I've been looking at some reference designs to guide my own design recently, and I noticed some possible discrepancies on one in particular.

This is the reference design:

http://www.ti.com/lit/ug/tidubq7a/tidubq7a.pdf (description)

http://www.ti.com/tool/TIDA-00848?keyMatch=cc1310&tisearch=tidesigns (page and design files)

In section 9.3 it states the feed line was modified to be 50-ohms with the 1mm thick PCB. The design files show a microstrip line width of 24 mils and a spacing of 6 mils from the ground plane on each side. However, doing the maths on these figures as a co-planar waveguide (with Saturn PCB toolkit) the impedance actually comes out more like 56-Ohms.

Is this correct? And if so, is it just a case of the feed line being so short they don't mind it's a bit mismatched?

If it's not correct, how was it calculated? By my calculations the feed line would have to be about 33 mils wide to have the correct impedance.

  • 1
    \$\begingroup\$ What dielectric constant did you use? It varies with frequency to a greater or lesser degree depending on the specific PCB material. Note also that 10% is about the best you can hope for in a practical PCB without special tweaking. \$\endgroup\$ Sep 21, 2016 at 9:46

1 Answer 1


The implementation is a coplanar waveguide with ground. See the layout - the bottom of the board is flooded in this area.

Coplanar waveguide with ground

This calculator yields 50.13 ohms with a dielectric constant of 4 (a decent starting point).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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