Tuesday, March 21, 2006

error analysis: drill press

Our design required quite a few large, precise holes to be bored through pieces of ice. Some of the holes were to accommodate freely turning axles, some were to be bonded to axles, but all needed to be perfectly perpendicular. The best way to bore perpendicular holes is with a drill press, but many of the holes, such as those in the middle of large gears, required a drill press with an incredibly deep throat. Thus was Lars's mobile drill press idea born.

He removed his drill press from its stock pedestal and mounted it on a custom carriage assembly above a steel table. This allowed us to position the drill over any part of a large slab of ice. While the concept was brilliant, the hurried construction resulted in critical failures that impaired our ability to bore truly perpendicular holes.

First, a weld broke on one side of the carriage, most likely because of the quality of the scrap steel it had been made from. The drill press could still be positioned and operated, but the drill pulled away from the carriage every time the bit was pressed into the ice.

Second, the carriage rails rested on top of vertical pins at the edges of the table but were not firmly attached to the pins. Even when holding the broken weld together, this allowed the entire carriage assembly to come out of alignment under pressure.

Third, even if we had solved the first two problems, there was no easy way to set up the carriage such that perpendicularity of the drill bit with respect to the table could be assured. We would have had to first level the table (not a simple task on hard packed snow) and then carefully adjust all four pins to level the bit. If we had ever changed the height of the rails, we would have had to readjust all of the pins.

The mobile drill press also had trouble with vertical travel. We weren't able to drill deep enough below the carriage rails, but we probably could have corrected this by shortening the pipe supporting the drill and/or using bit extensions. The total travel of the drill press was about three inches, but a few of the parts to be drilled were four or six inches thick. We figured that we could flip parts over and drill them from both sides, but we hadn't prepared any kind of centering jig to ensure correct alignment of opposing holes.

All in all, the mobile drill press turned out not to provide any benefit over a hand drill, but the original idea still has great potential. With several months to come up with a better solution, I'm sure that next year's model will be much improved.

Monday, March 20, 2006

ice garden

Sharon and Lars have completed a delightful Ice Garden in the Fairbanks Open Exhibition. I'm especially impressed by the sunflower. Wow!

After Lars returned from his extended job on the Kamchatka Peninsula less than two weeks before the Single Block Classic, he immediately enrolled in an ice carving class. Sharon joined him for the class because it was the only way she could spend time with him during his frantic preparations. It's great to see that they've put their new skills to good use.

Monday, March 13, 2006

error analysis: weather

We knew the ice would be brittle at cold temperatures, but we didn't expect the problem to be so severe. The weather during the competition (9:00 AM Tuesday to 9:00 PM Thursday) was unusually cold. Normal high temperatures in Fairbanks in early March are 15°F or so, but, out of three days, only Thursday had a high temperature above 0°F. Working the ice that day seemed incredibly easy after our experiences the previous two days.

At 20°F, carving ice with appropriate tools is like carving a block of cheese; at -20°F, it is more like carving glass. Miniscule impacts can result in huge fissures, and contact with liquid water yields a lattice of tiny surface fractures at best, more often causing substantial cracks. If you have ever heard ice cubes crack after being dropped into a beverage, you can imagine the effect on a larger scale. We learned a fun trick: Take a fair sized (fifty pounds or more) slab of ice at -30°F and toss a bucket of warm water over it. Spectacular shattering! Unfortunately the effect meant that we were completely unable to use waterglue at the extreme temperatures. Even around 0°F we had to be very careful, making sure the water was no warmer than 32°F and spreading it thinly.

Cold temperatures are hard on tools as well. Gasoline powered chainsaws are particularly difficult to operate, but we were able to minimize the problem by constructing a small warming tent for our tools out of a blanket, a steel chest, and a small space heater.

To a certain extent, we can chalk up our temperature troubles to bad luck, but there are a few things we could do to better handle the situation. Now that we've learned which tasks can't be accomplished at extremely cold temperatures, we can plan those tasks for the warmest times of day. We can also maximize the amount of daylight work hours. Spending the entire day of Tuesday in the shop eliminated a full third of our (relatively) warm temperature hours.

We've also talked about trying a larger scale warming tent, possibly encompassing some of the pieces of ice to be worked or even ourselves. Taking the idea to the extreme, we might try warming the entire 7600 pound block of ice or even constructing a tent around the whole work site. These efforts would certainly be overkill in normal temperatures but might be worth consideration in a bad cold snap.

next year

We learned a tremendous amount at this year's competition, but perhaps the most important lesson was this: it can be done. We had problems, mostly with tools and weather, that prevented us from reaching our goal of building a working clock out of ice, but we gained confidence in the medium. Counting our successes, we were able to

  • make working gears,
  • make working axles,
  • bore large holes,
  • lay a level pedestal,
  • waterglue a rigid frame, and
  • finish the competition not in last place.

