Verification of RELAP5/MOD 3 With Theoretical and Numerical Stability Results on Single-Phase, Natural Circulation in a Simple Loop (NUREG/IA-0151)

On this page:

Download complete document

Publication Information

Date Published: February 1999

Prepared by:
J. C. Ferreri, ARN
W. Amnbrosini, USP

Autoridad Regulatoria Nuclear
Av. del Libertador 8250
1429 Buenos, Aires, Argentina
Universita degli Studi di Pisa, Facolta di Ingegneria
Dipartimento di Costruzioni Meccaniche e Nucleari
Via Diotisalvi 2, 56126 Pisa, Italy

Prepared as part of:
The Agreement on Research Participation and Technical Exchange
under the International Code Application and Maintenance Program (CAMP)

Office of Nuclear Regulatory Research
U.S. Nuclear Regulatory Commission
Washington, DC 20555-0001

Availability Notice


The theoretical results given by Pierre Welander [1] are used to test the capability of the RELAP5 series of codes to predict instabilities in single-phase flow. These results are related to the natural circulation in a loop formed by two parallel adiabatic tubes with a point heat sink at the top and a point heat source at the bottom. A stability curve may be defined for laminar flow and was extended to consider turbulent flow. By a suitable selection of the ratio of the total buoyancy force in the loop to the friction resistance, the flow may show instabilities. The solution was useful to test two basic numerical properties of the RELAP5 code, namely: a) convergence to steady state flow-rate using a "lumped parameter" approximation to both the heat source and sink and, b) the effect of nodalization to numerically damp the instabilities. It was shown that, using a single volume to lump the heat source and sink, it was not possible to reach convergence to steady state flow rate when the heated (cooled) length was diminished and the heat transfer coefficient increased to keep constant the total heat transferred to (and removed from) the fluid. An algebraic justification of these results is presented, showing that it is a limitation inherent to the numerical scheme adopted. Concerning the effect of nodalization on the damping of instabilities, it was shown that a "reasonably fine" discretization led, as expected, to the damping of the solution. However, the search for convergence of numerical and theoretical results was successful, showing the expected nearly chaotic behavior. This search lead to very refined nodalizations. The results obtained have also been verified by the use of simple, ad hoc codes. A procedure to a issess the effects of nodalizations on the prediction of instabilities threshold is outlined in this report. It is based on the experience gained with the aforementioned simpler codes.

Page Last Reviewed/Updated Friday, March 27, 2020