I have the following circuit: enter image description here

which I need to implement in Microstrip. For now in order to get the correct results I was using lumped capacitors. But I will need to use Microstrip stubs when fabricating the board. How can I convert the circuit into one using stubs instead of these lumped elements.

Looking forwards to your suggestions.

  • \$\begingroup\$ What are you actually trying to accomplish? For example, are you trying to make a low-pass filter? Or some kind of coupler? Or a balun? Or what? \$\endgroup\$ – The Photon May 22 '19 at 16:56
  • \$\begingroup\$ I am trying to make a directional coupler. \$\endgroup\$ – hacker804 May 22 '19 at 17:04
  • \$\begingroup\$ What frequency band? \$\endgroup\$ – The Photon May 22 '19 at 17:13
  • \$\begingroup\$ 2-4GHz, with a center frequency of 3GHz. \$\endgroup\$ – hacker804 May 22 '19 at 17:15
  • \$\begingroup\$ For that band and ~0.2 pF, I would expect you could use discrete capacitors. You might get eaten alive by tolerances though. \$\endgroup\$ – The Photon May 22 '19 at 17:24

You don't want stubs, they are capacitance to ground. All of your capacitors are shown as floating. For this you'll want interdigitated capacitors, it's going to be hard to get the etching tolerances to control a small gap well enough to get fractions of a pF with any accuracy.

If on the other hand this is a balanced circuit and you can split each capacitor into two series elements, and ground the mid point of each, then you can use stubs, which being capacitance to ground are much easier to implement. Look for radial stub formulae, they will be easy enough to find with your favourite search engine.

|improve this answer|||||
  • \$\begingroup\$ Shouldn't they be able to implement the capacitors they have by reducing the spacing between the coupled lines for some distance? \$\endgroup\$ – The Photon May 22 '19 at 17:13
  • \$\begingroup\$ So I should take the coupled lines as equivalent to a parallel plate capacitor and find the distance? \$\endgroup\$ – hacker804 May 22 '19 at 19:31
  • \$\begingroup\$ @hacker804, probably not. There are formulas (I think) for coplanar plate capacitors, which would be a better fit. But you could just add another coupled line segment to your ADS model and optimize on the S parameter instead. Given that your separation is already down to 0.13 mm for the closely coupled segment, you might not be able to get much closer (maybe 0.075 mm) if this is a ordinary PCB. If this is a thin film circuit you can probably get much closer. \$\endgroup\$ – The Photon May 22 '19 at 19:57

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.