It seems strange that anyone would want to design a new 3D printer when so many options are already available. It is unlikely that you will end up with the design of a better "Mouse trap". Furthermore it is commendable to have the ability for a machine to replicate itself. This ability of all the various versions of RepRap ("Replicator Rapide" as I understand it) is commendable . Even with the ability of the RepRap to replicate itself a fair number of components require the use of a lathe to be accurately made. The problem with self replicating machines is that you first need to own the first machine. It is the age old chicken and egg story.
Having been left with the option of buying the first RepRap type machine the desire to build a 3D printer would be greatly diminished. The design of Fabrap therefore was to bridge this gap and to be able to build a 3D printer with the my available machines. The reason to document the build is the thought that others may have the same dilemma.
After a lot of searching and reading of information on the Internet, I decided to follow the Fab@Home approach to building a 3D Printer. I liked the interlocking method of construction and beleive it will provide a sturdy printer. This method suited my workshop and was within my potential. A sketch of the interlocking has been included.
The Fab@Home Project is an open-
The disadvantage of the Fab@Home is the motors are too expensive. Costing had them offered at $160.00 each. Multiplied by 4 or 5 and you have a high cost unit. Furthermore the software available is more restricted. RepRap on the other hand has a very large following and development participating community. The decision was then to construct the unit using Fab@home construction with the RepRap electronics and software. Hence the project and website name of FabRap.
The drawing and sketches shown in this section are provisional. As the build proceeds changes may be implemented and I am hoping I will receive constructive criticism which may aldo improve the quality of the overall project. The build log will reveal the final iteration of each part.
Now having decided the "what" project it becomes the question of "how". The first decision was that the axis rails would be 12mm stainless steel using SC12UU linear bearings for the X and Y axis and the SC12LUU long linear bearings for the Z axis. With the bed being a cantilever construction it was decided that this could use some additional support, hence the longer bearings.
An additional benefit of using the linear bearing blocks is that it will ensure the units are square and more accurate which in turn will improve the overall build quality.
The movement will be using 6mm wide GT2 timing belts. The pulleys will have 20 Grooves so that each revolution of the stepper motors will move the axis 40mm with a result each pulse will represent 0.2 mm movement. I believe this will give acceptable accuracy. The shafts of the pulleys will be 6mm whereas the shaft diameter of the motors will be 5mm. The change of diameter will have to be accomplished at the joining of the motors and the shafts.
The Y and X axis will be dual belts and drive pulleys whereas the X axis will only have the single drive belt.
The printer will have three distinct assemblies that will be joined by the axis rails. These assemblies are the printer case assembly, the hot bed machining table assembly and the X axis carriage assembly. Each axis will utilise an opto limit switch with the end positions defined by the machined assemblies.
The printer case assembly is the major unit into which the other assemblies all fit. The printer is to be fully equipped from the start with only a power cable and USB cable required to get the unit operational.
The first view clearly show the electronics mounted in the side compartment. The equipment is the power supply for the electronics, the Sanguinololu microcontroller board for the processing and 2 solid state relays. The first relay is for the heated bed or router spindle and the second is for the fan that will be fitted to cool the heated bed at the end of a printing cycle.
The holes in the front panel are for indicators, the on off switch and a fuse. The fuse and mains input plug are built into a single unit. Inside the printer case the bearing mounts for the belt shaft can be seen in the rear of the unit.
The rails for the Z axis are shown and will be duplicated in other assembly sketches whereas the axis rails for the Y axis have been omitted. In the final printer they will run from the front to rear panels just above the strengthening shelf. The DXF files for the machining of the front panel will have a closed as well as open front so that the choice of construction can be made at the time of preparation of the gcode. I have has advice to have the front closed for additional strength, but believe that the method shown in the sketch will have adequate strength. Should the strength of the unit be compromised by the open front, FabLab would have closed the front of their printer.
The Z axis and work table are a cantilever construction and therefore will he the long linear bearings for additional strength. Dual timing belts next to each axis rail will adjust the table height.
A fan will be fitted to the table so that it can be switched on at the end of the printing cycle to cool the hot bed. The hot bed is not shown on the sketch. These details still have to be finalised and will be addressed during the final checks that take place prior to matching of the parts.
Shown on the side of the unit is the opto limit switch. The end of the table has been extended to allow the switch to extend to the outer wall of the printer case. This slot will have the end stops fitted and will have to be adjustable to allow for single, dual extruders or spindle whichever is to be used.
The last assembly is the Y axis carriage. This unit is installed on the rails placed above the shelf. The rails are not shown in the sketches but their position can easily be identified.
The tool carriage for the x axis can be easily identified. The advantage of the FabLab approach is that it is easy to exchange extruders or spindles. This adds versatility to the machine. It must be born in mind that any milling on the machine must be of a light nature. It does, however, make the printer far more flexible.
The motor for the X axis drive is clearly shown on the front of the carriage. The axis used the same opto limit switch. Both the X axis and the Y axis limit switches are mounted on the assembly. The timing belt for the X axis couples directly to the motor shaft within the box structures either side of the carriage.
The single extruder is based on Wade's design. A dual extruder was drawn up but the design was not satisfactory so it will have to be redesigned. The design will have to use standard gears and will not use the herring bone design. This may follow once a 3D printer is operational.
The sketch give an impression of what the extruder will look like. The hobbed bolt and motor have not been shown in the sketch. Also the hot end has not been added. The hot end has been drawn but the sketches and drawings will only be finalised once the machining of the parts has been undertaken. This is an area of experimentation on most designs I have read about.
Although a lot of time has been spent designing the extruders, it is now possible to purchase these components complete and common sense dictates that this will be the route followed. The final quality of the 3D print is a direct result of the extruder and nozzle used. A proven design commercially available will help in this respect.
Research is being done to allow the use of a solid state laser to be used in place of the extruder. This will allow some laser engraving and cutting of thin materials. Plexiglass and balsa are examples of suitable materials. Jtech Photonics are a reputable supplier of suitable laser modules. The Chinese have a number of similar products.