The Electronics

So, now for the electronics element of this project.

Control Board

There are many 3D printer control boards out there. I wanted to go with one that has a good ecosystem and has good support. For this reason, I decided to go with the RAMPS 1.4. It is limited in it's processing power, but it is well established, and has wide support, not to mention a number of firmwares that can be used.

The RAMPS series of controllers are usually one part of a two part system. The main processing element of the setup is an Arduino Mega 2560. This board is responsible for receiving and processing the gcode and then calculating how the printer will move to match what has been requested. There are some custom RAMPS boards out there that incorporate the Arduino on one board. I decided to stay away from these due to the fact that if the 3D printer didn't work out, I could use the Arduino for another project.

The RAMPS board allows the Arduino, with the addition of a suitable power supply, to interface with the stepper motors, sensors and heating elements, of which will draw more power than the Arduino can supply.

Although  I would recommend purchasing an official Arduino and RAMPS boards (I already have an official Ardunio Uno),  I opted to order up a version from a well know Chinese supplier. This was mainly due to the fact that I had to keep the costs down. As a result of this choice, I was expecting to have to deal with some quality issues later on down the line.

As part of my order, I also decided to include a full graphics module. This seemed better that the 2x16 screen and also gave me the ability to print directly from an SD card.this seems a small point but it is worth noting if your PC crashes for any reason that 5hr print has just been ruined.

So after a few weeks my order arrived. Surprisingly it came in a nice plastic box and in sealed anti-static bags. It came with "almost" everything I needed to wire up and get going (We'll get onto the "almost" later). 

The package included: -

• 1 x Arduino clone
• 1 x RAMPS clone 
• 5 x DRV8825 stepper driver (1 for each axis and enough for 2 extruders)
• 5 x Heat syncs for stepper drivers
• 1 x Full graphic display
• 1 x Adapter to connect display to RAMPS
• 1 x USB cable

On first inspection, the quality seem very good. But while I was attempting to connect the RAMPS board to the Arduino, I found that there were issues lining up all of the pins. It later transpired that the headers on the Arduino were leaning slightly and this was obvious that this was done during manufacture. The cautious application of force in a strategic area sorted the issues and all went well from there.

As I said earlier, the deliver came with "almost" everything that I needed. For some reason the supplier decided not to provide me with the required jumpers that allow the RAMPS to do micro stepping. In short, microstepping allows the motor that drives the printer to be controlled with a greater level of accuracy. Thankfully my hording of old computer equipment provided me with a bounty of redundant jumpers from old motherboards and hard drives.

Now it was time to prepare the DRV8825 stepper drivers to be installed. This was relatively easy, but care still needed to be taken due to very tight tolerances between the top of the driving chip ad the back of the headers. To install the heatsync, it was a simple case of removing the thermal tape paper, aligning the heatsync with the chip and pressing down, while at the same time making sure the heatsync was not touching any other parts. 

Nnext up, the driver boards need to be inserted into the RAMPS board. Again, this was relatively easy, but care needed to be taken to orient the boards correctly. If these are plugged in backwards they fry the driver chip, so it's best not to get this part wrong. to install them, the small trim pot needs to be furthest away from the cower connectors. the picture below shows the green power connector on the left and the small round trim pots on the right hand side of each driver board.

At this point I decided to take a rest from all of the concentration. Unfortunately for you, this is where my lack of blogging skills showed, as I neglected to take pictures through the next phase.

I'll do my very best to recreate the process to complete the build, but i was having WAY too much fun at this point :)

Until next time

Build Update - The Printer Structure

Ok.. ok.. so looking though my photos of the build so far, I noticed one of two terrible things had happened: -

1. I had some how not uploaded pictures of the assembly process from my phone before it died
2. I had forgotten to take pictures of the build process.

I was sure No.2 was not an option, but it has been a while since this was completed, so I cannot be 100% sure. So i'm putting this down to Newbie Mistake No.1 - "Not taking enough pictures during the process!".

