String artwork is deceptively advanced. It appears fairly easy: it’s simply thread wrapped round posts (normally nails) that makes a sample, corresponding to textual content, a form, or a picture. However the math behind making a string artwork picture is extraordinarily superior, as a result of the accuracy of the picture comes from the density of overlapping strings and their angles. A gifted artist can try this by hand, filling in areas as they go alongside. However constructing a machine to provide string artwork is an enormous problem, as demonstrated by Paul MH.
On this video, Paul walks viewers via his journey to develop a string artwork machine from scratch. That got here with quite a few hurdles to clear and the mechanism to insert nails is an ideal instance. The machine itself is Cartesian, with linear X and Y axes similar to a CNC router. It operates underneath the management of a RAMPS 1.4 board, which is frequent for machines of this fashion. However grabbing nails and sliding them into the image’s base (foam board) was not a trivial process.
Paul initially developed a mechanism just like what you’d see on a nail gun, with a hopper on the shifting machine head that fed nails to an finish effector. However that was unreliable and in addition required the machine to hold across the full weight of all of the nails, which may quantity within the a whole lot. That weight made the machine sluggish, so Paul devised one other distant mechanism to feed nails for the top effector to choose up. That was nonetheless unreliable, so it required a webcam with a neural community that visually checks to see if a nail is current.
This machine additionally required a approach to feed thread, which was truly the best a part of the challenge. The thread comes off the spool via a PTFE tube out of a “nozzle” on the top effector. After attaching the thread to the primary nail, the stress is sufficient to maintain unspooling the thread as crucial.
The software program was Paul’s most spectacular feat. Utilizing the darkish magic that’s arithmetic, it takes a photograph as enter and calculates the thread paths that will create the very best visible approximation. It’s a little bit like dithering a monochrome image, however much more sophisticated. The software program then generates g-code for the machine to comply with in order that it may possibly wrap thread across the nails within the correct order whereas avoiding collisions.
We’re solely scratching the floor of the work that went into this machine, so be sure you watch the total video to see how Paul made this all come collectively.