With better tools and more time, I'm certain that we could have completed a ticking clock. As it turns out, we could have taken more time. We didn't know until the end of the event that it is not uncommon for teams to continue working on unfinished sculptures for a few days after the judging takes place. Unfortunately, we were incredibly exhausted after three days and two nights of hard work. We needed rest and already had plans for the weekend, so we decided to pack everything up and go home.

Another lesson: this is fun! Okay, that wasn't exactly unexpected, but the entire ordeal was a challenge we will never forget. Carving ice is fun, and the quality of ice, support from volunteers, talent of the competitors, and overall atmosphere of the World Ice Art Championships made the whole experience absolutely incredible. If they'll have us back, we would like to return next year and for many years to come.

We probably won't try to achieve our goal of a working clock next year, but perhaps the ice clock will become a reality in 2008. We have plenty of time to decide what to do in 2007, but right now we're thinking of trying something a bit less challenging that will allow us to improve our tools and skills before attempting the ultimate goal. Whatever we create, it will almost certainly be mechanical and/or interactive.

Tuesday, March 07, 2006

competition results

We attended the awards ceremony at the outdoor ice stage Friday night and were happy to see many outstanding artists win awards. In particular, I was glad that my favorite sculpture won the artists' choice award. After the event, we picked up our results and were pleased to discover that we placed 25th out of 29 teams in the realistic category. With as many problems as we had, I think it was a pretty good result for our first effort.

The sculptures had all been lighted by colored lights on Friday. These will remain in place until Ice Art closes for the year.

Friday, March 03, 2006


During the final minute before the horn sounded at 9:00, we brushed snow off of the sculpture and used a blow torch to make the clock face transparent enough to allow visibility of the interior gear teeth. We took a step back and noticed that the entire piece, with the exception of the pedestal and the little gear perched on top, was made on Thursday. Most of it had been assembled during the eleventh hour (actually the fifty-ninth hour).

The sculpture is over six feet tall but used very little of our original ice block. The gears engage and turned a little, but an alignment problem caused by poor drilling made it apparent that continuous turning would result in disaster.

Our idea for this alternative design was to create an interactive sculpture that allows people to use the crank to turn the hour hand on the clock, but we decided to waterglue the parts together to prevent rotation and subsequent breakage.

Quite a few sculptors dropped by during clean-up to congratulate us and express dismay at the removal of all of our insane equipment. One said that he already had some people willing to pool a few dollars a day each to pay us to stay and keep using our tools! Although the offer was tempting, total exhaustion convinced us to continue loading the truck and head back to Belfair for ice cream, Atari, and a long night's slumber.

the final day

The weather improved considerably on Thursday. Temperatures near 0°F allowed us to use waterglue to bond parts together and attach templates for gear tooth cutting without cracking the ice. Lars did a lot of chainsaw and lathe work, and I made gears. More and more people, including other sculptors, stopped to talk to us as the final hour of the competition drew near.

We encountered a number of drilling problems. The custom spade bits didn't hold up under extended usage; screw heads broke off, steel bent, and nuts came unscrewed, but we managed to get through everything that absolutely had to be drilled. Since the broken drill press carriage assembly wasn't able to make perpendicular holes, we had to use the hand drill, and hand drilling with failing bits resulted in some poorly aligned holes.

The gear tooth cutter had trouble as well. The first set of problems turned out to be caused primarily by poor drilling of spindle holes through the gear blanks, but a second set of problems resulted from the fact that the machine just wasn't well suited to cutting gear blanks weighing more than 50 pounds. I ended up using a manual method of positioning the gear blank for cuts but still taking advantage of the ability to spin the gear on the machine's spindle. This method worked incredibly well; had we turned to it sooner, we would have been able to make gears at a faster pace than was originally anticipated.

Thursday, March 02, 2006

the official page


This is the Ice Alaska page for our project.

the clock is ticking, but not ours

With the sun approaching the horizon on Wednesday, we came to the conclusion that we would not be able to complete the clock by the end of the competition Thursday night. Between our incomplete or missing tools, the extreme cold temperatures, and our lack of experience, the challenges were too great. Warmer weather during daylight was helpful, but too many tasks were left that could not be done at night without cracking the ice.

While other sites featured enormous, beautiful ice sculptures and a few tools, our site featured strange and enormous tools but no obvious sculpture. After we made the decision to abandon the original plan, we became more carefree and had as much time as we wanted to talk to spectators. It turns out that there are a lot of people who just love to look at and talk about the tools! I had always questioned the artistic merit of the clock in the back of my mind, but now I've realized that our project is a performance piece more than it is a sculpture. With this in mind, we reorganized our site with the gear tooth cutter and lathe front and center. People love seeing the things in action and learning how they work.

We haven't completely given up on the clock, but we are stopping work on this attempt. Instead we will focus on our performance, enjoy the moment, and build something out of ice. We don't know what it will look like, but it will have gears and will move.

