Kiln elements keep burning out

Question

So, I bought a medium kiln on craigslist for $100. It was old, and my friend and I wanted to upgrade it, so we bought some new heating elements, removed the entire controller-box, bought a temperature controller and a bunch of wiring. We only kinda know what we're doing - I took a sculpture class back in college, but that's it, and my friend has never used a kiln. We replaced the elements and connected them in parallel to ~220VAC. (We had some problems with the temp controller and the connected relay, but that's a different issue, and less important.) It worked for a few minutes; but then an element burnt out. Fast forward a few months, and we tried again with just the other two elements. The kiln lasted longer, maybe 30-60 min, but then another element burnt out. That's the brief overview - here's a heap of gritty details:

Kiln: Evenheat Classic Series Kiln - Model 1210, or a variation thereof, according to my records.
New elements: 3000W 220VAC FeCrAl Heating Coil

Now, here's the order of events, though I'm a little hazy on what happened several months ago.

1. To change the elements, we stretched them out to a bit longer than required (as we estimated), doubled the last few inches of each end back on itself, and twisted up that segment to be long and straight, so we could feed it through the kiln wall.
2. Then we put them in the kiln where the old ones had been, feeding the ends through the front as mentioned, to be attached in parallel to power.
3. We put a steel cylinder containing aluminum scraps into the kiln - we're a little more interested in melting metal than firing clay, though we may do some of both.
4. When we turned on the kiln (i.e., connected the elements to 220VAC), we let it heat up to a bit above the target temperature, as measured by a k-type thermocouple near the center of the upper half of the kiln, I believe. Then turned off the kiln for a minute or two, then turned it back on, etc., to heat up the whole kiln without overheating parts of it or something. We weren't sure it was ok to leave the thing on full power for too long. We left the lid closed. I think the kiln started steaming at some point; maybe the humidity/dampness evaporating out?
5. At some point, we looked into the kiln from a different angle, and noticed that while most of the elements were glowing a orange, one element had a ~3/4 inch segment that was glowing bright yellow. A minute later that element burned out. Our best theory is that there was a nick in the element, increasing resistance in that one spot.

A few months later, we tried again.

1. We didn't replace the burnt-out element, and just used the other two.
2. We on/off cycled the kiln on the way up the temperature scale, because we weren't sure about all the steam, and didn't want to blow up the kiln or something. Also, note, after 30 min there were some acrid fumes. Lid was still closed most of the time, though occasionally we'd open it to look. We didn't think to do a more thorough inspection of the elements powered on, though, unfortunately.
3. Over the course of 30-60 min, the kiln temperature reached a pretty solid 700-750°C. (On/off the thermometer would fluctuate by about 100°C, up to a peak temp of maybe 850°C at some point?)
4. We managed to melt the aluminum, which was exciting. Then, however, as we were melting more (immediately put it in, then power cycled every few min for about 10 more min) another element burnt out. We didn't see the spot before it burnt out, but I do have pictures of it after the fact:

(We think that's the burnt out spot.)

We measured the voltage over the elements at between 220-240VAC, one specific measurement being 236VAC. Amps per element was about 16A. This yields 3776W per element, rather higher than the listed 3000W.

According to my research, I'm pretty sure heating elements are supposed to last much longer than an hour.

What could have caused the elements to burn and how can we avoid it?

Is the voltage too high?
Is it purely a matter of nicks in the elements?
Are they just oxidizing away?
Do we need to keep the lid closed?
Do we need to cycle the power more frequently?
Do we not need to worry about sudden temperature change of the kiln, after all?
Is the power-cycling causing expansion/contraction damage?
Is the reduction in length from the folding-over decreasing the resistance too much?
Are these elements simply garbage?
Above all, how do we not burn out more elements?

Answer 1

When the elements are connected directly to power, the only limitation is the inherent resistance of the elements. Ohm's law states that current is inversely proportional to resistance. The elements are going to be low resistance, which results in high current, and high heat, of course.

Unfortunately, in your situation, the high current exceeds the capacity of the elements. I'm uncertain if you would be able to find elements capable of handling "unlimited" current, primarily because if you have such high current, you'll probably trip a breaker or burn something else.

I'm not sure about the old technology stuff but today's controllers are digital and will pulse the current on and off to accomplish the desired temperature and heating profile.

If you can find a rather large power resistor, you could place it in series with your current source and drop the current to a non-element-burning level, but that's a bunch of experimentation and possible expensive resistor purchases.

I re-read your comment. The elements should be in series, to increase the resistance over the full circuit. When in parallel, the resistance of the circuit is reduced, not increased. That's an important distinction.

Answer 2

The problem could be the Kanthal ( Fe, Cr, Al ) elements. Although they will tolerate a little higher temperature , they are "delicate" compared to the Nichrome type elements. Most consumer heat elements are Nichrome because it will tolerate more abuse ( eg. range elements). Kanthal becomes very brittle once it has been heated. Also it is weak when hot so must be properly supported. Also it is susceptible to deterioration when contaminated by silica and other agents. If Kanthal is the problem , the solution is easy but expensive - replace with a Nichrome ( Ni, Cr, Fe, Al) elements.