IRSN, Institut de radioprotection et de sûreté nucléaire

Search our site :


Contact us :

En Fr

All our expertise to protect you


Computer codes

DRACCAR computer software

Last update on September 2018


Deformation and reflood of a fuel rod assembly

during a loss-of-coolant accident



The DRACCAR computer code, developed by IRSN’s Cooling Accident and Uncertainty Modeling Laboratory (LIMAR), is used for the safety analysis of pressurized water reactors (PWRs) and to support research programs on the behavior of fuel rods in a cooling accident such as a loss-of-coolant accident (LOCA) in a pressurized water reactor or the dewatering of a spent fuel storage pool. During these accidents, for an extended period the water no longer provides sufficient cooling for the fuel assemblies, causing the fuel rods to heat up and potentially degrade. The pressure that builds inside each fuel rod as the temperature rises can cause the cladding surrounding the fuel to creep or even burst. The deformation of fuel rods can also cause major blockages in parts of the core and compromise cooling by the water injected by the emergency systems. DRACCAR can simulate the phenomena that characteristically occur in these accidents and can be used to assess the safety consequences.





The DRACCAR code is a multi-rod 3D simulation tool that models the thermo-chemical and mechanical behavior of water-cooled fuel rods, particularly to assess the peak cladding temperature  reached in a cooling accident and to evaluate the blockage rate associated with deformed fuel rods and the impact of this on core cooling. This multiphysics software couples the thermal (e.g. radiation), mechanical (e.g. cladding creep and rupture), thermo-chemical (e.g. cladding oxidation) and thermal hydraulics phenomena that occur during a loss-of-coolant accident.



In particular it can:


  • provide a coherent interpretation of the entire cooling accident experimental database, whether in a "single-rod" or "rod cluster" configuration, with real or simulated fuel, for separate effect, semi-integral and integral experiments;


  • transpose the phenomena involved to reactor scale by means of models validated on these test results and identify gaps in knowledge through exploratory studies;


  • be used for studies in support of safety analysis to check and consolidate the methods and approaches used for cooling accident-related safety analyses.



             Phebus LOCA test  © IRSN               1/8 of a 17x17 PWR fuel assembly © IRSN


The DRACCAR code is also used to simulate the uncovery (dewatering) phase of cooling accidents affecting spent fuel pools. However, the melting of materials that can occur because of dewatering is outside its scope.




Architecture and physics models



The code is based on a 3D description of the fuel rods in a nonconforming 3D thermal hydraulics meshing. The DRACCAR code is organized around an interface that couples a "structure" code, ICARE3D, and a system-scale thermal hydraulics code. The ICARE3D "structure" code is used to represent the core structures, fuel rods, absorber rods, support and mixing grids and the specific features of experimental devices, e.g. shrouds.

The system-scale thermal hydraulics (core, RCS loops, secondary cooling system and safeguard systems such as the safety injection systems) can be modeled by a system thermal hydraulics code coupled to the ICARE3D representation of the fuel rods.


To simulate the thermal hydraulics, DRACCAR offers coupling to two thermal hydraulics codes to represent, respectively, the flows in the fuel channels and the flows in the circuits:


  • CESAR, the thermal hydraulics module developed by IRSN, from the severe accident simulation code ASTEC. CESAR is a system thermal hydraulics code (5 or 6 equations) using a nonconforming 3D meshing of the core channels. CESAR also offers good numerical performance (optimization of the solver, parallelism). In particular, CESAR is used for applications requiring many calculations (uncertainty calculations) or very detailed meshes.


  • CATHARE-3[1], the thermal hydraulics code developed by the CEA (in partnership with EDF, Framatome and IRSN), which is an evolution of the CATHARE-2 code, considered to be a reference tool for simulating the thermal hydraulics of design basis accidents at pressurized water reactors (particularly cooling accidents). Using CATHARE-3 offers the benefit of a better level of validation than the CESAR code. It is therefore the thermal hydraulics reference tool for studies in support of safety assessments.



The DRACCAR models cover:

  • heat transfer by conduction, convection and radiation inside and between fuel rods;
  • the oxidation of cladding and guide tubes made from zirconium alloy in a steam atmosphere (and in air in pool dewatering accidents);
  • the thermo-mechanical behavior of the cladding (deformation and failure);
  • any contact between neighboring rods and its impact on heat and chemical transfers;
  • the relocation of the fuel through the collapse of the fragmented fuel pellets in the ballooned part of the rod;
  • the fission gas flow in the fuel rod;
  • the two-phase 3D thermal hydraulics at subchannel scale, including heat transfer between fluids and structures and a reflooding model.




Software architecture of DRACCAR © IRSN                


The coupling of the ICARE3D structure code with the CESAR and CATHARE-3 thermal hydraulics codes uses a generic computer interface, which is suitable for coupling with any other system thermal hydraulics code.


DRACCAR also uses a set of auxiliary modules developed at IRSN, such as the ODESSA database management system, which offers powerful pre- and post-processing tools. DRACCAR also uses the MDB (Material Data Bank) database of material behavior properties and laws. This database contains all the properties necessary to represent the materials in the core. Users can supplement or modify this material data when necessary.


DRACCAR can also be coupled with tools developed by IRSN for specific purposes, e.g. to study cladding embrittlement (SHOWBIZ) and to analyze uncertainties (SUNSET). DRACCAR's numerical performance and flexibility means that it can be used for parametric studies to assess the influence of certain modeling assumptions or assumptions about the interaction between phenomena.



