### 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

**Yield Measurements**

This example shows a simple microstrip transmission line to show several yield analysis concepts.

__Overview__

The “Microstrip_Line” schematic has 5 of its parameters setup for yield. The H, Er, and T of the line are set on the MSUB model block with each value set to vary by 2% using a Gaussian distribution. The W and L of the line are setup to vary by 0.5 mils using a Gaussian distribution.

Yield goals have been setup for Zin >= 45 ohms and Zin <= 55 ohms.

From basic transmission line theory, we expect H, Er and W to have a significant influence on the impedance of the line and T and L to have very little influence.

Run yield analysis by selecting **Simulate > Yield Analysis **from the menus and then selecting **Yield Analysis** from the **Analysis Methods** setting with 1000 iterations. This will get data on the graphs. What the yield dialog while the analysis is running to see the yield percentage and yield error.

Additionally, the yield data setup to be stored in a data set, so you can view the results as any time without requiring simulation.

__Yield Results__

The project contains the following graphs:

“Zin” - the input impedance of the line, used to base all the yield analysis in this project. All of the yield data is shown on

“Yield” - shows the yield percentage and yield error, same as is displayed on the yield analysis dialog.

“Sensitivity ..” - 5 graphs showing the component sensitivity. It is easier to look at this data on the “Histogram” schematic that uses window in window to show all the histograms in one display.

“Zin_Statistics” - Shows the Range, Mean, Median, 1 and 2 standard deviations of the yield run for the Zin measurement.

“Zin_Standard_Deviation_Absolute” - Shows 1,2 and 3 standard deviations of the yield run for the Zin measurement

“Performance Histogram” - Histogram plot with the measurement range on the x-axis and the percentage of results that fall in each bin.

“Yrank” - For each statistical variable, shows the difference between the yield percentage in the nominal value bin (can be found on the “Sensitivity..” plots) and the total yield percentage. It gives a feel for how much yield can be improved by removing variation from that component.

“Yrank Center” - For each statistical variable, shows the difference between the yield percentage in the nominal value bin (can be found on the “Sensitivity..” plots) and the yield percentage in the highest bin. It gives a feel for how much yield can be improved by changing the nominal value of that component.

“Pareto” - Shows the Pareto chart for the statistical variables for the Zin measurement. This measurement shows which statistical variable contributes the most to the chosen measurement. As expected, W, Er, and H contribute the most while T and L contribute the least.

__Yield Data Sets__

You can enable each project to store yield analysis runs in data sets. To turn this on, right click on the **Data Sets** node in the project tree and select **Options**. Then turn on the option for **Auto Save For Yield. ** With this option on, you will see a new data set added beneath the **Data** **Sets** **> YLD** node for each yield analysis run.

You can view the data from each data set by right clicking on each yield data set and select **Show Results. **

__Yield Optimization__

The yield of this overly simplified circuit can be improved. Because it is such a simple circuit, we could use TXLine to find the width of a 50 ohm line, which is 24.2 mils. However, in most circuits, the “right” answer isn’t this simple to determine.

One approach would be use the “Yrank Center” and “Pareto” chart to view the data. Both show that Er, H, and W most affect the Zin measurement and the circuit would benefit from centering these values. Er and H are not controllable by the designer in this case, but W is. The designer could change the nominal value of W and re run yield to see if the yield percentage improves. The first choice would be the value where the W sensitivity plot has a maximum (the “Sensitivity W” plot). Then the user could iterate to maximize the yield.

Another approach is to run yield optimization by selecting **Simulate > Yield Analysis **from the menus and then selecting **Yield Optimization** from the **Analysis Methods** setting. This analysis will find the optimum value of W after several iterations of optimization and this value will be close to 24.2.

The details of yield optimization are covered AWRDE Simulation and Analysis Guide.