PPP - Controlling Kiln Temperature
Applying microcontroller technology to classical, general-purpose temperature controllers has made the controllers less expensive, more accurate and more versatile. But more importantly, technological advances have made it practical to design special purpose controllers with features that make them especially suited to the small ceramic studio.
Impediments to the Use of TechnologyIn general, ceramic artists and small production studios have been content to use almost no modern temperature control technology, relying instead on pyrometric cones and infinite switches to manually adjust the heat of their kilns. To achieve a moderate rate of cooling, they usually depend on nothing more than the high thermal mass of the filled kiln.
While such techniques have a long and honored place in the folk-process of making ceramics, they also place limits on the final product.
One reason non-technical ceramists have tended to avoid modern equipment is the complexity and difficulty presented by the general-purpose process controllers. When the artist confronts one of these instruments, the result is usually total bafflement—and rightly so. These controllers are designed by engineers to be used by engineers.
General-purpose process controllers are intended to be used in myriad different applications, including oil refineries, steel and paper mills, chemical plants and factories—all places in which one can expect to find a “process engineer” who is highly trained in operating such equipment. These controllers are very far from plug-and-play. They need extensive setting-up, adjusting and tuning by an expert when they are installed and again when there are changes, even minor ones, in operating conditions.
Let’s look at the example of PID control. PID controllers use mathematically based algorithms to try to achieve optimal results. For example, as a kiln approaches the required temperature, a PID controller would start to lower the power to the heating elements. The alternative is to behave like your home heater, which runs at full tilt until the thermostat registers the desired temperature and then shuts off. Almost all PID controllers require the setting of various gains to tune their response. This is necessary because there is no specific knowledge of the type of equipment in which the controller will be installed.
There is a large cultural dissonance between the “go to cone twelve and let ’er cool down” approach and the use of a general-purpose controller. Even a ceramist who wishes to enjoy the benefits of using modern equipment does not necessarily want to become a computer expert. The trouble, to paraphrase Alan Cooper, is that when you cross a temperature controller with a computer, what you get is a computer. And the computer you get is not your familiar (if sometimes inscrutable) PC, but a gadget with limited human interface capabilities and layer upon layer of totally incomprehensible menus and submenus.
Meeting the Needs of ArtistsThe reason for all this complexity comes from the need to be a controller for all applications. However, if we know the limits of our intended application, we can simplify the operation of the instrument. While there is a great diversity among kilns used by ceramic artists, they are all really quite similar compared with, say, a steel mill or a chemical plant. In fact, they are similar enough that one can pick some value for the various PID gains that will pretty much work with all of them.
The key words here are “pretty much.” It is certainly true that for any given kiln, we could improve the response by adjusting the gains. Maybe there is some kiln so far from what we call “typical” that it would have poor performance. But on the whole, most kilns will be controlled “well enough” using standard settings. The main criterion is: will it affect the work? If not, then to insist on perfect tuning is a case of being dominated by technology instead of using technology as a tool.
Consider the following analogy. Do we want a Ferrari where a Volkswagen does an adequate job? Sure, that Ferrari is fun to drive: a mere thought will make it change lanes; another thought will pile on the Gs. But if we want to just stay in our own lane and reach our destination, it might be better to have a less responsive, less demanding vehicle. We would also gain an extra seat because we wouldn’t need to take our mechanic with us all the time.
Put another way, we have to contend with Frasier’s Fourth Law of Adjustments: if you give a user something to adjust that he doesn’t understand, one of two things will happen: (1) he will adjust it wrong, or (2) he won’t touch it at all. In either case, the result of a well-tested and well-thought-out standard factory setting will be preferable to what Frasier’s Law decrees. The widespread truth of Frasier’s Law can be seen in the infamous “blinking 12:00” VCR syndrome. Millions of VCRs all over the world are blinking 12:00 because the procedure to set the clock is too complicated for what the user perceives to be the value returned—a correct time readout. It is never adjusted.
Factory default settings that need never be altered and that provide reliable results are what most people want. Why should an artist who is focused on the form and function of an artistic creation have to be concerned with the technical aspects of the process? Insisting on this will call Frasier’s Law of Adjustments into play. Note that it is not just with PID adjustments that Frasier’s Law applies. Any form of alarm or limit adjustment is subject to it.
