Attempts 1,2 and 3. a Story About Failing and Learning Along the Way

This whole journey started back in 2005 when I accidentally found a little known website called sodaplay(unfortunately the soda constructor isn't available anymore). At the time there was a thriving community of very creative people making incredible animated models in this somewhat limited simulation environment. I spent many free hours designing all kinds of weird "creatures".

Skip forward 7 years, different study's and jobs and I was looking for a project to finish my graduation at Utrecht school of fine Arts. I was looking through my old soda play models one day and I remembered this strange worm like creature I made and after some quick cardboard mock ups I found a way to turn this simulation model into a physical object.

Step 1: Attempt #1

So, you have a cardboard model and a computer simulation, what's next?

An important part of my work as a artist has always been using technology in a way that makes sense to me. There are always better ways of doing things but for me it's the experiment that counts, not necessarily the result.

So I started out small, using laser cut plywood, cams and brake cables. The basic principle worked but of course there was way too much friction in the wood on wood joints and even a liberal application of grease couldn't make my carefully constructed model come alive. But there is no succes without failure and I learned a lot of valuable lessons.

Step 2: Attempt #2 the Real Deal

So with 3 weeks until my final presentation for my bachelors degree I decided to take the plunge. Go big or go home.

I took the lessons I learned from attempt #1 and designed a new version based on 1x1meter frames and a total length of just over 3 meters. Friction was the main reason #1 failed so I used steel guides and bushings on the critical parts but still kept the overall wooden aesthetic mainly because that is the material Im most comfortable with. I also ditched the external cable driven controls for a direct drive with gear motors and trapezoidal spindles. This did involve a lot more complicated electronics and another valuable lesson was learned. Motors with high torque and big stall current don't mix with simple motor drivers(L293D). So going back to the previous step I turned towards technology I was comfortable with and repurposed the cam system from attempt #1 into a sinusoidal switch switcher thingie. Time was running out for my final presentation and in the last few hours before the opening I managed to get a short 5 second movement until the motor couplers and power supply completely burned out.

Feeling a little disappointed and very tired I decided to leave it at that and just present it as is. That's where I started using the title Attempt #2, it's the attempt that counts for me, the result is just a accumulation of many different attempts.

Step 3: Attempt #3, Back to the Drawing Board.

So another 2 years passed, being busy with a lot of other projects and trying to survive as a beginning artist the attempt series got put on a hold. In the spring of 2015 I got contacted by KAAP, a art festival for children located in a old WW1 fort in Utrecht to make a new work. I immediately saw a great opportunity to get back working on the attempt series.

Both of the previous models showed great promise but ultimately failed on the friction of the linear guides and drive train. So I decided to completely redesign the frames from the ground up. I had done a lot more mechanical projects in the 2 years since my last attempt and also learned a thing or two about construction and electronics. The new frames used a much sturdier system of precision ground aluminium shafts and teflon plain bearings. I designed the frames in CAD and after many iterations I had a design I was happy with. The first tests were very promising and the new frames held up well to the big lateral forces of lifting the entire weight(30+kg's) of the ground.

Next up was the electronics. I started with a custom L298 driver board, able to supply 2A to each motor controlled by an arduino MEGA running a custom sketch I had written. It ended up working for a few seconds before the motors peaked at a huge stall current and I barely managed to stop the board from frying. So once again I took a step back and used technology I knew would work, relays! A complete redesign of the board and code and I finally had a semi working model. The issue now was the motors were under varying loads and after a few cycles the intended sinusoidal pattern turned into a mess. To try and synchronize the movement I added optical encoders to each motor and with the motors turning very slowly I got it to move forward. See the video for a short impression of the first baby steps.

Unfortunately after those first steps the hacked together arduino code I wrote got accidentally reset and the next day while I wasn't there it "attempted" to finish the half started cycle and literally tore itself apart at the joints. The main parts are still intact but with the exhibition in full swing there was no way of repairing the damage at the time. It was still a great experience and I hope I inspired a few kids to be creative and keep trying. Ofcourse I was sad that a simple overlooked line of code could have prevented this failure but this is not a story about succeeding, it's about "attempting". I hope this inspires someone out there to just keep trying, it's a cheesy quote but it's the journey that matters, not the destination.

And to end on another quote:

While(!(succeed = try()));

-Jochem

www.jochemvangrieken.com