Wednesday, March 01, 2006

Fairbanks sunrise

Compared to last night, -4°F and sunshine are incredibly invigorating. We got a good start this morning, but we've run into problems this afternoon. It is too difficult to align slab cuts from opposite sides of the block, so Lars is off looking for a bigger chainsaw bar. We found that a belt sander works very well on ice. We thought it would alleviate our slab cutting difficulties, but the sander started having trouble. A weld broke on the drill press carriage, a design flaw is preventing the gear tooth cutter from making the smallest gears correctly (although we can fudge the process pretty well), and some parts forgotten at Belfair are preventing us from making any gears other than the smallest.

It is supposed to be cold again tonight, but not quite as cold as last night. The high tomorrow is predicted to be 10°F.

first night

When we finally headed back to Ice Park, it was getting late at night. The sign at UAF said it was -28°, much colder than we had hoped. By the time we were on our way back to Belfair a few hours later, the temperature had dropped further to -38°F! Don't you hate it when your eyelashes freeze together every time you blink?

We arrived at the site to find that our neighbors had made amazing progress. This is what their sculpture looked like when we were just getting started cutting our block.

While I worked on making gears, Lars was able to cut some slabs with the chainsaw, using the hoist to move the bigger pieces. While setting up the hoist, we needed a small weight to take up slack in the cable, so we grabbed the bowling ball with its convenient hook. As Lars was testing the range of the hoist to see if it would reach the ice block, he barely tapped the ball against the block. The gentle bump caused a huge crack to zip through the block! We knew that ice gets brittle at extremely cold temperatures, but, wow, we were surprised by just how fragile it is. I guess my wrecking ball idea was prophetic. Not only does the ice break easily, but it can't be glued back together with water like it can at warmer temperatures. Lars had the good sense to try dripping a little water on a discarded piece of ice, and it shattered instantly.

By the time we left to get a little sleep, our sculpture consisted of a single slab (actually three separate pieces set like flagstones) on the ground. It isn't pretty, but it is strong, smooth, and level. This will be the pedestal upon which the ice clock will be constructed.

Due to the cold, we decided not to cut any more of the block for fear of making even more horrifying cracks. Hopefully it will be warm enough in the daylight to proceed safely.

at the shop

Back at the shop, much was accomplished. I finished a set of custom spade bits for the drill press, and Lars worked on his slab cutting bandsaw. It took several hours, but I finally finished the gear tooth cutter. The only pieces remaining to be done are a couple of differently sized guide rack and pinions, but those can be made at the site.

I was looking around for a counter-weight for the gear tooth cutter, and Lars had the brilliant idea of using a bowling ball he had in a junk pile somewhere (no, those of you who know Lars, it is not that bowling ball). I attached a big hook to it and hung it on the cable; It worked great! Hanging from the cable, it reminded me of a wrecking ball, so I suggested that we could use it as such, suspended by our hoist, in the event that we declare the project a total failure.

Alas, we had to give up on the bandsaw. The devices presented multiple unsolved challenges, but death was not pronounced until the motor started to smoke. It just didn't have enough power, or maybe it did have enough power but ran the blade too fast. We threw it into a snow bank where it will remain for several days at least. I am confident, however, that it will one day rise again, better, stronger, slower.

Instead of the bandsaw, we modified a (borrowed!) 20 inch bar chainsaw with a custom shoe plate for guiding perpendicular cuts. This will have to do for slab cutting, although it requires cutting from both sides and hoping for good alignment.

back to Belfair

We weren't able to accomplish much this morning because our custom tools were not complete. It was a good thing we were able to be on site in order to get our block re-positioned, but we didn't get anything else done apart from dropping off equipment and leveling the dirt upon which the clock with be erected. So, it's back to Lars's shop at Belfair to finish our tools. We are allowed to work at Ice Art around the clock until 9:00 PM on Thursday, so we should still have plenty of time once the tools are done.

Steve wasn't kidding about the roads being like skating rinks. On top of the sheets of ice, Fairbanks received over a foot of fresh snow just before my arrival. It was the highest single snowfall in years.

heavy lifting

Our block is 93 inches by 59 inches by 42 inches which means that the zoom boom had to lift about 7660 pounds. Our site (now #16 after a re-numbering that took place today) is surrounded by trees, so Connie, the driver, wasn't able to lay the block down on our timbers like we wanted. She did, however, deftly remove the block from our site, tip it on its side, and re-deliver it in a pretty good spot for us.

Most of the teams want their block placed in a particular way so that it can be sculpted in a good viewing position, but we chose to place ours in the back corner of our site. We are going to carve slabs off of the block, cut parts out of them, machine them, and assemble the clock at the other side of the site.


The chainsaw didn't want to start, so Lars warmed it up with a hair dryer while we were waiting for our block to be "tweaked" to our desired position.

The sun rose and revealed the depth of the cracks in our block. There are some major cracks, but the good news is that the overall quality of the ice is superb. With any other ice we've ever seen, we wouldn't have a chance at being able to see how deep the cracks go.