Finished the first prototype build, some things went well and some need attention. I was generally happy that the concept is sound, and that the threaded rod, rod and pipe created a rigid structure. The bearings, however, are not stable enough especially in the X axis. In general I have decided to double the bearings by stacking two of then inside the bearing assemblies. I have also separated the bearing rods as far apart as the design will allow for greater stability in each axis.
The Z axis worked nicely but was too big and did not allow for at least 300 mm of movement in Y, so I have reduced its size. The Y axis was also preventing the Z from moving 300 mm, so I have extend the T bar upwards to allow for the full range of movement.
After discussing the build with my mechanical engineer mate down the road, I have also changed the stainless steel rods for chrome plated as the bearings are chewing through the stainless and need to run on a much harder surface. All the design changes only took a day or so to make, and I can re-cut the timber parts next week, but I am now going to have to wait for the new parts to arrive.
My view on electronics has also changed a lot in the last couple of weeks. The WhiteAnt is no longer my preferred option. I now see two clear options
- RAMBo – The board has everything for a 3D printer in one, and can support dual extruders and a heated build platform. Would work with ReplicatorG or Pronterface for 3D printing, but may not work well for CNC work.
- Smoothstepper – The would work well for CNC, but would need extra boards for the motor drivers and extruder/heater. I would make a great CNC machine with Mach3, and would print with a Mach3 plugin. I have ordered both sets of electronics and will test with both.
12 Nov 2012
OK, what has changed in 2 weeks. Well I have been thinking about the build a lot, and I have some additional requirements to add to my wish list if you can believe this. So first I am simply going to add the extra items:
- Convertible to a large format 3D printer extruding some form of material (Foam, plastic, ???).
- Convertible to a hot wire cutter for cutting foam.
- Convertible to a rotating tangent knife for cutting foam board or corflute.
I also just spent 3 days at the Supanova Pop Culture expo with my Thing-O-Matic as a draw card, and the level of interest in 3D printing was massive. I had already though to prototype the large machine on a small scale as a 3D printer, and this is now the focus of my efforts. The machine will be based on cheap linear bearings on 8 mm stainless steel rod encased in MDF, but will have the same layout as the larger machine. I am looking for a build platform of 300 mm x 300 mm x 300 mm, and plan to ensure that multiple extruder or a CNC tool can also fit.
I am loving the new WhiteAnt motion electronics, so I think they are the brain, and I think I have found the right extruder. However, the build size will allow me to have multiple extruders, or even extruders and CNC Dremell all at the same time. How cool would it be to build a 3D object, and then CNC the details on the same machine without a tool change. Not sure that ReplicatorG will be the software option, as it seems quite limiting, but this is a work in progress and motion mechanics will come first, followed by motion electronics, and then software solutions.
Started the prototype build today with the base assembly. I used laser cut 6 mm MDF, 20 mm plumbing pipe and 0.25 inch threaded rod with nyloc nuts and washers. Not sure that it would not be better to CNC the end assemblies out of 18 mm MDF, and not sure that the motor and spindle mounts now scaled down will fit the NEMA 24, but happy with the rigidity and simplicity of the build so far. Waiting for my 8 mm stainless steel and linear bearings to arrive before the next step, but I am ordering the motion electronics and hot end and build platform heating pad.
OK, I have been inspired to design build my own CNC machine, something I may yet live to regret. Thought I should document my build, so this page will be updated as I go.
27 Oct 2012
Firstly, why am I doing this and what do I want.
Well, I owe a great debt to Patrick at http://www.buildyourcnc.com/ for giving me the idea that I could build a CNC. Why? Well I have already build one of his blackFoot 8′ x 4′ machines and in doing so have learnt a great deal. While I am happy with my blackFoot, it has taught me a great deal about what I want in a CNC, and it is for these reasons that I have been looking to design my own machine. So what is it about the blackFoot that I want to change:
Mostly the machine is not a rigid as I would like. I have noticed this especially in the movement in the Z axis when the machine is tracking in the X direction. The X axis is great because it tracks between two rails. This has made me think that the Y and Z axies should also track between two rails, as this should provide great rigidity. I was also not happy with my ability to adjust the machine until Patrick assisted me with an upgrade to V4.1 with the adjustable Z-Y plate. So I would like a design where I can make adjustments to all bearing assemblies easily.
I would also like to see a more universal method of mounting routers to the machine, and to easily change tools. At the moment I have to take the router out of its mount each time I have to change tools. I do not currently have a vacuum table, so I have also experimented with a push down foot and would like a simpler method of adding this when needed. It has also taken me a lot of time to solve my dust problems, so I also wanted to included both a push down foot and good dust extraction in the same design.
The last major issue that I wanted to address was the addition of future additional axies. The blackFoot is a 3 axis machine and it would be very hard to add any additional axies. I wanted a build that would allow me to add upto 6 axies in the future.
I also owe a debt to the Grumpy Geek http://grumpygeek.com/ as he has shown me a way to make a ridged machine with cheap rails. I love his build, but I can’t help but think his idea is able to sustian a much larger build, and that not just the plastic pipes should be used for rigidity, but also the rails themselves.
So, time to provide my wish list of what I want in my router:
- Very rigid in all axies
- Simple and cheap to build with easily available materials
- Chain driven in X and Y
- Lead screw drive in Z
- Scalable with at least 8′ x 4′ edge to edge
- Large movement in Z of at least 500 mm
- Adjustable bearings
- Good dust extraction
- Simple router mounting for multiple router types
- Use of plunge router mounting to be able to make physical adjustments to Z while routing
- Simple tool changing (possible addition of a tool changer later)
- Multiple hold down options (Vacuum, clamp and hold down foot)
- Multiple bed options including rotary, and wet bed for use of coolant
- Expandable to 4 axies (rotary fitting)
- Expandable to 5 axies (A and B)
- Simple cable and dust management
- USB driven using the Soothstepper
- Expandable to include options for position camera, coolant, future expansion
- Easy material loading
- Option for slide on slide off at each end for fast sheet changing
Thats probably enough for now. You may think that this is an impossible list to satisfy, but I have been working on this for some weeks now, and I have a concept that I think will meet all these requirements. My basic concept is simple, and that is that each axis X, Y and Z should track between two set of Grumpy Geek inspired rails. Given this the Z mover can be a basic box with a flat bottom that any router can be mounted to, and given the cheap rails the Z mover can be large enough to allow for future expansion into the A and B axies.
I plan to start with a basic 3 axis machine, and add to this later. Stand by for the next installment.