With the help of Anthony Gallia I designed the first prototype. Together, with the help of some experts we constructed a virtual model in Pro Engineer. Running VirtualBox on my Mac with some shareware called SolidWorks I can view the assembly. It looks like this:
So, after a few weeks, all the parts that were designed in CAD were milled. This gave way to the first prototype, which looked like this:
So, here, the pillow blocks that the pen slides on is made out of Delrin, which is a plastic. The whole plotter is mounted on a sheet of aluminium.
The belts used are designed to be used on model helicopters. At first I was going to use belts from printers but I couldn't find ones big enough, I need A3. Helicopter belts are very cheap (~£3.00) and so are the pulleys, so it seemed a perfect choice. The pulleys at the other end of the motor were machined from plastic.
With this version I tried basic movement with the motors to see how everything worked. It is at this point I noticed a few problems. The main one being that the drilled holes in the Delrin were not accurate enough, and so there was some movement, this movement was magnified through leverage and gave way so some pretty nasty effects. The second problem was that the block would jam. This was pretty serious.
In this prototype there was no pen action at all. That would need to come later, but it was clear that the mechanics of this was not going to work, so it was back to the drawing board.
My father, who had a keen interest in the hardware side of the project, (he is an engineer at heart...), suggested to put a cylinder with ball bearing in it for the bar to slide on, this would then be housed in the pillow block. The cylinder that I am talking about looks like this:
Adding this made a difference, it did't jam any more, but there was still movement and it wasn't working. Also, the side of the shorter axis which was not attached to the motor was dragging behind. More research needed to be done.
By this point I had become quite good at seeking out printer/scanners that had the kind of thing that I was looking for. I need to see how it was done professionally. So, while rooting around my family home I came across this old scanner:
Now, unfortunately I don't have any images of what it looked like inside because things move so fast. (The above image was taken after I ripped it's insides out. So, when I looked in I saw two things, one, was the the moving head had a wheel on its opposite side, it did not slide on a bar. I also admired it's sliding mechanism, it was super smooth with no play at all. This was done by the use of brass discs.
I couldn't help but notice that the bar in the scanner was very similar to the ones I was using. So, I found I pair of callipers and measured. By some coincidence, they were exactly the same. So I took the whole head out and took it away to chop it up.
The scanner then looked like this:
So, I then chopped the bottom off the old pillow block and simply mounted it on the one I recycled from the scanner. I then took the wheel it used on the opposite side and screwed it on below the other block, with a bit of metal for height adjustment. As the height of the bar has to be the same for the pen to touch the paper at the same point across the page.
So up to this point, every thing was looking good, but there was no pen action. I had read on the internet, where other people had made their own plotters, that this was the hardest part, so just 3 weeks before the deadline, I was quite worried.
It was decided by now that I was going to use a relay for the pen movement, but how I was going to do it was undecided. All credit for the final design goes to my Dad, who came up with it.
It uses a piece of MDF with a hole in the middle, this is where the relay goes. On the end there is a small hinge, bought from a normal hardware store. There is then a bolt that goes through the wood to keep what moves up and down in the same place horizontally and to stop it going to far up with the use of bolts. The piece of wood looks like this:
Then an L shaped piece of metal is glued to the hinge. Then, my idea was to bolt two bits of wood to this piece of metal to hold the pen. The whole block is then mounted on a cylinder of bearings shown above.
See the Pedro Gallery for detailed images and captions.