How to generate EM(electromagnetic) Simulation in ADS

[btv_murthy]

Dear Sir,
I want to generate the Electromagnetic(EM) Simulation for the Branch Line Coupler. I have created the schematic of the Branch line coupler using ADS2009, and generated the layout for the same. Later in the Momentum, I went to substrate > Create/Modify Substrate and went to the substrate layer initialized the thickness of substrate, loss tangent and dielectric constant. Similarly I initialized the Strip conductor and alumina thickness in the same Momentum. Later I set up the frequency parameter and simulated. After Simulation I am getting the error as "no citi files could be found " , it is showing and it is not generated the .dds file for Branch line Coulper.
I have attached the Branch line coupler Schematic and layout and error generated file and also attached the .ZAP file of Branch Line Coupler.
I am new to the EM simulation in ADS2009, Can you suggest me , what mistake I am making and help me create the EM Simulation using ADS2009 .

 

[btv_murthy]

Dear Sir,
For EM simulation , I used Momentum setting up for the substrate and strip conductor declaration and used EMDS for the frequency set up and execution of simulation. I am new to the EM simulation , can you guide me in proper way for EM simulation.

 

[BigBoss]

I think that comes from layout/schematic unit corruption.You used mm in schematic but layout units are inch/mils. there might be a tranlation error as mentioned in simulation status..
I will check it for you..

- - - Updated - - -

Yes, I supposed well.. The layout units are the problem because database exceeds the defined value while translating from schematic to layout.
I recommend you to check "Layout Units" before translating schematic to layout otherwise some such problems may occur.
I have corrected your project file, you can continue with it if you would..

 

[btv_murthy]

I think that comes from layout/schematic unit corruption.You used mm in schematic but layout units are inch/mils. there might be a tranlation error as mentioned in simulation status..
I recommend you to check "Layout Units" before translating schematic to layout otherwise some such problems may occur.

BigBoss Sir,
You told me check the "Layout Units", while translating schematic to layout. Where is the "Layout Units" in ADS2009, where I have to go for getting the "Layout Units" . I searched ,couldn't get "Layout Units". Please where I have to go and search for changing the "Layout Units" in the schematic view.

Few doubts I have

(1) I have done the same schematic for Branch Line Coupler in the AWR Microwave office , I have got the different results at the designed 3GHz for Branch Line Coupler. MWO results as follows

f(GHz) S11dB S21dB S31dB S41dB
3 -13.206 -3.8736 -4.5117 -13.166

Same Schematic with same values of width and length of all MLINs and same substrate values for Branch Line Coupler, when generated in the ADS at 3GHz

f(GHz) S11dB S21dB S31dB S41dB
3 -14.302 -4.812 -4.2 -14.048
I am confused , which one I have to believe. I have attached both AWR MWO and ADS Branch Line Coupler files, can you clarify it.

(2) Whenever we go Momentum > Simulation > S parameters, I have got error same as like earlier no citi files could be found. I didn't get the S-parameter simulation, when I went and pressed in the Momentum >Simulation > S parameters.
But When I go to EMDS > Simulation > S Parameters , then I am getting the above shown ADS generated S Parameters.
Which I have to choose, I am in confusion, weather I have to choose Momentum S parameter Simulation or EMDS S Parameter Simulation. Which one is correct for EM simulation.

(3) Whatever the declaration of substrate parameters in the Momentum is it correct.

 

[SpaceDust]

Are your above attachment of error after the corrections suggested by Bigboss??
You have large variation in results from MWO to ADS .. I believe you have some setting not right in one of the simulation tools. Check for Mesh density, I can also influence your results.!!

 

[btv_murthy]

Dear Sir,
Yes, I have simulated for BigBoss .ZAP modified file, have attached the AWR MWO office file in the posting #4, and ADS file used same as BigBoss modified .ZAP file and simulated the both MWO and ADS file. I am getting results at 3GHz as shown in the posting #4. But I am confident about the AWR MWO file results, since I was extensively using EM simulation in MWO. But I am new to the ADS EM simulation, am not confident about the settings in ADS and generate the EM simulation. Can you anybody try to find, where is the mistake and incorrect settings in ADS , can you suggest me.

 

[BigBoss]

Dear Sir,
Yes, I have simulated for BigBoss .ZAP modified file, have attached the AWR MWO office file in the posting #4, and ADS file used same as BigBoss modified .ZAP file and simulated the both MWO and ADS file. I am getting results at 3GHz as shown in the posting #4. But I am confident about the AWR MWO file results, since I was extensively using EM simulation in MWO. But I am new to the ADS EM simulation, am not confident about the settings in ADS and generate the EM simulation. Can you anybody try to find, where is the mistake and incorrect settings in ADS , can you suggest me.

The differences between ADS and MWO are normal even tough they use Methods of Moment.Every EM simulator has its own way and computation method.
For instance AWR has 2 simulators called as Axiem and EMsight... Axiem gives closer results to ADS but EM sight not.Because computation methods are pretty different.
The "simulating" word is coming from there."Simulating" does not mean "exact and absolute" response, no.. Simulation is just an close-in approximation.
I'm personally confident to Sonnet but not in every case.There are some cases in which Sonnet is giving nonsense results where ADS is much more accurate regarding to measurements.
Consider these simulators in this way of thinking..

 

[btv_murthy]

Dear Sir,
O.K Sir, I have EM Simulation generated Branch Line Coupler created earlier and attached , made mistake of not setting the "Layout Units" as mm , when schematic I used mm as unit. BigBoss Sir corrected the file and attached in the posting # 3.
What he made the correction other than layout unit , now I am getting the different results for both he corrected # 3 file and my original file.

