One of the most time consuming and tedious parts of any project is taking digital CAD files and turning them into something physical
Modern CAM such as Fusion 360 are awesome but there is still a lot of steps required to even make a simple part such as a square with some accurately drilled holes in it.
I will often completely avoid using CNC's and do it by hand because after all, it is just a few holes? right, but even this can take a painfully long time and more often than not the holes don't even line up, thanks to the not so straight edge of the work piece ......... (Nothing a bigger drill bit can't fix )
The Idea behind the M1 is that anyone with a smartphone and the device will be able to quickly and easily fabricate 2D parts without needing a full blown CNC machine.
Though I've been working on the device for a while now I thought it might be nice to document the process for anyone that might want to build their own.
The Inspiration for the M1 came from thinking about the difference between polar and cartesian coordinate systems and how they are applied to real world machines.
Cartesian coordinates are relatively intuitive both conceptually and from an application standpoint, any point in the plane is defined by 2 discrete coordinates on the X and Y axis.
In application cartesian machines are super simple. If you want to move say a drill or pen to a point X = 50mm, Y = 50mm from a known origin all the machine needs to do is move 50mm along the X and Y axis.
For more info about cartesian machines this is a great starting place.
Where cartesian co-ordinates define a position on a plane using X and Y points, polar coordinates utilize angle and distance to define position.
In terms of application machines that utilise polar coordinates strictly in terms of angle and distance are not very common. This is primarily for 2 reasons.
High precision of the encoder required to measure the angle (I’ll talk about this in a later log)
Polar machines are essentially very long levers which means even a very small force at the end of the arm requires a very large counteracting torque at the point of rotation.
In summary a purely polar machine would need a massive, very accurate motor……. Not an easy combination.
But what if we could remove one of these two requirements? What if we remove the motor and instead move the arm by hand while we measure the angle. We could accurately go to any point simply by setting the distance and rotating the arm until it was perfectly aligned with the point. Easy right?
Not exactly. Though setting the distance is easy with something simple such as a belt drive. Very precisely positioning the arm angularly is another matter, even a very small angular misalignment of 0.1° means we will miss the target by a significant margin. This misalignment also gets worse the longer the arm gets.
For example if we were trying to position the arm over the point X = 300, Y = 300 in Cartesian coordinates we would need to rotate the arm to exactly 45°. But if we were off by even 0.1° (45.1°) that would mean an absolute misalignment of 0.74mm..... not cool.
To solve the misalignment problem I am going to add an axis (B axis) perpendicular to the A axis that can move slightly back and forth. With this small amount of additional play in the B axis we will be able to compensate for any angular misalignment.