uhf rfid tag antenna simulation for help

[hylinuxer]

hi guys,
i have used HFSS for a long time,and i also designed and fabricated two or three RFID tag antenna for various applications.but now,i confuse about some setup about simulation,beacause i had seen many different viewpoint to this,but i believe some of them must be wrong or fallacious,i hope u can help me and make clear why. my questons are as follows:
1,we all know the impedance of rfid tag antenna need be designed to match the microchip impedance for max power transfer,but how to setup in HFSS?i saw four different ways to "full port impedance " and "port renormalization",pls look over attachment named "port setup"(chip impedance is 20-j160).i need know which is right or more accurate? i use the second way,but other designer use other ways also have good design and fabricate tag which also have good performance.
2,many thesis and paper refer to S11 parameter in HFSS for tag antenna,expatiated the reflection coefficient matching complex chip impedance is T=Zc-conj(Za)/Zc+Za(is this formula right?),the power reflection coefficient is then abs(T)²,so is it reasonable directly adopt default S11，VSWR of HFSS in results?i saw someone leave 50ohm in full port impedance of excitations,and directly use S parameter of HFSS in result,but it has nothing to do with the chip's impedance,so i don't think it's reasonable,but how to setup output variables for this?i never saw others sample or model,so if possible,a simple or model including all parameter(VSWR,GAIN.S11,return loss and so on) is very useful for me. best wishes!

 

[flanello]

Irrespective of HFSS, I leave the port impedance on 50 Ohm and use post processing to calculate the reflection coefficient between the chip and the antenna according the formula you have mentioned.
When using a renormalization to a complex impedance, some programs have a different behaviour thus I avoid these functions.

 

[hylinuxer]

thank you flanello, if you use the formula T=Zc-conj(Za)/Zc+Za to calculate the reflection coefficient ,then how to define VSWR,according to the theory,VSWR=(1+|T|)/(1-|T|),but the result of VSWR curve is very bad,on the side,the result of reflection coefficient curve seems also not good, although the antenna design is very good,the S11 just above -15dB to the resonance frequency,but many paper's S11 result is under -25dB to the antenna simulation,and the vswr is also good,so i doubt my review ways is not corrcet,someone set full port impedance is conjugate impedance of chip,and not renormalization,this setting can give very good result for the same antenna and accord with the papers,so i think the conjugate matching S11 is not suit for display antenna result in HFSS,because it is not "no reflection matching",so S11 and VSWR all are not good, just set full port impedance is conjugate of chip's impedance ,and then this impedance and the impedance of antenna form a no reflcetion matching circuit,the closer impedance of antenna is to the conjugate of chip's impedance,the better the VSWR and S11,so this mode and circuit is accord with HFSS's defination.so conjugate matching is a goal for antenna for max power transfer,but can't use it to review the S11 and CSWR.if you have better point,pls discuss with me,thank you

 

[flanello]

If you are matching to a complex impedance, then S11 and VSWR to 50 Ohm cannot be good.
You need to look at the S11 and VSWR between the chip and the antenna.
The S11 (=reflection coefficient) and the VSWR have a direct connection, thus if one is good the other is also good.