Coaxial shielding in layout

Hi all,

How we do coaxial shielding in layout? What are the disadvantages of that.Can anyone give an explanation from layout perspective?

Coaxial cable is, by nature, shielded. The electric field that carries the desired signal resides between the center conductor, and the outer conductor (braid). Any signal that terminates on the shield from the outside, should be diverted back to its source (ground) at the cable's termination point.

Perhaps you can clarify your question a little more?

Hi enjunear,

I wanted to know how we protect the critical siganl using coaxial shielding in layout design.I think as in coaxial cable we are surrounding the critical siganl with metal routes.Please correct me if I am wrong. I think I posted in a wrong forum.

Maybe coaxial shileding word is confusing here. We use coaxial cable until the signal arrive the board. On pcb layout, you can use shielding, microstrip techniques, for example if you use coaxial to transmit an rf signal like gps,
there are some tecniques and softwares to match the impedance etc to eliminate coaxial-to-microstrip signal discontinuity. So onboard we use shiledings, microstrips, splitplanes, and signal guidelines to protect our signal.

annelizabath said:

Hi enjunear,

I wanted to know how we protect the critical siganl using coaxial shielding in layout design.I think as in coaxial cable we are surrounding the critical siganl with metal routes.Please correct me if I am wrong. I think I posted in a wrong forum.

Click to expand...

The term, coaxial, is incorrect for the context you are talking about. Coaxial means that you have multiple layers that are symmetric around the same axis... as in the small circular cross-section that defines a metal center conductor, and the metal ring that defines the braid, or shield, on a piece of coax cable (think about the coax cable that goes to your TV, if you cut that in half and look at the cross-section, that is coaxial in design).

Now, to discuss signal shielding on a printed circuit board. From a general concept, signal traces that run in parallel will couple energy between each other. In order to reduce this effect, you need to increase the separation between the conductors (E-field strength falls off as distance increases, so less energy can travel from one signal trace, to another). Additionally, you can provide a point of termination between the two signal traces for those stray E-field lines. This can be done by running a grounded trace between adjacent signal conductors. The E-field lines will tend to terminate more on the ground strip, than the further-away adjacent signal trace.

If you are operating at higher frequencies, you can use a transmission line structure to make the ground-signal-ground traces and spacing. On the board surface, you can use **broken link removed**, or pseudo-coplanar, where the gap between signal and ground is equal to the substrate thickness. For internal layers on a PCB, you can use **broken link removed**.

A good microwave engineering/design book (or many websites) can give you the fundamental equations you need to calculate the characteristic impedance of your transmission lines. Microwaves101.com is a great starters resource.