Antenna Simulation tool which includes matching with L R and C components

[hedie]

Hello @ all

Im quite new here to this board.
I have heard a lot about it. Only good things
So i hope that anyone from here can answer my question.

Im an electronic technician who is really interested in the antenna technology.
escpeacially printed antennas on PCB's.

Therefore, i buied some book. I bought "Antenna Design for Mobile devices" and a ook about smith-diagrams.

The smith-diagram book was really helpfull but the antenna design book not.
So i have ordered 3 more books about the antenna design but thei didnt arrived until know.

Therefore my first question to you... Can you recommend me some books where i can learn the tehnique how to design
such pcb antennas? I mean the real design. that means: why does the antenna looks how he looks.
why are there corners and how can i estimate which change of the geometry will take which effect on the resonance.

This was question one...

But the much more important question comes now:

In the meantime, i played around with such simulation tools like Sonnet or HFSS from Ansys.
They are quite well. Especially Sonnet is really easy to handle.

But after the reading of my smith-book, i know what it means when the antenna is not matched tho 50 Ohms
and what the reflection or the vswr factor means to it.

And i have learned a lot about matching circuits.

And here comes the point... I can simulate many geometries with these tools.
But all of my designs are not matched to 50 ohms. So i thought to use such matching circuits to change these.

But i want to directly implement these circuits into the simulation.

So my idea is the following:

[ANT-Geometrie] <--- [matching-circuit] <---[RF-Port] <---- Simulation

So is there a Tool out there, where i can simulate my own antenna geometrie (design) with an specific matching circuit?

What simulation tools can you recoomend and are there books out, where i can learn how to use these tools?


Thank you!
Claudio

 

[RiffMaster]

For name of good microstrip antenna books check the references books at the bottom of the link:
http://www.qsl.net/va3iul/Antenna/P...Ideas_for_Printed_and_Microstrip_Antennas.htm
For printed antenna simulation you can try these freeware programs:
http://www.sonnetsoftware.com/products/lite/
http://www.feko.info/product-detail/feko-lite

 

[hedie]

Thank you for your answer.

when you have read my post, then you know, that I already use sonnet.
Feko is new to me.

My question was:

Can I add some matching circuits into the feeding line from my antenna and then simulate the whole antenna - solution including the matching circuit?

 

[BigBoss]

Thank you for your answer.

when you have read my post, then you know, that I already use sonnet.
Feko is new to me.

My question was:

Can I add some matching circuits into the feeding line from my antenna and then simulate the whole antenna - solution including the matching circuit?

Simulate your antenna standalone then obtain s-parameters.
After that put these s-parameter into schematic of any RF simulator ( ADS, Microwave Office, Ansys Designer etc. ) as a black box that will represent your antenna then apply your matching circuit with lumped or distributed elements then simulate whole system with matching circuit+antenna to get the performance.
If you use ADS or Microwave Office, you can do all those in a single environment because both have EM simulators ( like Sonnet) too.

 

[hedie]

Simulate your antenna standalone then obtain s-parameters.
After that put these s-parameter into schematic of any RF simulator ( ADS, Microwave Office, Ansys Designer etc. ) as a black box that will represent your antenna then apply your matching circuit with lumped or distributed elements then simulate whole system with matching circuit+antenna to get the performance.
If you use ADS or Microwave Office, you can do all those in a single environment because both have EM simulators ( like Sonnet) too.

Thank you for your Answer...

How can i put in a blackbox with S-Parameters in it into Ansys HFSS?

Thanks..

 

[BigBoss]

Thank you for your Answer...

How can i put in a blackbox with S-Parameters in it into Ansys HFSS?

Thanks..

You cannot do it in HFSS
Use the simulators mentioned above..

 

[E Kafeman]

I design cellphone antennas for living. It becomes more demanding for each year. It was years ago that it was possible to first design an antenna and then implement a matching circuit as smaller final adjustment for a tri-band GSM antenna.
Antenna nearfield environment have now become very complex and available antenna volume becomes smaller for each year. Worst are the smartphones. They allows very low height above ground and big part of existing height is covered by a LCD and remaining available space is crowded with, besides antenna, a lot of other stuff.
As phone is very thin will also hand-effect be much more cumbersome, it is not possible to guarantee more then maybe the plastic cover as a distance between hand and antenna. Bandwidth that an antenna need to cover is also steady increasing.

