Welcome to the CRÈME site
It has been almost a decade since the introduction of CREME96, the current state-of-the art tool for SEE rate prediction. CREME96 uses phenomenological models to predict SEE rates. These models were based on two assumptions. First it was assumed that the ionization trail left by the particle was much narrower than the minimum feature size in the microelectronic circuits. Second, it was assumed that the SEE sensitivity of individual microcircuits could be idealized as being due to a single sensitive junction. The cross section of this junction versus the linear energy transfer (LET) rate of the ionizing particle could then be measured and used to estimate the SEE rate. Since CREME96 was developed the minimum feature size has shrunk by more than a factor of 100. As a result, the interaction between track microstructure and device characteristics can no longer be ignored. This assumption in CREME96 has been shown to have significant shortcomings when applied to new and emerging technologies like advanced CMOS, SiGe HBTs, photodiodes, and IR FPAs. The solution is to replace current models in CREME96 with a physics-based model that correctly accounts for the distribution of energy deposition about the track and the possible existence of multiple sensitive junctions in each microcircuit.
The need for a comprehensive and extensible complement to CREME96 that is widely accessible is now apparent. The approach chosen for this new SEE model is modular, uses standardized or widely-adopted computer languages, and is based on a core of open-source material, the Geant4 Libraries, for the basic radiation-computation engine. This core Monte Carlo engine can be supplemented in an extensible way with specific models relevant to new technologies, as these models are developed.
Overview & references to Geant4 Documentation
The graphics for the site are generated using the very nice Grace plotting package, which is open source, free, and very well debugged. It produces high-quality PDF output. Users are encouraged to download this package so they can directly manipulate the Grace files from the site, rather than reverting to the CSV files, and having to reformat the plots. Note that Grace is excellent for publication plotting, since it is capable of saving style files, allowing one to format all graphs for a paper in a consistent manner. The conversion of the PDF output of Grace to high-quality antialiased PNG is carried out using ImageMagick combined with Ghostscript.
We use ReportLab to generate the pictures of RPP stacks, and other line drawings.
Upon the public release of this site in November 2010, the user can expect to be able to run the CREME86 and CREME96 modules without modification. The novel features of this site will include:
- an update to the model for the galactic cosmic ray model
- multilayer planar structures to evaluate the effects of semiconductor materials near the sensitive volume
- Monte Carlo radiation transport modules to
- simulate protons, alphas, and heavy ions
- capture the effects of high-energy particles including nuclear interactions
- capture energy loss variation, straggle, and stopping of particles
- reproduce ground-based test data with a mono-directional flux including at-angle irradiation
- produce error rate predictions for isotropic space environments
- evaluate coincident upsets in multiple devices
- model devices with multiple weighted sensitive volumes, both rectangular parallelpiped and ellipsoidal volumes
If you are new to the site, please read Getting Started
If you are coming to the site as a CREME96 user, you should find all of the old codes available, with a much newer user interface.
Notes to users migrating from the NRL CREME96 site:
- You have to create a new user account on this site. We have not ported over the CREME96 accounts from the NRL site.
- Only port over files which contain information which is not easily recomputed. We support import of the .rqp and .rqh files into our PUP and HUP interface, to allow you to retrieve the rather complex information in the device descriptions. Other request files are not supported for direct reading, since they tend to only have a tiny amount of information in the forms.
- Do not import any graphics. Our site has a graphics management tool which allows you to plot the Creme96 output files, and efficiently caches the results of such plots for viewing or download. We provide high-resolution PNG files and scalable PDF files of the graphs.
- The site allows you to create a nice, hierarchical folder-based filesystem to organize your results. Use it! Keeping each folder to a small number of files greatly improves the speed of the web interface, since it doesn't have to display a massively long list of information every time you change something.
- If you upload files manualy, make sure the extensions are lower-case (i.e. upload xxx.gtf instead of xxx.GTF). The tools only look for files with lower-case extensions.
- See this news item for information about a simple way to transfer multiple files. Note that this tools automatically converts file names to lower case.
- The CREME96 part of this site is not part of the beta testing. As a CREME96-only user, you are not bound by the beta agreement presented below.
Please use the following references when publishing results that have been produced by this site. Additionally, we would like to hear from you regarding when and where your paper will be published.
- A.J. Tylka, J. H. Adams, Jr., P. R. Boberg, B. Brownstein, W. F. Dietrich, E. O. Flueckiger, E. L. Petersen, M. A. Shea, D. F. Smart, and E. C. Smith, "CREME96: A Revision of the Cosmic Ray Effects on Micro-Electronics Code", IEEE Trans. Nucl. Sci., vol. 44, no. 6, pp. 2150-2160, Dec. 1997.
- R.A. Weller, M. H. Mendenhall, R. A.Reed, R. D. Schrimpf, K. M. Warren, B. D. Sierawski, and L. W. Massengill, “Monte carlo simulation of single event effects,” IEEE Trans. Nucl. Sci., vol. 57, no. 4, pp. 1726-1746, Aug. 2010.
- Marcus H. Mendenhall and Robert A. Weller, "A probability-conserving cross-section biasing mechanism for variance reduction in Monte Carlo particle transport calculations", Nucl. Inst. & Meth. A, Volume 667, 1 March 2012, Pages 38-43, doi:10.1016/j.nima.2011.11.084.