<h1>Computer Controlled Machining</h1> <h4>Wednesday March 07, 2018</h4> <div style="width:30%; float:right"> <img src="media/week8/plastics-engineer-darshan.jpg" style="width:100%; float:right; padding:5px 5px 5px 5px;"/> <legend>"Plastics Engineer Darshan evalating my use of plastics. <br>He approves."</legend> <img src="media/week8/fab-soccer-how-to-play-soccer-with-almost-anything.jpg" style="width:100%; float:right; padding:5px 5px 5px 5px;"/> <legend>"Fab Soccer - How to play soccer with (almost) anything. -At the SuperFabLab"</legend> <video width="100%" autoplay loop controls muted><source src="https://i.imgur.com/vtCnGvo.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> <legend>"I'm so sad to leave this amazing place!"</legend> <img src="http://www.dailyicon.net/magazine/wp-content/uploads/2009/05/nasa01dailyicon.jpg" style="width:100%; float:right; padding:5px 5px 5px 5px;"/> </div> ##What's on my mind "To go places, you first have to leave places." -something Darshan once said to me. I hadn't mentioned yet, but I plan to head to Iceland in April. I've been generously offered a place to study in Vestmannaeyjar with Frosti (I think Bas is there as well). In my time in Dubai, I've had the privilege of meeting some of the great names in Fab Academy; Wendy, Francisco and Fiore. I'm torn about when I should go - last weekend I had an incredible time in Al Ain with Abdulla, and in two weeks time Zahrah is taking us to Fujeira with some of the other classmates. (edit: Zubair also took us to his town Ras Al Khaimah that weekend too!) We've been having such a great time and spending a lot of time together, but all good things do end. I'm just not feeling like it has to be so soon! In the spirit of making tough decisions, I've also decided to leave a few things behind in my final project. Considering how busy I've been with the FabAcademy work and just generally trying to enjoy myself in Dubai, I consulted a genius friend Brett Downing. Brett and I met back in Perth at an ancient hackerspace called the Artifactory. We happened to go to the same university campus and worked on different projects within the same building. I'd always think that what I was working on that week was pretty cool until he showed me what he had been up to! 😂 He's now been living in the Netherlands for the last year or more, and it's because of him I managed to get a job as a lecturer and fund my FabAcademy adventure! We went over options of scaling back the power expectations of the project. Given the timeframe and how arduous it is to get parts in we figured it was reasonable to take it back a notch. Brett suggested either using a stepper or significantly lower kv motor than the one I was planning. Also we considered swapping LiPo's out to something more forgiving and could charge with more current like NiMH's if possible. In order to make up the extra current requirements we could use 10F supercapacitors perhaps. After reconsidering and letting go of some of my initial performance expectations, I decided to take the opportunity to harvest some expertise around the lab. Gabriel from Brazil had just started working in Abu Dhabi and was visiting the lab a few days of the week. He's an experienced electronics engineer who showed us the basics of Altium and using oscilloscopes to analyse a circuit. <div style="width:30%; float:left"> <img src="media/week10/oscilloscope-gabriel-game.jpg" style="width:50%; float:right; padding:5px 5px 5px 5px;"/> <img src="media/week10/oscilloscope-gabriel2.jpg" style="width:50%; float:right; padding:5px 5px 5px 5px;"/> <img src="media/week10/oscilloscope-gabriel.jpg" style="width:50%; float:right; padding:5px 5px 5px 5px;"/> <img src="media/week10/oscilloscope-gabriel3.jpg" style="width:50%; float:right; padding:5px 5px 5px 5px;"/> <legend>"Gabriel taking us through oscilloscope games!"</legend> <video width="100%" autoplay loop controls muted><source src="https://i.imgur.com/KAfnVkx.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> <legend>"N-FET controlled constant voltage boost converter temps holding steady."</legend> </div> He came up with this cool game where you make a random function on the function generator and muck up all the knobs on the oscilloscope, and the goal was to (as quickly as you can) adjust the oscilloscope to show the output of the function generator. Zahrah and Zubair in particular loved this game, and it was a great way to find your way around the device and methodically dial in on a signal without knowing what it's meant to look like. We got excited about the output stage of my project and so we took the time to breadboard a switching CC/CV boost converter for charging a capacitor. The circuit we built was capable of up to 30V and around 1A, and I learned SO much about circuits that university really never prepared me for! This circuit would not only be useful in my project but in many other projects I have on my backlog. I told Gabriel about my concept of makers being like wizards who practise a whole bunch