Archive for the ‘Microstrip’ Category

The HP Half Wave 50 ohm transmission line trick

Sunday, October 20th, 2013

If your 50 ohm microstrip transmission line varies somewhat in impedance when you are connecting 2 50 Ohm elements together you can use a 1/2 wave length of line in between the blocks if space is available and the design is narrow band. What will happen in a full spin around the Smith chart bringing the load end of the transmission line back to where is started at feed end.

Using WinSmith Program to Design Matching Circuit of 900MHz ISM Band Power Amplifier

Sunday, October 20th, 2013

WinSmith is a Smith chart match calculating program.  The starting point is 8 – j16 ohms @ 915MHz.  Both the blue and the green curves start at this point at the lower left of the curves.  Click on the image to see a bigger version

DrainMatching_MW7IC915NT1_WS

What is interesting about the above plot is that the elements work orthogonally.  That is to say for any point selected by component of move #2 or move #3 work independently.  

The matching circuit is shown below:

DrainMatchingCircuit    

Nominal values

  • TL1:  45 degrees of 50 Ohm transmission line @ 900MHz
  • C2:   25pF 
  • C1:   7pF

The left side is 50 Ohms match and the right side is what presents a conjugate match to the FET IC drain circuit.  Move #1 on the Smith chart is the 45 degrees of 50 Ohm characteristic impedance transmission line.

Tuning Procedure- theoretical

  • The output load of the FET MMIC amplifier will vary somewhat.  In spite of this the 45 degrees will transform the load to a region in the charge where the two caps can move the load to the close to the center of the chart.
  • C2 has one job only:  move the load to a 0.02 conductance circle on the Smith chart.  This is the conductance circle that corresponds to 50 Ohms in shunt.  Once you know the value of this capacitor you solder in place and leave it.
  • C1 is used in shunt to bring the whole thing to the center of the chart.  A calculated value can be used for a first try.  You may need to cut and try to get absolute optimum on your lab model.  

Input Match

The input match is the same topology.  It starts much closer to 50 Ohms.  The nominal values are

  • TL1: 60 degrees of 50 Ohm line at 900MHz
  • C2: 22pF 
  • C1: 0.5pF – This is placeholder in case of surprises.  The match looks accomplished without it.

MW7IC_InputMatch

 

Links

 

Microstrip Filling Factor versus Relative Dielectric Constant Given fixed Line Impedance

Sunday, October 20th, 2013

Research Links

The question arose given a fixed microstrip transmission line characteristic impedance how does the percentage of the field in the substrate change? This is called the filling factor:

Keff=ER*FF+1x(1-FF)

It is the fraction of the field in the substrate.  I calculated the following values:

  • FF=(Keff-1)/(ER-1)
  • Dk=4 Z0=51 Deff = 3.07 which yields a filling factor of 0.69
  • Dk=10 Z0=50.5 Deff = 6.72 which yields a filling factor of 0.635

Thus it appears that the filling factor goes down with Dk as Z0 is held constant.