Glaze chemistry is mostly about understanding and controlling functional glazes to create surfaces that are durable. But it also helps explain extreme behavior such as that shown on the piece on the left: Severe running (with danger of glaze compression), severe crazing and pinholing. "Art language" is OK to describe these glazes visually. But, objectively this recipe’s cute name, Juicy Fruit, just diverts attention from what it really is: A Gerstley Borate, lithium and soda ash addition that melts the crap out of a high feldspar glaze, producing crystallization and phase separations. It is all a delicate balance of unbalance, an aesthetic developed by someone with only non-functional priorities. Moving this cone 10 effect to cone 6 involves moving its problems also (because they come with the effect). Since it is already highly melted, melting it even more could only be done with more boron, perhaps from pure Borax. A better approach would be to start with an existing cone 04 glaze and add the colorants and titanium to that. The recipe on the right is a first attempt at "moving chaos"! The next xyep is more flux, like B2O3, to increase melt mobility even more.

The old one inside is in bad condition (a new one is sitting on top ready to install). In 2022 these cost about $35 CDN. The temperature-measuring part of a thermocouple is the join of two dissimilar metal wires, these are 8 gauge. The junction produces a temperature-dependent voltage that a pyrometer can convert to a reading. Thermocouples can degrade into pretty poor condition yet still work, notice the one in this kiln is separating in two. Thermocouples generally need replacement more often than elements. Notice the elements in this kiln are laying down, they also need replacement.

In bisque kilns ware is being fired for the first time. If pieces are thick and the rate-of-rise is too fast then water (which is turning to steam) cannot escape fast enough. The internal pressure will fracture a piece like this that has not gone through a thermal drier. The schedule to fire this test brick was 150F/hr to 250 and hold 90, 200F/hr to 1640 no hold, 120F/hr to 1888 no hold. The fracture likely happened because the kiln schedule did not allocate enough hold time at 250°F or proceeded too fast to 1000°F. 250°F might sound like too high a temperature to do water smoking at because it is over boiling point, but in practice it does work in industrial kilns and driers with good airflow (which this kiln does not have).

It is often said: "No good deed goes unpunished"! This can happen when doing a clay session with children. So there is a need to enter, with "both eyes open" to avoid a negative experience. If you can succeed they will get a unique tactile experience in a social setting. And they will experience the anticipation of waiting to see what the kiln will do with their creation. If the kiln gifts them a brightly colored durable piece, especially one that they can drink or eat out of, the experience and the object can stay with them for a lifetime. If you are an instructor inheriting a kiln and clay program for kids it can seem overwhelming, especially if your technical knowledge is limited. But there are some simple things you can do to assess the condition of the kiln in your facility and whether it is practical to attempt some clay sessions with the children in your trust. Click the link below to learn more.

The only difference between these two cone 6 glazes is the silica. Both are the G2926B recipe, both were thickly applied and fired in the same kiln. The left one employs the 90 micron (200 mesh) grade silica and the right one uses 45 micron (325) mesh. These test tiles are about 6 months old. There was no crazing out of the kiln. The porcelain recipe is 25% silica, 25% nepheline and the remainder kaolin and bentonite. It appears the finer particle size silica is dissolving in the melt much better, this narrows the difference between calculated and actual behavior, especially relating to coefficient of thermal expansion.

This 12 inch test kiln has done 910 firings. The element loops are laying down and nearly touching each other. If they are not changed soon the coils will touch the kiln will have hot spots. And the coils are expanding and getting tighter in the grooves, the longer we wait the more the grooves will be damaged when removing them. Although elements seem expensive, when costed on a per/firing basis they can be surprisingly inexpensive. Most hobby kilns service two elements with each relay and relays generally need to be replaced more often than elements. Consider, for example, replacing the elements on a Skutt 818. Being a smaller kiln it is well-powered in relation to size and elements can last up to 1000 firings (assuming 50:50 bisque and cone 6 firings). It has 4 elements and 2 relays (relays cost $65/ea, elements $95 each). The labor to replace is ~4 hours or $250 - total cost is about $750 (that is ~75¢ per firing or 32¢ per ft³). How about a larger kiln? An 8 ft³ Model 1222 has 5 elements and 3 relays and replacement is ~$1100. But its elements are only likely to last 200 firings. That yields a per firing cost of ~$5 and per ft³ of 65¢. But there is a much greater cost to consider: Old elements increase power consumption. An 818 uses 6.4 kwH and a 1222 uses 11.5 kwH - at our electricity cost of 14¢/kwH a firing costs ~$7 for the small kiln and ~$13 for the large one. But that is the cost when elements are new. When they need changing those numbers can more than double! An additional cost of old elements is ware consistency, the kiln cannot execute the firing schedule in the time programmed and this will likely affect the appearance of bodies and glazes.

