Mov'en Along

Auto Auction needs to have a hand with articulated finger (similar to Think Machine).  The pose of the hand is very specific so I took some time to make sure my hand was positioned correctly. The block at the end is a proxy for the bell so I could get my finger at the correct height

I used a scrap of MDF to position my hand...

...then built a box around it with 1x6 boards

I put small brads at each fingertip so I could feel for the location even when the box was full of alginate

I'm sorry, can you get the phone?

After removing my hand I tilted the mold upright and filled it with plaster

It needs some cleanup but this will work

Once I had my hand cast, I set the plaster model aside for a few days. It needs to dry out thoroughly before I can continue. The plaster is only an intermediate step. Eventually it’ll be cast in polyurethane. On to other things, like….

Making a new bell
The base for the bell I’ll be using is pressed sheet steel. Kinda flimsy and really no way to attach it to the table top. I doodled around with a couple different solutions but eventually I just re-made the bell using better materials.

The sold modeling program let me experiment with subtle changes in geometry

The new bell next to the stock bell

Assembled and working

Relatively straight forward but it took me about a day-and-a-half to design and machine the parts. From there it was time to move on to:

Wiring
I installed all the solenoids in the new steel frame. Of course the solenoid wires aren’t long enough to reach the Arduino so I had to solder longer leads and connectors to each. Next I installed the seven magnetic sensor used to zero each tumbler. Lot’s more fiddly soldering, crimping, and heat-shrinking.

Solenoids wired and installed

Sensors wired and installed - what a mess

Getting better

The sculpture needs 12 volts for the solenoids and 5 volts for the BeagleBoard. I specced a power supply early in the week and got it on order - arrived this afternoon. Looks like I’ll have to move it to a different location. Solenoid wiring interferes with the location I originally planed. I’ll pick up here next week. There's still a lot of work left, but it's starting to take shape.

Hammering Away

I began by welding the steel frame I cut last week. I intentionally made the tabs a bit long so I could grind them flush. The joinery I cut into each piece made it simple to line everything up for welding.

Welded and ground

Onto making the covers. I don’t have much experience working with sheet metal. I’ve been vaguely dreading it. I started with the bottom cover - the easier of the two. I spent some time running 22 gauge steel through my slip roller and it started to take shape.

Once this was roughed out, I moved on to shaping the upper cover. This one was much more challenging. I began by cutting out square windows for each tumbler.

Quality time with a hand file 

Then back to the slip roller and some fine tuning with a panishing hammer. This upper cover came together very quickly - just a few hours. That almost never happens. I was rather pleased with myself… until I realized I had folded the piece backward.

Not a bad fit straight from the slip roller

I closed up these gaps with a panishing hammer

The better part of a day lost because I lost focus and rolled the part backward.  That's really unfortunate.  Frustrated, I knocked out another.  Scrapped that one inside a couple hours.  At least I was getting faster.  I managed to work out all my troubles on my third attempt the next day.  I cut both covers a little over sized, rolled the edges slightly with a panishing hammer to tighten up the curves, then ground the edges flush with a die grinder.  


The windows for the tumblers began as rectangular. While making the part, I decided it might be more interesting to make them round. I also experimented with different tumbler and number colors. This is where the solid model starts to payoff.

My original design with square windows

Round windows with black and gold dials

Round windows with the original white and black dials 

I decided I preferred the white tumblers with round windows which has the added benefit of making the part slightly easier to fabricate.

Oversized panel rough ground to size

Finished cover with two scraped attempts

With the covers done, I pressed sleeve bearing onto each of the seven tumbler ratchets. I haven’t been able to get the solenoids installed yet, but the ratchet mechanism works very well when I run it by hand. Exciting.

Ratchets and tumblers installed

Utility plate for various controls

Finally, I made a small plate for mounting all the electronics. Next week I’ll mount the electronics and finish wiring.

Back Blogging

I’ve been away from my blog for a while. There’s a lot to catch up on but, rather than making one egregiously long post, I’ll just work recent progress in my future posts.

ToDo was installed at the San Jose Institute of Contemporary Art. It ran for about a month and developed an impressive pile of paper. I’m please with how this piece turned out and it was very well received. I installed it about a month ago at the home of a collector in Los Altos.

