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.

Difference between ADS simulation results and experimental results

Status
Not open for further replies.

modelstudent

Newbie level 6
Joined
Aug 12, 2010
Messages
14
Helped
2
Reputation
4
Reaction score
2
Trophy points
1,283
Activity points
1,370
I have designed a 3 band antenna and it gives good results in the ADS S11 simulation. However, when the antenna is fabricated and measured, one of the band only gives -6dB, but in the simulation results, it gives -20dB.
Are there any reasons about this problem so I can solve it ?
Many thanks~:cry:
 

Hi
One of the reasons can be the dispersion characteristics of the substrate in practice. The second reason can be your inaccurate simulation such as choosing broad frequency band, simulation type,... and finally since ADS uses moment method for analyzing and this method have some problems in edges, if your antenna structure have more edges you should use higher density meshes.

I'll its be helpful
 

3 band antenna is not an easy thing, you should be happy that you got this much in the first prototype. Here are some recommendations.

-Try to keep the antenna away from surrounding objects during s11 measurements.
-Try modifying the antenna by cutting off bits and adding to it.
-What kind of frequencies are you working at? ADS gets pretty inaccurate at high frequencies. try a different software if you can.
-It is probably the high frequency that is not working. Check you feed, make sure it is suitable for that high freq.
-I can give you more feedback if you give more details.
 

Hi
One of the reasons can be the dispersion characteristics of the substrate in practice. The second reason can be your inaccurate simulation such as choosing broad frequency band, simulation type,... and finally since ADS uses moment method for analyzing and this method have some problems in edges, if your antenna structure have more edges you should use higher density meshes.

I'll its be helpful

It's a good policy to enable edge-meshing when using modeling radiating/gap-coupling circuits in ADS Momentum. Start without edge meshing to tune your design. Once you get close, turn on Edge mesh in the Momentum control window and re-optimize. If you have the time after optimizing the design with edge meshing, then also use a Thick Metal model for the conductors (takes a lot longer to run). That will help improve your calculations at the edges of the conductors, getting you closer to reality. Still, some hand-tuning is to be expected, so get out your exacto-knife and copper tape.
 

since i am using a partial ground plane, I am simulating with 2 ports , one is the internal port and one is the reference ground port. In the manual of ADS, it states that there will be some difference in simulation results between using the internal port and single mode port, is it right for me to use an internal port? For the simulation results , the internal one is more similar to experimental result than single mode one~
Thanks~
 

since i am using a partial ground plane, I am simulating with 2 ports , one is the internal port and one is the reference ground port. In the manual of ADS, it states that there will be some difference in simulation results between using the internal port and single mode port, is it right for me to use an internal port? For the simulation results , the internal one is more similar to experimental result than single mode one~
Thanks~

An Internal port will act like you are feeding the structure at a single point. Think of this like running the center conductor of a coax up through a via, and soldering it to the copper of the patch antenna... the input signal enters the EM structure at one point, and propagates out radially from there.

Using single mode excitation will act more like a transmission-line feed (use this if you are using a microstrip to feed your patch). If I remember correctly, single mode will do a port extension and de-embedding automatically. This would be like adding a section of lossless transmission line (at the same impedance as your specified port impedance) to the circuit, simulating it, then de-embedding the solution to the reference plane of the port pin.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top