Each of these bags is batch-numbered and time-stamped, this is an impressive product-tracking effort by US Silica. It appears the time changes once per minute and from our calculation, their plant was producing about 600 bags/hour. They stack silica 60 bags on a pallet, 3000 lbs! The only material we receive that is stacked more densely than that is plaster, 80 bags per pallet at 4000 lbs. We stack our clay at 40 boxes/pallet giving 1760 lbs.

Ware is not turning out as expected and a potter needs to verify the temperature in the kiln. The standard cones on the upper right are misleading. The cone 7 is telling one story but the cone 6 and 5 another. On the lower right is a better way: Self supporting cones. They are always at the right angle and this set of three is bending as expected. To be a full cone 6 the middle one needs to bend just a little more until the tip is even with the top of the base (maybe 2 or 3 degrees). On the top set, the cone 6 is clearly totally flattened and the 5 is a pool of glass, this firing went way beyond cone 6.

This is G1214Z1 brushing glaze (it has 5% titanium added). For a 340g powder batch (to get a pint) my target is 5g CMC gum and 5g Veegum. CMC controls drying speed and Veegum the amount of gelling. I already mixed the CMC with the powder, and shook the whole batch in a plastic bag. Then I added it all to 440g of water in the blender jar and mixed it really well, being sure no agglomerates remained (that was stage 1). Stage 2 is adding the VeeGum slowly, while blender mixing on high speed, this enables tuning the degree of gel. Because this recipe has little clay, it took all 5g of Veegum without gelling too much (too much means the entire mass is not moving freely in the mixer jar). Under-gelling is a good idea since slurries will almost always gel more on sitting over night. This recipe could have taken a little more CMC to slow down drying a bit (this amount enabled applying three coats to a bisqued piece in only a couple of minutes but it did not brush on quite as nicely as it could have). Sure enough, it gelled more overnight (so the next time I used 4 grams of Veegum, each recipe has its own optimal amount). Both of these gums are very difficult to mix into water without agglomerates forming (also the titanium dioxide), the high water content enabled the stage 1 slurry to be mobile enough in the mixer jar that the blade has access to all particle surfaces. And it enables mixing in enough stage 2 VeeGum to achieve a consistency that will never settle out.

This recipe is from page 2 of the booklet: "15 Tried & True Cone 6 Glaze Recipes". Click the following code, G3955, to see more information on how we compare it with G2934 and another matte, G1214Z1. This flow test and these test tiles were in the same kiln, fired at cone 6 using our PLC6DS schedule. Obviously, the defining characteristic of N505 is its extreme melt fluidity, clearly it is not a native cone 6 glaze (it's a lower temperature one being used above its range). Still, the surface on the N505 tile is arguably more interesting, that's why it's popular with potters. But is it functional? Some felt pen marking helps reveal one big difference: The micro surface of the G2934 is much smoother. From the chemistry shown on Insight-Live side-by-side screenshot very low Al2O3 and SiO2 are evident. This also reveals it should fire glossy, so it is a "crowbar matte", forced to be such by 6% addition of magnesium carbonate. I was understandably suspicious that this glaze would have more issues than it actually does. Although the surface is rough and it does mark and stain, it can be cleaned with effort. The low SiO2 suggests it would cutlery mark but it does seem quite hard. However on the matter of leaching the jury is still out (a stain needs to be added for testing in an acid). Crazing is another possible issue. Our G3924 recipe, although more boring, excels on all four of these tests.