In putting together the X axis of my repstrap machine, I encountered some hurdles. Take a look at a picture of the design I am following. In this picture you can see the X axis has just been bolted down to the frame.
In building my repstrap machine I am hoping to achieve the greatest possible amount of accuracy in all three axis of movement in order to provide the most functionality to the final machine. My approach to this is, in Computer Science terms, a "greedy algorithm." In other words, If I do my absolute best at making each and every part as accurate as possible...I am hoping that the overall machine will turn out more accurate.
On the other hand I am trying to build my repstrap machine on a tight budget. You might say that I am trying to find the sweet spot between accuracy and cost: a tightrope balancing act.
My first challenge was to overcome spending $30 each on aluminum U-Channel for the axial slides. Sure, its nice and smooth, but its ridiculous to spend $30 for six feet of the stuff. At my local hardware store the shelving hardware is on the same isle but opposite the metal stock and so I noticed the polished steel shelving brackets and I think that they are going to be a workable alternative. They cost 84% less for a six foot section than the plain aluminum channel and come with holes pre-drilled at two inch intervals.
I bought one and figured that I'd use it to make the X axis slides with it as a test before buying enough to make all of my axial slides.
When I started working on drilling the holes in the iron piping to mount the slides I quickly realized I would have a problem: how do I get two holes drilled at the same degree of rotation (around the outside of the pipe), 35cm apart? I posted to the reprap forum and while waiting for replies I set out to mount the innermost rail and ignored the outer set of holes.
Then I ran into another problem: how do I get the X axis direction of slide parallel to the back plane of the device? I am not entirely sure that it is necessary, but I know that the X axis slides need to be parallel to each other for the greatest slide freedom and I figure it will be easier to get the second slide parallel to the first if it is itself parallel to the back-plane.
Here you can see the method I used to get the first set of holes together and parallel to the back plane of the device. There is a rubber band wrapped around its legs, which allows me to measure the distance from the back plane (the front of the 90° angles) to the first hole I drilled.
Next up: putting two holes rotationally symmetrical in one section of pipe. On the forum somebody mentioned that I could use a section of my U channel and lay it on the pipe to draw a line down the length of the pipe, parallel to the length of the pipe. I did this exactly, first centering one edge of the U channel on the hole I had already made in the pipe for the first rail. I pulled a line out to where I needed it and punched a hole into the pipe, found its counterpart on the opposite leg and mounted the rail.
The rig I made to help achieve parallel slides.
My first attempt at this failed. My rails were not parallel and over my 40cm span I had about 14mm difference in width (yea, it failed well). I re-measured and found that I must have just mis-marked the pipe. Measure twice, drill once.
Warning:
I should also point out that if you are planning on building one of these with this shelf-bracket U channel, I'd steer clear of a traditional drill bit when putting holes in the bottom of this channel. I shattered a drill bit into ~2mm sections when the bit caught on the slot I was drilling into -- I was lucky not to have damaged anything, including myself. Instead I used a unibit which worked well. Unfortunately it brings up the cost of using this building material. I happened to have one on hand from previous projects, but they cost about $30-70 new.
The rails mounted. I measured the distance between the rails and found them parallel in the neighborhood of 0.1mm, but I don't have the proper tools to measure beyond that. In passing a math-major friend of mine mentioned that I could put a ruler across the two slides at an arbitrary angle and measure the opposite interior angles to verify parallelism.
Update:
I verified the accuracy of my parallel rails using opposite interior angles. Assuming that the rails are parallel, laying a straight edge across them at an arbitrary angle will produce two pair of equal angles. I anchored my straight edge with rubber bands and measured the angles using a speed square. In order to compensate for the inaccuracy of the speed square I just marked a line on the measuring edge where the square crossed the straight edge. Flip the square around and measure the opposite side to verify the angles match.
No Comments Posted.