typically progress, or rather digress.

On with the testing. A cute way of testing can be found on Instructables, using only components that I already have (yay electronics conspiracy!)...
Or so I thought. Turns out my big mess of resistors did not in fact contain the required one. Now, at least my resistors are sorted so I don't spend an hour every time I start a project finding out that I don't have the resistor (except I haven't labelled them all yet for lack of good label).

Actually, I didn't have the resistors. Instead, I just took a larger resistor onto my breadboard, gave 5 V through it, and tried all combinations until a light went on. From then on, it was easy enough to map out the segments. All but one worked nicely.

Problem now is that it would require 7 pins per display to run it, or a driver (e.g. 4511) to decode and latch them. I don't have enough pins to drive more than one, and I have no 4511. So I can't use that for a display. Instead, I'll just read output on the Arduino, adding a potientometer if needs be.

Fiddling around a bit, I now have a little board with a 220 ohm resistor going to a high-power LED that points at the light detector. There's a 1K ohm resistor on the other side of the reading wire. The analog reading with full light is 800ish, with light blocked it is 550ish. Enough difference that I can use it! That's enough for tonight, it's time to go to bed.

It moves!

2010-04-11T14:40:00.000-07:00

Just got the pieces fitted together for the Z stage for the first time. It moves! It lives! Surprisingly, the stepper motor driver worked right off the bat, unlike the two previous ones, both of which needed some fiddling. Probably the better soldering iron is to blame.

One bolt is about a millimeter too long, I figure I can dremel it down. Bigger problem is that one of the two inner pieces that hold the rail on is so short it can flip to the wrong side, plus the nail is loose. It will have to be redone, and I'm not sure I have a proper piece of acrylics for it. Also am missing two tube flanges. Once that is done, I am ready to mount the first tool!

My tool mounting is with some tie-down strap I got for DKK 20. I have 2x4 long holes they go through (nicely unsharpened with dremel and sandpaper) and they can be used narrowly or widely as befits the tool. It looks primitive, but I think it's good for flexibility and pretty damn solid.

Progress in the 3rd dimension!

2010-03-22T05:21:00.000-07:00

I have declared March to be "get the RepStrap running in 3D" month. With some help from Buro Detours, I have assembled the third stepper motor driver, and subsequently cut and drilled the pieces for the Z stage. It helps a lot to have a good hacksaw and a drill press, this one has gone a lot faster than the first two.

One difference between the Z stage and X/Y stages is that there is a plate under the rails that anything sticking too far down will hit. I had to change the spring-loaded wheel holders a bit to be as flat as possible, including cutting the bolts down to minimum size.

I should have a flange today or tomorrow so as to mount the whole thing together. The next interesting part is what to put on in order to hold a head, and how to avoid the whole movable part falling out when there is weight on it.

My first drawing is complete

2009-09-16T12:27:00.001-07:00

In an act of utter self-glorification, I made the first thing to be written by the 2D repstrap be the first letter of my name (also, it was only orthogonal lines). After some as just unexplained problems with the Y stage refusing to start, and a slightly wobbly fastening of the Sharpie, it indeed made a letter L (in Blocky Outline 83 point or so). I have now successfully programmed and output something with my RepStrap! Woohoo!

The next thing should be a fractal. Just because.

It moves! It lives!

2009-09-13T22:59:00.000-07:00

Having mostly dealt with the problems from last post (Flimsy wiring? Get the right jacks. Overheating driver? attach driver directly to frame. Loose coupling? Tie down coupling with hose tightener (which required dremelling a hole for the screw to rotate through), I have been able to make both X and Y stages move! The whole setup also looks much tidier now.

The stages can move for a seemingly arbitrarily long time, but they still overheat if standing still for about 30 seconds. Have to find a solution for that. I don't know if thermal paste will change much when I've already bolted them tightly together.

Major problem: Both stages have a tendency to switch between running smoothly and getting stuck, depending on exactly how the rod is aligned. I'm guessing the guiding bolt is a little loose, and can change between perfectly aligned with the rod and not quite perfectly aligned.

After cutting off superfluous edges and cutting a notch on the Y stage for the tube tightener screws, I have about 11x16 cm of writeable area. Not a lot, but enough for tests and for a RepRap, should I go that way (requires a plastic extruder, though. First some 2D tests!)

Softwarily, I have refactored the little test program to where I have two functions: move(dir, axis, len) and pauseSeconds(seconds). The move() functions is almost calibrated to move in cm.

For playing around, I am tempted to hook up three three-way switches so each axis can be controlled independently. Or maybe get a mini joystick controller:)

It's alive!