A common problem is that adding a simple matching network afterward the actual antenna design, degrades more or less whole bandwidth, as already achieved reasonable impedance in middle of antenna bandwidth will be negative affected when adjusting impedance for band-edges.
Instead must whole antenna together with its matching network be designed to reach system impedance, whatever antenna impedance actually is. It is a system-thinking that is rather different then what was possible some years ago, as impedance network must be included in design process, before antenna design is mechanical finished.
This allows much more freedom as bigger part of Smith-chart is acceptable for the resulting antenna impedance as long as the combination matching network+antenna results in low VSWR.

Many RF labs have built their own software tools to make this kind of antenna design possible, as no one can look at a Smith chart and then in their head recognize that a certain curve-shape will fit well together with an certain impedance topology.
Some of these software tools are relative simple, having predefined Smith-chart areas where they know by experience that a certain type of impedance matching will result in a possible system impedance. Real complex solution when also EM solvers are included do also exist but as this often is too slow solutions must some shortcuts be taken.

Even better, some RF labs have a software installed direct in VNA, that dynamically calculate impedance matching network in combination with current measured antenna structure impedance.
That makes the job much easier. It allows you to spend more time to design an effective radiation structure, whatever its impedance it is.

If not using any software is it an infinite trial and error loop, save S-parameter for an ongoing design and import in a simulation tool and try to find an optimized topology, and then back to lab bench for a new try.
For a simple antenna design job do it maybe not need that many iterations but it is in any case time consuming.

It is also possible have this kind of tool installed as an PC software. VNA is then connected with PC over GPIB/LAN/USB and in PC is an optimized combination antenna+network calculated. Some of this type of software are however a bit slow too be able to use as a lab-bench tool. I prefer an update rate of at lest 5 times a second to give a useable real-time feeling when actually sitting there with scalpel cutting CU tape.
Important is also that the tool can calculate matching network based on S-parameters for the actual type of inductors and capacitors. It can be an huge difference comparing with lossless components.
A plus if there also are tools for including hand-effect as a parameter and allowing to design against an irregular system-impedance for optimal matching against RX and/or TX circuits actual impedance as a part of TIS/TRP. The later is almost a most these days as battery efficiency have become very important.

All these thing are possible to do in most common simulation tools, but an normal antenna job that takes hours to finish, do then takes weeks. I know, as that was how I previously worked.
Export S-parameters from VNA, importing these parameters in an external software and try to find an optimized network topology for best fitting current measured impedance curve in combination with an assumed handeffect impedance and how this converge with RX/TX impedance, it is maybe 15 minutes to hours.

If alternative is to get same information 10 times each second, is it a huge advantage.

Have been involved in software projects for several RF labs, both as antenna designing guinea Pig and as software developer since more then 10 years. These software are however not public as they in most cases have special functions that results in a twist that represent actual antenna brand and in some cases are all previously done experience for old antenna projects saved as an AI database, which then in an early stage can predict final result for a current project.

Separate PC software that can present live data from VNA that I know about, and is public available, is BetaMax and AnTune. In my opinion is BetaMax a more general RF tool and AnTune is more specialized as an antenna development bench tool with minimal menus and settings and extreme fast. I am currently partly involved in AnTune both as user and developer, and recommend it as it is free to try, fully functional. Less functional demo-version is available for BetaMax.
Have problems with internal VNA software as internal processing power is too slow to get a fast enough response, but it is maybe just me that have too old and slow VNA. In any case, what I have seen is a bit to simple tools, calculating with ideal components and requiring predefined network topologies.

An alternative is to more or less do whole antenna development in an EM simulating software that can optimize for above requirements but have so far never seen anyone succeed with a complex antenna. Simulation takes days and final result is a bit worse then what is reached within minutes at lab-bench. Optenni is else a such software that in combination with CST can do complex antenna tuning while doing mechanical development.
A simulation tool is always useful in parallel with other tools as it better can visualize what the problem may be if antenna have a deaf range or if RF currents results in unwanted ground-currents that cause EMI, but which to select is more or less a question about taste and cost.