Saturday, February 14, 2026
Why PIKA Matters
PIKA Is the proposed V0.05 XY Stage for RepRapMicron, and with luck will replace the V0.04 design. I've just updated maus_pika.scad on github, and have reasonable expectations that it should work. This isn't a beautification exercise, there are a number of important milestones wrapped up in it.
First, it's print in-place. This means less fiddling around to put a RepRapMicron together, which makes it more accessible to anyone who cares the make one. That's an important milestone, because it marks the point where μRepRap stops becoming a collection of prototyped Metriccano parts standing in close formation and starts to become an actual complete design. Much less of the mechanism relies an manual alignment and levelling because it's all printed in the right place.
Another issue is stability. Because there aren't so many fasteners, things that should have been linked before can now be joined up. They weren't linked earlier because adding all the fasteners would have taken up too much space and added far too much complexity. Look at the structure below:
Each set of axis flexures are all rigidly linked at the top and at the bottom. This reduces the tendency for the axes to twist when taken to their extremes of movement. The Z Axis Driver is integral to the XY framework, removing the need for a base board. Extra adjustable brackets are not needed to join the Axis Drivers to the axes. Every step there is one less thing to go wrong.
And finally, not only is this going to be easier to assemble, but it's going to be easier to remove the Axis Drivers from the main framework, which means it is going to be easier to transport without the weight of the Axis Drivers distorting or breaking the frame. In theory, if you screw it all together at the destination, you won't need to realign anything.
There are some disadvantages: If I need to make a change or break a flexure, I have to print out a whole new framework, much as per the old prototyping days on the OpenFlexure microscope. The framework will take about 12 hours to print, which is quite a bit. But hopefully the last couple of years of prototyping will have paid off though, and there won't be too much of that going on. Fingers crossed.
