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.

A basic doubt on Antenna resonance

Status
Not open for further replies.

svarun

Full Member level 5
Joined
Mar 31, 2002
Messages
271
Helped
22
Reputation
44
Reaction score
15
Trophy points
1,298
Location
In the warm company of God's most exquisite creati
Activity points
2,746
Hi all,
Sorry for this trivial question but I thiink asking trivial questions is better than knowing the things wrong.In terms of the magnitude and phase of Z11 with respect to the frequency ,how would you define the resonant frequency of an antenna. Phase must be zero but what about its derivative ? Similarly what about the second derivative of the magnitude ?

-Arun
 

svarun said:
Hi all,
Sorry for this trivial question but I thiink asking trivial questions is better than knowing the things wrong.In terms of the magnitude and phase of Z11 with respect to the frequency ,how would you define the resonant frequency of an antenna. Phase must be zero but what about its derivative ? Similarly what about the second derivative of the magnitude ?

-Arun

it depends on the feed too. in definition, when image(Zin)=0, then the antenna is in resonance; but some times it is not! becouse of the feed. the best way is to consider Im(Zin) and Re(Zin). the antenna is in resonance when the Im(Zin)=0, and d(Re(Zin))/df=0, in other words, when the Re is in a peak, and the Im is zero. if the Re is in a peak, but the Im is not zero, then the resonance is accured, but the feed has an imaginary impedance, affecting the Im of Zin.

marti
 

the antenna is in resonance when the Im(Zin)=0, and d(Re(Zin))/df=0, in other words, when the Re is in a peak, and the Im is zero.

Hi, Marti and Arun.

I think the condition given above (i mean word and) is valid only for microstrip antennas because it's alternate scheme (parallel RLC).
For example if you consider dipole's first resonance, there's no peak in Re(Zin).

My opinion is that the condition is Im(Zin)=0 but with reference point at the antenna, not including the feeder (parasitic inductance).

Rgz

eirp
 

svarun said:
So if I need to find the true resonance , I need to compute compute Zfeed , subtract it from the overall Z11 and then look at the resonance condition/conditions.Is this correct ?

-Arun

yes, and the peak condition is for RLC like resonators, such as p@tch antennas, as eirp sed.
bytheway, sometimes other things should also considered, for example, if two resonance are accured nearly, in these cases, we have, say, two serial RLCs connected in parrarel, or two parallel RLCs connected in series. well, another easy way to define the resonance frequency in these structures is to define them as the average of the frequencies of the start point and end point on the impedance loop on smith chart. this method is used in two-segment dielectric resonator antennas, and other structures too. i used it in my fractal p@tches succesfully, and the results are really near the res, freq. defined by the other methods.

marti
 

Hi friends,

In the case of antenna matching, we are looking for a maximum of power transfer between the feeding port and the radiation resistance of the antenna.
At the frequency at which Zin is purely real (i.e. Im(Zin)=0) you can get exact matching at that port (eventually with an impedance transformer). If there are no losses, all the power is radiated.
Reactive components (feeders, stubs, etc) shift the resonant frequencies or can produce multiple resonances.

For d(Re(Zin))/df =0, I don’t agree that this is a necessary nor sufficient condition of resonance, and let me give an example.

One could think at an analog condition for the admittance Yin, say that it should be d(Re(Yin))/df=0 at resonance. (After all, the two dual models are equally valid). But the two conditions are very different: If one is satisfied at a given frequency, the other is not necessarily satisfied. What is guaranteed is that at a frequency at which Zin or Yin is purely real, the other is real as well.

Imagine a series RLC circuit. At any frequency Re(Zin)=R. At the resonant frequency Im(Zin)=0 and Re(Yin) has a maximum.

[Now the dual:
Imagine a parallel RLC circuit. At any frequency Re(Yin)=G (=1/R). At the resonant frequency Im(Yin)=0 and Re(Zin) has a maximum.]

Now consider the series circuit at the end of a lambda/8 line with Zo=R: there is a maximum of Re(Zin) at a frequency above resonance [and the maximum of Re(Yin) is at a frequency below resonance]. Nevertheless, at the resonance frequency the value of Rin is unchanged [the same happens for Gin], and the reactive part of Zin [and of Yin] is zero. There is mismatch at the frequency at which the maximum of Rin is located [the same applies to Gin].

The same happens with other line lengths, but the fact that at the resonance frequency Zin and Yin are purely real doesn’t change for any line length.

This example shows that d(Re(Zin))/df [and d(Re(Yin))/df] change although the resonant frequency (at which the circuit is matched) does not. The derivative can be different from zero at resonance, and it can be null far from resonance.
I hope this is clear. :?

Regards

Z
 

Great posts on the topic everyone...
Like most people suggested in the discussion, the Im(Zin) should be equal to zero for resonance. The condition on Re(Zin) is not necessary.
I am currently simulating an antenna in HFSS and have gotten my antenna to resonante but matching is proving to be a real problem.

For matching purposes, one could use the feed line directly to feed the antenna and find Zin and hence resonant frequency(at Im(Zin)=0) and then impedance transform the real part of Zin to the desired value...
If anyone has any other suggestions on matching, they are most welcome to enlighten us..
Thanks again for the great posts guys,
Ananth.
 

I agreed with eirp,

the resonance condition is Im(Zin)=0
with reference point at the antenna.

Thanks all for your great posts,
 

hi, see great discussions here,

how about Re(Z)? Most textbooks talk about it is really a resistive loss + radiation resistance, so when designing an antenna, should we design an antenna with as large Re(Z) as possible? assuming patterns are ignored.

Thanks!
 

Some naive questions about Im(Z) =0. Assume we can always add a passive element to tune off the reactive impedance of the antenna. Does it mean the antenna is in its resonant frequency? If this works, why do we want to put the antenna in resonant frequency? Just tune the Re(Z) to what you want and add a passive element to tune off the Im(Z). I know a lot of times it is hard to do it, but at least theoritically.
 

kyjackchan said:
hi, see great discussions here,

how about Re(Z)? Most textbooks talk about it is really a resistive loss + radiation resistance, so when designing an antenna, should we design an antenna with as large Re(Z) as possible? assuming patterns are ignored.

Thanks!

I think, matching would be a problem if the antenna impedance is too big.
 

I would like to add that, the gain of the antenna at this "fabricated" resonance might be small to be usable.

-svarun

Added after 6 minutes:

kyjackchan said:
hi, see great discussions here,

how about Re(Z)? Most textbooks talk about it is really a resistive loss + radiation resistance, so when designing an antenna, should we design an antenna with as large Re(Z) as possible? assuming patterns are ignored.

Thanks!

Re(Z) and Rr (Radiation resistance) are two very different concepts. Your Re(Z) is the actual input impedance seen by the line feeding the antenna. However Rr is a purely fictitious quantity which just relates the input current into the antenna feed lines and the power radiated by the antenna. One should have Rr as large a possible while Re(Z) plays an important role in matching the antenna to the feed and thereby reducing mismatch losses.

-svarun
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top