2009-09-10T13:26:00.000-07:00

My mad genius neurons are all a-twitter with excitement! My creation moves! It's alive! (For very small values of alive.)

I've hooked up the new power supply and both of the mounted motors to the stepper drivers, and the Arduino to a solar powered (!) USB power supply. Pins 4 and 5 drive one stepper and 6 and 7 drive another, with the standard test program that merely moves back and forth. And the Y stage moves! Happily, happily back and forth. The X stage, not so lucky, the tubing is not tight enough to move the weight that's on there, so I'm currently glueing it on. Should be ready to run tomorrow evening.

Issue 1: The wiring is flimsy at best. I need some jacks that fit the Arduino, and maybe a little board that I can mount the various things on so they're less random. Just mounting the drivers next to the motors would help.

Issue 2: After 10-30 seconds of power, the stepper drivers start to stutter audibly, whether actually moving anything or not. I suspect the power transistor is getting too hot and starting to flake out. It certainly gets hot. I thought my heat sinks would be enough, but maybe not. Must go to Aarhus Radio Lager and get better.

Issue 3: As I mentioned, the X stage coupling is not tight enough. Not sure what the best way to make it be tight and stay tight is. Maybe just tie a lot of string really tightly. Or tie some string that will shrink when wet.

Working power supply

2009-09-02T13:50:00.000-07:00

Briefly tested the new power supply, and it works with a big resistor on 5V. Lights up the stepper motor driver bicolor led somewhat prettily. Tomorrow: Rewire the stepper motors according to some hint I got from another Sanyo user.

New power supply coming up

2009-09-01T08:21:00.001-07:00

Thanks to my brother's people-wrangling abilities (surprisingly good for someone who's pretty Aspy), I have a new old power supply on the way. The two big ones didn't work out. One of them spins up, but doesn't deliver power no matter what I try to put on for load and shorts, the other one doesn't even make a sound.

2009-07-26T12:54:00.000-07:00

I got a new power supply to replace the one that got broken by jitter (two, actually, but one of them doesn't seem to work). This one is from an IBM PC 330 server that I got for $5 from the local second-hand store. I can tell that it works, as it actually spins up a little even when no power is drawn. This is the label:

However, it's a big and special power supply with connectors I haven't seen described in the usual literature.

This connector P3 looks a bit like the two old AT style connectors rolled into one, but the wiring color is all wrong.
Front look of P3.

There are also two huge connectors P1 and P2 (not connection to bananas) that I have never seen before. Serious mofos with thick wires (standard harddisk connector shown for comparison - look at the size of those things!). Probably can't use these easily, but I feel certain that it can give the output I need. It also has not 4, not 6, but 8 standard harddisk connectors plus assorted small bits.

Here is a better view of the wires for the non-standard 10-pin connector. The white would be -5V, the gray a power on indicator, the black ground, the yellow 12V, the blue -12V, but I don't know what the brown would be, and the green is a power on input. Maybe the brown is 3.3V sense (only brown I've seen mentioned), but that makes no sense with no 3.3V output.

Update: Tried with shorting the green to black, and with a 7W 1 Ohm resistor between a 5V and ground. The power supply hums merrily, but there is nary a blink of power output. Strange.

X and Y mostly done

2009-07-19T14:48:00.001-07:00

Got the last bolts today, so the X and Y stages are pretty much done. The Y rod has been lined up, and the motor attached. When rotating the motor, the driver LEDs light up. Still need to get the endstops to work, and to get a working power supply - how hard can it be? The new server has one that works, so either I can get power out of that, or I have three supplies I need to do something special to.

Some pictures of the X and Y stages

2009-07-10T13:22:00.000-07:00

The Y stage at its current form. Here we see the two main kludges I have made: Gluing together two layers of acrylics, and making the rail bearings from pieces of acrylics with teflon tape around them.

Two of the stepper motors and their drivers.

The top of the X stage. Markings are for aligning the Y stage rails.

The full X stage underside, with motor attached.

The roller skate bearings and threaded rod (underside of X stage). The teflon kludge is even clearer here. I will probably get the right stuff if these don't hold up (unless I switch to magrail:)

The roller skate bearings aligning the X stage. My macro lens appears to front focus?

Improved X stage movement

2009-07-09T13:00:00.000-07:00

I had noticed that the X stage threaded rod was a little off from the motor drive shaf, and it didn't move very easily. To correct this, I moved the nut to a new position, carefully measured against the actual rail when mounted between the roller skate bearings. After adjusted the nut (it can rotate slightly due to a stupid bend in the holder), it is now near-perfectly centered and movement is much smoother. This is one of the few things where accurate measurement is actually important.

