Dear Munish Kumar

The attached formula is used to calculate the characteristic impedance of coaxial probe, where:

                 (d) is the inner diameter of coaxial probe 

                 (D) is the outer diameter of coaxial probe 

                  (er) is the dielectric constant 

                  (Zo) is the characteristic impedance 

The characteristic impedance (Zo) should be 50 ohm which is also available for fabrication of antennas. Then, u can find the diameters if u know the dielectric constant (er)

The link below makes sense if you want to calculate dimensions of coaxial probe:

http://www1.sphere.ne.jp/i-lab/ilab/tool/cx_line_e.htm

Regards 

Thanks Rammah for sharing this. It means whatever the impedance of the patch, the dimensions of the patch should be as per 50 ohms. Isn't it?

Most coaxial cables and connectors in common have a typical value for industrial standard lines of 50 ohm or 75 ohm characteristic impedance. Which is required to be properly matched with the patch impedance of the designed antenna. (Read more about Matching)

Regards  

Thanks Rammah.  I'll design according to 50 ohms. It means I have to change the feed location for matching with my patch? I am using FR-4, with dielectric constant=4.4 and height=1.6mm. Can you please calculate which dimensions of probe and coax should I take while simulating it? I'll forward my design to you.

Can you also give the accurate feed location formula for equilateral triangle?

Thanks again.

Hello Munish Kumar,

Yes, you need to change the feed location. The feeding point could be placed in any desired location inside the patch in order to be matched with (50 ohm) the desired input impedance. [except at the edges or center of the patch due to E-field =0 at the center which causes zero input impedance while at edge is maximum E-field which causes high input impedance] 

You need to be aware of the location of feed point which determines the patch's input impedance at the resonant frequency. If you use the feeding with a 50 ohm standard SMA connector, the input impedance of the patch should be adjusted to approximately 50 ohm for optimum matching. 

The link attached below used the same specifications as yours, you may refer to it. It also includes the formula that you needed in order to calculate the position of the coaxial cable. 

http://ijettjournal.org/volume-4/issue-4/IJETT-V4I4P340.pdf

Thank you thank you so much Rammah. Actually I have used the formula that you have shared for calculating the dimensions of the 50 ohm coaxial probe. But still I am not able to find the exact feed location for equilateral triangle. I have tried this for the rectangular patch, but this time I am using equilateral triangle for fractals and unfortunately, not able to find the exact feed location. But still I am trying, and if I find any solution for that, I'll definitely share this with you and others.

Thank you so much.

Thankssss Bahman... :) Ya its working slightly.

Bahman can you also confirm for FR-4 substrate having permittivity=4.4 and height=1.6mm?

Generally, easily available 50-ohm connectors are having inner conductor diameter of 1.5 mm and Teflon 2.1 epsilon diameter of 5 mm. For finding 50-ohm point of equilateral patch you can use parametric sweep or use its design equations available in Chapter 6, Microstrip Antenna Design Handbook By Ramesh Garg.

Thanks Dibinmary... That's what I actually asked. Thanks for understanding my problem correctly.

Regarding the coaxial probe dimensions, check it.

Thanks Ayan. Actually, my question is not about how to make coaxial probe in HFSS. It is about what are actual calculations behind it. Moreover, thanks for sharing this guide with us.

Hello friends,what if the antenna is PIFA. What are are the formula for the location of the coaxial feed

How to give coaxial feeding in cst with known feed diameter of 1mm for Epison-r of 2.5?

i am working on mmwave patch antenna design....what should be the outer and inner diameter of a coaxial cable feed??

I have found for my frequencies 1.85mm and 2.92mm k type connectors are suitable...Is that good one...