How about remote infrared sensors, you'll have to turn the rf off temporarily. The sensors have focuses of various lengths, if this is a microwave oven, you may be able to make it work.
kch
Anything at 800 degrees C is going to be radiating a lot of energy, including microwave energy. Build a simple radiometer, and you can probably sense the temperature just fine, especially if whatever you are heating up remains constant.
Radiometer requires access to the raw emittance of the object. Either the radiometer needs to be placed within the cavity, which is a non-starter, or it needs to have a window, which would leak microwave radiation, again, a non-starter.
The OP needs to post more information about object or material is being heated, how often the measurements need to be performed, allowable latency, is heating allowed to stop temporarily, etc.
Why? You mean you can not design a horn structure that would pass, say 60 GHz energy, but block 2.4 GHz heating energy? How about 1 foot of 1/2 inch tubing? Easiest thing in the world!
I guess it's a question of how much attenuation you're willing to tolerate, since it will most likely affect the performance of MMW radiometer. A commercial microwave oven uses a grid with about 2 mm openings to ensure negligible leakage.
I personally don't have enough expertise to design a radiometer for this short a workign distance, nor do I think the OP has that experience. I also don't think that calibrating such a device is a trivial matter.
If you're offering to do the design and calibration, then by all means...
There would be no attenuation of the millimeter wave signal. It would propagate normally. You might wish to study the properties of a cutoff waveguide.
I too need temperature monitoring inside a microwave cavity (oven) and have considered the Luxtron. I also need multiple depths to be monitored at the same time. What about using stainless steel shielded thermocouples where the shield (only) is grounded to the cabinet? The probe diameter can be selected from 0.010 to 0.1 inches when purchased. Too small a diameter may self-heat but too large a mass/diameter might have the temperature distorted since the material I am heating is a dielectric powder. The TC element can be read out by battery operated and thus isolated liquid crystal display. The diameter can be easily changed by experiment.
Sorry, bad grammar. Regarding depths, I meant that temperature measurements at multiple depths were the goal. I envisioned 2-4 thermocouples with metal sheaths inserted in the material and not having any interference between them nor at the readouts with the microwaves on. The sheaths would be grounded at the cavity wall but I worry that the probe-like geometry might still cause arcing between the probe sheaths or to the cavity wall. They come with rounded tips but electrically, they may still be sharp.
