Preamble (or is it Pre Ramble)
I’m experimenting with the format of my blog entries. I want it to be usable and consistent and if at all possible useful, educational and interesting and contribute to ending hunger and creating a lasting world peace but I’d be happy with interesting. I think a section approach is is a good way to go with specific content in each section. I’ll try to keep what I cover in each section related in some way to the main topic of the post.
For this post I’m breaking it into 5 sections, this one, one on my print experience because wow, it was not good this time, one with something interesting I learned in Fusion 360, one with my next steps and a closing remarks. I’m sure the sections will vary from post to post depending, and I won’t always give an inventory at the top, unless people want me to.
Web Cam Positioner – The beginning
I decided on my first 3D project. A system for mounting my web cam better than hanging it off something or duck taping it to the wall (or more accurately the closet door.) So, I jumped right in. I started off by opening Fusion 360 and making an arm. My thinking was that you need an arm for any kind of hand to grasp something. After a few attempts and modifications I had a nice looking I-Beam style arm with neat rounded edges, very cool. I was ready to print. There was a problem though. I had not included a way to attach the arms together in a way they could be used to position something. Basically, I had designed what would be plastic sticks if I printed them.
I spent the next hour or so going back and forth with different attempts at adding joints to the end of the arms which adjusted but I didn’t like any of the designs I came up with so I cleared the slate and thought about what I needed.
What I need is a way to connect 2 longish things (call them arms) together in a way I could adjust how they were angled relative to each other. There’s a lot of ways to do this, elbow joints (which is what I had already tried) and ball joints are 2 examples. After thinking about it a bit I thought a ball joint would probably be easier to create and more flexible. It can be angled in any direction regardless of the orientation of what it’s attached to and it can rotate. Ball joint it is.
A ball joint is basically a socket of some sort with a ball in it where not quite 1/2 the ball is exposed on one end. The ball usually has something attached, I’m going with a threaded rod, and there is some way to lock down the ball when it’s pointing in the right direction. My idea is to have a cup the ball sits in, on the back side of the cup is a threaded rot to attach it to things. A ball with a threaded rod will go into the cup, threaded rod facing out, and a threaded cap with a hole in it will slip over the rod and screw onto the cup holding the ball. To set the position, you loosen the cap a bit so the ball can move, set the position you want it, then tighten the cap which will lock the ball in place with friction.
Before I print balls with rods sticking out (look up, get your mind out of that gutter and I don’t want to see any inappropriate comments) and hoping the fit well when I insert them (this is getting worse and worse) in the socket I thought I’d try printing a simple container. A cylinder with threads on the outside and another with threads on the inside that should hopefully screw together (OK this is starting to get ridiculous.)
I exported the design from Fusion 360 in STL format, imported to Cura to slice (See the Fusion 360 – Threads section about my first attempt) saved the resulting code, logged into OctoPrint uploaded the file and hit print. I’m watching the print when I wonder why it’s printing a big flat interconnected looking thing. Raft! Cura was set to add a raft. I’m planning to do a series of posts on support and adhesion helpers where I will go into detail but a raft, as the name implies, is a flat platform a mill or so thick on which you print something. In my case it was a small flat item with some holes in it. Because it builds a multi layer raft to build the item on, it’s kind of wasteful and for this print totally unnecessary. I stopped the print.
Stopping a print is easy in OctoPrint, just hit the stop button. But it requires cleaning off the printer making sure the filament is off the extruder and that everything is copasetic. So, I did this, updated the model in Cura removing the raft and tried again. This time, no adhesion and I started to get a birds nest.
Stop. Clean. Reset.
I added a brim to the print, this causes the printer to print a couple lines outside the area of the items to print. I think it’s to ensure extrusion is working and sticking and gets things flowing nicely because it doesn’t touch the printed objects. I hit print at almost immediately had to hit stop. It was again trying to print a birds nest or tumble weed.
Stop. Clean. Reset. Extrude some filament to make sure it’s coming out as expected, check. Clean the bed with alcohol in case I got potato chip grease on the bed. Back to OctoPrint and hit print. Immediately started getting a birds nest.
