Problem with building air bridge at the CPW T-junction

[cuty81]

Hi,
Anyone has experience with CPW T-junction, please help me! I'm trying to design a SPDT at the center frequency is 20 Ghz in HFSS with the CPW configuration! the problem is that, I put 2 MEMS switches at the output arms with quarter wavelength distance from the T-junction, but I don't know how about the input port for the T-junction?? The ouput ports and input port are 50ohm. Do I have to match the input port??? Even though I match the input port, the result is not good! (S11 is about -6dB, so high for a SPDT using MEMS switches!)
If anyone also has experience with building air bridges at the CPW T-junction, please help me how to work with that!
Thanks in advance!

 

[snkhan]

Re: CPW T-junction

HFSS has one tutorial regarding T-Junction and its optimization. Have a look on it

 

[cuty81]

Re: CPW T-junction

Hi,
thanks snkhan for your reply! But CPW is very different from the waveguide like that! Anybody knows how to work with CPW in Ansoft, please let me know!
Thanks

 

[snkhan]

Re: CPW T-junction

Look for Ansoft HFSS cook book in the forum. I hope its good to start from that

 

[sinatra]

Re: CPW T-junction

Hello.

There is also another tutorial where they build a microstrip hibrid.
I think you can modify that tutorial and simulate your CPW.
Try first with a simple CPW transmission line and verify if the results agree with the formulas available for the CPW transmission line.

You can verify the impedance of the CPW line using the free trl85 program from ansoft that you can download at the following link:
https://www.elektronikschule.de/~krausg/Mikrowellentechnik/Stripline_Calculator_TRL85/

After you can copy the checked transmission line and make your Tee. Add another port and then you will be able to simulate it.

Probably the attached CPW tutorial can also help.

Hope it helps
S.

 

[cuty81]

Re: CPW T-junction

Hello,
thanks for help! Please let me know where I can find and download the tutorial for microstrip hybrid, as you mentioned! '

 

[madengr]

Re: CPW T-junction

I assume you wire bonded or bridged across the T junction like so:

 

[cuty81]

Re: CPW T-junction

yeap,
that the one I want to have! Anybody has designed this kind of CPW, please help!

 

[Azulykit]

Re: CPW T-junction

At K band I would be concerned about the bends and connector foot print as well. Look at all the elements in your RF path carefully. I imagine that you are considering SMA connectors and I would be a little concerned as they are at the upper limit of their performance (frequency) band. You may want to look at 2.4 mm or K connectors as an option. You may also want to consider grounded CPW and via's as an alternate to CPW with a couple of bridges. Your choice of substrate material is another big consideration.

 

[flyhigh]

Re: CPW T-junction

Hi,

The pictures shown above are really good and valuable! You must have air bridges around T junction or the current in direct path (left to right) will be broken for the bottom arm. Likewise, both of the bottom arm of T junction excite parasitic microstrip-like mode at discontuity unless this mode is blocked out by adding symmetrically the airbridges so that the potential is equalized around the discontinuity.

To conclude, if airbridges are not arranged as on the picture, you will have some undefined circuit that will look like CPW T junction but will not function as one.

My advice is as Sinatra's: Do T junction first and make sure that it works more-less as sugested from general transmission line theory (3x50 OHm lines connected in a common point), then add the switch circuitry. You will, of course, see the effect of discontinuities, but generally you should see something like:

S21, S31 ~-6dB
S11, S22, S33 ~-6dB.

flyhigh