of spells to make cool things happen like "Convert Voltage" or "Interpret Serial". It's this awesome sharing of our experiments among spellcasters that I find most enjoyable about learning electronics. A week later I jumped in on question time with Neil at the lecture and he mentioned that the project was wholesome and to not build things that don't provide much value (like lithium charge controllers that can be sourced on Digikey). He also said that there had been a few wind turbine projects in the past but none of them were able to generate an appreciable amount of power. That was awesome because I absolutely love excessively high powered electronics, the question remaining whether I could build a turbine that could do it. He did say to drop the gearbox, but that's a thing I don't think can be avoided given the motors that I have access to. Wendy and I had a moment of frustration not having the right tools for machining the shopbot bed (read below) and for a few hours after that I was inspired by another potential final project. I've always been inspired about being able to do what I do anywhere at any time, which could be a subset from the FabLab concept as "How to make (almost) anywhere". One of my key inspirations is machining and manufacturing things in space, for example the Moon, an environment where you would have very limited tools and you'd need to be EXTRA creative about making things happen. In particular it is an environment where 3D printing would be an indispenable technique, where the only materials you would have are soil and the occasional stock shipped up from Earth. I wouldn't forget about subtractive manufacturing however, which in general produces faster, stronger and more resilient parts. On board the ISS they use a <a href ="https://www.popularmechanics.com/space/a24639/space-drill-used-by-nasa-astronauts/">pistol-grip rotary tool</a> (i.e. multi-function power drill, more pictures <a href="https://www.dailyicon.net/2009/05/nasa-space-tools-designed-for-hubble/">here</a>) that is versatile and presumably worth it's weight in rocket propellant. In those few hours of distraction I conjured a concept for a hand-held recoilless CNC machine, in a similar form factor to that "Space-drill". The idea is that you pin or magnet it to the side of a satellite, wall or whatever, and it can drill any size hole, or machine almost any shape. However in order to minimise the footprint of the machine it wouldn't use traditional Cartesian layout, it would make use of polar-coordinate systems similar to the <a href="http://www.pocketnc.com/">PocketNC</a>. So if Wendy and I ever have that situation again, we could machine (of course it would take time) those square holes we needed. I'm calling it the Space-mill, and I'll do up a rough CAD model to explain how it works soon. Kind regards, Carl <div class="row"> </div> <div style="width:100%" align="middle"> <img src="media/week8/wendy-shopbot.jpg" style="width:50%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/under-shopbot-bed.jpg" style="width:50%; float:left; padding:5px 5px 5px 5px;"/> </div> ##The menu for this week: This week's special dish was Computer Controlled Machining. In the Dubai SuperFabLab we have a brand new full-size ShopBot, which in classic fresh-out-the-packaging style hadn't been set up yet. The instructions this week were to quite literally 'make something big' so most people make some kind of furniture. ##This week's designs This week's design is a bit of a sad one, since I am moving away I'll need to be able to bring my tools and project with me on my travels. In order to do this I've decided to try to machine a rack for my luggage to wheel around on. It didn't make much sense for me to be making furniture since I neither live permanently in the Emirates nor am I staying foreseeably for much longer. The luggage has a few requirements, and because of the time I've had to compromise on how well I expect it to perform: + It needs to be folding, having a long mode for rolling it around like luggage and having it folded out will make it easier to use, and a folded mode for taking onto an aircraft and into a taxi. In the folded mode it will be less susceptible to damage (i.e. snapping at the hinge) + In typical Carl style, it will have to have an element of ruggedness to it. This is self explanatory - I go into places where the sun doesn't shine sometimes and so I'll need my luggage to keep it together. + All high performance features need an upper limit, and for this case (pun intended) the design will need to minimise weight. Of course, because this is supposed to be portable and go into an aircraft it will need to take up as little precious weight allowance as possible. Ruggedness and lightweight-ness is generally a Pareto tradeoff so I'm gonna need to think smart about how to work this. + Lastly, it will have to be simple and reliable to use. I don't want to be fraffing around at the airport arguing with this device about which mode it wants to be in so it will have to work when I ask it to and stay standing up when I'm doing something else. <div style="width:30%; float:left"> <video width="90%" autoplay loop controls muted><source src="https://i.imgur.com/zEUja4Z.