Validation and prospects for development


The first version of the code, DRACCAR V1, delivered in March 2008, was validated on the experiments looking at the phenomena that occur during a cooling accident (EDGAR[2], PHEBUS LOCA[3], PERICLES[4], REBEKA[5], HALDEN[6], etc.).


In 2017, the first "industrial" version of the code, DRACCAR v2.3.1, was finalized. This version could be coupled with the reference thermal hydraulics code CATHARE-3. It was validated on nearly 200 tests on the phenomena involved in loss-of-coolant accidents. The validation process includes comparisons with separate-effect experiments (i.e. experiments in which each physical phenomenon is studied separately, such as EDGAR), multi-rod out-of-pile experiments (on fuel assembly reflooding, such as PERICLES, SEFLEX[7] and ACHILLES[8] and on thermo-mechanics, such as REBEKA), in-pile experiments on irradiated rods (i.e. in an experimental reactor like HALDEN) and integral experiments combining all the phenomena (e.g. PHEBUS LOCA). This version of the code is also used to support the interpretation of results from the OECD SCIP-3[9] out-of-pile program, which aims to characterize the fragmentation, relocation and dispersal of irradiated fuel in a cooling accident.


The PERFROI and DENOPI experimental programs run by IRSN and supported by the National Research Agency (ANR) as part of the Investment in the Future program following the Fukushima-Daiichi accident, are specifically contributing to the improvement of the DRACCAR models. For the next version of DRACCAR, the challenges associated with simulation mainly concern improvement of the prediction of cladding failure (PERFROI ELFE experiments), the modeling of contact between rods (PERFROI COCAGNE experiments) and evaluation of the cooling of deformed rods (PERFROI COAL experiments).


Other developments within the DRACCAR code aim to make multi-scale modeling possible so that researchers can focus on the fuel rods of a particular assembly within a whole core connected to the loops and systems in the primary and secondary circuits of a nuclear reactor. These multi-scale developments will enable complex studies to be carried out using a single tool, in support of assessment activities.




Scientific community and collaborations


DRACCAR was developed by IRSN in partnership with EDF and Framatome.


DRACCAR is a tool for capitalizing on the knowledge acquired about cooling accidents, benefiting from experimental programs run by IRSN and other international organizations. Its simulations also help to identify gaps in knowledge about this type of accident and to specify the experiments to be performed to fill these gaps, as in the case of the PERFROI and DENOPI programs. The development of DRACCAR benefits from a high-level scientific environment, particularly through collaborations with academic research partners: the Theoretical and Applied Energetics and Mechanics Laboratory (LEMTA) in Nancy and the Institute of Fluid Mechanics in Toulouse (IMFT), which contribute to the acquisition of experimental data and to the development of models on a local scale. Finally, benchmarking (simulation tool comparison) and validation are carried out in the context of European and international projects supported particularly by the OECD.






  1. Emonot, P., Souyri, A., Gandrille, J., Barré, F., 2011. CATHARE-3: A new system code for thermal-hydraulics in the context of the Neptune project. Nuclear Engineering and Design 241, 4476–4481
  2. Forgeron, T., et al., "Experiment and modelling of advanced fuel rod behavior under LOCA conditions: α↔β phase transformation kinetics and EDGAR methodology", in proceedings of the 12th International Symposium on Zirconium in the Nuclear Industry, ASTM STP 1354, (2000), pp. 256-278
  3. Scott de Martinville, E.& Pignard, M., "International Standard Problem ISP 19. Behavior of a Fuel Rod Bundle during a Large Break LOCA Transient with a Two Peaks Temperature History (PHEBUS Experiment). Final Comparison Report". OECD/NEA CSNI Report n°131, 1987.

  4. Veteau, J., Digonnet, A., PERICLES programme: boil-up, boil-off and reflooding high pressure experiments in a PWR assembly. European Two Phase Flow Group Meeting, Paris. 1989

  5.  Karwat, H., et al., 1985. International standard problem ISP 14: Behaviour of a fuel bundle simulator during a specified heat-up and flooding period - REBEKA experiment - final comparison report. Tech. Rep. CSNI report 98,OECD/NEA, Committee on the Safety of Nuclear Installations.

  6. Wiesenack, W., et al., Safety significance of the Halden ifa-650 LOCA test results. Tech. Rep. NEA/CSNI/R(2010)5 - JT03292495, OECD/NEA, Committee on the Safety of Nuclear Installations. 2010

  7. Ihle, P., Rust, K., 1987. PWR reflood experiments using full length bundles of rods with Zircaloy claddings and alumina pellets: results of the SEFLEX program. Nuclear Engineering and Design 99, 223-237

  8. Dore, P., Pearson, K., 1991. ACHILLES ballooned experiments. Tech. Rep. AEA-TRS-1060, AEEW-R 2590 (included in OECD/NEA CSNI-1015/01), Nuclear Electric (former CEGB), AEA Technology, Winfrith Technology Centre

  9. SCIP-3 OECD/NEA project, NEA Studsvik Cladding Integrity Project (SCIP-3).2014., accessed: 2017-12-06, project period: 2014 to 2019




Send to a friend

The information you provide in this page are single use only and will not be saved.
* Required fields

Recipient's email:*  

Sign with your name:* 

Type your email address:*   

Add a message :

Do you want to receive a copy of this email?