www.freepatentsonline.com/6836247.html

Hi All,

Now I got stuck in my project, so I really hope someone could give me a suggestion.
The problem is, I want to join/merge two patch antennas which operate in different frequency using one single feed to yield a dual frequancy antenna.
At first the two antenna is work in acceptable parameter independently, but when the two antennas are mergerd, the characteristics is changed.

So what I sould do, to make it a successful dual frequency antenna.
I know the problem is mutual coupling between the two patch, so what should I do to make the antenna work fine in both frequencies.

I really appreciate your help.

Thanks.

Can u give a little details abt ur project ?

ACtually merging 2 antennas side bt side is generally not acceptable. So if u can give the resultant requirement thn i can hlp u .

I am total begineer to RF but IIRC, you can design a single antenna to operate at more than one frequency i.e. with some trade-offs. Check the following link for an idea.
http://www.cst.com/Content/Applications/Article/130

Thanx mainak001,

OK, the two patch is printed in the same layer.
And is not side by side, but more than inside, because the first antenna for the higher frequency is square antenna, and for the lower frequency is a printed ring antenna. At first I'm gonna feed it using couple feed by microstrip line below the two patchs. The requirement is good matching in the both frequencie.

So how, I can make somekind of "isolation/barier" between the two patches, so one patch will not (or only made small) interfere to the other patch.



Added after 1 minutes:

Hi kishore2k4,

I know one of method to yield dual frequency is by adding a slot, but I want to use this merging technique.

I think we should use some printed diplexer circuit between these two antennas.

Should be at least 20dB isolation between antennas (rule of thumb).
In this way they behave independently and can be tuned separately for each frequency.

you can use two cirular slot antenna and feed them together. i hope this would work.

Hi, khan:

I think there are two possible reasons affecting your designs:

1. There are some strong coupling between the 2 patches and the 2 patches are affecting each other. If the distance between the 2 patches is not very small, the coupling may not be very strong to change the performances the two antennas substantially.

2. The impedance behaviour of one antenna is affecting the matching to the other antenna. I think this is quite likely to be reason. The reason is the following: You have antenna A working perfectly at frequency F_a with Za = 50 ohms and antenna B working perfectly at frequency F_b with Zb = 50 ohms. If you shunt them together, you will see the impedance of the connected antenna to be:

Zab = 50 * Zb_at_F_a / ( 50 + Zb_at_F_a ) at F_a

The |Zb_at_F_a| may not be much bigger than 50-ohms. In that case, the Zab may be much different from 50-ohms and you do not have good match at F_a. For example, if Zb_at_F_a = 0, the Zab = 0 and it is a bad matching at F_a.

Regards,

I think if u place the patch inside e a rectangular ring and locate a coupling strip in between the rectangular ring and the patch . The coupling strip can be used to excite them through a 50- coaxial cable. This proposed antenna can be built on a two-layer (FR4 and air) substrate .Thesea antennas can be used in 2.4- and 5.2-GHz bands .

ucan also see the document . It can be of hlp

www.freepatentsonline.com/6836247.html

Hi jian,

Thanks a lot for your detailed explanation.

For the 1st reason, I have thinked about it and tried to make the distance far apart as possible, but it seems not give better results. By the way, do you know how to calculate the minimum distance so the strong coupling won't occurs?

For the 2nd reason, you did give a better explanation what is cross in my mind.
Yeah, I do think when the two patches is combinen together, then there will be a change in the input impedance, said in parallel.

As you said in the equation, so I should take account the impedance of one frequency, from the other frequency point of view. How could I calculate it?
The problem is I use a proximity coupling feed, form which point the input impedance should calculate.

Do you have sugesstion how I could get good matching for the both patches?
If we take the 2nd possible cause as the problem, then an impedance transformer could it be the answer?



Added after 3 minutes:

Hi mainak001,

Thanks for the suggestion, I've also think about it either, the coupling strip between the two patches, but how I could calculate/figure out the right coupling strip (I'm sorry the link you give not give a good answer). Do you know any good reference about it, thanks.


Added after 2 minutes:

Thanks vfone,

but how I could make this "barrier" of 20 dB isolation? I've tried distance, but seem not works as expected. Any suggestion?

Intersting discussion,

anyway somthing is missed:

what are the frequencies ? too mach close frequencies sre difficult to be integrated in single structure still preserving good isolation.

Have you check the radiator proerties when only one elment is excited ? I mean when you merge patch and ring are you sure that the ring and patch are still working correctly at least if separately excited ?

could you provide a sketch of you structure ?

bye, bye

Hi, Shadow Khan:

If you combine 2 antennas together and you want them not to be intefering with each other, you should try to do the following no matter whether you are shunting them together or using coupled feed:

1. You should avoid any strong coupiling between the 2 antennas.
2. Assume you have antenna A at frequency Fa and antenna B at frequency Fb. You should make the A has very large input impedance at Fb and B has very large input impedance at Fa. Then, when you shunt them together, you will have almost the same good matching of A at Fa and B at Fb.

If you can't meet both of the above conditions, you should try to optimize the final feed network and both antennas for good match at Fa and Fb simultaneously.

Regards,

Hi jian,

Thanks again for the suggestion.
But let me ask some question regarding it,
- to avoid strong coupling between the two patches, is distance the only way?
- Is there a method to make impedance small (e.g. 50 ohm) in one frequency and higher impedance in another frequency. or simply by using feeding network?

Last thing, quit differ from the topic, I'm currious is there possible if we merge patch antenna and slot antenna also using the same method (feed by proximity couple microstrip feed line). So the strip line gonna feed the patch and also the slot, do the same analysis could applied?
Sorry if its out of the topics, but I thing its till related.

Thanks and regards,

Higher the frequency difference between antennas, higher is the isolation between them.
Greater the distance between antennas, higher the isolation between them.
Using different polarization for antennas increase the isolation between them.

There is no other ways to increase the radiated isolation between antennas.