About three weeks of paper

Once everything was finished with ToDo, I started working on finishing the edition for Gimme. I’ve make great progress but that will have to wait for a future post. For now I’m focused on...

Auto Auction
It’s been a while since I worked on this sculpture. It’s a machine designed to sell itself on eBay and there have been half finished pieces of the machine laying around my studio for months. Time to finish this piece. In my last post for this sculpture, I had just assembled my custom electronics - a sensor/solenoid shield for the Arduino. A couple of minor tweaks were needed to get everything working the way I wanted, but the electronics have been working very well since. Each tumbler has a magnetic switch so that the tumblers can be zeroed on power-up.  I machined a bar that positions each of these sensors and modified each tumbler to hold a small magnet - then started feeding wire around the machine and soldering connectors.

Each hole holds a magnetic switch

I solder each crimp connector so I'm sure nothing will come loose

The round dot on each tumbler is a small magnet

Solenoids installed and wired

Almost ready for a test run

Programming
My friend John has done all the programming on this piece, which is lucky for me. It’s become far more involved than I initially anticipated. I’ve had numerous conversations with both John and Martin about how to approach controlling this piece and we decided to use an Arduino and BeagleBoard. The BeagleBoard connects to the internet though WiFi so it should be very clean and transparent for anyone using the machine. I mounted the temporary masonite prototype, got everything wired together, and left it with John for a few weeks so he had hardware to test. While he was working on programming, I moved on to fabricating the frame in steel, but first...

CAD
I’ve never been really happy with my CAD (Computer Aided Design) software. I’m using a 2D program on a Mac. It’s cutting-edge circa 1990 and I’ve been interested in moving into a 3D package for several years. I’ve done some 3D modeling for my sculptures but never designed anything from the ground up in a 3D package. Well, my CAD seat expired in November and I decided it was time to make a radical change. I’ve been determined to stay with the Mac and I was excited to find that AutoDesk has finally brought AutoCad to Apple. Of course, my particular computer is to old to be supported so I’d be looking at upgrading my whole system. More involved and expensive than I had hoped. After many frustrating days with various trial software, I decided it was time to cash it in on the mac. I went and bought a PC and installed AutoDesk’s Inventor. It took me about three weeks to get comfortable with the software but it looks very promising. I’ve had to rebuild parts of Auto Auction in the new software. Not particularly efficient but a good way to learn.

Water Jet
I’ve been having parts for my steel frames cut at a local vendor for several years. I email them CAD files and they call me to pickup parts. It’s been great but it’s also been expensive. Believe me, they’re worth every penny. I’m just out of pennies. A local facility, TechShop, opened down the street and I was amazed to discover I can have access to a huge water jet. This particular machine is a beast. I’d guess it’s about 10 by 15 feet. It blasts water mixed with garnet at over 50,000 psi. I’m told it can cut steel up to 6 inches thick. I signed up immediately. So did everyone else in Silicon Valley. It took me about a month-and-a-half to get cleared on the equipment because it’s been so backlogged. The day after I was cleared to use the machine, I dropped a quarter inch plate of cold rolled steel on the bed and cut it to pieces.

It’s a complicated machine and it’s going to take me a while to get comfortable using it but I’m making (mostly) usable parts. Last week I went back and cut the ratchets from eighth inch steel. I’m still trying to figure out how to hold raw stock down. Cold rolled steel seems to warp during cutting. I’m assuming there are internal stresses from fabrication that manifest during cutting. This is problematic. If it warps more than a quarter inch it’ll catch the head of the machine. Best case, it moves the material and scraps parts. Worst case, it crashes the machine and damages something expensive. How to hold it down isn’t immediately obvious to me or anyone at TechShop. I’ve been Googling around. Figure I’ll machine some kind of clamp fixture next run.

Everything lines up without clamps

Oops - something moved

Next time I'll find a better way to hold down my material

I leave this week with all my steel cut and ready for welding. Next week, I’ll weld the frame together, make some covers, and hopefully start painting.