In spite of this I will attempt to recant the process and submit new pictures to fill in the gaps.

So the time has come to put all of this together, and hopefully get a printer at the end of it.

First I needed to cut out the printer bed. This is where all of the printing will take place and needs to be as accurate as I can make it. Thankfully I had some 12mm plywood left over from some DIY work in the house that could be re-purposed for this. I downloaded a drawing of the base which had all of the measurements.

I then printed this out on paper at a 1-to-1 scale and measured it for accuracy. Once I was happy, I stuck it to the ply and proceeded to cut it out with a jigsaw.

Once cut out, I drilled the 6 holes out with a 4mm bit on my drill press where the motor mounts would be fitted. I then took a similar sized board and screwed it to the cutout though 3 of the holes I just drilled. I then setup a router table with a straight cutting bit with a bearing. Using the first cutout as a template, I used the router to cut out the top. I then drilled out the 3 of the holes, moved the screws one by one, and drilled the remaining 3, again using the first cut as a template to ensure they were both exactly the same. 

For the very bottom, I decided to trace around the bed and extend by 25mm all the way around and extend at the front. this gave the motor mounts and the LCD something to sit on.

Now all of the wood cutting had been completed it was time for assebly. One by one, I fixed each of the 3 the NEMA17 motors to the motor mounts with 4 10mm M3 hex screw and washers. I then inserted each of the 6 8mm smooth rods into the motor mount and tightened with 6 15mm M3 hex screw, a washer and nut. Each of the 3 assemblies were then screwed through the base giving the first signs of the size and scale of the printer.  

I then assembled the 3 idler brackets with a bearing that would hold the timing belts that would run the printer. This was done using an M8 hex screw and nylock nut. These were then attached at the other end of the M8 smooth rod at the top of the printer and tightend with 4 15mm M3 hex screws, a washer and nut. 

Things had come together quickly to this point so I decided to get an imopression of the final size . I attached the GT2 drive pulley to the NEMA17 motor shafts and placed the top on the printer. I knew the top and idlers would have to come back off to attach the end effector assemble, but I wanted to get a feel for the final look.

So far so good, but I knew the next step would be challenging. Getting the electronics installed and working.

Minor update

Well, needless to say I have been a little preoccupied since my last post.
Alot has happened and I have not had very much time to post about progress.

Currently the printer is running, and I have had some good practice printing in PLA and ABS materials.

I have however, run into some stumbling blocks that could be down to my my poor fabrication skills, or due to my lack of research (something i will go into in more detail in future posts.)

for now I leave you with a montage of images of work I have done in over the last month or so.

"First Drawing"

"A little bit more accuracy"

"Extruder designed in OpenSCAD and printed on ply wood"

"Roughly cut out :s "

"Hot End holder traced on Ply and Cut"

"Waiting to be drilled and cut to take the hot end"

"Extruder Assembled.... sort of"

"It Works.... First Print"

"Quick... Print some replacements before it all falls apart!"

"Thats Better!"

"First Stress test. Skull next to my Fitbit for scale"

"Not Bad for a first go"

"First fix. broke a pulley in my food mixer (left) OpenSCAD replacement (right). And best of all. it works :D"

"Larger Skull (5hr Print time"

Replacement rod ends in ABS

"Ginger bread house cookie cutters. much fun was had with the kids :)"

"Decided to be lazy and print these so I don't have to tie my shoes :)" 

Until next time........

The Build

So, after much researching and hunting for parts, I managed to assemble the majority of the items that are needed to assemble the printer. For purpose of keeping the cost down, I will be making some of the parts, such as the hot end mount and the bowden extruder. These only need to last long enough for the initial calibration and then these will be replaced with printed parts. The main question at this point is where to start the assembly.

