Week 8 Lab Summary: Pottery Firing

Introduction

On Thursday our class finally convened in the Arboretum art yard to fire the loom weights that we shaped in Week 3. We had a great time firing the loom weights with tremendous help from Kelly Connole, the master of pottery, and Austin Mason, our professor. We came to the art yard knowing that the smell of smoke would follow us for the rest of our lives (or this week). We set up in a gravel-covered area of the yard, close enough to hear and see one another, but still separate enough to work in our own spaces.

The hypothesis of this experiment and lab is: “A lifetime of experience in building and maintaining a hot fire to transform wet clay into hardened ceramic ware would have influenced practices of fire construction, temperature maintenance and retrieval of bone from the ashes in the cremation rite, and vice versa,” according to the lab manual. We also compared our experience keeping the cheese fire steady to keeping these fires hot. From our discussion, we think it’s really interesting how the ability to maintain the fire heat in these activities could be transferable and relevant to each other. The differences between the activities boil down to how the heat is transferred: for the cheese, it was in a vessel over the fire, but for both cremation and firing, the heated element was directly in the fire.

We also discussed the purposes of this lab: “To gain an experimental understanding of the final firing step…, to appreciate the inherent properties of the raw materials, to better understand the constraints those properties imposed on the desired finished product…, to assess the effectiveness of different materials for smothering the fire to protect pots from thermal shock and produce a black finish, and to compare the practices of pot firing and open air cremation.” These goals were met both during the data collecting, the lab processes, and the post-lab discussion. 

There are a number of research questions one can ask about this lab because there are a number of different factors starting from the weights, thickness, length, and elements of the loom weights to the various materials used to smother different piles of fire. One question we came across is how possible is it that we can accurately conclude the information from the experiment given that we had a high level of uncertainty in the decisions we were making throughout the process. Similarly, we had a very low success rate with the weights that had not been bisque-fired, implying something was amiss with our process.

Process

Steps:

When we arrived at the Arb art yard, Kelly gave us a quick rundown of the process and of the resources we had available (mostly wood and more wood).  After that, it was time to get started.  The three groups each staked out a spot in the yard and began to gather fuel for the fire.  We picked up armfuls of cut wood from the art yard’s supply as well as stacks of raw branches that Austin had brought, and we started to build fires out of the former.  Group A opted for an open triangle shape, while B and C used a log cabin structure.  While the variation was interesting, it wasn’t at all crucial to the results of the lab, and we were able to build our fires however we pleased.  Our objective at this point was to allow plentiful airflow through the structure, as we were attempting to produce hot flames quickly.  Though we had gotten some practice in fire building during the cheesemaking lab, many members of the class were slightly at a loss as to how to build a successful fire— but by 1:30, all three groups had figured it out and had moved on to the next step.

While the flames were crackling away, we used the wait time to measure our loom weights.  We failed to standardize what terms we were using for each dimension throughout the groups, but generally speaking, all of the groups measured the weight and some groups measured full diameter, diameter of the hole, and height/thickness of each weight’s side.  At this stage, it was important to note which weights had previously been bisque fired by Kelly and which had not.  With a few exceptions, all of the weights labeled “1” were not bisque fired prior to the start of the lab.  We could tell these from the unfired weights by a slight difference in color as well; the fired weights took on a faint orange-pink tone while the unfired weights remained fully grey.

It took a fairly long time for all three groups to finish measuring all of their weights, but as soon as we were done, we arranged them in neat circles around our fires.  We were attempting to warm the weights up somewhat gradually, in order to give them a fighting chance at remaining intact during the firing.  From this point onward, all of the groups were meant to take regular measurements of the temperature of the flame and/or loom weights.  We all varied in how we approached this, which makes analyzing trends in the resulting data somewhat difficult.  Group A recorded the temperature of each individual loom weight in intervals ranging from 3-10 minutes; Group B measured mostly the heat of the flame itself, but in precise 5 minute intervals; and Group C also measured the fire, but in 10 minute intervals (with some unaccounted-for time — probably spent stacking the weights — between 2:17pm and 2:53pm).  While we recorded temperatures, we also tried to rotate the loom weights and gradually move them closer to the flame at each time interval.

