Coax Balun Hfss and Genesys

I′m trying to simulate in HFSS a Balun cable , and try to get the same results as i have in Genesys , i alredy test it in real world and Genesys result are very consistent , so i′d like to undestand what im doing wrong in HFSS.


Below is the circuit in Genesys ( it is a line 20mm 75 ohms )

The S paramitters of this circuit ,

Now i prepare a simulation in HFSS using a cable ( calculate the inner and outer to get the same 75 ohms ) and i use a board to do the unbalaced measurment.
This is the ciruit

I already try the terminal and Modal solution and i got the same result, and its very different from Genesys linear simulatio.

could someone give me some idea about it ?

The Genesys model is incomplete, it omits the common mode line impedance, assuming it to be infinite. In fact it is rather low, effectively shorting port 3, as HFSS reveals.

Thanks for your reply .

In fact im trying to simulate a compleat wilkinson combiner that use the regular 2 lines 70 ohms from port 2 and 3 and use 2 cables whit the shield and inner crossed and a 50 ohms load to ground , it workd fine in real world and i can get over the UHF band a veru good return loss and isolation between ports , but when i try to simulate is in HFSS does not work , so i decide to split the problem and check where was the difference between genesys and HFSS .
All Other parts works fine but in the part i have to ground the inner from one cable and conect the shield in the inner of the other cable and in to load , it does not work in HFSS.

i cant see the reason , and even if genesys is not complete model , the result match very good whit reality .

But of course if i can simulate it in to HFSS propertly will be bether.

Thanks

Lucas

What you describe sounds completely different from what I now as Wilkinson divider. Can you show a diagram?

B.t.w., I would expect TL4 to have λ/4 electrical length, but it's much shorter.

Hi FvM , this is a veriety of regular wilkinson, and about he length in my model in HFSS is a variable, so the 20mm is just for reduce the simulation time .

The wilkinson diagram is the one below , ( sorry is just a pain diagram , because im out of office now).

As i told you in real tests it work very well and in linear genesys simulation as well .

Lucas

I see that the modified Wilkinson behaves equivalent to regular type in a restricted frequency range (< 2 Fc), but only if the common mode impedance of the flipped transmission line doesn't disturb too much.

As previously explained, the basic SPICE (or Genesys) transmission line doesn't model common mode impedance, it always behaves ideal. You can supplement the simulation by adding a second TL component for the common mode signal (coaxial sheet to vicinity). The basic problem is that the electrical length of the common mode line will be usually shorter than the coax line with dielectric, thus it's not λ/4 in length and the shunt impedance won't be eliminated.

HFSS can be expected to model the transmission lines completely, if the result should comply with the real circuit if all parameters are modelled realistically.

Could you help me to setupt propertly the hfss to simulate this flipped transmission line ?

I could't find any example of this situation, as you can see in my 1 message i try to setup a cable conect to a a pcb board in oder to measure the flipped cable, one side conect direct to board GND and the other side i conect in 2 lumped ports .

The cable impedance i had calculate and using and check the dimentions using a 75 ohms Huber suhner to see if the dimentions was ok .
The trace 50 Ohms of the board i had calculate and simulate in Hfss before, jus to be sure.

As i undestood it is ecencialy a balun using cable and i should have the power splited in the 2 output ports , but seams that the shield of the cable are compleat grounded . I set all conductor as cooper , not as perfet E just to try.

all the lumped ports i had set as 50 ohms impedance .

Thanks

Lucas

What's the actual technology used for the flipped transmission line?

Why are you using this topology instead of standard Wilkinson?

Today o use cable 75 ohms TO flip it and reason is insted TO have a 100 Ohms resistor between the 2 inputs o Have a load 50 ohms TO ground

If you say, the real design has acceptable behavior, then a HFSS simulation of the exact cable geometry should give similar results. A sufficient large air box is needed to model the common mode impedance.

Dear Fvm , i have the simulation on the file, could you take a look and let me know what is the problem ?

Thanks.

Lucas

No access to HFSS, unfortunately.