# Thermal runaway in resistive elements

I'm working with some 230 VAC resistive heating elements which are switched by a solid state relay (https://uk.rs-online.com/web/p/solid-state-relays/9045728) and driven by a 4-20 mA signal from an external PID controller. There are 4 x resistive elements and all of them are in parallel. The resistance of the elements (before the initial test) were as follows:

R1 = 86 ohm R2 = 85.5 ohm R3 = 85.7 ohm R4 = 86.9 ohm

The test was then conducted with all 4 x resistive elements inside a stainless steel cylinder bonded with thermal compound and the temperature raised to 600°C and controller via the external PID controller.

I'm having some problems with the resistive elements themselves. For some reason, after around 10-12 hours of testing one of the elements either short circuits or goes open circuit (I've run this test multiple times and I continue to get failures). My questions therefore are as follows:

1. Is the reason the element is going open / short circuit because the SSR is too harsh on the elements themselves? Should I instead be using a Silicon Controlled Rectifier that will allow a gradual voltage change across the resistive element? Something similar to this: https://uk.rs-online.com/web/p/signal-conditioning/8513563

2. When one of the elements goes open or short circuits, the resistance of the other elements jumps from around 80 ohm to >200 ohm (Note that I'm not measuring in parallel, I take each element away and measure the resistance). Is the reason for this similar to the thermal runaway effect found in diodes and transistors in the respect that if one element breaks, the other elements will try to compensate by maintaining the same level of current?

For information, the resistive elements are 230 VAC devices and are capable of handling up to 700°C. The chamber in which the elements are installed only reaches 600°C and the maximum temperature the elements themselves reach is around 610°C so the elements themselves aren't overheating.

The elements after their 600°C test and after returning to room temperature changed to the following values:

R1 = 85.7 ohm R2 = 84.8 ohm R3 = 85 ohm R4 = 85.9 ohm

The jacket of the chamber is water cooled by a booster pump that provides more than enough cooling capability to keep the entire chamber at ambient. I would suspect that we're overdriving the elements rather than overheating them.