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> </div> ##Prep for work In order to do shopbot week we <a href="media/week7/reminder.jpg">needed to set up the shopbot again</a> (You can do it without this, but your design is pretty much limited to a rectangle measuring 1220x2440 across.) The shopbot had actually been used before, but the bed installed was only ever a temporary solution. Understandably, the shopbot had been moved and was in a temporary location. Not so understandably, the workbed was glued in with <a href="https://www.google.is/search?q=world%27s+worst+plywood&oq=world%27s+worst+plywood&aqs=chrome..69i57.4479j0j4&sourceid=chrome&ie=UTF-8">absolutely ZZ-grade plywood</a> and not even surfaced flat. Because the virgin shopbot hadn't been adequately bedded yet (yes, pun intended) we needed to grab a couple things from the local: + Drill bits + Fixing screws + End mills + 12mm 9-ply 1220x2440 birch + 18mm 13-ply 1220x2440 birch for bed And yes you read that right. We didn't even have drill bits yet. <div style="width:30%; float:right"> <video width="90%" autoplay loop controls muted><source src="https://i.imgur.com/wbfqMDS.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> </div> ##Setting up the Shopbot! After prying up the old, glued in workbed, Wendy and I measured out the holes for the new 18mm base sheet. To do this Wendy drilled the holes from underneath and we measured out from the top. After a few checks we agreed it would be safe to assume the machine is straight enough, so I drew up a dxf in Fusion 360 for the holes. Then we put the holes layout into V-carve, which is the shopbot toolpath managing software. Initially we "pecked" the bed to cross-check my dxf measurements and found out it was a little off (circa ~2mm X-axis, ~1mm Y-axis). But after some very fine guesstimation we managed to line up the holes using V-carve. Since we didn't have the right hand-tools for machining the squares or countersinking the dome-head bolts that were available (which lead to the incipient concept of another final project, see the Preamble up top) we used the shopbot itself to locate and machine the holes. Once the first 18mm sheet was in place we laid down another 18mm sheet as the plunge bed and then we could secure the sheets we intended to use (12mm) on top of that. This bed was the one Wendy surfaced, because if we surfaced the first sheet there was the possibility of hitting a dome-head bolt if the shopbot's levelling was too far out of whack. Fter surfacing it appeared that someone had crashed the spindle previously since there were small ripples in the X-axis. This could affect the fitment of any joints cut along the X-axis Later on I did a test for this with press fit joints along the X- and Y-axes and it wasn't noticeably problematic. <div class="row"> </div> <div style="width:100%" align="middle"> <img src="media/week8/surfacing-bed-shopbot.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/shopbot-fun.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/surfacing-bed-shopbot-angle.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/tool-config%209.35.00%20am.JPG" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> </div> <div style="width:40%; float:left"> <img src="media/week8/shop-bot-plywood-bracket-test.jpg" style="width:90%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/x-y-joint-test%209.35.00%20am.JPG" style="width:90%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/shop-bot-x-y-calibration.jpg" style="width:45%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/shopbot-press-fit-mortice-tenon.jpg" style="width:45%; float:left; padding:5px 5px 5px 5px;"/> <video width="45%" autoplay loop controls muted><source src="https://i.imgur.com/WznhR16.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> <video width="45%" autoplay loop controls muted><source src="https://i.imgur.com/3KmJ0VJ.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> </div> ##Setting up V-Carve for the Shopbot The shopbot is a machine I have particular respect for, given how powerful it is and also how fiddly stepper motor drivers can be. I'd completely cooked out two stepper motor drivers on a large machine I was working on at FabLabWA (a project I took over from Brett) and so I was particularly careful about sticking to the recommended loading for this machine. Here are the steps for setting up files to go to shopbot + Open yer file + Check the following: + Origin (since there's no limit stops on the machine if you get this wrong you could blow **everything** up) + Page size/orientation (landscape preferred) + Thickness + Find and acquaint with the 'Join Open Vectors' button + Just in case the shapes aren't closed. This will join any open vectors, like it says on the label + Make a new layer for screws + Use circles to