Sunday, July 31, 2011

First print!

First print...
The 'repstrap' prints... it actually prints! :) I know that's the whole purpose of the device I've been building, but it is still quite exciting and amazing when everything starts working in harmony and it actually comes to life in a productive way, for the first time!

After months of construction, assembly, alignment, calibration and testing, this past week saw my resprap pass a new milestone. I took a set of STL files (Greg Frost's adaptation of Wades Extruder) and turn them into a useful set of three dimensional objects, right in my own home! It may have gone unnoticed, but up until this point, the objects that these printers produce and the components that make them up, had been completely virtual to me, only existing in the photos and videos of the various RepRap related sites I visited. They're not exactly 'down the shops', or it's not like someone else in the neighbourhood has a 3D Printer spitting out interesting bits'n'bobs that I could get to see and touch.

 I've been working in wood and metal exclusively up to this point, and today everything changed! It was like taking delivery of a StarTrek Replicator! Just magic! :)

(Pict showing general overview of my bench as things came toghther in the last few days.)

I'm really just an enthusiastic follower of this concept, but hopeful that the observations I make and obstacles I overcome along the way, and document here may be of some benefit to other like-minded followers. The real credit and thanks goes out to those who have pioneered this concept over the past few years and kindly shared their work on the site and many associated blogs.

The last mile...
The biggest hurdle for me was getting a stable extruder working, and the current design is hanging in there. The next biggest challenge was the software. Things are not exactly at stage where one could expect to double-click some setup file, take all the defaults, and be up and running in 5min! Our expectations in relation to software have changed greatly, and you need to brace yourself when entering the software world that is used to enable these little devises. My chosen path, through much blog reading and good guidance of others, was Skeinforge for STL processing to gcode, and currently RepSnapper for sending the gcode to my Gen6 hardware, and controlling the temperature.

RichRap's blog has good practical tips on getting started with Skeinforge. Also, once you install the software click on the help button. There's a tone of good links from within the package. The number of settings within Skeinforge' is a bit daunting, but if you follow the links on it's Help page you will build a picture. e.g. Help will link you to it's main Wiki:  From there you can find a link to a "Configuring Skeinforge: basic terms" page, or even a list of Skeinforge settings. I found all of these links of some value, but my key point is that they are all anchored back to the help button within Skeinforge.

I'll conclude this post with some very amature video clips of my repstrap printing it's first real objects...


Thanks for viewing!

New build platform

I had reservations about the suitability of my original MDF (6mm) build platform and this past week proved it wasn't up to the job. With my latest extruder incarnation working consistently I was ready to start laying down some plastic, but no matter how much I fine tuned the level there was inevitably some part of the surface that was too high or too low. It proved too much of an obstacle so it had to go! I'm also dubious about the four-point spring loaded levelling approach that's in common use but reworking that is for another day.

I got lucky in my junk collection and dug out a 12mm thick rectangular piece of Perspex (acyclic). What ever it's previous use, it had four domed rubber feed, nicely positioned, one in each corner. They were threaded in place which made them potentially ideal should I need to level the new platform in any direction. I figured it had sufficient weight on it's own to not need much bolting down, so I came up with the following way of positioning it on the Y-buggy.
Above you see the 12mm thick Perspex and it's rubber feet. I've put blue-tape on the top surface.

To seat the new build platform on the Y-buggy I cut a "H" from some MDF and bolted it down (no adjustment). I drilled cone shaped holes using a large countersink bit, in each corner, in positions corresponding to the feet on the new Perspex platform. Then I simply sat the platform feet in to these recesses. It's own weight held it in place. See below photo of my new build platform in place!

Initial alignment testing, by moving the extruder to various locations, showed this new arrangement was much more satisfactory, even without any adjustment of the feet. I tweaked two of the feed a tiny bit and away I went. I laid down a test print with a raft and it took to the surface nice and evenly!

The added bonus is that the platform can be easily removed, and returned, if desired, without any apparent impact on it levelling.

Next post is on how well the printing has in the past week... It has gone well as hinted in the above photo! :)

Thanks for viewing!

Tuesday, July 26, 2011

Extruder Hot-end v2

The previous extruder hot-end served me well as a 'proof of concept'. It showed that a stainless steel down tube, attached to a fan-cooled heatsink, could ensure the filament remained solid and ridged until it entered the hot-end melting chamber, and my single resistor hot-end had the heating capacity required to melt the PLA moving at a steady, if not slow, flow rate. It also allowed me to exercise my cold-end feed mechanism. It worked well initially in a clunky sort of way, but due to it's hand-built nature is soon suffered from repeated stalls. I put this down to the imprecise nature of my hand drilled holes in the SS tube and aluminium nozzle. I felt the introduction of an inner PTFE tube liner would help matters, so over the past few weeks I designed and build a new extruder. Here's some detail:

The stainless down tube is made from a stainless socket head bold. It's head was drilled and tapped with M8 thread, it's shaft drilled out with 6mm hole. The nozzle is an M8 threaded brass rod, tapered cone tip with .4mm hole. The nozzle is drilled out with a 3mm hole to within about 2mm from the tip. The back end of the nozzle has a cone lead-in to the 3mm hole, visible in the next picture.
(A big thanks goes out to a friend for drilling out these components.)
PTFE tube is from   I put the PTFE in the freezer for 10min and tapered the tip with a pencil sharpener until it matched the cone in the back end of the brass nozzle. The heater block is aluminium and unchanged from the previous extruder. I like the heat conductivity capability of the aluminium so am sticking with it for now. Many other designs use brass heater blocks.
I assembled the unit by first threading the brass into the stainless steel tube. There's about 8mm of thread in the SS. The brass was wrapped with plumbers PTFE tape and tightened home into the SS tube. I filed two flats on the brass nozzle to get a good grip. It's a good hand-tight. I had to widen the hole in the heatsink and clamp block to 8mm to receive my new SS tube. I inserted the ss tube into the heatsink as far as it would go and tightened the clamping block up (clamp nuts visible in picture above).

The last assembly step was to thread the heater block onto the brass nozzle. I fitted a Teflon sheet between the heater block and the ss shaft to act as a heat shield. This is off-the-shelf baking tray protector (rated to 260deg C). It also provided a nice air duct for the little fan which I moved to the face of the heatsink, keeping the cold air away from the heater block. The fan is held on with a dab of hot-melt. In testing, the heatsink remained cold to touch at all times once the fan was running.

Detail note: Initial testing of this new design showed that some method was needed to ensure the PTFE tube was held firmly against the cone seat at the top of the brass nozzle. Initially the PTFE tube moved upwards within the SS tube and plastic leaked between the two, jamming the hot-end. I have addressed this issue by chilling the PTFE, inserting it and trimming it flush with the top of the SS tube, and then placing a small washer to cap the PTFE flush with the SS tube. The wooden block was then slid on over the SS tube to hold the washer firmly in place.
Here you see the wooden block that sits on top of the SS tube, (washer not in position). This wooden block pushes down of the washer that caps off the SS tube and is bolted in place with two bolts through the aluminium heatsink bulkhead.
I've had limited printing success with this new extruder, as you can see from the scrappy objects in the photo above. I do need to step back and get the extruder calibrated properly. On the other hand my x and y axes is square and general dimensional accuracy is good. The 40mm square is 40mm! I also need to learn a whole lot more about GCode. The above tests were printed by just downloading the STL files, converting to GCode in RepSnapper and hitting Print!

I'm reading a lot about people using Skeinforge instead of RepSnapper. Should I get into Skeinforge at this point? I don't know... opinions welcome.

Thanks for viewing!