waveguide attenuator - waveguide attenuator theory?
Post27 Apr 2007 5:01 waveguide attenuator theory?
A rectangle waveguide with a resistor film putting in middle of wide edge. This resistor could rotate,thus control the attenuator value. Book said only electical field parallel to resistor film would be absorbed. Could anybody explain it? Thanks.[/img]
I have the same problem, exactly with frequency independent. What is the thin resistor film ? How can I do it ? For example: waveguide R120 - cca 10 GHz - 15 GHz. I have dielectric slide and I spray on graphic spray (Rsq = 1000-2000 ohm/sq), Its OK, but: 10 GHz attenuation is 17 dB, and 15 GHz is 22 dB, so . professional attenuator has about 1dB difference from between 10 and 15 GHz
Thank for your answer
(And sorry for my english :D )
Well, the shape of the resistive card is, or course, important. I would vary the length until it is more uniform in loss over frequency. I would play with the taper on both ends until it has a good return loss over all frequencies.
If this is intended to be a "variable" attenuator, you need to move the card in and out of the waveguide thru a slit in the middle of the top of the widest wall. Your drawing implies you are moving it from left to right, which is not the way to do it. Also, those two holes in the side small wall will leak RF energy.
Thank,
I know, but problem is in the frequency independence S21 - transfer, if you like attenuation. What parameter have the resistive card ? For example resistance (Ohm/sq), size (length and thickness),.. ?
Taper on both is good for return loss, but what is good for frequency independence transfer ?
When you measure S21, does the attenuator have a standing wave upon it, or is the generator and load well matched?
I have dielectric slide, rectangle 8x50x0.4 mm (its for waveguide R120 19.05x9.525 mm), without he taper on both ends. The resistive film was graphit spray. I gave in it to waveguide (about center) and measured it with Scalar network analyzer (SNA), measuring schema was: SNA - bridge - BNC conector - waveguide with slide- BNC - SNA. S11 was OK, about -20 dB, but S21 was frequency dependent. Attenuator didnt has standing wave upon it, generator and load was well matched.
I think, that there is problem with resistive film, but i have not article with theory about this problem :(
And I have problem with simulate CST MWS, the simulation is different with fact.