Wow, it's been over a week since my last blog post and probably a day or two before that. I've fallen off the blog wagon hard and need to get right back on.
Why did I fall off? Well I blame the boy, had about three weeks where my fantastic son George decided to wake up every few hours, combine this with my fantastic wife going back to work as a midwife left me sleep deprived and only really able to perform the basic functions of this daily pursuit I've come to call work. Orders packed, emails replied, much coffee drunk.
That's not quite true though, another pursuit managed to fill the cracks and become an obsession. The MeArm. The MeArm is a low cost robot arm, build for my frustration with the uArm Kickstarter for an Open Source robot arm. Why the frustration? Well it was $185 for a product already available for $115. Touted everywhere as Open Source and yet a month on from successful funding to the tune of $251,887 and there's still no files. Not to say they won't come, but for me until it's open it's closed.
Never one to allow the energy of frustration to go to waste I set about building the MeArm v0.1 (files on thingiverse). In construction I borrowed a lot from the plot clock, actually starting with their files and stretching out the arms. There was no gripper but I was quite pleased with the results.
Not really a finished article and lacking in gripper - the best bit about a robot arm right? It did at least inspire Jack Howard to get involved in the project. Jack is a mechanical engineer with a lot of experience in CAD. He come to phenoptix fairly often on our mostly unpublicised Open Office Fridays. Jack has turned up with designs for the laser cutter before and I've always been impressed by the fact that they work first time. I do everything in 2D and mesh them together in my mind and by rotating the work on the page, relying on my access to the laser to tweak down to a finished product.
Very quickly Jack built was was to be considered the v0.2 (files on thingiverse). It was heavy weight and had a gripper, unfortunately too heavy for our choice of servos, but it was so solid compared to my v0.1. I could still add improvements so considered running my own development alongside Jacks. Setting aside a couple of days of hacking time where I could learn CAD to improve my design I ultimately ended up frustrated by my slow progress. I managed to add a centre piece for stability but it took me so long. I made a short video of the v0.2 and where improvements could be made for starters.
Sorry it's recorded hand held on my phone.
Really where the project gained momentum and started to take shape was after the feedback I received at the Linux User Magazine Pi Jam in Poole. Tim from PiBorg asked if me learning CAD was worth my time, when I was working with someone who was clearly expert in it. I'm too used to doing things myself, learning a new skill when its needed then moving onto the next problem that needs solving. Silly that it took this comment but it opened my eyes and I felt able to feedback what I felt was needed for the next iteration to Jack, who was able to add his own genius to come up with the v0.3 (files on thingiverse).
The v0.3 is the first kitable version and ticks all of the boxes I set out to tick. It can sell for a retail of around the £25 mark (available here) and could be made by a savvy school or maker for around £8. You can cut it from a sheet of A4 sized acrylic and build from common fixings (everywhere but the USA with their stupid Imperial measurements!). Hopefully it will stand up to 3D printing too. Kevin Osborn of Wyolum is currently printing off a first draft.
If he can track down some metric (M3) fixings it should work, it would be interesting to see how imperial screws fit though.
Last night I took it home to work on some code, but ended up just playing - I did at least make a video!
There was a little more product testing this morning too. This time from my wonderful daughter, who was quite impressed with it and wasn't able to break it! She load tested it more thoroughly than I did discovering it will lift spinning ballerinas and plastic ice cream cones but will not stretch to Mummy's keys.
So there you have it, why I'll been so quiet this last few weeks. Also new product - £25 robot arm! Let me know what you think!