mmcp42:
as the probe is inside a metal sleeve, there does seem to be some inertia in the measurement now
"Dead-time" in PID jargon.
I may well add PID processing - which should stop the overshoot
Not necessarily. There is a conflict between getting the response you want, the dead-time, and overshoot. If possible, you should first strive to reduce the dead-time as much as possible. That will make working with PID much much easier.