Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

RF choke Implementation using Microstrip Lines

Status
Not open for further replies.

Maryam45

Newbie level 2
Joined
Jan 25, 2012
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,301
Hi everyone,

I'm designing a mixer for C-band. I need a help regarding the realization of RF chokes using microstrip lines. Can anyone suggesst me.
 

hi maryam
you know for microstrip RF choke you need lambda/4 line that shorted to the ground using capacitor or some thing like that.for lambda/4 line,considering the substrate parameters you can use calculators of AWR microwave office or ADS and calculate the lenght.for the parameters that you need to consider for this line are S21 and Zin.you should tune these two parameters in a way that you can relay on them.for example Z1n should be greater than 500 ohm and s21 as small as possible.for the capacitor you have two choices.first mounting some SMD cap (0603 or 0402).the value should be choosed in a vay that the real part of reactance smaller than 2 ohm.the other method is using radiall stubs.you can use CADs and tune the stub so that can be considered as short circuited in the place that situated
HOPEFULY IT CAN HELP YOU
 

young's got the two methods I've used for this type of circuit. The "shorted" quarterwave line is a good, but very narrowband solution. The radial stub will provide a wider bandwidth, but it's still a narrowband solution when compared to an inductor. Keep that in mind when determining your feed network's performance and the load impedance it presents to your mixer at frequencies away from your desired signals. Remember that mixers need to see good load impedances at the harmonic frequencies, lest you want to cause yourself some major headaches down the road.
 

hi
In the case of inductor,you should consider the SRF of inductor.the SRF should be close to the operating frequency
 

Thanx every1 for Ur timely reply.
I'll check your approach. In mixer I'm using two RF chokes to give path to DC current. I just want to realize it using transmission lines.my design is in ADS. please suggests me some reading stuff that can tell the steps to make it possible using ADS.
I've a 1 more question.Actually I want to plot the conversion loss versus RF power graph in ADS. I'm not interested to use the build in equations of mixers templates in ADS design guide. I've used this equation ConvGain= dbm(V_out) + RFpower using measeq. But plots are not correct one inotherwords there are no results. please guide me how to write this equation.
 

hi
I'm using AWR microwave office.I you like I cat tell you about that
 

Thanx every1 for Ur timely reply.
I'll check your approach. In mixer I'm using two RF chokes to give path to DC current. I just want to realize it using transmission lines.my design is in ADS. please suggests me some reading stuff that can tell the steps to make it possible using ADS.
I've a 1 more question.Actually I want to plot the conversion loss versus RF power graph in ADS. I'm not interested to use the build in equations of mixers templates in ADS design guide. I've used this equation ConvGain= dbm(V_out) + RFpower using measeq. But plots are not correct one inotherwords there are no results. please guide me how to write this equation.

It's not that hard, if you understand how a quarterwave transmission line moves you around the Smith chart.

Start with a simple, long 50 ohm microstrip, put an MTEE in the middle, then add another 50 ohm line out the other side (you'll have a long 50 ohm line with a tap/DC feed point in the middle). Put TERMs at the ends of the 50 ohm line. If you measure S21, you should get a very low loss, and S11/S22 will be a high return loss (around your freq of interest).

Next, add a high impedance (narrow) transmission line to the open port on the tee. Make this line 1/4 wavelength long at your center frequency (90 degrees). At the end of the quarterwave line, add a capacitor to ground. The capacitor should appear to be very low impedance at your frequency of operation (looks like a short circuit). When you transform a short circuit around a quarterwave line on the Smith chart, it will look like an open circuit. This effective open circuit will appear at the tap point on the tee, so overall, the 50 ohm line will have an "open circuit" hanging off the tee. This is only the case near your frequency of interest, since a quarterwave line is only 1/4 of a wavelength at one frequency.

Next, plot S11/S22 and S21 vs. a broad frequency sweep. You can tune the location of the notch in S11/S22 by adjusting the length of the quaterwave line. Once tuned, you should have a very low insertion loss and a large return loss (just like a good 50 ohm transmission line should have, except that it only looks good around your freq of operation).

For some good reading on how quaterwave lines work, check out Microwves101.
**broken link removed**
 
  • Like
Reactions: ferdem

    ferdem

    Points: 2
    Helpful Answer Positive Rating
In additon to what enjunear has suggested: the next optional step is to replace the capacitor by a radial stub. Design the radial stub with radius = 1/4 wavelength, so that it represents a "short" at your frequency of interest. Together with the narrow 1/4 wavelength line, this is then transformed into an RF "open" at the circuit.

 

In additon to what enjunear has suggested: the next optional step is to replace the capacitor by a radial stub. Design the radial stub with radius = 1/4 wavelength, so that it represents a "short" at your frequency of interest. Together with the narrow 1/4 wavelength line, this is then transformed into an RF "open" at the circuit.

Volker is spot on with this recommendation. The radial stub approach is especially useful at high frequencies (low freqs and low Er dielectrics will make your quarterwave lines and radial stubs very long... and often, vias at these high frequencies are notoriously crummy, i.e. slightly inductive). The radial stub makes a nice AC-ground for RF.

To see the difference between a radial stub and an open-circuited transmission line, look at the Quarter-wave Tricks link I posted; see about 1/3 of the way down the page for the section "Radial Stubs". Most stubs work well with an angle of 60-90 degrees (the wider, typically the better broadband performance you get... if you have room). The radial stub's response is based on Bessel functions, so the easiest way to tune them is to sick your favorite optimizer on them and let your PC do the heavy lifting.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top