George Vendelin,in his book(MICROWAVE CIRCUIT DESIGN USING LINEAR AND NONLINEAR TECHNIQUES 2nd Edition Page no. 249 & 250) tells that shunt stub is the only viable option for stub design.The question is why. I guessed it is because we get the ground plane readily available while implementing shunt design using microstrip lines.

Now the choice narrows down to whether it should be open or short configuration.

On discussion, my friend pointed out that there's no standard for what will be defined as short because any 'short' termination will not exactly be short (for eg,by drilling a hole in PCB & soldering will be of a finite length & a finite impedance)

So we concluded that Open Shunt is the most viable choice for shunt design.

Are the views & reasoning expressed here correct?

  • \$\begingroup\$ There are other geometries available for stubs. Hopefully the book goes over them. \$\endgroup\$ – tyblu May 11 '12 at 20:49

Assuming you are designing a microstrip stub (not coax cable), the exact electrical length is quite tricky to figure out beforehand. An open stub can be manually trimmed for adjustment, whereas a short stub can't.

More in general, both open and short stubs can be used and will have different lengths in each case. The geometry of your design may better accommodate for one rather than the other. Not always you will opt for the shortest: sometimes the shortest stub will be too short for it to act as a proper microstrip line (i.e. comparable to the strip width).


added for emphasis Vendelin's thinking is rational. Although PTH shorts are not ideal, open cicuit tracks can be even less ideal with stray coupling to near conductors, also IMHO, shorted stubs are more reliable for EMI concerns. Use multiple plate thru holes (PTH) where needed in poured ground planes for microwave... every one (1) cm and use multiple holes for shunt stubs to reduce PTH inductance.

I believe shunt stubs are more immune to stray EM fields. It is for this reason they use standard 0 ohm shunts for calibrating a vector impedance network analyzer. I used these for getting -60dB return loss from 0 to 6GHz. Now an open stub would be more pray to stray EM fields than a short stub to EM currents.


Design references; enter image description here https://electrondevices.grc.nasa.gov/wp-content/uploads/R-12-June1997-MTT-Trans5.pdf enter image description herealso shows design of tuning stub in a shorted cavity for a microwave power transmitter. enter image description here http://www.jmpee.org/jmpee_site/Vol_44(4)/JMPEE44-4-178Bilik.pdf

  • \$\begingroup\$ Thank you Juancho & Tony Stewart for your response. 1. But Sir,how do we implement a short in a microstrip line? 2. From what I have read & seen(in this tutorial video: youtube.com/watch?v=z9XicSuk97Y) a hole is drilled & connection is made to the ground plane.Won't the solder cause any kind of parasitic effects? Pardon me for my inexperience. \$\endgroup\$ – WhizKid May 11 '12 at 18:29
  • \$\begingroup\$ you can experiment with fringe effects from bad soldering, also keep in mind a wire or plated thru hole is also an inductor, so 2 in parallel is better. If you keep irregularities a few% smaller than the 1/4 wave length, then it won't make much difference. Consider a solder blob as stub. Usually ok to ignore. \$\endgroup\$ – Tony Stewart EE75 May 11 '12 at 18:46
  • \$\begingroup\$ @WhizKid, If you use plated-through holes, there's no solder involved. The connection is made by copper plated onto the walls of the hole. \$\endgroup\$ – The Photon May 12 '12 at 16:17
  • \$\begingroup\$ @all:May you direct me to some definitive resource on Microwave Design(apart from Vendelin) or Ckt Design in general, which explains such considerations. \$\endgroup\$ – WhizKid May 13 '12 at 9:19

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.