waveguide pressure window

Anyone know how to go about designing a waveguide pressure window for high power (say 20 KW)? What type of glass/ceramic would you use? What is the window thickness (quarterwave, half wave, or as thin as possible)? Anything else to consider? 2450 MHz.

For obvious reasons, wavelength matching doesn't work at 2.4 GHz. So "as thin as possible" is next best option and low Er (glass rather than ceramics) would be preferred. Of course, you can examine the exact impact of introducing the window with an EM simulator.

Does wavelength matching means landa/2 ?
I dont see how it wouldnt work.

I understand it as a practical question. Practically, it doesn't work because of the windows dimension at 2.45 GHz. To mention some detail aspects related to a "thick" window for high power: Bulk absorption, losses, material heating, thermal expansion.

Let say, material dielectric constant is 2.
A half wavelength thick slab would be approx 4cm thick and with a low loss tangent material. probably rexolite...

but i suppose there are more question need to be ask

1) what is the pressure the waveguide need to withstand
2) 20KW peak, average, CW, pulse ?

Average power 20 KW, and 100 PSIG so I want the window to be thick enough to take that pressure.

I am just trying to figure out how the standard design goes for something like that. Probably do not want half wave of glass because of the length, but if I had say 1/8 wavelength of glass that would probably be a pretty big VSWR. I could match it out, with say an inductive iris further down the guide, but I have never seen one made that way. Luckily, it is very narrow band 2450 +/- 15 MHz.

well, I worked on pressure windows and I've seen and used some pressure windows.
Many of the commercial ones was designed for the radar market, while I've used them on another context.
The WR187 windows was made by a composite material similar to a woven based epoxy while the WR90 and WR28 was made by a Quartz.
The epoxy was firmly "clamped" between two metal flanges.
The quartz windows was attached with "metal to ceramic" brazing technology.

But from a EM point of view, an homemade window, a very thin PTFE or Kapton tape (i.e. 25 microns for WR22 to WR75), is much less lossy, even without chooke rings.

My experience is based mainly on vacuum environments ( i.e. 1 Kg/cm^2). But I know , that in hi power , the pressure of the gas HF6 is typically less than 1 Kg/cm^2

Yes, as I found out, most of the pressure windows out there are designed for vacuum chambers--15 psig pressure. I have some of those. My problem is that I need higher pressure--4 x tha amount. The fact that the WR430 waveguide opening is so big makes it even more interesting.