They worked but were rather ugly, so I decided to design a new bracket that would mount inside of the instrument panel and behind the ScanGaugeII so that it would be hidden. I designed the first version of the brackets using Google SketchUp, but I found it too limiting when I wanted to design the new bracket, so I decided to try Blender instead. Blender is an excellent app for this, but it is extremely complex and not intuitive.. even the layout of the application's interface has little resemblance to any other app I've ever used. Fortunately there are tutorial videos online and I was able to learn enough in a couple of days to design my bracket. (I think I'll continue learning to use Blender.. it's interesting.)
The orange outlined object is the bracket, well, more of a block really.. the black outlines are a very rough model of the ScanGaugeII. Once I was happy with the model I exported it to an .stl file and then ran ReplicatorG, which is the app used to connect to and control the Cupcake. In which I opened the .stl file.. it interpreted each Blender "Unit" as 1 mm.
Then you use "Generate GCode" to create the instructions that are sent to the Cupcake to print the part. To generate the code the 3D model is sliced up vertically, like a loaf of bread sitting on one end. Then for each slice it generates X and Y co-ordinates that form a continuous path that the tip of the extruder must move along while the extruder pumps out a thread of heated plastic. The GCode also contains controls for turning on the extruder's heater, for turning on the heated build platform and also for printing out the intial "raft" of plastic which forms a stabilizing base onto which the object is printed. If you choose to print a raft that is.. it's not required, but I've never had success printing without one.
Here's what a bit of the GCode looks like.
Once that has been generated it's ready to print.. nearly. The Cupcake needs to be physically set up first, which involves moving the print platform so that it's centered under the extruder tip, thus zeroing the X and Y axises, and moving the extruder tip down until it's just touching the surface of the build platform. I slip a piece of paper between the tip and platform and move the extruder down until I can just barely pull the paper out from between them. Doing this zeros the Z axis.
Then it's click "Build" and first it will move the extruder up 15 mm and then pause while the extruder and heated platform warm up. It'll then squirt plastic from the extruder for a few seconds. Once you clear that out and the heated platform is hot enough (the red LED switches off) it's ready to print. In the early stages of the print it's best to keep an eye on it and be ready to click "Stop" if something starts to go awry.
The following video is a quick overview of my Cupcake. It's more-or-less in my bedroom closet, with a coil of plastic hanging from a hook above it. At this stage it's just finishing up printing the raft and starting to print the bottom of the bracket. I tried to show the filter fan behind the Cupcake, but it's too dark to really see it. It's a fan with an activated charcoal filter, which helps cut down on the melted plastic smell.
You can see it a little more clearly in this photo. It's intended to be used while soldering, to get rid of the flux fumes.
In this next video it has printed a few more slices and you can, hopefully, see how it prints the interior of the object as a grid of lines, leaving a lot of empty space between them. This is part of the GCode creation. It means that any thick parts printed this way will have a solid skin on the outside, but be mostly hollow space inside. This makes for a lighter part and saves a lot of plastic and time. However it could cause problems if you're expecting to drill holes through the part later on or otherwise cut into it.
This video is further along in the printing process. You can see that the sides of the part are slowly angling inward, so that the top of the part will be smaller than the base. This isn't a problem, but doing the opposite, printing an overhang, is difficult. I've read that you can print up to a 45 degree angle, but I've not tried it and I suspect you might run into stability problems for any big over hangs. I've also seen a few experiments with printing open cell "foam" support structures, similar to the hollow insides of a part, which are removed afterwards, but I have not tried that yet.
In the next video it's printing the upper face of the part, sealing over all of those open cells. Once it's done the extruder rises up out of the way and it and the heated build platform start cooling off. I leave things to cool off for a few minutes, with the fan running, before removing the part. The plastic is still quite hot.
And here's the printed part. It took about half an hour to print. You can clearly see the raft here. After printing I cut the raft off and sand down any lumps in the surface of the object. There are many different ways of finishing the surface of printed parts; coating them with a mix of glue and melted plastic, painting, electro plating, etc. Since I'm using ABS plastic I can use the same glue that is used for ABS water pipes to stick printed parts together. There's also a variety of colours of plastic, though you can only use one colour at a time, at least for now.
Here's the part after finished, stuck to the back of my ScanGaugeII using double sided tape, with another strip along the top to stick it into my car's instrument panel.
And here it is in my car. It does slightly cover the speedometer, but I have the ScanGaugeII display my speed along with my fuel efficiency. The car's speedo is 10% "optimistic" about how fast the car is going.. if it reads 100 kph I'm actually only doing 90 kph. -.-
I'll be building at least one more version. I'd like to make one that uses less plastic and that the ScanGaugeII attaches to using velcro instead of tape. It's a real pain getting the cable to plug into the back of the gauge the way it is now, and very difficult to adjust the way it's sitting.