BigBoss Sir corrected posting #3 file obtained result as follows for Branch Line Coupler

f(GHz) S11 dB S21dB S31dB S41dB
3 -14.302 -4.812 -4.2 -14.048

Earlier Same Branch Line Coupler obtained results as follows

f(GHz) S11dB S21dB S31dB S41dB
3 -12.039 -22.896 -1.688 -39.055

For my earlier file even though, I changed the "layout units " as mm ,am getting the above results.I couldn't find,where I have done the mistake, Please can you help me to find, where I have done mistake in the settings.

And also , I am getting the same error when I went Momentum > Simulation > S parameter " no citi files found",it is showing again, how to avoid this.

Please go through the posting #1 .rar file "Layout Unit has to change as mm, please go through my file and clarify me whether setting are correct or not for EM Simulation.

 

[RealAEL]

The error about "no citi files could be found" is nothing to do with the layout units, it should work with mm or mils as long as the layout is drawn the correct size.

The problem is that you are running a very old unsupported version of ADS on a much newer PC. I see what looks like Windows 7 and I presume therefore a multi-core processor. At the time ADS 2009 was built for Pentium 4 processors which only had typically one or two cores. The code would not always be optimum for the newer configuration. Try limiting the simulation to fewer cores. This will slowdown some operations but it should run.

With ADS 2009 running and a Momentum simulation attempted but still open start the Task Manager (right click on taskbar and select Task Manager...). Find the hpeesofemx.exe process, right-click on that process, select Set Affinity... and select only CPU 0 or CPU 0 & CPU 1. Make sure any other CPUs are unchecked and OK out and close the Task Manager. Now restart ADS and the simulation should be successful.

 

[BigBoss]

The error about "no citi files could be found" is nothing to do with the layout units, it should work with mm or mils as long as the layout is drawn the correct size.

The problem is that you are running a very old unsupported version of ADS on a much newer PC. I see what looks like Windows 7 and I presume therefore a multi-core processor. At the time ADS 2009 was built for Pentium 4 processors which only had typically one or two cores. The code would not always be optimum for the newer configuration. Try limiting the simulation to fewer cores. This will slowdown some operations but it should run.

With ADS 2009 running and a Momentum simulation attempted but still open start the Task Manager (right click on taskbar and select Task Manager...). Find the hpeesofemx.exe process, right-click on that process, select Set Affinity... and select only CPU 0 or CPU 0 & CPU 1. Make sure any other CPUs are unchecked and OK out and close the Task Manager. Now restart ADS and the simulation should be successful.

I'm working with 2 versions (2009U1 and 2014.01) together on a same machine that has recent processor ( i7 Win7 64bits).I don't think that was the problem.
I checked his layout, I guess he changed layout units because layout dimensions had discrepancies compare to schematic view.

 

[btv_murthy]

RealAEL Sir,
As I told you earlier, I am new for EM simulation in ADS. I request you to check the attached file of BranchLine_Coupler_Zap_File.rar in the posting #1, whether it is in the right form of settings in the ADS or not. If it is not in the correct format, what mistakes I made it, what are the additional setting do it for ADS file for EM simulation. It is crucial for go ahead to my work, please answer for this.I have designed the Branch Line Coupler for 3GHz for example EM Simulation.
Can you attach the simple example of from Schematic to layout generated Momentum EM simulated .ZAP file and also design details, I can make confirm for my design clarification. I have searched in the ADS2009 example of Mometum Simulation, I didn't understand those.

 

[RealAEL]

Can you attach the simple example of from Schematic to layout generated Momentum EM simulated .ZAP file and also design details, I can make confirm for my design clarification.

Not necessary.

The original project posted in #1 can simulate completely without error even when left exactly as posted. No changes are required. Try adjusting the Affinity as I listed earlier.

The only comment I have about the settings is that you list comments about the performance of the coupler at 3GHz but the s-parameter sweep for the Momentum simulation is 1 - 2 GHz.

 

[btv_murthy]

With ADS 2009 running and a Momentum simulation attempted but still open start the Task Manager (right click on taskbar and select Task Manager...). Find the hpeesofemx.exe process, right-click on that process, select Set Affinity... and select only CPU 0 or CPU 0 & CPU 1. Make sure any other CPUs are unchecked and OK out and close the Task Manager. Now restart ADS and the simulation should be successful.

RealAEL Sir,
As you said , I clicked on the task bar and found the task manager and hpeesofemx.exe, right click on that procsess, and select set Afinity and selected CPU0 or CPU0&CPU1.Then I restarted the ADS and simulated S-parameter momentum. I am getting the same error like as before "could create the data set,since no citi files could be found".

My system details as follows
Processor: Intel(R) Core(TM) i5-2410 CPU@ 2.30GHz 2.30GHz
System Type: 32 bit operating system
Installed Memory(RAM):3.00GB RAM
Operting System Using: Windows 7 Ultimate.

I am here with attaching the error message



RealAEL Sir,
As you said , I set the Affinity and ran it. Now it is working very nicely without error, and results also nearer to the AWR MWO earlier generated EM simulated values. But results of ADS are not exactly same as AWR MWO results, little difference results are obtained. Thanks god, I am really thankful to you sir. Without you I will be away from the ADS so far. You have helped like anything, I really appreciate your generous , timely, patience and elaborate help. Thank you ....... thank you very much...........