Need your suggestion regarding 3D EM simulator

[saravan]

Hi all,

I am new to this forrum. for the last 2 months i am working on IE3D for the design and simulation of planar inductor and I am fabricating the device too.

I want to know why the simulator is taking so much time to simulate my design at higher frequencies of meshing.

I am in confusion whether my design is wrong or something else.

Please give me suggestions.

thanks in advance.

 

[jian]

Hi, Saravan:

For most spiral structures, IE3D is in fact very fast compared to other EM simulators. However, I would like to suggest you to mesh the structure fine enough for it. The rule of thumb is that you should mesh the inner most segment (assuming rectangular with 4 segments for each term) to be meshed into 2 cells. A spiral inductor normally has much detail. There might be many vias and multiple layers. If you do not try to reduce some un-necessary detail, the problem can be huge. You can send me the file at jian@zeland.com and I can try to help you on it. Regards.

 

[maxwellian]

Hi Saravan:

Maybe you just need to try a different simulator. Two months is a lot of time to lose fighting with software. Your time would probably pay for a large part of the software cost.

On the planar side, I would recommend you look at ADS Momentum or Sonnet. In my opinion (and it is just an opinion) IE3D does not have any large advantage over either of these. At least it is a good idea to see for yourself if you have the best tool for the job. Both Sonnet and Momentum now can do thick metal in some fashion well enough for this application.

On the 3D side, you might consider HFSS or CST Microwave Studio. However, if you must contend with very thin oxide layers, then you might want to stay with a planar EM solver; the full 3D solvers have a lot of trouble with very thin dielectric layers in semiconductor simulations.

--Max

 

[jian]

Hi, Saravan: If you want to be able to solve your spiral problems easier, I do suggest you to try our IE3D Special Flow Edition with Automatic GDSII to IE3D Flow (AGIF). AGIF can automate the model creating automatically from GDSII file. It can even do a batch simulation directly from a set of GDSII files with the same layer configuration. AGIF is in the design flow of some major semi-conductor companies and they are getting very good results (both in accuracy and efficiency). AGIF is temporarily enabled for all IE3D users. AGIF will be available for Cadence Virtuoso and Cadence Allegro very soon for automatic model creation and simulation. Thanks!