1
\$\begingroup\$

I built a PCB filter in Class "Band-pass" with center frequency of 2.4 GHz. The board has two SMA's (input 1 & output 2).

S(1,2) looks like this:

enter image description here

The frequency is good, However, the attenuation is bad. My problem is that my Reflection signal S(1,1) looks like:

(I thought I would see a valley going down at the 2.4 GHz frequency)

enter image description here

(Note: I am using a transmission line design)

The PCB Circuit:

enter image description here

Specifications:
Filter Type: Chebychev bandpass
Ripple: 0.5 dB
Ctr. Freq.: 2.437 GHz
Att.: 50 dB @ 2.459 GHz
Tgt. BW: 20 MHz

Substrate: Rogers RO4350B
εr : 3.48 +/- 0.005
Tsx: 1.524 mm
TCu: 35 μm

Board + Layout: enter image description here

enter image description here

\$\endgroup\$
10
  • \$\begingroup\$ Please share the datasheet of the device you're testing. \$\endgroup\$ – The Photon May 1 '19 at 15:29
  • \$\begingroup\$ What is the spacing between the elements of your filter? \$\endgroup\$ – analogsystemsrf May 1 '19 at 15:45
  • \$\begingroup\$ It is a PCB board I designed, should I attach an image of it? \$\endgroup\$ – Anwar Elhadad May 1 '19 at 15:46
  • 3
    \$\begingroup\$ The answer is that you've not built a PCB filter, you've built a PCB thing which is almost totally mismatched. If you want help with the design, then you need to post what type of filter you're trying to design, what order, what specs (bandwidth, attenuation depth at what offset), what design program you've used, and a picture of the board. \$\endgroup\$ – Neil_UK May 1 '19 at 16:35
  • 3
    \$\begingroup\$ Those look to be humungeous great gaps, what synthesis program came up with those? A gap of 4 or 5 substrate thicknesses is essentially not coupled, unless, perhaps, you have extraordinarly high Q resonators, which you don't have with copper. Are you targetting a very very small bandwidth? \$\endgroup\$ – Neil_UK May 1 '19 at 18:23
1
\$\begingroup\$

Thanks, everybody. I think I solved it. The reason was the wires (cables) resistance was high so it was causing the attenuation. I replaced one of them and I was able to see better spectrum.

\$\endgroup\$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.