Hi,
When I am analyzing a coplanar waveguide with and without ground plane I get some strange results.
Just for example:
- dimensions are in mils, Er=4.3 (FR-4)
- W=20, G=6, T=2.1, H=58, L=1000 (L is not important for impedance calculations)
- with groundplane I get Zo=53.9 Ohm (1), and without ground plane Zo=41.7 Ohm (2)
- still "with groundplane", when I change H to be H=58000 (equivalent to no ground plane), I get Zo=54.7 (!!!, compare to (1) and (2)). It means that groundplane below has almost no influence on Zo, which is reasonable because it is far from the signal line. But Zo is totally different from "no groundplane" option. Looks strange. Which one is correct and why there is a difference?
I guess that there is some problem with calculation for "no groundplane".

Any comments will be greatly appreciated.

Sincerely
Pavel

Greetings PavelM:

I get the same results as you do when using the latest version of AppCAD (3.0.2).

AppCAD:
Z = 53.9 ohms for H = 58 mil
Z = 54.7 ohms for H = 1000 mils

I also plugged the numbers into AWR’s TXLine program (http://web.awrcorp.com/Usa/Products/Optional-Products/TX-Line/) and got these results:

TX Line: CPW w/ground plane
Z = 55.7 ohms for H = 58 mils
Z = 56.6 ohms for H = 1000 mils (same as for 5800)

TXLine: CPW w/o ground plane

Z = 55.0 ohms for H = 58 mils
Z = 56.9 ohms for H = 1000 mils

Yet another online calculator (http://chemandy.com/calculators/coplanar-waveguide-with-ground-calculator.htm) gives the following (I converted from mils to mm when entering):

chemandy.com:
Z = 63.5 ohms for H = 58 mil
Z = 64.7 ohms for H = 1000 mils


Part of the difficulty with calculating an answer lies in the iterative approach needed to calculate an effective dielectric value, as well as simple models neglecting fringe fields, conductor type and thickness, etc., as the chemandy.com model evidently does. The others are pretty close to each other.

Once the height is larger than the gap/height ratio gets small, the results do not change much with increasing height. The approximation used in simple calculator test gap/height ratios and apply different equations accordingly. It would take a field solver to hone in a really precise answer.

There is a good paper in the IEEE Library, but you need to sign in to access it:
“Accurate and simple closed-form formulas for coplanar waveguide synthesis”
T.Q. Deng, M.S. Leong and P.S. Kooi

_________________
- Kirt Blattenberger
RF Cafe Progenitor & Webmaster

Hello, Kirt
Thank You for your time and informative comments.

I will try to explain my needs and then to summarize my understanding of your comments.
I am developing a very low noise analog front-end with low pass filter for high speed signal acquisition board.
The corner frequency is at few hundred MHz (sorry, I can't be more specific). It is a Bessel (constant group delay) filter. I try to do it with discrete components (L-s and C-s). Those are quite good components with high Q (>10) and low production tolerances (<=2%). My problem is to take parasitics of PCB layout into consideration in selection of discrete L and C values. I plan to calculate the L and C of PCB and reduce values of correspondent discrete components by the same amount. Usually, PCB calculators provide transmission line characteristic inpedance Zo and wave propagation speed Vp. I use them to calculate L=Zo/Vp and C=1/(Zo*Vp) per length's unit. So, correct prediction of PCB absolute equivalent electrical characteristics (Zo and Vp) is quite critical for me. Let say, my design is sensitive to difference in trace inductance of 0.5 nH vs. 0.7 nH and I want to be sure that my prediction is reliable.

Now, to your comments.
1. As I understand, you agree that AppCAD is misleading in prediction of impedance for CPW w/o GNDplane.
Quite surprising for me, but I will take it as it is.
2. You provided results from two more calculators, and they are quite different. Zo of 55.0 ohms vs 63.5 ohms is a big difference for me, as you already understand. You also wrote that "chemandy" calculator is "worse" (or not ?) by definition, because it does not take (or does ?) in account the fringe fields etc. etc.
Which one would you use for my specific need?

Can you or anybody else provide any suggestion, especially at the higher level of the task.
For example, may be, to use significantly lower value of H will be better (i.e. to put a GND plane closer to signal layer). It will increase the Cpcb, which may be compensated by appropriate selection of Cdiscrete with good !!! (0.1+-0.05 pF) resolution, but, simultaneously, it will decrease Lpcb, and it will make selection of Ldiscrete (which are available with not so high resolution, but only with 10+-2 %) much easier.

Any other proposals/advices will be appreciated.

Thank You for your time.


Sincerely
Pavel

Hi pavel and Kirt (and anyone else)
I have played around modelling the closed and open stripline/microstrip situation, and compared a number of formulae and EM software.
I ran the parameters through my resulting excel calculator and for the closed case and 58mil spacing
i) 53.9 ohms
ii) 53.38 ohms
iii) 53.28 ohms (I think linxtechnologies
.com?)
iv) 53.41 ohms (Richardson's formula)
v) 54.35 ohms (a rival website to this one)
but for several of those models I had to use a fudge-factor to allow for the fringing fields.
To derive that I used fairly recent work by the University of Taiwan that has resulted in design graphs from Finite Element Analysis of the accurate fringing fields. All of those checks were for a symetrical groundplane relative to the center trace.
For the open micro-strip case I get figures around 100 ohms for the width stated (20 mil) and that agrees fairly closely across a number of models.

_________________
At bottom, life is all about
Sucking in and blowing out.

Hello,
After long silence I think it will be quite informative to provide an update about my experience from building a practical circuit, using of AWR TX-Line.
The circuit's purpose is to work as high speed differential ADC driver with differential anti-aliasing filter. After some advices (see discussion above) about TL calculators, I used an AWR TX-Line in order to calculate the Zo and Vp of transmission line, knowing the PCB stack-up and TL geometry. Then, having line fragment's length, I calculated a fragment's equivalent L and C.
Then, I used those calculated L and C as parts of the filter. Since filter has its pass band of few hundreed MHz, each 0.1 pF of parasitic capacitance and 0.2 nH of parasitic inductance may influence circuit performance. Of course, filters configuration and order were selected in order to use those parasitics and not to fight against them.
Finally, I made PCB (FR-4), assembled and measured it.
It works just fine from the first time, as it is designed to do it. All characteristics are by design without any need to change any component.
So, looks too good to be true. But two more circuits behaves the same way.
So, this way works. And I want you to know it.
May be, you will use this experience in your work.

Good Luck!
Pavel