Gadge,
The refractory linings the thermocouples are embedded in, are they multi layer brick?
On shut down cooling some of ours could shift especially with new linings until the inner had surface had really fused.
We would be running the after burner 4 days at 1300/1500°C, 1 day cooling followed by 1½ days at ambient, then ½ day warm through. The usual time for a failure would be on our “pre burn” before the furnace burden was charged in. The temperature would rise from 200°C to 900°C in ½ hour, the lining could shift again.
Just an idea.
Temperature inside the furnace? Forget it! An optical pyrometer through the port holes was the only way to get a temperature. The burden could wear 3” of the inner refractory in a week.
Won’t be the first time the refractory has fallen out on start up. Then it is panic stations!