Page tree
Skip to end of metadata
Go to start of metadata

Where To Find This Example

Select Help > Open Examples... from the menus and type either the example name listed above or one of the keywords below.

Or in Version 13 or higher you can open the project directly from this page using this button. Make sure to select the Enable Guided Help before clicking this button.

Design Notes


This example shows both a circuit model

and EM simulation of a balun

     . The graph "Transmission Characteristics"

shows the input return loss and through loss. The circuit model was designed and optimized with discrete variables, so that transferring the layout to the EM simulator is a straight-forward process, since the layout dimensions are already on grid.

Circuit Model Details

The schematic "Balun_Ckt_Model" makes use of several EM-based models. The coupled lines (M3CLIN) are modeled with an internal cross-sectional quasi-static solver. The MTEEX, MCROSSX, MSTEPX and MGAPX elements are table-based models derived from EMSight simulations.

All of the optimization and tuning variables are kept on a 5um grid in the y-direction and a 25um grid in the x-direction with the use of discrete variables. The array "N1", for example, ranges from 10 to 100 with a step size of 5. Variables for the width and spacing of the coupled line sections are set equal to a value in this array. When tuning or optimizing a variable that references an array, you do not need to set the upper and lower constraints on the variable, as these are automatically set to the limits of the array.

The circuit model was optimized to minimize S11 as well as the "Differential Output Error".

This is defined in the "Output Equations" section, and it is the sum of the two vectors S21 and S31. In an ideal balun, these two vectors would be of equal amplitude but 180 deg out of phase. Therefore, if you take the sum of the two vectors they should equal zero. The magnitude by which they differ from zero is the balun error.

The port impedances are 50 ohms on port 1, and 25 ohms on ports 2 and 3. The variables "Zin" and "Zout" are defined in the Global Definitions page.

EM Simulation Details

The schematic has an extract block to use AXIEM to simulate the design. Enable this block to see the layout extracted to a new EM document and simulated with AXIEM.  This AXIEM simulation has been configured to simulate all the metal as thick metal.   The EMSight thick metal shows an approximation to thick metal.   

Two EMSight EM structures were created, "EM_Thin_Balun" and "EM_Thick_Balun". The difference is that "EM_Thick_Balun" uses an extra layer of metal where the 'top' of the metal would be (3um above the substrate). Vias connect the top and bottom metal layers about every 300um.

Note that the separation from the structure to the wall is at least 3 times the substrate height on all sides (including the sides with the ports.) This is important, because the boundary of the enclosure is a perfect conductor, so if it is close to the structure it will add extra shunt capacitance. On the sides with the ports, a long distance between the port and the walls separates the port discontinuity from discontinuities in the structure. If these discontinuities become close, they can interact and produce erroneous results.

Because the port impedances are not all 50 ohms, extra schematics are set up with the correct port impedances so that the s-parameters can be displayed correctly. These schematics are "EM_Balun_with_Thick_Metal_and_25_Ohm_Zout"

and "EM_Balun_with_Thin_Metal_and_25_Ohm_Zout".