My notes :
Our tutor has set up a lunix laptop, so we could use the Modela to start milling our board.
note: the board were making is a 'programming' board.
When installed it's useful to program other (new made) boards. Aka, programmer.
Start up the modela program by entering the right command in the terminal. (make_cad_rml)
Step 1 : load the .cad file the want to mill
Step 2 : remove the '#' symbol of the step you want to execute
(so first traces and exterior) (make sure the others in the list are turned on > with '#')
Step 3 : set the pop down menu (on top of screen to 'mill traces (tool: 1/64)'.
Step 4 : make .png
Step 5 : in the path parameters, set the error margin to 1.0
Step 6 : click the make path button to generate the toolpath. (use rest view to reset the view :)
Step 7 : set the xmin and the ymin (the starting point of the modela's milling head)
Step 8 : make .rml (then send it)
Step 9 : While the machine is running, replace the protective plastic hood.
When the machine is done, remove the hood again and use a brush to wipe away the dust.
(it's paper and copper dust, don't try to blow it away).
Step 10 : Next, write a list for the 'stuffing' of the board.
(add a piece of upside-down sticky tape to the 'shopping list'. This will prevent your components of getting lost or mixed up).
ok, now soldering for the first time...
Note: You need to have 2 steady hands!
So try to plan soldering in the early morning, after a good night's sleep, no coffee and with proper breakfast etc. ;)
Step 1 : place a bit of tin on one of the two copper squares.
Step 2 : place your component on top of it and pin it down with your tweezers.
Step 3 : reheat the drop of tin until the pinned component sinks in it.
Step 4 : add tin to the other side. (place a drop, keep heating it till it starts to nicely flow out).
Step 5 : add some extra tin to the first side (so both sides are nicely flowed out).
Step 6 : check if your happy, before continuing. (the more components that are already soldered,
the harder it gets to make small adjustments)
Step 1: installing winavr (this program includes 'avrdude' which we need to do the 'make' step)
Step 2: open cmd and execute 'make clean' and 'make hex'
What is 'make clean' ? It removes the old 'object and hex' files. (like throwing away a old ziped file in the bin).
What is 'make hex' ? a hex is the 'machine code' for the programmers board. (like making a new zip).
Step 3: installing Atmel studio.
We need this for the diver of the 'blue box' This 'box' is also a 'programmer'.
The new board I made is still unaware of what it is, so it needs to be installed first, hence the second programmer
(aka. blue box).
Step 4: Libusb downloaded. (this is to install the 'driver filter' for the blue box.)
To get this to work on my windows 8 laptop we need to do a 'workaround' because windows 8 doesn't allow it.
the how-to we used is from http://fabacademy.org/archives/2013/students/nielsen.michael.hviid/class04.html
almost there...
Step 5: test in cmd if avrdude work correct. (no point in trying to make the fuse,
if something unknown went wrong with all the installing steps).
Test the command line: avrdude: which programmer (-c), which port (-P) and which part to use (-p).
Getting a 'failed' response back from avrdude is only logical. No worries, this is suppose to happen.
You have know confirmed that avrdude responded to your request, but because the request is not executable.
(because the programmer board still isn't installed yet) it says failed.
Step 5: Run cmd line 'make fuse'. (a installment window should appear).
Step 6: Run cmd line 'make program'.
Last step: de-solder SJ1 and SJ2.
Since we installed the program to the programmers board we don't need these components anymore.
The 2 components can be reused for an other board.
And create a flat cable with 6 wires (the colors don't matter)
Put the colored cable in the grey connector and then use the vise (the same little one used to hold the board while
soldering) to close the grey connectors.