I was initially overwhelmed with the amount of work that was ahead of me, not knowing where to start first. So I decided to focus on making the assembly as modular as possible. That way it's broken down into smaller jobs which are more manageable. With that in mind, I chose the smallest 3D printed part to start with, the push rod ends.  

The push rods connect the printing platform to the three vertical axis. These need to be strong as they are the main moving part that carries the weight of the print head or "hot end as it is being moved around. For the connecting rod, a number of different materials can be used, such as aluminium tube, carbon fibre kite rods or wooden dowel. I chose to use the wooden dowel, as the carbon fibre rods need to be turned down to fit the rod ends and I felt more confident that I could achieve acceptable results with the dowel. The internal diameter of the rod end is 5mm, however I was only able to find 6mm hardwood dowel. This meant that the dowel needed to be cut down to fit.

For my specification, the rods need to be 250mm from the centre fixing points. to achieve this I made a jig by drawing 2 parallel lines that are 250mm apart on a piece of scrap plywood, putting a mark on each line 25mm apart to mark the position of each assembled rod. The use of a jig allows me to get consistent results for all of the rods. In addition to this I marked the end of the printed part where it connects to the rod. I then measured the depth of the hole in the arm end with a digital calliper and transferred this depth onto the jig. This made an easy visual reference where to mark the dowel for each cut. Once cut I used a sharp craft knife to turn down the ends to fit.

To finalise the jig, I hammered nails into the plywood on the makings for the pivot points in each rod end, taking care to make sure the nail was correctly placed and as straight as possible. Once all of the nails were in place, I used a dremmel tool with a cut-off wheel to remove the head, so the the rod ends would fit. The arm ends were glued to the dowel and placed into the jig to set for 24hrs. After this time, the joints were taped and the arms placed back in the jig to ensure they didn't distort due to moisture etc  

 
Once completed the time consuming job of assembly could start. Each of the 3 axis (X, Y and Z) has 2 push rods that are connected at one end to the delta platform, and the other to a sliding rail attached to the linear rods by a pair of LM8UU linear bearings, the connection is made by screwing a 3D printed block to form a universal joint allowing the platform to move feely.

With the carriage assembly completed, my attention turned to the main structure of the printer.

The Rostock Printer

Following on from my previous post, I have decided to build a Rostock style printer.
So what is a Rostock printer and why is it different to other FDM X/Y printers.

Traditional XYZ machines, such as CNC milling tools and routers, including 3d printers use 3 motors to control each axis of movement One for left and Right, one for forward and backward and finally one for up and down. The Rostock is a delta robot printer. This type of printer has a simpler design with 3 motors working in constant coordination for movements across the print bed. this allows the delta printer to move the print head freely and rapidly during the printing process, in theory making the printing process quicker.

In addition to this free range of movement, all of the delta printers axis are aligned in a vertical configuration. this means the build volume is only limited by the length of the linear rods and drive belts allowing for some extremely tall prints by comparison.

This image illustrates the difference between XYZ (Cartesian) and Delta printers

Image source: https://printspace3d.com/cartesian-vs-delta-printers-work/


This style of printer is derived from the delta robot. Quote from Wikipedia  - "The delta robot was designed in the 1980s by a research team led by professor Reymond Clavel at the École Polytechnique Fédérale de Lausanne (EPFL, Switzerland). The purpose of this new type of robot was to manipulate light and small objects at a very high speed, an industrial need at that time. Industries that take advantage of the high speed of delta robots are the packaging industry, medical and pharmaceutical industry. For its stiffness it is also used for surgery." 

This type of printer is well documented within the reprap community with a number of spin-off designs. Reprap Rostock Page.

Where do I start
The Reprap page has very good documentation, including an extensive bill of materials section.
After taking notes of all of the components I scoured the internet for parts. Many of my projects have stalled due to lack of funds. I often embark on the project and do the "free" bits and worry about the cash when I get there. Inevitably when I get there, there is no cash so the project stalls and gathers dust until the next new project comes along and the process repeats. I did not want this to happen on this project, so the list making began.

