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.

Patches input impedance with HFSS

Status
Not open for further replies.

EnricF

Newbie level 6
Joined
May 21, 2009
Messages
14
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,283
Activity points
1,372
how to calculate imput impedance in hfss

Hi guys,

I want to build a 4 patches array antenna working at 2.4 GHz designed with HFSS v10. In the first place, I simulated a single patch with a 50 ohms line feed (lambda/2 lenght, because a lambda/2 line doesn't modify input impedance, I think). I've obtained a patch impedance of 235 ohms for a Return loss of -3.8 dB.

My first doubt it's if line length is important in order to obtain patch input impedance. Could someone answer it?


After that, a 4 patches array simulation has to be done. Initially, input impedance of any patch has to be found(because an array modify impedance of a single patch), but how I could do it with HFSS? I was thinking on connnecting four patches with four 50 ohms line with the length necessary to connect them (isn't a lambda multiple), but it's the length of this line important? This is so important for me, because in a future I will design a more complicated 10 GHz array antenna.

I attach the two simulations (4 patches simulation isn't correctly finished, but I attach only in order to show how, more or less, it will be connected).

Regards,
Enric.
 

impedance matching with hfss

Please, could someone help me?


Enric.
 

patch impedance

aaa
EnricF said:
Hi guys,

I want to build a 4 patches array antenna working at 2.4 GHz designed with HFSS v10. In the first place, I simulated a single patch with a 50 ohms line feed (lambda/2 lenght, because a lambda/2 line doesn't modify input impedance, I think). I've obtained a patch impedance of 235 ohms for a Return loss of -3.8 dB.

My first doubt it's if line length is important in order to obtain patch input impedance. Could someone answer it?


After that, a 4 patches array simulation has to be done. Initially, input impedance of any patch has to be found(because an array modify impedance of a single patch), but how I could do it with HFSS? I was thinking on connnecting four patches with four 50 ohms line with the length necessary to connect them (isn't a lambda multiple), but it's the length of this line important? This is so important for me, because in a future I will design a more complicated 10 GHz array antenna.

I attach the two simulations (4 patches simulation isn't correctly finished, but I attach only in order to show how, more or less, it will be connected).

Regards,
Enric.
 

lambda/4 impedance transformer

hi Enric,
the line length is very important to obtain the desired input impedance,
for designing the array case we have to divide the power using power splitters, normally corporate feed network is used,

for example if u r designing the 4 element array, each one individually having 200ohm, then combine those into ,,, and get 100 ohm.
the next 2 elements r also same like this ,,,from this two 100 ohm and combine those to get 50 ohm characteristic impedance in final,,,,thats d procedure to design the array,,,,

regards
abhi
 
  • Like
Reactions: EnricF

    EnricF

    Points: 2
    Helpful Answer Positive Rating
50 ohm characteristic impedance hfss

Hi Enrinc,

You did some mistakes in the HFSS files.

1) The length of the transmission lines that feed the antennas is relatively important (only in terms of phase/delay). The most important is to feed the antenna and match the input impedance. Patch antenna usually have input impedances (at the edges) of 100 or more ohm; then, if you use 50 ohm transmission lines, you may need a lambda/4 transformer (the so called inverter) to transform from 50 ohm to the required input impedance of the patches.

2) You are not feeding the patches at the middle of the edge. If there's no reason for that, you have to center the feed line with the patch.

What you can do is:

1) Start with the simulation of the single patch antenna, and plot the |S_11| parameter. And then try to renormalize the port impedance to a higher value than 50 Ohm. You will see that when renormalizing to 100 or 150 ohm, the matching is better.

2) Calculate the required width of the lambda/4 transformer with the best value of impedance renormalization. So, you have to set the port impedance to 50 ohm, since the begining of the feed line is with 50 ohm, and since you use the lambda/4 transformer, you should obtain a pretty good antenna matching (|S_11| < -10 dB).

For the case of the patch array, you have to take into accound the dividers. When you add in parallel two feed lines, ath the intersection you have half the impedance. So, you may need additional lambda/4 transformers in order to work with 50 ohm feed lines, and at the same time, match the patch antennas with the last lambda/4 tranformer, or just play with 100 ohm feed lines and the dividers.

Bye,

Pere
 
renormalized impedance

Hi there EnricF, I see you have taken on an ambitious project, especially if this is for school. I am currently designing a planar antenna array myself. There are a lot of things to consider when doing your design.

First off the length of the microstrip line to the patch itself won't have much effect on the impedance matching however it's width will. The quickest way to get a decent match with the feeding method you choose will be to cut off a square/rectangular portion of the patch so that you can have your feed line insert itself ­"into" the patch. Your gonna have to look at some books or papers to get a good understanding of what I mean by that. You can use HFSS to play around with the dimensions of the part you cut off and find how deep you microstrip line should go into the patch to get a good match.

Next thing to consider is the feed network. What governs this is the number of element you plan to have in your array and the amount of space you have available on your board. If your gonna have a 4 elment array then you can keep your current design, however if you have plans for a bigger array say a 32 element array then you might want to consider a combination of a corporate and series feed. This part will require some planning. One thing to keep in mind is that you'll want all your line lengths, from the point where you connect your coax to the patch, to be equal.

Good luck with your design.
 
  • Like
Reactions: EnricF

    EnricF

    Points: 2
    Helpful Answer Positive Rating
hello

i am new to antenna array design...i want to design a 1 by 2 array of patch antenna..i am facing problems in designing the power splitter..please can anyone suggest me in detail the way in which i can design the power splitter..

apurv
 

hello

please can anyone help me out on this...??

apurv
 

im getting correct rectangular and polar plots but simething is wrong with the radiation pattern when i compare it with my matlab routine results..
can someone plz post a screen shot of radiation pattern of 4 element array..plzz guys i really need tht...thnx

Added after 2 minutes:

i have made some adjustments mentioned above, is it possible if someone can post simulation file 2 so i can check where im making the mistake..plllzzz
 

Thanks all, you were helpful. I 've finished the design following your ideas (2 years ago, but never is to late to reply). I've the simulation in HFSS if any require it (not in this computer, but next week I will post it, this is an open source project).
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top