The response to Frasier’s Law is: if the scope of the application is limited enough so that “one size fits all,” then offer only that one size, even if a few cases need a shoehorn or extra socks. A well-designed controller makes choices for its user rather than requiring the user to make choices for the controller.
A series of programmable, ramping temperature controllers has been developed for glass and ceramic artists with Frasier’s Law in mind. The whole point of such a device is to control a kiln according to a time/temperature profile. The only thing the user should enter is the profile itself. It is very important that this interaction be as clear and intuitive as possible. Submenus are not invited.
BenefitsNow that we’ve discussed some of the impediments to the use of high-tech equipment by artists and some ways to mitigate them, what positive benefits do they provide?
The most obvious is automatic operation. If no one has to be present to ensure that the infinite switches are periodically adjusted, there is no chance that such adjustment will be forgotten. Since no one is burdened by having to do the adjustment, it can be done much more frequently, even every minute. This allows you to have your kiln start slowly and then, after a certain time, heat up more quickly. You can have it perform an automatic soak at any point and for any duration. This technology has allowed at least one pottery maker to eliminate her bisque firing. Ramping the temperature up slowly allows the work to dry without a separate firing. This saves steps, time and money.
Another manufacturer had recently purchased a large kiln with three-phase power. This kiln had too much power and mass to be adequately controlled by means of infinite switches. Adding a programmable temperature control tamed this beast, so that it became one of the most useful kilns in his studio.
Of course, there are some processes, such as growing crystalline glazes, where careful temperature control is critical. But even in many less obvious areas, ceramists are discovering the benefits of better temperature profiling. These benefits include less breakage and cracking of the clay and glaze, more uniform glaze quality, more uniform glaze color, and lack of bubble formation and other defects.
Ceramists are also discovering that they can get an increased range of effects from their standard glazes by varying the temperature curing cycle.
For the more methodical experimentalists, many controllers can be furnished with a data-output port that can be connected to a PC. Times and temperatures can then be recorded for future reference and even graphed for easy visual inspection. Above, you can see a sample of a graph made from an actual firing using this type of controller.
Uniformity of heat is another concern. For the very common type of kiln that consists of three sections stacked vertically, temperature stratification can be a problem. The top is always hotter than the bottom, so work on the top shelf is either over-fired, or work on the bottom is under-fired, or both.
A zone control system is the cure for this. Each section’s heating coil is controlled by its own relay. The temperature controller has at least two thermocouple probes, one for the top and one for the bottom. The controller can then turn each kiln section on or off as appropriate. In this way, a uniform temperature can be maintained in spite of the rising hot air.
Another problem that a certain number of studios face is insufficient electrical service to run all their kilns at once. This is another area where modern temperature control equipment can be helpful. Generally, when a kiln is ramping or soaking at a moderate temperature, the heating duty cycle is actually quite low. But with on/off control, when a kiln is on, it’s on full. So if it can be guaranteed that no two kilns are on at once, it will seem to the electrical service that just one kiln is going. The controller forces the two (or more) kilns to agree to share the available power.
An area of concern is reliability and the handling of fault conditions. One of the questions most frequently asked is: “What happens when power fails?” Many older kiln controllers simply shut off—and stay off—when power is removed even momentarily. It is very important to users not only that temperature profiles be maintained in the absence of power, but also that the controller continues its operation after power is restored.
For short power outages (and most are short), the user should not even notice any effect on the operation of the kiln. For long outages, the controller may not be able to decide on exactly what to do—that requires human intelligence—but the controller should behave in a way that tries to stabilize the situation until the artist returns and can decide on the best course of action. It may even try to contact the artist to tell her that it needs her urgent attention. Some commercially available controllers* include an alarm output relay that can be used to trigger a light, a bell or even an automatic phone dialer in the case of various fault conditions.
Other exceptional conditions that modern controllers should be able to understand and deal with include a burned-out thermocouple, loss of profile information or a stuck external relay.
One seemingly obscure fault condition that actually happens much more often than one would think is when the thermocouple probe is not in the kiln—either it was removed and not put back, or someone tripped over the lead, or gravity has worked its will. One controller* features a special algorithm to reliably detect this situation.