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.

coaxial cable length in antenna feeding

Status
Not open for further replies.

ghasem_008

Full Member level 4
Joined
Feb 9, 2012
Messages
221
Helped
11
Reputation
22
Reaction score
10
Trophy points
1,298
Activity points
2,990
Hi.
I want to feed my antenna by coaxial cable.But I don't know how set a proper length for coaxial cable.
in fact,I have a question.How coaxial cable length has effect on resonant frequency of monopole antenna?
I used from "Wave Port" in left hand side of coaxial cable (out of radiation boundary) and connect
inner probe of the cable to a ring (that sorround my antenna).
in fact my structure is as attachment.I have to apply two signal to my antenna by coaxial cable.
Untitled.png
how set a proper lenght for coaxial cable?
my frequency range is VHF band in simulation.
thanks
 

Hello Ghasem,

Can you provide a more detailled image/drawing so that I can have a better idea of the physical structure?

What is your goal with this structure/simulation? More info may give us the opportunity to help you better.

Generally spoken, when a coaxial cable sees a load that causes reflection (that means |S11| <> 0), the impedance at the input of the cable will mostly not be equal to the load impedance. You may know that adding a lossless cable section results in a clockwise rotation of the Ref. Coeff. when plotted on a Smith Chart (2*(electrical length of cable). When the cable section has electrical length of 180 degrees (that is 0.5 lambda), you make a full turn on the Smith Chart and the impedance doesn't change.

You may also know that when an increasing length cable section has loss, S11 makes a spiral. Due to the increasing loss with increasing length |S11| reduces.

Adding even the best cable to a structure may result in so-called common mode current. That means the net current through the cable isn't zero. If so, the cable will radiate and will be part of the radiating structure. Whether this occurs or not, depends on the structure
 

Hi WimRFP.how are you?thank you very much fro your reply.
yes,of course.my structure is a monopole antenna that I want to apply a signal to it.
you suppose "signal 1" is on and "signal 2" is off.and suppose I want to excite it,according with figure.
I don't want use from a simple lamped port at bottom of monopole or a coaxial cable that feed antenna from bottom.
my structure is like as figure.
I have a serious problem WimRFP.
I want to calculate "radiation efficieny" of my antenna by FDTD.I don't know where I have to place input port.do I have to place "input port" in left hand side of coaxial cable (in connector Location)?or do I have to place it in right hand side of coaxial cable (in connection location to my antenna)?
when I place input port in left hand side,my efficiency is low.do you think that it is total efficiency?
thanks...
 

Ghasem,

For me the picture isn't clear. For example I don't see a coaxial cable. I assume it are the horizontal lines, but where is the screen connected and where is the center conductor connected? A clear picture would help me to guide you better.

Do you want to include losses in coaxial sections into the radiation efficiency? If not, you better not simulate coaxial sections, but you should leave the screen in the physical structure. Are you sure you want radiation efficiency only? I would like to have reflection coefficient also.
 

Hi WimRFP.
yes,horizontal lines are pin of coaxial cables.and small rectangulars in left side of coaxial cables are N connectors.
I updated the figure:
Untitled.png
in fact,Pin of coaxial cables are connected to ring,that rings are hollow metal cylinders.
Also,my metal monopole antenna is sorrounded by a thin layer from a lossy dielectric ans this dielectric is sorrounded by the rings,too.
I don't want to include losses in coaxial sections into the radiation efficiency.But my coaxial sections are Perfect conductors and there is not any loss in pin sections.if I place input port in N connector Location,and calculate efficiency as:
efficiency = radiated power/ input power
this efficiency formula is radiation efficiency or total efficiency?I think that this is total efficiency.
How obtain radiation efficiency only (without considering of Losses)?
 

Ghasem,

I am very sorry, but I still don't understand your structure. Please make us a better drawing that clears any doubt so that we can get a good idea of what is in your mind. Make sure that it is clear where the center conductor and braid are connected.

Radiated power / input power is called radiation efficiency. Radiation efficiency does include loss. If the radiation efficiency is 60% and the net input power is 1 W (that is NOT the forward power), then 0.6W is radiated into space and 0.4W is dissipated into conductors and dielectric. If you don't want to include loss, radiation efficiency has no meaning.

Total efficiency = radiated power / forward power Forward power is also called maximum available generator power.

in dB's: dB(total efficiency) = dB(radiation efficiency) - dB(mismatch loss).

Input power is sum of all net powers flowing into your structure (that is how I would define it in case of a multiport network).
 

ok WimRFP.
I attached a new figure that include in "signal 1" only.
تحریک 3.png
explicity,I have a generator that is connected to BNC connector.therefore I can measure a voltage and current in connector location.it's true?
Now by inner probe ,I apply generator's signal to ring and therefore to my antenna.
Now I want to calculate radiation efficiency of the antenna.I can calculate radiated power,simply.
BUT,my problem is that how define input power port.
there is two case:
1- I meausre input power in connector Location.
2- I measure input power in location where inner probe and ring are connected.

Which of these two methods should be used for radiation efficiency calculation.
when I use from "case 1", my result is lower than reference (for example my result is 60% but reference has 75% efficiency).
I don't know how include "reflection loss".because when I include "reflection loss" by s11,namely:
input power = generator power (1-|S11|^2)
in this case,in "efficiency veruse frequency" diagram, there is two point in VHF band that magnitude(S11)-->1.So my input power become zeros and radiation efficiency is very big.while in other frequencies in VHF band mag(S11)<1 and my result is true.
because of this error in S11,I don't include S11 effect and I use from "input power = generator power".so my efficiency become lower than reference.
What should I do?
 

Ghasem, this makes more sense. Just one question: where is the coaxial screen (or is it not present and only connected to the ground box via the BNC connector)?
 

yes,coaxial cable is not presented here and only we need to consider it's inner probe.in fact outer shield connected to the ground box via the BNC connector and inner probe in connected to ring.
can you help me please?
 

Assuming that there is no braid present (braid removed) around the inner conductor, just make a localized feed (shown in red) as shown in the graph below:



There is no need to model a coaxial transmission line.

The input of your structure is the BNC connector, so in my opinion you need to simulate the radiation efficiency (Prad/Pinput) when feeding at the BNC (as shown in the figure).

In many cases |S11|>0.33 (that is VSWR>2) is not accaptable. If this is true for your situation, a matching network needs to be included, or you need to change the design to get a descent S11. The matching network will have loss and will therefore reduce the radiation efficiency, but will also reduce the mismatch loss. So in the end total efficiency (Prad/Pforward) will increase.
 

thanks.
I've simulated like as you.but I don't Know why my radiation efficiency is lower than reference.I will try and tell you the result.
ghasem
 

You talked about a lossy dielectric. You may set tan(delta) = 0 for the dielectric, so you can see the loss in the conductors only.

Is your reference from measurement, or simulation? Are you sure that your model matches the reference system, etc?
Measuring radiation efficiency is difficult, as you need to measure the complete radiation pattern
 

yes,I'm sure.
I have a question. inner probe length how affect on resonant frequency of my antenna?
when I calculate input impedance from N connector location and connection location of inner probe with ring,my resonant frequency are differ.I know that this shift is due to inner probe,but can you explain it perfectly?
 

depending on your frequency of operation the coaxial cables are different. A coaxial cable has a impedence of 50 to 75 ohms in the mhz to a few ghz range. and then the impendence will vary if u use it further up. Also the input port has to be matched with the port of the coax.
 

I know your explanation well.my purpose from this question is that how much coaxial length change resonant frequency?
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top