Computer Programs

DOT-3.5, 2-D Neutron Transport, Gamma Transport Program DOT with New Space-Scaling

NAME OR DESIGNATION OF PROGRAM, COMPUTER, NATURE OF PROBLEM OR FUNCTION, METHOD OF SOLUTION, RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM, TYPICAL RUNNING TIME, UNUSUAL FEATURES OF THE PROGRAM, RELATED AND AUXILIARY PROGRAMS, STATUS, REFERENCES, MACHINE REQUIREMENTS, LANGUAGE, OPERATING SYSTEM OR MONITOR UNDER WHICH PROGRAM IS EXECUTED, ANY OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS, NAME AND ESTABLISHMENT OF AUTHOR, MATERIAL, CATEGORIES

[ top ]

[ top ]

To submit a request, click below on the link of the version you wish to order. Rules for end-users are
available here.

Program name | Package id | Status | Status date |
---|---|---|---|

DOT-3.5 | CCC-0276/02 | Tested | 01-MAY-1979 |

DOT-3.5/E | CCC-0276/07 | Tested | 01-MAY-1979 |

Machines used:

Package ID | Orig. computer | Test computer |
---|---|---|

CCC-0276/02 | IBM 370 series | IBM 370 series |

CCC-0276/07 | IBM 370 series | IBM 370 series |

[ top ]

3. NATURE OF PROBLEM OR FUNCTION

DOT solves the Boltzmann transport equation in two-dimensional geometries. Principal applications are to neutron and/or photon transport, although the code can be applied to transport problems for any particles not subject to external force fields. Both homogeneous andexternal-source problems can be solved. Searches on multiplication factor, time absorption, nuclide concentration, and zone thickness are available for reactor problems. Numerous edits and output data sets for subsequent use are available.

DOT-3.5 improves the space-scaling algorithm.

DOT-3.5/CAB contains group by group UPSCATTER scaling method.

DUCT calculates perturbations to the scalar flux caused by the presence of ducts filled with coolant.

VIP is a program for cross section sensitivity analysis using two- dimensional discrete ordinates transport calculations.

DGRAD calculates the directional flux gradients from DOT-3 diffusion theory flux tapes. In conjunction with VIP and TPERT, it allows the use of diffusion theory fluxes to obtain exact and first-order perturbation reactivity changes. In order to calculate the reactivity associated with changes in reactor compositions using diffusion theory, it is necessary to fold not only the scalar fluxes with the appropriate cross sections, but also the average flux gradients with the diffusion coefficients. Since DOT diffusion theory does not directly calculate these gradients, it was necessary to calculate the needed quantities external to the DOT code.

TPERT is a perturbation code to obtain exact and first-order reactivity changes. TPERT is coupled to VIP which generates adjoint forward fux tables using DOT-3 scalar flux tape information.

GRTUNCL calculates an analytical first-collision source for subsequent use in DOT.

DOT solves the Boltzmann transport equation in two-dimensional geometries. Principal applications are to neutron and/or photon transport, although the code can be applied to transport problems for any particles not subject to external force fields. Both homogeneous andexternal-source problems can be solved. Searches on multiplication factor, time absorption, nuclide concentration, and zone thickness are available for reactor problems. Numerous edits and output data sets for subsequent use are available.

DOT-3.5 improves the space-scaling algorithm.

DOT-3.5/CAB contains group by group UPSCATTER scaling method.

DUCT calculates perturbations to the scalar flux caused by the presence of ducts filled with coolant.

VIP is a program for cross section sensitivity analysis using two- dimensional discrete ordinates transport calculations.

DGRAD calculates the directional flux gradients from DOT-3 diffusion theory flux tapes. In conjunction with VIP and TPERT, it allows the use of diffusion theory fluxes to obtain exact and first-order perturbation reactivity changes. In order to calculate the reactivity associated with changes in reactor compositions using diffusion theory, it is necessary to fold not only the scalar fluxes with the appropriate cross sections, but also the average flux gradients with the diffusion coefficients. Since DOT diffusion theory does not directly calculate these gradients, it was necessary to calculate the needed quantities external to the DOT code.

TPERT is a perturbation code to obtain exact and first-order reactivity changes. TPERT is coupled to VIP which generates adjoint forward fux tables using DOT-3 scalar flux tape information.

GRTUNCL calculates an analytical first-collision source for subsequent use in DOT.

[ top ]

4. METHOD OF SOLUTION

The method of discrete ordinates is used.

Balance equations are solved for the density of particles moving along discrete directions in each cell of a two-dimensional spatial mesh. Anisotropic scattering is treated using a Legendre expansion of arbitrary order. Convergence can be accelerated by several optional schemes, including a pointwise rescaling technique.

DOT-3.5/E: Differs from DOT-3.5 in that exponential supplementary equations, as well as the usual diamond and weighted schemes, may be used to find the mesh-centre flux from the fluxes at the faces of the mesh.

