< back home

Machine Design

Fabacademy : Week 15 May 02/2018

Assignement - Mechanical design - schedule

Multiple purpose machin

  • 3D files : fabRack.f3d

  • Machin files : machinDesign.f3d

  • Stocks Materials

    • 2 Delrin : 28€ / 500 mm² --> 12mm height
    • 1 DHCP : 26€ / 500 mm² --> 10mm height

Final Images

machin Front 1 machin Front 2 machin Back
machin Front 1 machin Front 2 machin Back

Design thinking :

Process

Inspiration :

To begin with this machin we choose to take two machin as reference. The first one is the one from Hector and the second one came from Nadya.

Nadia's machin
Go to the page
Jacob's machin
Go to the page
the Chamfer Rail from fellesverkstedet repository
nadia's machin jacob's machin rack and pinion metal
Designing :

The main goal we wanted to succed in making this machin so that anybody who have access to a milling machin could do this. We choose to aboard this design with one main constrain, replace screws by japanese joinery and reduce the necessary supply you need to have for this kind of machine.

Rack and pignion test

I made wood test to validate the rack format and the faisability on our CNC machin. | rack and pinion mill | wood rack| |---|---|

Parametric design

I know that I would have to make a lot of back and forth to succed to make this machin so I worked on making the joinery parametricaly depending on my stock height and my the smaller mill I have that could go that deep.

After working in fusion on parametric joinery

joinery parametre joinery param
joinery step joinery setup
print emboitement piece emboitement piece
first  piece print test emboitement
Think about the stock and your machin :

There is two main constrain that you couldn't avoid when doing a machin like that it's your stock, what size and weight is it.

stock

milling prototyping

In order to know more about rack and pinion, I made a first rack and minino with a junk of wood we had here in the lab.

Then I wanted to be sure of the joinery I made, so I made a simple milling test on our shopbot.

milling test pom export setup
ramp conparatif
pocket contour
contour contour
contour contour
contour pocket

On the testing phase, I break a ballnose mill cause I didn't made my pads high enough. broken mill

3D prototyping

To be sure about the join I used I made some rapid prototyping with the milling machin first, and then with the 3D printer.

parametric element parametric element
3D test print test
Montage

contour

mounted machin triptic machin pocket

Stock comparison, from the store to the lab

In every design process, there is junk generated by your creative process that could be more or less important du to complexity of what you are trying to do. Here is everything we tryed and fale to do this machin.

dhcp pocket

And here is the comparison of our product. On the left everything we didn't made ourselves, and on the right, what we made in the fablab, including electronics.

Milling :

Preparation

prepa cam

Toolpaths
Parrallel 3D

surfacing

!

2D Pocket
pocket pocket milling
pocket milling rack engraving
ramp pocket
pocket
2D contour
The Machine works by hand:

Electronic Design & production

The PCB uses the DRV8830 ICs. An I2C motor driver. There is 9 of them on the PCB.
I designed the PCB using KiCAD. The sources files are here. Look into the
It can be controled by I2C with any MCU. Exemple sketchs for Arduino can be found there.
If all the ICs are working together at full current (1A), the PCB's temperature stabilises at 90°C.


Testing

It works using the PCB and stepper-motors: