Tuesday, April 26, 2011

Tick Tock Tiles

Silhouettes of the Joy tiles cast their eerie shadow on my broken kiln shelves, two reminders that working in ceramics really is playing with fire, in more ways than one!

The past couple of weeks have been a tempest of tile tribulation to create, package and ship 140 custom ceramic tiles for the Michigan Model Madness Spring Show 2011. Ordinarily, this would have been a routine project if not for the fact that it lacked the two most important components to ceramic production: predictability and time. 

Why? Because this commission required the use of entirely new clay and new glazes, so I had no idea how they would behave or what to expect. In terms of time, Pop's situation essentially consumed all of March and a goodly part of April. Translation: the ample time I thought I had when I accepted the commission unexpectedly and quickly flew out the door like the Concorde on crack. 

Predictability and time also combine together in ways important to ceramic production. For instance, new materials require a series of tests to identify their quirks and how to work around them to engineer some degree of predictability, and that can take a goodly amount of time.
In particular, new glazes are especially test-worthy since any number of variables can affect the outcome, from how they're applied to the individual idiosyncrasies of each kiln to how the shelves are stacked to where the piece is located in the kiln, among many more. Remember, glaze is a combination of minerals (with glass) and so it doesn't just melt onto the bisque piece, but undergoes a series of chemical reactions during the fire - and we all know how touchy chemical reactions can be. 

In fact, Lynn, Lesli, Joan and myself ran a series of tests a few years back with each of us firing the very same glazes applied the very same way (even to the very same tile) in each of our respective kilns. Lo and behold, each test tile came out differently! For my purposes, I discovered that Big Al loves the colors blue, teal and green, bringing those tones out of any glaze consistently (compelling Lynn to coin the term "Big Blue" for him!). Even Big Al and Maury fire the same glazes quite differently, with Maury favoring the brown/orange/red end of the spectrum.

So after all was said and done by April, I was left with an April 25 deadline to get these award tiles in the mail, which is about as tight a deadline for ceramic tiles as imaginable in a bad nightmare, due entirely to the time component in creating them. Let me break the panic down for you:
  • 1-2 days: pressing 140 tiles
  • 7-10 days: tiles drying
  • 2-3 days: glaze fire and cooling
  • 1-3 days: tying on ribbons, packaging and packing for shipment
  • 5-7 days: shipping
Altogether, these numbers meant only one thing: there was no time to test anything - it was full bore or nothing. But when it comes to ceramics, that's playing with fiery fate. Should anything go wrong, there wouldn't be time to repeat the time-laden process of making more, and so the show simply wouldn't have its awards and my name would be mud - and not in a good way. So I set out to work like the wind, making extras as "just in case" back-ups.

I like to use bisque tiles as pressing plates for the stamped clay. Don't use the glazed ones since the clay will stick to them!

Here are the regulars in the process (clockwise from top left): baby powder, "Sir Squish" my tile press, the custom stamp made from my design, and the cut-out clay on the pressing plate.

There's Sir Squish pressing the stamp into the clay. Thanks to his gear box, it takes very little pressure on the lever to exact a lot of force onto the clay. Trust me when I tell you that a tile press is required for this process - attempting to do so with your strength or body weight not only produces a lousy result, but your body will hate you the next day. While you think the clay is soft and gooshy, it's remarkably resistant to stamping on this scale.

Then I pull off the plate and stamp the back. Clearly, I need a new stamp for the back. Drat. It's interesting that stamps behave in different ways even with each design or clay. Note to self.

Repeat 140 times and voilĂ ! The clay tiles dry sandwiched between drywall boards to keep them flat and to suck out the moisture evenly. Yet the both the predictability monster and time monster chose to raise their heads at this stage, since this particular clay presented an unwanted surprise: it took 14 days to dry, even after being placed in a room made hotter by a space heater! The clay was so wet and dense, it nearly doubled the drying time! Argh. In the end, I was forced to fire about 25% of them still semi-damp because I just couldn't wait any longer. The good news was that they fired just fine, but lemme tell ya...it was pins n' needles! Thank goodness for the "slow" fire program on Big Al! I also believe the clay being low-fire earthenware clay worked in my favor because it's more porous and forgiving. I'm not sure they would have turned out okay had the clay been high-fire stoneware or porcelain.

All 140 had to be cleaned and bisque-fired, after which they needed to be glazed. The order called for 70 glazed green and 70 glazed violet, but since I didn't have these colors on hand, I had to buy them. In other words, I had no experience with these new glazes. Even more worrisome, one of these glazes - the violet - was a lead-free glaze.

Anyone who has worked in low-fire glazes knows that leaded glaze produces a superior result, having better refractive qualities and "pooling" abilities. In short, it looks better and behaves better than unleaded glazes. This is because the lead in the glaze forces  the components to move, smoothing them out and making the minerals "pool" inside the crevices, a feature that beautifully amplifies a stamped, sculpted or carved design in the clay. Being leaded, the green glaze ordered for this commission was a perfect example of the benefits of leaded glaze:

 Amaco F-47 Christmas Tree Green, a leaded glaze. It just doesn't get any better than this, folks! Notice how glassy, smooth and rich the color is, and how the pigment has pooled into the crevices left by the stamp, making the design "pop." The fun thing about this glaze, too, is that it's probably named after its intention, that of the quintessential "mudhen" project, the ceramic Christmas Tree!