Photography

With my deadline rapidly closing in, I put the finishing touches on ToDo and dove straight into documenting the piece. Good photography is a devil. I usually shoot on a black background which, I’ve been told, is particurlarly difficult. Here’s my thinking: I don’t want to shoot on a colored background - too distracting. That leaves me with black, grey, or white. Objects on a white background always read “internet” to me. I don’t want it to look like my sculpture is on Amazon. So...that leaves me with black or grey. I’ve shot on grey before (Auto Masochist for example) but I’m always frustrated with it. The paper backdrops I’ve used always seem to have some kind of color cast - often magenta - and then you have to worry about how the light falls on the background so you get a nice gradient. I really don’t want people thinking about the space surrounding the sculpture. Okay, so that’s leaves me with black. That’s how I got here initially and every time I rethink background options I come back to black. To me, it’s the best way to isolate the piece and really focus attention on the sculpture.

Now, getting the background to completely drop out is surprisingly difficult. For years I’ve shot on black velvet. It sucks up light well but even the slightest wrinkle reflects light and becomes really distracting. You can tweak around in Photoshop but it’s always better to just get it right in the camera. A couple other problems with velvet: it’s expensive and it’s difficult to get in large pieces. All this was a huge problem for me when I was shooting my last piece, Auto Ink. Someone turned me onto duvetyne. I can buy it at a local theatrical supply and it’s dirt cheap. I bought a huge piece to shoot Auto Ink and it worked really well. I figured I had found the perfect solution so when it was time to shoot ToDo I setup a large piece of plywood in my studio, draped it with duvetyne and mounted ToDo to the wall.

My original plywood backdrop

Turns out mounting to the wall is a problem. I couldn’t get any separation between the wall and the piece and whenever I got a good exposure one the piece the background became a washed-out black. I fought with this for a day and eventually made some spacer blocks to float it off the wall. Still crapy, washed out photos.

Then things just spiraled out of control. I figured the best way to get the background to drop out was to really get some significant space between the piece and the backdrop so I pushed the plywood aside and mounted a beam to by workbench. I figured I’d mount the piece to this beam and poke the end of it through the backdrop.

Adjustable beam lets me move the piece after the backdrop is hung

I attached the duvetyne to a piece of electrical conduit I hung from the ceiling of my shop. I wanted the beam to stick out so much that I had to make a special bracket to offset the piece so I could keep the beam behind the sculpture when I was shooting a three-quarter view.

Backdrop hung from the ceiling

The sculpture mounts on the left side so the beam is hidden during shooting

This got so ridiculously complicated but I really don’t know what I could have done to simplify things. Probably should have stuck to the velvet.  Next time.

Anyway, I finally got the shots I needed - stills and video. You can see them here on the website.

Home Stretch

ToDo is almost finished. I need this finished and photographed before september 9th, so I’m right on schedule.

I began the week with a problem on the Y-axis. The paper roll is 128 feet long, so if it doesn’t track properly I can develop some nasty problems.

It starts as an innocent looking bubble of paper...

...but gets ugly

The knurled drive roller just wasn’t gripping the paper well enough. I began by increasing the weight pressing on the roller – which improved things but didn’t eliminate the problem. Eventually, I decided I had to scrap this roller and make a new rubber driver roller. It took a bit to sort out exactly how to make this. In the end, I machined an aluminum core on which I pressed a piece of rubber tubing. A little polyurethane glue helped the tube slide on and helps keep the rubber in place. It turns out that the tubing isn’t at all consistent in thickness which makes a wobbly roller. Luckily, rubber machines well with a sharp lathe cutter so it was easy to true the outside diameter. It does grab the paper much better but it still didn’t eliminate the problem. (Obviously, since you can see the new roller in the crash picture above.) I finally tracked the problem down to the brass bushings that support the paper roll. I found that when the pen was drawing an up-stroke the roll could slightly climb out of it’s supports and allow the paper to slip. I changed the angle on these bushings and everything came to together. Whew.

A couple of posts ago, I described some problems I had making the pen holder. Time to sort that out. The new design uses stainless tubing with and upper and lower piece pressed into each end.

New pen parts

New pen pressed together

I also wanted a better looking jam nut for the pen. Functionally, there’s no problem with the nut pictured above but I can make something more aesthetically harmonious.

Easier said than done. I kept machining the flats wrong...

New pen & nut installed and working

A sample from the machine - looks just like my writing

Everything seems to be working very well. Time to break it down and paint everything.

There are a couple of loose ends on the wiring. I added an on/off switch and cleaned up some of the motor and switch wires, then began final assembly.

Should be done next week. Look for good photos and, hopefully, a video of the completed piece.