Stop. Clean. Reset. Check settings in Cura. I had already set the initial layer to be printed with the cooling fans at 50% so the filament will be hotter and theoretically stickier and set the speed to 30 mm/s instead of the normal 60 mm/s. Hotter and slower should give it time to adhere the first layer and I already had it set at 1/2 speed and low cooling but I dropped the speed to 20 mm/s just to be sure. I saved the file, uploaded the changes to OctoPrint and hit print. I’m not sure how to type a long pause of frustration, but if I could it would go here. Birds nest.
Stop. Clean. Reset. Cry.
I had been in my bedroom on the second floor using my big 32 inch monitor while my printer is downstairs in the guest room. I unplugged my computer and went down to the guest room. I was tired of going up and down the stairs and I was going to get this print to work. In frustration I re-leveled the bed, which was pretty good, although I did make some minor adjustments and I cleaned the bed with alcohol again and restarted the print.
Sto… NO, it was working. Halleluiah!
I let the print run overnight. In the morning I had a nice looking print. Almost no stringing or artifacts. Given the difficult start I was half expecting to wake up to a partially threaded set of orange shot glasses with afros, so I was very happy.
There was a problem, however. The threads were so tight I couldn’t get them to screw together. I did a little light sanding, scraped between the threads with a knife and was eventually able to force them together most of the way. I’d call this a success. With a little more work sanding and scraping I could get this to work, but I don’t want to do that for every print. So, I though I’d try scaling the outer threads down or the inner threads up a bit to make a little more room.
Since I work from home right now and don’t have to commute I had figured I had time for a quick change and start the print before I start. I opened Fusion 360, opened the project selected the part which has threads on the inside, lets call this the cap, and scaled it by just a little bit. I think it was 1.05, which should 5% and saved just that piece as a new project, then exported to STL and into Cura, uploaded to OctoPrint and hit print. To my surprise it printed without any issues. I watched from my desk via the webcam streaming in OcotPrint as I worked and when it was done I got it and it screwed toether like they were made for each other. Which, they were.
I later noticed a little issue. By scaling it up a bit, you would expect it to be bigger, which it is. But, the scale option in Fusion 360 has 2 modes, uniform and non-uniform. Uniform scales all axis the same, non-uniform let’s you control the x, y and z axis scaling seperately. For my test I scaled only the x so the lid is bigger but it’s also not exactly round any more. Something to keep in mind when scaling round things.
I included pictures of the the results. I now have a round threaded box.
Fusion 360 – Threads
What’s next
As I write this I’m already printing a first attempt at the next phase in my efforts, a ball joint. My plan is to make a system of components which can be connected together, screwed together to be precise, to create stands and supports for not so big things. And if they scale, maybe bigger things? My next entry will talk about my first attempt at a ball joint..
After thoughts
I’m finding the process of design both exciting, interesting, challenging and frustrating. I spent several hours in Fusion 360 making things, destroying them and starting again until I felt like I had something usable. But when it comes together and the thing I thought up, designed and finally printed works it’s quite rewarding.
I’m not totally inept at making things, although my track record is a bit hit and miss. However, so far with this I feel like I’m able to do good work. I think it’s because I can build something several times and measure, remeasure and if I mess it up, delete and start again.
I’ve been in IT a long time and a software developer on and off for most of that time. I remember something a mentor of mine said back when they still had leaded gas.
If engineers built bridges like programmers built software, they’d build 5 feet and drive a car off, build 5 more and drive a car off until they got across.
Someone whos name I don’t remember but gave good advice…
Maybe I’m comfortable with this because I can do just that. Build something, drive it off a cliff (figuratively I live in Virginia there are no cliffs, big drop offs maybe, but you need mountains and there are none east of the Rockies. And don’t let anyone tell you there are.) and if there’s a problem, make an adjustment and try again. I find myself doing this virtually in Fusion 360 then, as with the threaded cups, with the printer.
Great timing, it looks like my next print is just about done stay tuned to find out how it went…..