 Some of the bisque tiles with the violet glaze, before the glaze fire. What you see now isn't the actual glaze, but the dye added to the glaze so you can see where you applied it.

In contrast, the violet glaze, Amaco F-70 Violet, ended up being a fiendish headache, creating the second unexpected hiccup. Why? Because it's a lead-free glaze. This means that rather than moving around during the fire, the pigment stays put, resulting in a flat, dull finish that does nothing to flatter the stamped design. In fact, it can obscure it completely! For example, here's the unleaded violet out of the glaze fire (quite a contrast to the leaded green, huh?)...

While the color itself was pretty, the glaze's effect was...in a word...disgusting. Yes. Disgusting. To go through all that work and expense creating the stamp design, and then stamping and cleaning and glazing...to then end up with this was definitely an unpleasant experience. There was no way I was sending this to the show. So I had to fix them. All 70. And the clock was ticking.

Luckily my years of experimenting with glazes paid off because I now had an arsenal of potential fixes for these 70 deeply offensive violet tiles. I knew that Duncan CR 823 Celadon Crackle Glaze could fix just about any low-fire glaze problem. However, I also knew it would create a greenish-blueish pooling of color that veered from the violet that was ordered. What to do?

I finally settled on creating a wash of leaded glaze in in the same color family as the violet, but with a contrasting leaded color added to provide "pop." This wash would force the unleaded glaze to move, and the contrasting color would create an artificial pooling of color. In this case I mixed Mayco AG-263 Purple Iris with Mayco Exotic E-103 Midnight Mist together in an 80/20 ratio and thinned it with water to about the consistency of skim milk and brushed it on, right over the fired glaze. I then put them through a second glaze fire. 

After I pulled them out, I could see I was on the right track, but I wasn't aggressive enough - that Amaco violet needed a major smack down! So I repeated the process, but added more of the Midnight Mist to the mix and applied said mix thicker...then ran them through a third glaze fire. The situation was dire at this point, since I knew this would be their final fire - there just wasn't any more time. It was Saturday, and they all had to go out on Monday, and a glaze fire takes a minimum of two days between firing and cooling.

Lucky for me, that bolder treatment did the trick! I pulled them out of Big Al with kitchen mitts to cool them faster so I could package and pack them for shipment that day. You think hot cookies are tricky? Try pulling seventy 192˚ tiles out of a gaping, hot kiln!

Here's the final result of those fixes. Quite an improvement, eh? Leaded glaze, you are my bestest low-fire friend! Unleaded low-fire glaze, you are my sworn enemy! Get thee behind me!

Interestingly enough, however, the problems with unleaded glaze don't really exist with high-fire glazes because all that added heat forces the glaze to move despite the lack of lead. The issue only presents a problem for the low-fire folks, especially since most of the leaded low-fire glazes are now switching over to unleaded formulations. This may be because low-fire glazes are used more for craft rather than "real" ceramics, so manufacturers want a more benign formula.

But this puts me in a pickle for show awards. I chose low-fire earthenware clay for this project because it has a near-zero loss to warping, which means that all the stamped tiles will be usable, and I know they'll be usable before I apply glaze to them (because low-fire clay is brought to maturity first, then glaze fired at a lower temperature afterward). As such, I can keep costs down for the show holder who commissions them. In contrast, high-fire stoneware has about a 33% loss rate to warping, but I find that out only after I've put them through the glaze fire (because high-fire clay must be brought to maturity with its glaze). So I now have to decide which is more feasible if leaded glaze may no longer be an option with the low-fire clay.

Anyway, despite all this scrambling and stress, I have to tell the truth - I've enjoyed this project immensely. That may seem counter-intuitive, but most of my studio is run by fuzzy deadlines of my own making, and being so, they tend to get fudged as I dodge the monkey wrenches life randomly heaves my way. Also the unpleasant upheaval caused by my Dad's situation has been a tornado of anxiety, but this project gave me something to cling onto as a goal. Something firm. So in many ways, this project's hard deadline kept me centered and was a refreshing "reality check" of sorts. Plus, the fruits of my freak out were well worth it:

70 lovely green tiles and 70 lovely violet tiles were put in the mail on Monday. Now I pray they arrive safely! Thank you Melissa for this opportunity, not only to create awards for your show, but to learn a great deal about the media and myself in the process!

But I can't relax for long - now is just the eye of the storm. This weekend, Laurie Jo decends to judge a local show and she's staying with us. I expect by Monday I'll have lost my voice and my face will ache from laughing. So before she arrives, I have to madly clean my house because, since March, it's sunk into an audacious catastrophe! 

Mind you, I still have to finish Mr. Pony to boot, and then start the BOYCC swag bag tiles. No wonder then that I have that Jurassic Park scene running through my head.

"You build up resistance to stress by learning, acquiring and practicing skills needed to go forward and cope." ~ Dr. Donald Meichenbaum

Related Posts with Thumbnails