The model:

1. always gives positive solutions and does not require any fixup techniques provided that the source is non-negative;

2. improves convergence rate in most neutron deep-penetration problems and, for any practical spatial discretization, always requires CPU times not only smaller than those required by DOT-3 mixed (linear + step fixup) model, but also shorter (generally 10-20%) than the times required by DOT-3.5 weighted difference model;

3. increasing spatial mesh size supplies solutions which are always reasonable overestimates of the exact solution and its numerical behaviour is more stable and coherent than the mixed mode.

Experience up to now from several deep penetration problems in (r,z) and (x,y) geometry shows that, while for neutrons the exponential model almost always works very well, for gamma rays its behaviour may be critical and in some cases there is lack of convergence.

The method of discrete ordinates is used.

Balance equations are solved for the density of particles moving along discrete directions in each cell of a two-dimensional spatial mesh. Anisotropic scattering is treated using a Legendre expansion of arbitrary order. Convergence can be accelerated by several optional schemes, including a pointwise rescaling technique.

DOT-3.5/E: Differs from DOT-3.5 in that exponential supplementary equations, as well as the usual diamond and weighted schemes, may be used to find the mesh-centre flux from the fluxes at the faces of the mesh.

The model:

1. always gives positive solutions and does not require any fixup techniques provided that the source is non-negative;

2. improves convergence rate in most neutron deep-penetration problems and, for any practical spatial discretization, always requires CPU times not only smaller than those required by DOT-3 mixed (linear + step fixup) model, but also shorter (generally 10-20%) than the times required by DOT-3.5 weighted difference model;

3. increasing spatial mesh size supplies solutions which are always reasonable overestimates of the exact solution and its numerical behaviour is more stable and coherent than the mixed mode.

Experience up to now from several deep penetration problems in (r,z) and (x,y) geometry shows that, while for neutrons the exponential model almost always works very well, for gamma rays its behaviour may be critical and in some cases there is lack of convergence.

[ top ]

[ top ]

6. TYPICAL RUNNING TIME

Problems of practical value may range from 0.05 minutes to many hours in running time. A formula given in the input description allows estimation of central processor time for a particular problem.

DOT-3.5/E: In the most deep penetration problems, DOT-3.5/E with exponential model is about 10 to 20 % faster than DOT-3.5 with weighted difference model.

Problems of practical value may range from 0.05 minutes to many hours in running time. A formula given in the input description allows estimation of central processor time for a particular problem.

DOT-3.5/E: In the most deep penetration problems, DOT-3.5/E with exponential model is about 10 to 20 % faster than DOT-3.5 with weighted difference model.

[ top ]

7. UNUSUAL FEATURES OF THE PROGRAM

Flexible dimensioning is used throughout so that only the total storage requirement for a given problem is of concern. The program has two levels of external storage utilization in order to adapt to various problem requirements efficiently. Cross sections can be input on cards, from an ANISN-type nuclide-organized data set, or from a special group-organized data set which is valuable for very large cross section sets. A free-field data input format facilitates problem preparation. Special features which make DOT-3 well adapted to shielding problems include albedo boundary conditions, the capability to output fluxes at exterior or interior boundaries for use in 'bootstrap' problems, and an analytic first-collision source calculation that mitigates ray-streaming effects in certain problems. Upscattering is allowed, and special routines are able to synthesize two-dimensional flux guesses from one-dimensional data produced by the ANISN code. Diffusion theory is also available. The convergence test can be restricted to specified zones.

DOT-3.5/E: Exponential equations to compute mesh centre fluxes.

Flexible dimensioning is used throughout so that only the total storage requirement for a given problem is of concern. The program has two levels of external storage utilization in order to adapt to various problem requirements efficiently. Cross sections can be input on cards, from an ANISN-type nuclide-organized data set, or from a special group-organized data set which is valuable for very large cross section sets. A free-field data input format facilitates problem preparation. Special features which make DOT-3 well adapted to shielding problems include albedo boundary conditions, the capability to output fluxes at exterior or interior boundaries for use in 'bootstrap' problems, and an analytic first-collision source calculation that mitigates ray-streaming effects in certain problems. Upscattering is allowed, and special routines are able to synthesize two-dimensional flux guesses from one-dimensional data produced by the ANISN code. Diffusion theory is also available. The convergence test can be restricted to specified zones.

DOT-3.5/E: Exponential equations to compute mesh centre fluxes.

[ top ]

8. RELATED AND AUXILIARY PROGRAMS

ANISN - produces one-dimensional flux guesses.

SPACETRAN - calculates the neutron or gamma flux at arbitrary points in space exterior to a two-dimensional right circular cylindrical system which as been calculated by the discrete-ordinates code DOT.

DOQDP: A computer program to generate level symmetric quadrature sets.

ADOQ: combines asymmetric quadrature data with symmetric quadrature data to form "biased" quadrature sets.

ESTOQ: a program to compute first-collision sources for eliminating ray effects in two-dimensional Sn calculations.

