Antenna Gain of Patch Antennas in HFSS

[Paulpen]

Dear Fellows,

Please, I need some help with my design of an equilateral patch antenna in HFSS . I have designed a probe-fed triangular patch antenna for 960Mhz operation using a waveport for excitation. The problem is that after simulation my total gain is always negative. However, in most literature they have positive gains.

After simulation I got 945.9Mhz as the resonant freqency and return loss was -20.1242db. The value of the Z parameter was 49. 04ohms, which I feel is close to the "perfect' 50hms and not bad. WIth this value of Z I am assuming that the energy reflection is very minimal. However, when I comptuted the antenna parameters at this frequency I got the following values:

a. Incident Power=1W
b. Accepted Power=990.33mw
c. Radiated Power=165.67mw.
d. VSWR=1.2194
Please, I believe something is not right, especiallly given that so much power is being lost. What could I be doing wrong and what can I do to obtain postive gain values? From the look of things, it appears that some setting might be wrong. Could my far field radiation sphere setup be responsible? Peradventure anyone is interested in taking a look at my design, I could forward the project file.

Thanks in anticipation of your feedback.

 

[volker@muehlhaus]

Run a test where you set substrate losses to zero.

 

[Paulpen]

Run a test where you set substrate losses to zero.

Thank you for the suggestion. Please, could just give me an idea on how to go about this? I am currently researching how to go about this. Any guidance provided will be highly appreciated.

 

[volker@muehlhaus]

You have created a substrate material, set loss tangent (tand) to zero.

Patch antennas with low gain and low radiation efficiency are often discussed here, and the typical reason is that a lossy substrate like FR4 was used. You can test that by enforcing lossless substrate material in simulation. If this has a big effect, you know that your substrate is not approriate, and you need a better substrate material with lower loss.

 

[Paulpen]

Run a test where you set substrate losses to zero.

Thanks for your feedback. Honestly, I have tried to find out what the problem was but didn't succeed. Going by your advice, I changed the substrate to Rogers (RO4350B) but still got simillar results. I also set the loss tangent of FR4 substrate to zero and nothing changed really.

I am so sorry for bothering you. Please, I have attached the project for you to help me by taking a look at it. Perhaps, there is mistake that I made but didn't notice. I perceive it could have something to do with the way I designed my coxial probe, bug in my software, or settings. I have watched countless videos and seen some variations in how the probes were designed.

In this specific project, I have an Equilateral Triangular Patch Antenna designed for 0.96GHz operation. But it resonnated at 0.9536GHz with a return loss of about -18db. The Z impedance is almost perfect (49.61ohms), which suggests good impedance matching. Unfortunately, the gain is -2.15db. I feel this is abnormal since almost all the power transmitted gets received.

To find out if I did my coaxial right, I took an exisiting design of a probe-fed microstrip antenna from HFSS (ACT) and discovered that the ground plane had no metalic coax material connected to it. Instead the material was teflon. Could I be missing something? I thought that the ground plane must always be connected to external coax copper while the probe goes to the patch itself. Or is there a new convention in HFSS for exciting patch antennas without connection to their ground planes?

Please, help. I have been stuck at this stage and will appreciate any assistance to help me resolve this. Thanks.

 

[volker@muehlhaus]

I have nothing to add to post #4.

I am using Empire XPU solver, not HFSS, so I cannot help with HFSS specific details.