I listed all of the parts that I needed in a spreadsheet, along with the URL for the part and the cost (including postage). After getting a ball park figure I knew I had funds to cover, the real search began. This involved looking for the parts that I needed and find the best deals.

For the structure of the printer, I managed to salvage 2 pieces of plywood left over from a DIY project some years ago. For the 3D printed parts I found what I though was a good deal on Ebay along with 4 NEMA17 motors, a heated bed, E3D v5 hotend fully assembled with Bowden tube, GT2 toothed belt and pullies and flanged bearings.

For the electronics  opted for the well established RAMPS1.4 system that plugs directly into an Arduino Mega. I found a good deal on Banggood for one that came with a full sized LCD and SD card reader. I have heard some good and bad reviews on these,but for the money it was worth the gamble.

In my next post I will be going through the parts in more detail and going through the build process starting with the endeffector

Machine Selection

OK, so I know what I want to build, and why I want to build it. The next step is to decide what printing technology I want to use. The first thing to look at is the different technologies that a home builder can achieve with good results and within a reasonable budget.

There are many different technologies used in 3d printing, some more exotic than others. For this reason I'm going to limit my choice to 2. This is mainly to ensure ease of construction, at the same time stretching my abilities and learning. The technologies I will look at are: -

- Stereolithography (SLA)
- Fused deposition modelling (FDM)

How all current 3d printers work

So, a 3d printer constructs a real world object by building it a layer at a time. Once the layer is complete the printer moves by a fraction of a mm in hight. The next layer is then built directly on top of the previous layer. This process is repeated until the 3d object is complete. How the layer is constructed and the material the object is made from is where the technologies differ.

Stereolithography uses UV curable resin as its build material. It is placed in a bath where it comes in contact with the build platform, this is where the object will be constructed. A UV laser traces the layer using a series of mirrors that are attached to galvanometers allowing them to move in the X and Y axis. The printer then uses the process described above until the object is complete. Once completed, the object is removed from the build platform and rinsed to remove any excess resin which would spoil the print if left behind. Finally the object is placed in a UV oven to harden completely.

Image Source: http://www.solidsmack.com/fabrication/stereolithogrphy-110-micron-old-world-laboratories-nano-3d-printer/

Another method involves using a a modified DLP projector. In this method instead of the laser drawing the layer, the projector exposes the whole layer at a time. This speeds up the printing process considerably. The construction of this machine is also simpler as there are fewer moving parts. The complexity comes in the form of software.

Firstly an application is needed to divide the object into its individual build layers, also know as a slicer. Most slicers that I have investigated convert the object to gcode, a format mainly used in CNC and FDM machines. Secondly, controlling the projector and the timing of the movements of the build platform seem to be a bit of a barrier to me.

Image Source https://code.google.com/p/lemoncurry/wiki/main

Fused deposition modelling or FDM uses a thin plastic type material, also known as a filament as the build material. This is fed into a heated nozzle forcing a string of melted material to be placed on the build platform. The heated nozzle, also called an extruder is then moved back and forth to draw the layer very much like an etch-a-sketch. As before, the build platform moves a fraction of a mm and then next layer is drawn on top until the object is complete..

Image Source: http://www.hotcopier.com/fdm-printer

A variety of materials are available to print in at substantially less cost than UV curable resin.

Although the print quality is not as good as SLA, FDM printing has a well established ecosystem, which is good if you need help fixing something.

For this reason I have decided to go down the route of an FDM printer, specifically a reprap Rostock delta printer. I have selected this printer for it's simplicity in construction, large build area and support in the reprap community. Additional benefits of this route also include faster time when compared to traditional X/Y configurations and the reprap's ability to replace many of its moving parts due to clever design. For more on the reprap community visit www.reprap.org

Image Source: http://www.reprap.org/wiki/Rostock