Working on Y stage

2009-07-08T10:50:00.000-07:00

As expected, doing the Y stage goes a lot faster, now that I've done the X stage. In less than an hour of work, I cut, sanded, teflonized and mounted the four bearing pieces, countersunk the holes for the roller bearings, and mounted the nut for the thread. Currently missing three 8mm flathead bolts, more rail (one piece has its hole too far towards the side, grumpf), an 8mm drill bit that's not dull as all get-go, and 4 more roller-skate bearings. Improvements in this stage: Rounded the sides of the rail bearing acrylic pieces, hopefully lessening the wear. Snipped the corners of same instead of dremelling them down, much easier. Rollerskate bearing bolt holes are now fully countersunk (not that it's really necessary) due to diligent testing and hand-improvement.

This is going to be the ugliest RepStrap evar:)

Progress, and a setback

2009-07-06T12:37:00.000-07:00

Today I soldered plugs onto the wires from two of the motors, and tested them with two of my boards. The result: Success! The motors turned, at a rate of about .5 Hz or so, both back and forth. Woo-hoo! The first step towards an army of robots!

However, when I tried to set both motors up at the same time (to check power usage and Arduino connections, and to get a picture of them both moving, they started acting erratically. At first, I thought it was the Arduino connections, cause they had a tendency to pop out real easy. However, I soon found out that the switch I mounted on the power supply was the problem, causing glitches and jitter. Finally, one of the switch wires turned out to just have broken off, so I stripped some more to reattach it. But now, the power supply doesn't do anything anymore, not even with just the car bulb attached. I guess all the twitter must have blown the supply -- possibly just a fuse, but I'll have to open the supply for the first time to check.I might want to hardwire the green line and use a switch on the main power.

Lessons learned:

Gotta get some proper plugs for the Arduino connections.
The heat sinks get warm quite quickly, I'll have to keep a good eye on them.
Power supplies are fragile to loose connections.

Motors figured out

2009-07-03T13:16:00.000-07:00

Followed a link from Wikipedia to a page on stepper motor wiring, from which I found the following about my Sanyo Denki 103F771-0611: There are 6 wires, and they are connected with a black center to the pink/pink stripe wires and a white center to the blue/blue stripe wires. So now I should be able to run a test!

Almost got a stage!

2009-07-03T11:58:00.000-07:00

Got the teflon all nice and on. Cut the threaded rod shorter so I don't have to fiddle with the hole in the other end. Put the spring-tightened wheels on (instead of countersinking on the bearing arms, I put an extra nut between them and the plate, allowing the head of the bolt room to be there. Much easier. Don't know what the whole countersinking deal is, except to make it harder). Got some 9mm tubing from Dragefisken for free (thank you!). However, my drive shaft on the motors is 6mm, so I'll have to extend it somehow. Best bet is these little drill stops that happen to go from 6 to 10 mm, with a tightening screw. The alternative is to get a 6mm threaded rod, but that would require finding some 6mm tubing, which Dragefisken does not have. Trying with the drill stops first, even though they're 50kr apiece.

Slick!

2009-06-30T15:35:00.000-07:00

Today I wrapped teflon tape around the sliding blocks on the X stage. I doubt that it'll be as durable as using teflon blocks as such, but it's what I could find. If it becomes a problem, I'll look online for real blocks. It may be a good idea to round the edges on the blocks, I'll see if they tear.

Also marked where the hole in the other end should go. I have cut the threaded rod long enough to go through the other end, which means the rod will not be hanging from the drive nut. I considered adding a drive nut in the other end, but I imagine it could cause problems if they are not perfectly aligned.

There are already some problems with the omega-shaped holder for the drive nut, as it has a groove that the nut can slide back and forth along. Not sure what to do about that. Glue might work, but that should be only on the holder, I want to be able to swap things out.

I'll need to grind down or cover the part of the rod that goes through the hole in the other end, and the fastening method to the motor is uncertain, probable duct tape for a proof of concept.

Good news for Europeans

2009-06-27T06:09:00.000-07:00

There is now a European source for RepRap pieces: http://reprapsource.com/en/shop has kits, PCBs, screws and various other useful things. Finally, we don't have to worry about whether we'll end up paying $30 worth of fees on $20 worth of parts. Yay!

Progress: Stepper driver, X stage, endstops

2009-06-27T06:03:00.000-07:00

With some help from a Norwegian friend, I have made significant progress.

