Cartesian Robot DevelopmentBack to Main Page
12/05/2010 - End of Blog
Work has now ceased on the cartesian robot designs. The robot appears to be no better or an improvement on any designs that haven't already been completed. Work has now started on the Delta Robot project. Here is a view of the complete assembly before it was dismantled.
Z axis is not performing too well as there is currently no vertical guide rails. Due to the awfull straightness of the M8 threaded bar the tooling head carriage moves around everywhere! The timing belts and pulleys are also very expensive. The, belt is around £9 and the pulleys are about £8.50 each. So £34.50 for this mechanism, this is just a plain rip off from RS. With that in mind over the weekend I did a little more CAD work and came up with the following design.
The design is not finished, but it does remove the troublesome z axis lead screws and the expensive belt and pulleys. Fortunately I will be able to reuse some of the components I have already repraped so the previous design won't be completely wasted.
The Z axis belt drive is now finished complete with adjustable tensioner made from two 608 bearings. It works very well and much better than the Makerbot pulley system which constantly fails (I already have upgraded this with another design on the Makerbot). Not much to do now before I can have all the axes running. I might quickly design a pen holder so at least I have a 2D plotter!
I have completed rapid prototyping the X carriage parts and assembled them together. The mechanism is very smooth running so I will be retrofitting bearings to the Y axis as soon as possible. The more I design this machine, the more I realise what a good design the Reprap machine is. Often I seem to be going full circle and finally coming to the same design conclusions as the Reprap team did. The next job is to complete a belt tensioner for the Z axis toothed belt, complete the X axis lead screw and design a clamp mechanism for the Dremel tool head.
I have been trying to make progress to finish the rapid prototype machine design. My initial thoughts were to use plain bushes for the linear bearings and I made some using my lathe. The friction was still too high so I have now completed a design using roller bearings for the X axis. If its OK I will probably have to retrofit some to the Y axis. The design required some intricate design to fit all the necessary components into a small space whilst still giving room for the lead screw. The real benefit of designing on CAD is that you can squeeze the design space right down. Other progress includes debugging the code for the Arduino Mega to control the machine. I found some problems with the software when it is working in metric units. Some variables go out of range. I informed the software author and these will be corrected in the next issue. At least soon I will have a complete machine ready for improvements. In order to get the machine doing something useful I purchased a Dremel so this should make the machine capable of some light milling work quite soon. Perhaps a few name plates for the Helium Frog workshop!
I completed the single motor Z axis with a upper belt drive. Rather than using conventional roller bearings I tried to use Rod eye bearings (also called Rose Joints). The reason is that they are more forgiving of mis alignment and also remove the need for an accurate bearing housing, something which is difficult to achieve in a rapid prototyped part. The rotational resistance of the seals and bearing (one at the top and bottom of each vertical shaft) proved too much for the stepper motors and the mechanism would not turn when asked. I have begun a new design with roller bearings. The design is complete so I will rapid prototype the parts hopefully for another trial this weekend.
I have completed the construction of the electronics and uploaded the firmware. I now have the machine moving all axes to CNC G codes script. I had some problems with the motors stalling / missing steps, but I dropped the supply voltage of the motors from 12V to 5V and this seemed to improve things. Whilst running the machine it was noticed that the Z axis mechanism which used two motors seemed a little more trouble than it was worth, particularly when adjusting the start point by hand. I'll convert this to a single motor mechanism driven by a belt to the two vertical lead screws. Now I have a spare stepper motor I may have a look at using it for a stepper motor driven extruder.
After searching the net for some Arduino Mega firmware to control my machine, I read about a similar project called the Hydra MMM ProjectThe firmware loaded immediately to the Arduino and the PC software ran my G code files with the stepper motors moving in what seemed to be a logical manner. It stumbled a little on some G codes that are not yet implemented, but the Author Clayton Webster is making continual improvements. I hope in a small way to be able to contribute to debugging the code when I have my machine running. His code style seems similar to my way of thinking which is also a bonus. This find has saved me a lot of time developing my own firmware so thanks go to him.
I have nearly completed all the rapid prototype parts for the XYZ axes so will soon have the main machine built and working. I then need to concentrate on the electronics.
Work continues on the development of my own machine. I am printing the parts on a MakerBot, so I am spending a lot of time trying to get a reasonable quality print. I have got the Y axis built and working on a Arduino Mega microcontroller. The movement is very good with very little backlash. The only concern is the speed of travel which because of the use of lead screws is very slow. I may have to rethink this machine and use it as a milling machine, or convert it later to belt drive, still its an interesting project to get my teeth into. Even if the machine is slow, if it is well built it should give good reliability if left for a long time unattended during long part builds.
My Makerbot rapid prototype machine is up and running and I have built a heated bed for it to reduce warping of the parts. Its working quite well now. I have also begun to design my own machine and completed a test assembly to try out the Y axis to see if the mechanism wil be accurate enough.
I wasn't using my two screen computer setup much, so I built a second PC from a few parts I had lying around. I installed Ubuntu Linux on this one and was very impressed with how easy it was to setup. I only had a few issues with the graphics driver but not too much problem to sort it out. I now use it for browsing on the net. I would love to do a version of helium frog for Linux and this may be possible.
I ordered a Cupcake CNC rapid prototype machine from Makerbot in the USA. I hope to be able to make my own 3D parts from ABS plastic in the near future. I've been following the RepRap and Makerbot projects for a few months now and decided to make a purchase. I hope I can spend a few happy hours over Christmas putting it together.