Rectangular Patch - RF Cafe Forums

 Post subject: Rectangular Patch

Posted: Tue May 29, 2007 4:12 pm 

hi everybody,

I would like to decide about the length, width, substrate and spacing of the rectangular patches of a 4-element phased array at ISM. could u help me with it?

thanks, Panagiotis

 Post subject:

Posted: Wed May 30, 2007 3:55 am 

Hi Panagiotis

Which of the ISM bands? 2.4GHz?

I have some design data. Let me know which band, and I will either send formulas or do the design 4U.

BTW: one factoid/rule of thumb I seem to have gleaned is that the lower the k of the substrate, the greater the bandwidth for a patch antenna.

 Post subject:

Posted: Wed May 30, 2007 10:55 am 

The array is at 2.4GHz. I design a coupled oscillators array for my diploma thesis.I have the outputs of the 4 coupled oscillators and i want to take the radiation patterns of the phased array. So, I should know the dimensions of the rectangular pathes,spacing and etc...

Thank you in advance,

Panagiotis

 Post subject:

Posted: Thu May 31, 2007 4:23 am 

Ahh, when I found the design equations they were for a coaxial feed circular patch at 2.8GHz, with a complete analysis given for similar frequencies 1 to I guess 8 GHz. (Liang C. Shen Proc IEE Vol 126 No 12)

I will continue the search of my design database for a recatangular patch. Most of what I have found is about arrays of rectangular patches with a combining corporate feed network, which is not exactly what you want.

If you are expected to build one as part of your project, you will need to have a knowledge of likely substrate permittivity and thickness.

Perhaps while I am searching the database you could research what substrates are available for you to use for the antenna model.

 Post subject:

Posted: Thu May 31, 2007 10:00 am 

Hi panagiotis

I am now ready to "rock n roll" with either a rear-fed (from coaxial) or an edge fed (from microstrip) rectangular patch.

We also have it in round circular patch if you prefer.

All I need now is the material thickness and epsilon value.

For all this, do I get to use half of your diploma??

 Post subject:

Posted: Thu May 31, 2007 11:00 am 

First of all, thank you nubbage.

Second: My work is to design and materialize the 4 coupled oscillators(4 VCOs coupled by a coupling network) which will give 4 outputs for each element(patch) of the array and not to materialize the array.

So, I do not know the exact material thickness and epsilon value of the substrate for the patch.

Can u decide about these values or what should u know from me to decide? Also, I do not know either a rear-fed (from coaxial) or an edge fed (from microstrip) rectangular patch will be appropriate..

It would be very good for my work to have the design of the patch

Of course u can use half of my diploma

 Post subject:

Posted: Fri Jun 01, 2007 3:42 am 

kali maira Panagiotis

Great, thanks for the use of 50% of your first class diploma.

Your application is now much clearer. But to decide how the patches are fed with RF we need to determine if the antenna array is immediately integrated with the oscillator sources or is placed some distance away (more than a few cm). If it is integrated, then a microstrip line and an edge connection through a quarter wave transformer is called for. If it is at a distance, then a semi-rigid coaxial line and a rear feed through the ground-plane is better.

I will assume an integrated structure, and I will look into what specifications of strip-line board would be best. The choice is determined mostly by the band-width you require for the signal. This will be the fine-tuning range or modulation bandwidth of the oscillators. Bandwidth and substrate height from groundplane to trace are directly and almost linearly proportional in patch antennas.

So, what is the full tuning range of the oscillators? If fixed in frequency but they will be frequency modulated, what is the likely frequency deviation in % of center frequency?

Do you need to produce a predicted radiation pattern for the array for the thesis report?

 Post subject:

Posted: Thu Jun 07, 2007 4:16 am 

Hi panagiotis

The design is ready. Four rectangular patches on Rogers Duroid 5870 board, each 50 ohm imedance, centred on 2.45GHz. The bandwidth will be about 50MHz, so I hope that is enough, as you did not yet respond to my query about this.

Are you still interested in this antenna?

Do you need a polar radiation pattern computed?

It has been a week since we heard from you.

 Post subject:

Posted: Thu Jun 07, 2007 6:01 am 

Good morning Nubbage,

Excuse me for this week, but i had exams, so, I had to study.

Of course I am still interested in this array.

I was ready to send to you that the desired bandwidth is 50MHz but you send to me first .I would be grateful If I had a polar radiation pattern computed. So, can you send the design to my e-mail?

My mail is pmourtopallas@gmail.com. Thank you very much Nubbage for all your interesting.

 Post subject:

Posted: Thu Jun 07, 2007 6:14 am 

and something else,

the purpose of my application is to have a beam scanning of some degrees. By detuning the end elements (end VCOs) the 4 VCOs are locked to a new frequency and they create a new phase difference between elements. With this technique you can succeed a beam scanning of some degrees.

 Post subject:

Posted: Thu Jun 07, 2007 7:37 am 

Hi panagiotis.

I hope your exams were not too painful

I will sketch the layout in a Word document and email it to you as an attachment.

BTW: not sure if that scanning method will work. It might if the outer left VCO were "up-tuned" say 1kHz and the outer right were "down-tuned" 1 kHz.

Even so, the scanning rate would be 360000 degrees per second. There are always problems if the elements of a phased array are out of phase by more than a limited amount. The pattern resembles something like chewed dates.

From your comment, it is now clear you wish the elements to be arranged in a line and not in a "box" format.

 Post subject:

Posted: Thu Jun 07, 2007 10:05 am 

Hi Nubbage,

Ok, I wait for your e-mail.

I have one more question..

With what program have you designed the array?

It would be very good for my diploma thesis to conclude the design!

Also, I have found many papers, which have applications like this. The maximum beam scanning with this method is about +-30 degrees, because the maximum inter-element phase shift is 90 degrees.

thanks

 Post subject:

Posted: Mon Jun 18, 2007 9:11 am 

Hi panagiotis

email was sent off to you last week.

Please confirm if it was received OK.

It had a sketch drawing in a Word file attachment.

I am currently looking at the radiation pattern envelope.