By this point, we were no longer feeding the fires— quite the opposite.  We were waiting for them to consume the wood we had initially fed them and subside to beds of coals.  It didn’t take too long; all of the groups began to move their weights onto their coal beds at roughly 2:35pm.  Some poking of the fire, getting it to finish burning the sides of pieces of wood and lie flat enough to stack things on it, was necessary.  We used tongs provided by Kelly to pick up the weights, as by this point, they were too hot to touch with bare hands.  In our stacking of the weights, we didn’t do anything very fancy; we were generally just trying to fit them all close together so that they would be easily contained by the structures we would build in the next step.

This part was a little tricky.  After a period of waiting, we were under some time pressure to build a “teepee” structure of Austin’s branches that held itself together and contained the fire within with few to no gaps.  Here, we wanted to prevent too much air from getting into the structure.  Many of the branches we were using were fairly long, so we had attempted to break them down to a more useful size while we were waiting.  Some members of the class took to this with more pleasure and ease than others.  Using the axe we had brought was challenging for many, and it was ultimately easier to attempt to snap the branches underfoot.  Though some groups were low on suitable branches because of their difficulties in breaking down timbers, they were able to supplement with Kelly’s stock of cut wood and fill in the gaps in their teepee structures successfully.

The wind was strong, which caused a slight delay in a few groups’ fires catching to the whole structure.  When it did, though, it was a dramatic flare, tall, hot, and bright.  This is the moment that I suppose would have been the emotional and theatrical climax of cremation rituals: the great blaze of hungry flame consuming the body.  Popping sounds could be heard from all of the fires almost immediately as our weights began to blow out despite our best efforts.  We stood back a little ways and watched the branches burn up and collapse.  Again, we tried to utilize our waiting time to gather materials that we would need in the next step: smothering.  Once the fires started collapsing, they were dying down and smothering themselves slightly.  That was the time to act.  Group A used sawdust, Group B used bark, and Group C used straw to pile atop the flame as quickly as possible and choke it.  We all poured water on top of the material to ensure that the fire wouldn’t just consume that too.  Group B also ended up using the broad green leaves growing in the art yard to plug some of the spaces left by the bark, aiming for a consistent seal.  Group A’s sawdust covered up their fire very effectively; there wasn’t even any smoke visible.  The other two groups still had plumes of smoke emerging from the mounds they had created.  Once we were satisfied that the fires were sufficiently smothered (minimal airflow, generous mound, and thoroughly damp material), and had tentatively stepped and stomped on the edges for good measure, we gathered up our materials and left for the day.

We did come back on Friday to retrieve our weights and take another set of weight measurements.  Almost no non-bisque-fired weight had survived the experience, and we saw a range of blackening.  Every group ensured that the remnants of their fires were completely put out, and then we packed up our weights and departed the art yard.

Results

We were looking at the following variables in our research:

• Firing status (bisqued versus unfired)
  • We fired thirty-seven loom weights, of which twenty-five had been previously bisque fired by Kelly.
• Type of smothering material
  • Group A smothered with slightly damp sawdust, Group B smothered with large leaves and damp wood shaving scraps, and Group C smothered with slightly damp hay.
• Temperature of fire
  • The temperature of the fire ranged both by time and by group widely, as did the carefulness of taking the temperature with regularity. The measurements can be split into four types: the initial or drying fire, the firing fire, the smothering temperature, and the day after. The highs (or onlys*) of each group for each type are as follows in Table 1:

Type	Group A	Group B	Group C
Initial	627* °C	780 °C	986 °C
Firing	893 °C	814 °C	808* °C
Smothering	813* °C	784* °C	722* °C
Day After	19* °C	156 °C

• Group A also took measurements of the clay weights themselves every five to ten minutes in the initial stage (Table 2) and Group B took fire measurements every five minutes (Figure 1). It is hard to be certain of the measurements of the temperatures of the loom weights, since they vary so much even within the same time and loom weight, but it gives an estimate for about how hot they were when they went into the firing fire. Group B’s dip in fire temperature around 14:35-14:55 likely has to do with the process of building up the teepee for firing.

