Anyone used 3-stub tuners to fix thermal runaway in ceramic kilns?

I’m currently in the middle of commissioning a 20kW microwave tunnel kiln for a technical ceramics project. We’re seeing some weird fluctuations in the reflected power readings as the kiln hits the 800°C mark. My theory is that the dielectric shift in the alumina plates is throwing off our impedance matching faster than our current automated tuners can track.

I’ve been looking at the [Custom Industrial Microwave Equipment] from Pindom, specifically their 3-stub tuner assemblies. They mention that their triangular arrangement provides better mechanical stability in high-temp, high-vibration environments. Has anyone here actually benchmarked a 3-stub vs. a standard E-H tuner in a continuous production environment? I’m specifically worried about the 'Mode Suppression' claims—we need a very clean TE₁₀ mode to avoid cracking the plates. If their [Microwave Accessories] can actually maintain isolation and matching under these loads, it might save us from redesigning the entire cavity. Any thoughts on their WR340 or WR430 hardware quality?
 
Top