plot screw holes + Place fillets + T-Bone - there's two types of T-bone, concealed and flared + Dog bone + Pass depth roughly diameter of cutter unless you're comfortable with the type of wood + Other defaults (These were starting points. We actually moved to even slower feed and RPM in the end because this particular machine was making noisy noises) + Stepover ~66% + Spindle RPM 12000 + Feed 60mm/s + Plunge 30mm/s + Like the laser cutter there's a few different types of operations (and you should do them in this order) + Drill - drills + Pocket - cuts a pocket from a shape. Useful for removing large internal sections of a vector, but takes the longest. Equivalent to the 'raster' function of a laser. + Profile - Cuts along the vector, and you should make sure you get the correct side of the line. You should also add tabs here to keep your workpiece secured otherwise the cut could end badly. + When doing through-plunge - give it +0.5mm through the bed because the thickness of the wood is only nominal! + We ended up halving the pass depth, feed and plunge after listening to the sound the mill bit was making. + For my cuts we also used a 6mm two-flute down cut end mill. <div class="row"> </div> <div style="width:100%; float:left"> <img src="media/week8/hard-case-handle-test.jpg" style="width:17%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/hardcase-bracket.jpg" style="width:17%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/hardcase-handle-test-2.jpg" style="width:17%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/hardcase-wheels-test.jpg" style="width:17%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/laser-test-handle.jpg" style="width:30%; float:left; padding:5px 5px 5px 5px;"/> </div> <div style="width:30%; float:left"> <video width="90%" autoplay loop controls muted><source src="https://i.imgur.com/laWCE9u.mp4" type="video/mp4"/>Your browser does not support the video tag.</video> </div> ###Single-sided machining rack with handle To design the part I used Fusion and did test-fits with the laser cutter. The main part I needed to test was that the swivel-clip-handle idea would work. It actually turned out quite satisfying. Next I drew the main body that would attach my hardcases together. I'm not sure if everyone in the class has access to Fusion's generative shape optimisation function but since I wasn't super sure what pattern I should use I decided to see what Fusion thought. It gave me some pretty funky stuff which I cleaned up manually, but my self esteem wasn't high enough at the time to rock such an organic pattern. So I decided to leave it as hollow and if it breaks I guess that would be one way to learn. <div style="width:70%; float:right"> <img src="media/week8/fusion-load-optimisation-7.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/optimised-part-generatino.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-optimiser-bodies.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-generative-cad.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-optimiser-part-2.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-opti-part.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-optipart3.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/fusion-generative-cad-2.jpg" style="width:25%; float:left; padding:5px 5px 5px 5px;"/> </div> ##Things I've given up on (for now) ###Cutting a chamfer I really would have liked to chamfer all the edges on the design - chamfers give a really good 'deliberate' feel to machined parts, and they also hide dents and scratches quite well. We did end up getting a V-cutter for the shopbot, but by the time we actually got the shopbot operational again it was about 10 days after the week ended. This, and there were six other people wanting to use the machine as well, so I decided I would maybe try it another time. ###Two-sided machining by Machining a Fixture I still have a hard time understanding two-sided machining so I was kinda hoping I would get the chance to flip the part, only as a learning exercise and open up to the possibilities of 2.5-axis designs and some 3-D machining. As mentioned above there wasn't really a lot of machine time available and I was a little demotivated by the design in the end so I didn't make it to this part. ##Update: Finished product in Week 11 Finally completed it when I got the right bolts! It worked almost perfectly as it was designed, however there were some things I didn't quite like about it. This lead me to almost entirely abandon the project, but I had a renewal of faith in <a href="assignment11.html">Week 11</a>. <div style="width:100%; float:right"> <img src="media/week8/luggage-rack1.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/luggage-rack2.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> <img src="media/week8/luggage-rack3.jpg" style="width:33%; float:left; padding:5px 5px 5px 5px;"/> </div> <div class="row"> </div> ##Design Files Design files for this week can be found at <a href="design-files.html">Design Files</a>.