I have the rails for the X frame bolted onto the end pieces.
I have the X stage acrylics mounted with nut and sliders (though I need to wrap some teflon around the sliders).
I have tested and fixed a soldering issue on one of the stepper drivers, it now works for all I can tell.
I have gotten an Ethernet cable that can actually communicate with the opto endstops.
I have tested the opto endstops, and found that they work... on and off. And not in the manner I expected.

I still need some way to connect the threaded rod to the motor, I figure duct tape will do for testing purposes. I am also out of solder.

Useful heads

2009-06-23T07:50:00.000-07:00

Collecting in this post various types of head that might be useful once the 2D part is actually running:

Fabric painter (might be able to get one from the US)
Wood engraving burning tool (seen at local crafts store for DKK ~500)

Some things to make

2009-06-23T06:00:00.000-07:00

While I'm trying to get things done, here's some things I'd like to do once the RepStrap is actually working:

Electronics organizing cabinet thingy.
Resistor lead shaping tool.

My work yesterday was trying to put the frame together. It was foiled by 1) The narrowness of the rails made it really hard to turn the bolts, instead I should enlarge the holes slightly to allow the bolt to be thrust through, and 2) the holder for the moving rod was a different size than the layout used, I'll have to redrill.

Further work

2009-06-21T12:25:00.000-07:00

Todays work:

Epoxied together the X and Y stage acrylics. It's been a lot of unnecessary extra work to grab too thin sheets. Can't recommend it, but I think it'll work.

Drilled the X and Y stage attachments' holes for fixin' to the ends. Using a dull drill bit to drill aluminium is interesting: Rather than cutting, I had to run it at full speed so that it eventually melted the aluminium. Apart from being extra work, it also meant that there was a rim of molten metal around the hole afterwards, which took a fair amount of dremeling to remove. And my dremel is now stuck with a half-used grinding ball on it, as it has shown its crappiness (it's off-brand) and worn itself to where I cannot remove the bit. Bugger.

Tested the one stepper motor driver that seems to work. The Arduino could talk with it, and the exercise program showed that only one side of the motor connectors works.

Success!

2009-06-14T10:07:00.000-07:00

So I swapped out the power supply I hadn't been able to get to work for one my brother had lying around, and voilá -- success! My car bulb lit up, as did one of my stepper motor driver boards. The other board has a short, it lit up briefly and then went out, after which the 12V line was dead until I unplugged the power supply and plugged it back in. However, the 5V and 3.3V lines were fine and under load from the 10 ohm 7W resistors put in -- which got quite hot in the process, as my fingers can attest to after trying to remove one:)

Pictures of the board with the short:

http://www.flickr.com/photos/51741439@N00/3625937412/
http://www.flickr.com/photos/51741439@N00/3625124543/
http://www.flickr.com/photos/51741439@N00/3625122231/

On the down side, I found that the acrylics should be about 1/4" thick (which is 6mm, mine are 3mm). To not waste the drilling I just did, I'll try to glue two pieces together.

Tip for drilling pipes

2009-06-13T08:33:00.001-07:00

I don't have a drill press yet. I want one, but can't afford to get one that I trust to be of OK quality. So drilling into pipes is tricky, the drill moves off the spot even worse than when drilling on flat surfaces. At the last hole (of course:) I hit on the trick of taking the metal saw and making a little X centered on where the holes should be. Voila, no more drill movement.

Imperial vs. metric

2009-06-13T06:14:00.000-07:00

It's kinda annoying that the McWire bot is designed partly in imperial units, partly in metric, and always described in metric. Some parts are in really odd imperial lengths that turn out to be very nice metric sizes, while others are nice in imperial units. And the printouts for cutting the acrylic pieces for the stages? Made for 8.5" x 11" paper, which means it doesn't quite fit on A4. Grumpf. Everybody know that metric has won.

Bits and pieces assembled

2009-06-13T05:25:00.000-07:00

I now have most of the bits and pieces assembled for making my RepStrap (Seedling). I have:

Two finished but untested stepper motor drivers
Three stepper motors
An Arduino
1m 10mm threaded rod (just enough for all three stages!)
Piping galore
Rails and angle pieces, cut and with holes drilled as appropriate
Nuts, bolts and other misc hardware
A 1m x 30cm acrylic sheet (perhaps not thick enough, but it was cheap:)
A PC power supply that refuses to stay on for more than a second:(
Teflon tape (couldn't find teflon pieces, so I'll cover some bits of acrylic with teflon:)

Forwards build!