This group project is going to be a real challenge. We are in the middle of the semester and everybody is a bit tired because of the Fabacademy, the classes we are teaching in our own university and projects in our parallel profesionnal life. We met all together on Friday afternoon and brainstormed about what machine we could make. After the brainstorm, we wanted to make an origami (pre-)folding machine. The very difficult part is that we have asynchronous agenda and it is difficult to have long period of time where we are all together. So we decided as a start to divide the work into parts and assigned them to each of us.
|Who ?||What ?|
|Victor||Research on the end effector to pre-fold origami.|
|David||prepare the 2D/3D files to be cut.|
|Jeremy||get the materials and laser cut the stages.|
|Michele||build the linear stage.|
|Denis||In charge of the documentation and the team and tasks organisation.|
As a start Victor made some sketch of the machine we could make.
Victor also looked into the different end effectors that could be used to pre-fold cardboard and paper.
This cardboard has the good thickness (3 mm) but we had a lot of difficulty to fold it.
Victor and Jeremy then used another cardboard that was easier to fold but too thick for the linear stage design (we couldn't change the thickness parameter easily has the parametric file was borken).
A group is like a living organism. Energy is entering in the group and energy is dissipated/lost out of the group. When input energy is higher than output energy, the group is healthy. But when output energy becomes a bit too high it's good to stop and reflect on it. That's what we did on Tuesday morning. We used post-its too identify "good" energies and "bad" energies. It was really interesting has everybody could express himself, we got to know each other a little bit more and debated on ways to improve the way we work together.
Here is what came up after this meeting:
On Tuesday afternoon, Victor and Denis cut and assembled two linear stages with a 3mm cardboard. Michele came to help the team also.
The kit was quite easy to assemble.
Here are the linear stages.
Here is a movie of the moving linear stage (with Victor's hand).
By building this cardboard linear stages, we spotted a few difficulties and problems:
At the end of the week, some of us where not so happy with the idea of building a machine with this linear stages. So we thought about another project. More risky but that we enjoyed a bit more.
On Tuesday night, Victor and David discussed about this idea of making a drawing machine. On the morning of the next day, we decided as a group to build such a drawing machine. Here are some inspiration machine that Victor and David found online.
During the weekly regional review, Aldo showed us this video:
And we were also very inspired by this kickstarter project : the Maslow CNC machine.
Denis took in charge the organisation of the tasks. We divided the drawing machine into smaller parts and we assigned a leader for each part. We will work in small group on different part of the project depending on our availability. Here is the trello that Denis has made.
For each general tasks, we divided the work on a checklist to help anticipate the work. Denis has also assigned priority to the tasks. "Priority 1" is a task that has to be done in order to have a working machine, "priority 2" is a task to improve a working machine. The green, orange and red labels illustatrate the status of each task. Green, the task is on track. Orange, the task is critical and efforts should be put on it. Red, the task is really treatening the completion of the project.
In summary, here is a general view of how the tasks are divided :
|Who ?||In charge of what ?|
|Victor||the end effector design and fabrication.|
|David||general design improvement and integration of the (motor, spools, cables, end-effector).|
|Jeremy||improving the machine.|
|Michele||building the prototype and in charge of the electronics.|
|Denis||the documentation, the team and tasks organisation, launching the first protype and the programmation of the microcontroller.|
On several occasions, we discussed in group about the design of our drawing machine. We liked the idea of having two stepper motors on the floor and just some very light spools that we could hold on any wall, or glass window. The idea being that we could draw anything on any surface of any size...
Denis and Michele discussed on how we should program our drawing machine. We looked how machine and graphic card were working. We ended on Bresenham's line algorithm. So if we give to points, it is possible to interpolate a line in between the two points.
That is Michele's sweet spots. He builds prototypes at the speed of light. For that he had the help of helpful student, Thierry.
Based on the morning discussion with Michele, Denis set the problem. On the figure here below, the two upper spools are separated by a distance of W. The position of a point (red point) can be expressed in two different sets of coordinates: in cartesian coordinates (x,y) and in the two wires coordinates (a,b). When we control the stepper motors, we control the number of steps which corresponds to the length of the wires a and b. The relation in between the tow sets of coordinates are:
Here is an illustration of a point represented in the 2 sets of coordinates:
Here is an illustration of the Bresenham's line algorithm. The alogrithm finds the closest path to the theoretial lines connecting two points (x_0,y_0) and (x_1,y_1).
The stepper motors always move one after the other. So if we want to move form one point to the another, the end effector will produce a drawing that looks like half a rectangle (see the figure exercice 1 down here). If we want to draw an approximately straight line in between the 2 points, we can use the Bresenham algorithm to calculate the serie of steps that each motor as to make. Each motor will move one after the other depending on how far the end effector is from the theoretical line.
Moving between two points back and forth produce something close to a rectangle.
For this exercice, we asked the machine to draw straight lines in a zig-zag fashion using the Bersenham algorithm.
For this exercice, we asked the machine to draw a square, still using the Bersenham algorithm.
We have switch the motor in the kit for 2 Nema17 with 20 tooth wheel for GT2 belts.
Each motor is now embedded in a piece of Hi-Macs (thanks David!) and then fixed in a smaller drawing board
We are now using two GT2 belts from a delta 3d printer made by ourself and now scavenged for this project
We were using different end effector for our tests and it still need improvement, especially if we want to add a step motor to move up and down the pen.
For now, we are testing this one
After the cans, we tried the conterweighs from our window blinds and it's was working pretty well! We can open them to adjust the weigh
We have changed a lot from our initial setup. The base is now a Ramps 1.4 with an arduino MEGA and 2 Pololu a4988 configured for 1/16th microstep. To power all of this, we are using a power supply 230VAC to 12VDC This setup allows us now to be much more precise and to reach a greater approximation value for our drawing
This is of course the trickiest part. Thanks to Denis, we have now a solid piece of code to start but a lot of things need to be changed (new electronics, size,...)
For this part, a friend of mine, Benoit Sobrie, help us a lot, thanks again!
For now :