FAQ 2010

This is a list of questions we have received relating to the version 2010 release. Some were asked during the webcast, others in the course of normal email correspondence. This list will be updated regularly.

General

How do we get version 2010?

Everyone with valid software maintenance or an ongoing rental contract will receive CST STUDIO SUITE 2010 automatically in January 2010

Is Antenna Magus distributed with CST STUDIO SUITE as a bundle?

Antenna Magus is a product of Magus Pty. It is a standalone software package that offers the capability to export antenna models to several 3D tools in addition to our own. Because of the powerful pairing, our customers are often interested in trying and purchasing the two tools together, however, this is not required.

What is new in CST EM STUDIO?

Just to give you some short news on CST EM STUDIO 2010: All solvers will have 2nd order tetrahedral capability. The TET-meshing is also featuring true geometry adaptation. Our transient LF solver (fully supporting non-linear materials) got some many additional features and more monitors.

Which operating systems are supported?

The CST STUDIO SUITE 2010 runs on Windows (XP, Vista, Windows 7, Windows Server). The computations can be run remotely or using distributed computing on Linux RHEL4, RHEL5 or CentOS workstations. We will soon have a beta version of the entire CST STUDIO SUITE under Linux soon.

What machine you need to address the different problems shown using mixed mesh and sensitivity analysis?

This depends a bit on how complex and electrically large you device under test is. We have used standard workstations with 4GB, 16GB, and 64GB RAM are. The new features mixed mesh and sensitivity do not require extra RAM.

What computational platform used for navy simulation?

We used a powerful workstation with 64GB of RAM, however not all simulations required all of the RAM.

Applicability

Can we do some simulation in the THz frequency range?

Yes, this is possible, especially the transient solver in CST MWS is capable to simulate electrical large structures, up to about 50-100 wavelengths in size. We have customers working on optical devices.

Is it possible to model Carbon Nanotube size material where carbon tube lines are 10-50um radius and net space 100-500um?

The geometric dimensions should not pose any problems. Concerning the nanotube material, CST has many options including dispersive and anisotropic behavior to describe materials. We cannot however describe the tubes on an atomic/molecular scale.

Can CST STUDIO SUITE simulate pcb and flip-chip die interaction in the 10-80GHz range?

We do have measurements/simulations for flip-chip ball design for frequencies in the order of 0-40GHz, however we don’t have data for higher frequencies. We think that it is possible to analyze at higher frequencies as well.

Will it be efficient to simulate an extremely UWB antenna from 4 to 70GHz using frequency domain solver instead of time domain solver and with stable results?

While it may be possible to do such a simulation in the frequency domain, it would probably not be the most efficient solver choice. For that very wide bandwidth, a lot of frequency samples are going to be required which will lengthen the simulation considerably – on the other hand in a transient analysis you would obtain broadband results in a single pass. Additionally, I am guessing that at 70 GHz the structure is going to be fairly electrically large which may cause problems in regards to the amount of memory required.

EDA

Which formats are supported for PCB design import?

The import functionality into CST MWS includes direct imports from Mentor Graphics PADS and Expedition, Zuken CR-5000, ODB++ and Gerber. Version 2010 will also feature a direct import of Cadance .brd files. It is also possible to access Altium P-CAD, Altium Protel, Cadence Allegro, Cadence OrCAD, Mentor Graphics Hyperlynx, Zuken Visula and the CST/Simlab internal formats.

How do you model the chip itself? Do you have its input/output capacitances and resistances?

There are practically 2 different ways to model chips: 1) 3D model or 2) equivalent model (circuit models SPICE based, touchstones..etc) and in CST MWS and CST PCBS there are both options. Furthermore from a 3D simulation of the chip an equivalent SPICE model can be extracted via MOR and used in a subsequent circuit simulation. The PEEC method, implemented in CST PCBS, allow to directly generate a SPICE model directly related to Maxwell equations.

Postprocessing

Which farfield representations are available?

Besides the standard spherical representation CST MWS 2010 additionally supports Ludwig II and III representation of farfields. These can also be accessed during parameter sweeping and optimization runs. A new feature of 2010 version is the support of SLANT-polarization.

Is there a possiblity to overlay farfield data from different ports to achieve an overall graph (including phase shifts between the ports)?

Yes, there is. In the Postprocessing you have a comfortable “Combine Results” option, in which you can specify Am/ph for every port. In case you are working with feeding networks, you can construct it in CST DESIGN STUDIO, afterwards the correct am/ph at each port are automatically transferred to CST MWS to get the correct combined (broadband) farfield patterns.