	13:51	13:54	13:59	14:09	14:17	14:32
Hali’a	22 °C , 27 °C , 19 °C		35 °C , 35 °C, 33 °C	100 °C, 80 °C, 63 °C	213 °C, 261 °C, 280 °C	394 °C, 163 °C, 113 °C
Elliot	50 °C , 40 °C , 45 °C		83 °C, 65 °C, 94 °C	88 °C, 68 °C, 59 °C	101 °C, 94 °C, 96 °C	222 °C, 374 °C, 320 °C
Margie		26 °C, 25 °C, 24 °C	107 °C, 100 °C, 62 °C	282 °C, 283 °C, 200 °C	430 °C, 320 °C, 150 °C
Phoebe		21 °C, 24 °C, 25 °C	63 °C, 41 °C, 46 °C	75 °C, 68 °C, 271 °C	386 °C, 138 °C, 103 °C	230 °C, 329 °C, 357 °C
Dane		25 °C, 21 °C, 30 °C	58 °C, 74 °C, 63 °C	148 °C, 115 °C, 101 °C	150 °C, 80 °C, 127 °C

Our results were in the following areas:

• Blackening
  • There was a wide range of blackening. Many loom weights were partially blackened. Those that were most blackened were in the fire that was smothered with the damp straw. The sawdust firing saw some blackening, but not consistently. The weights that were the least blackened were in the fire smothered with leaves and wood shavings.
• Shattering/exploding
  • There were thirty-seven loom weights, of which ten exploded. All of these were the unfired loom weights, regardless of hole status. The twenty-seven which did not explode had all been bisque-fired with the exception of two. When looking at the temperature data from Group A, it seems likely temperature shock also played a role many of the loom weights were still much colder (~500 degrees Celsius) than the fires they were put into.
• Weight (possibly)
  • Unsurprisingly, there was a change in weight from the day after to the post-firing stage. There was an average of about forty-one grams lost between the day after forming and the day after firing and about fifty-eight grams lost between the mass of wet clay on the day of forming and the mass after firing. However, most of this seems to have been lost in the bisque firing; if the weights of the loom weights differed between the bisque and pit firing stages, they were only a gram different—and sometimes they were one gram heavier rather than lighter. This suggests either uncertainty with the scale or ash in the holes poked through the weights. See Figure 2.

• Size
  • Not enough data for comparison. Only had data from one group.

Conclusion

In this lab, we experimented with several factors in the firing process.  Some of these questions began in the loom weight formation lab in the ceramics studio, as many of us tried out different strategies to promote successful firing.  Poking multiple holes throughout the weights, for instance, seems not to have had enough of an effect to help the unfired weights survive.  Whether it allowed for a higher-quality finish on the pieces that did survive is unclear.  Some of us also tried out different ratios of sand to clay as we shaped our weights, and class members have reported that the weights with very high amounts of sand exploded without turning black.  We can at least fairly conclusively say that bisque firing before fire-firing effectively ensures a higher survival rate.

We attempted to create a reducing environment by smothering the fires with the pottery inside, which would ideally produce a black finish in the final product.  Nobody seemed to achieve an attractive and uniformly black finish, but the variation between the different groups in how dark their loom weights became suggests a potential influence from the different smothering materials used.  Group B’s bark-and-leaves smother resulted in very little blackening, while Group C’s straw produced the most and Group A’s sawdust struck a balance between the two.  Group B’s failure might be put down to their extensive watering of the flame and fairly low temperature going into the smothering process, but Group C also used copious amounts of water and was successful.  It has also been suggested that Group A’s sawdust, which wasn’t even smoking after the smothering occurred, might have been too successful in smothering the fire before the weights had a chance to turn black.

In any case, the class as a whole certainly produced enough fired loom weights to be of use when we finally begin weaving our garment!  Though we lost some weights along the way, we gained a more nuanced understanding of how to create different environments for firing and how to create (somewhat) black, useful, intact loom weights.

---

Group Data Reports