Saturday, September 02, 2006


Introduction of another reprap builder

I've been following the reprap project for quite awhile, and started working on my own repstrap type thing a few weeks ago. I don't actually have any part of it working yet, but since I went and mentioned my no-rp extruder in the comments of the previous post, I thought I'd go ahead and show what I do have so far.

First, my take on the Afghan lathe concept. I had a drill guide laying around, so I figured I'd try to incorporate it. It sort of works, but not very well.

Next, the extruder that I've been working on. It's mostly the same parts as the plastic based extruder, but cut out of MDF and some thin plywood. I don't have the electronics done yet, so I don't know if it's capable of actually extruding (I figure it would not be safe to heat up the nichrome wire with no temperature sensor attached). It does at least feed through 3mm rods of polymorph, though not very smoothly: I didn't get the grooves in the brass bolt where it rests on the half washers cut very straight.
Also, there's an extruder video.

Last weekend I started on the frame for the 3-axis part of the reprap. I had already bought a bunch of drawer slides on impulse, thinking that they might make good linear bearings. This is what I've come up with so far:

Other tidbits: I'm going to use Atmel AVR microcontrollers in the electronics (I have parts and dev tools for those already, and access to AVR related advice from an engineer at work). I also plan to see if I can use an optical mouse as a position sensor, though I would not be surprised if that doesn't work out.


Great work, Jim! I just love it! :-D

Keep on documenting with those pics and telling us about how the Atmel programming work is going. The more that's up on the blog, the more inspiration others will have.
Just a detail question. I've looked at those drawer sliders several times myself but never actually tried them in a design like you have.

How are they for movement perpendicular in the horizontal plane to the slide direction? I'd worried that they might rattle from side to side a little too much.
That's pretty cool.

If you can lay your hands on an infra-red thermometer (they cost about £20 from online gadget shops in the UK) it should be OK to fire it up without the electronics. When it gets to about 140 C put about 5v through the motor and it should extrude. You'll find you can control the temperature crudely just by switching the heater current on and off by hand.
On the drawer slides: Good idea. this is what I've done myself. You can reduce any lateral wobble to zero by using 3 slides per axis (rather than two, as you have on the bottom panel). Two slides are mounted such that they oppose each other: when one is fully open, it is supported by another that is fully closed. I used slides with ball bearings and two inner rails, and they appear almost identical to the ones you have in these photos -- they are extremely stable when fully closed, hence, my reason for pairing two together to compensate for the wobble they have when fully extended

To answer Forrest's question, I could not measure any wobble in *any* direction, laterally or vertically, on the piece I created to test the concept. It is very easy to move in the slide direction -- The concern I have right now is there is a 'click' point where one of the inner rails hits the other, and I can feel a diffinite change in force needed to continue motion.. It's not too bad; I'm not sure how much this is going to affect a threaded drive.

One drawback to my 3 slide method is it is a bit expensive: I think I'll be paying about $15 per axis for the 'bearings'... buying the slides online should bring it down to about $10 for the slides per axis... however, if you can afford it, it is one of the easiest and fastest ways to make a solid axis with bearings; It only took me about 30 minutes to assemble the 3 slides to 5 pieces of popalar.

I'll post pictures of my contraption, and a parts list, when I get back to finishing it (I ran out of wood and though I have more now, I hadn't gotten back to building it before the Z8's arrived -- I'm having more fun with them right now.)
*** I'm going to use Atmel AVR microcontrollers in the electronics ***

It was a close tie between the Z8 and the Amtel AVR when I was looking at alternatives to the PIC. You'll have to let us know how it works out. The EZ8 chips have been great so far.
A most impressive running start! It also gave me a pleasant vibe to see the exact same approach to the afghan lathe that I myself took a while back (though, because I had to make my extruder nozzles out of mild steel, it was barely accurate enough to do the job without eating drill bits)

I haven't updated my own weblog in a while, but it's centred almost entirely on my foundry construction at the moment, which isn't directly connected to reprap, except that it can be used to cast feed rods for my proposed eventual experiments with high temperature metal extrusion.

I'd make one suggestion about the drawer runners - be especially careful not to get any swarf, dirt or sawdust into them. They're the very devil to clean out without disassembly, and that's a nightmare too because the geometry makes it incredibly difficult to reinsert all the ball bearings on both sides without dropping them all over the place.
Jim, great stuff! I'm also interested in the AVR MCUs. Good to see you've picked those.
I finally found a codex that would play that little video clip you made FMP4 format. It's wonderful!

We've been talking making a Mk II from scratch for months. You just went ahead and did it!
Thanks for the comments everyone! The feedback motivates me to go get more done on this. :)

About the drawer slides: There is a little bit of flex in oen direction (if used on a drawer, this would be moving side to side), but not much. This is due only to the metal in the slide flexing, and not any looseness in the bearings. That is why I positioned the slides on the top perpendicular to each other, though since the weight-bearing parts of the slide will never go past the supported base it's probably not necessary. If you picture these used in a drawer, I plan to keep all the weight on the back of the drawer, so any flex on the open part wouldn't matter.

I don't have too much in the way of plans or drawings. I find that if I start thinking or planning about something I tend to get bogged down and not finish everything. For instance, when I made the frame with the drawer slides, I spent maybe 2 hours scribbling quick sketches and figuring dimensions on a piece of paper one night, then started cutting and assembling the next morning. Very bad engineering, but at least I accomplish something. I do tend to have to throw away some parts and re-arrange things as I go, but I've only used 1/4 of a 4x8 sheet of MDF so far, so there's lots of raw material left.

About keeping the drawer runners clean: I already learned that the hard way myself. I just washed them out with soap and lots of water, then re-greased them. Not perfect, but they still slide well enough.

On the AVR microcontrollers: I have a few kinds on hand already: ATTiny13, ATTiny2313, ATMega8, ATMega88. I'll probably start with the Mega88. These are 20MHz/20MIPS, 8K flash, 1K ram, 512 bytes EEPROM, 23 I/O lines, 8 of which are 10 bit A/D, and cost only $2.90 each from Futurlec. They can be programmed using C (gcc can target avr). I haven't done any reprap-related electronics yet, but I did get a microcontroller blinking an LED and talking to a serial LCD.
Post a Comment

Links to this post:

Create a Link

<< Home

This page is powered by Blogger. Isn't yours?

Subscribe to
Posts [Atom]