DASH: a void tracing and Sn Monte Carlo bridging code. Using angular angular fluxes from cylindrical geometry calculation, it calculates angular fluxes on arbitrary cylindrical geometry surfaces outisde the DOT boundaries. It is useful to reduce ray effects.

DOMINO: a general purpose code for coupling discrete ordinates and Monte Carlo radiation transport calculations.

APSAI: a computer code for plotting fluxes and absorption densities generated by the ANISN code.

ANISN - produces one-dimensional flux guesses.

SPACETRAN - calculates the neutron or gamma flux at arbitrary points in space exterior to a two-dimensional right circular cylindrical system which as been calculated by the discrete-ordinates code DOT.

DOQDP: A computer program to generate level symmetric quadrature sets.

ADOQ: combines asymmetric quadrature data with symmetric quadrature data to form "biased" quadrature sets.

ESTOQ: a program to compute first-collision sources for eliminating ray effects in two-dimensional Sn calculations.

DASH: a void tracing and Sn Monte Carlo bridging code. Using angular angular fluxes from cylindrical geometry calculation, it calculates angular fluxes on arbitrary cylindrical geometry surfaces outisde the DOT boundaries. It is useful to reduce ray effects.

DOMINO: a general purpose code for coupling discrete ordinates and Monte Carlo radiation transport calculations.

APSAI: a computer code for plotting fluxes and absorption densities generated by the ANISN code.

[ top ]

Package ID | Status date | Status |
---|---|---|

CCC-0276/02 | 01-MAY-1979 | Tested at NEADB |

CCC-0276/07 | 01-MAY-1979 | Tested at NEADB |

[ top ]

10. REFERENCES

- W. W. Engle, Jr.:

'A User's Manual for ANISN'

Computing Technology Center, Union Carbide Corporation

Report K-1693 (March 30, 1967).

- N.M. Greene, and C.W. Craven, Jr.:

'XSDRN, A Discrete Ordinates Spectral Averaging Code'

ORNL-TM-2500 (July 1969).

- W.A. Rhoades:

'The ALC1 Program for Cross-Section Library Management'

ORNL-TM-4015 (December 1972).

- Unpublished Notes for GIP, GRTUNCLE, SPACETRAN, FALSTF, ISOPLOT, PLOTTER,

PERT-2D, FACT, and DOMINO.

- R. Douglas O'Dell and Raymond E. Alcouffe:

Transport Calculations for Nuclear Analyses: Theory and Guidelines for

Effective Use of Transport Codes

LA-10983-MS and UC-32 (September 1987).

- NEA-CPL:

Note on DOT3.5 (12 January 1977)

- OECD/NEA:

Note from the NEA Data Bank (May 1984)

- L. Erradi:

Corrections Apportees a DOT-3.5/E - Adaptation sur la Machine IBM 4361 (4341)

VM.

- W. W. Engle, Jr.:

'A User's Manual for ANISN'

Computing Technology Center, Union Carbide Corporation

Report K-1693 (March 30, 1967).

- N.M. Greene, and C.W. Craven, Jr.:

'XSDRN, A Discrete Ordinates Spectral Averaging Code'

ORNL-TM-2500 (July 1969).

- W.A. Rhoades:

'The ALC1 Program for Cross-Section Library Management'

ORNL-TM-4015 (December 1972).

- Unpublished Notes for GIP, GRTUNCLE, SPACETRAN, FALSTF, ISOPLOT, PLOTTER,

PERT-2D, FACT, and DOMINO.

- R. Douglas O'Dell and Raymond E. Alcouffe:

Transport Calculations for Nuclear Analyses: Theory and Guidelines for

Effective Use of Transport Codes

LA-10983-MS and UC-32 (September 1987).

- NEA-CPL:

Note on DOT3.5 (12 January 1977)

- OECD/NEA:

Note from the NEA Data Bank (May 1984)

- L. Erradi:

Corrections Apportees a DOT-3.5/E - Adaptation sur la Machine IBM 4361 (4341)

VM.

CCC-0276/02, included references:

- W.A. Rhoades:Comments on the DOT 3.5 Version of DOT III (November 1975)

- Utility Network of America:

Program Additions to DOT 3 (July 1975)

- W.A. Rhoades and F.R. Mynatt:

The DOT-III Two-Dimensional Discrete Ordinates Transport Code

ORNL-TM-4280 (September 1973)

- P.A. Read, W.E. Selph and R.J. Cerbone:

DUCT Code Manual, Gulf-RT-10654

- J.T. West:

SORREL (November 1975)

- J.P. Jenal:

Common Symmetric Quadratures and the DOQDP Computer Code (August 1975)

- R.L. Childs:

VIP Input Instructions (November 1976)

- R.L. Childs, D.E. Bartine and W.W. Engle, Jr.:

Perturbation Theory and Sensitivity Analysis for Two-Dimensional Shielding

Calculations, ANS Transactions, Vol. 21, pp. 543-544 (June 1975)

- J.P. Jenal et al.:

The Generation of a Computer Library for Discrete Ordinates Quadrature Sets

ORNL/TM-6023 (September 1977)

- E.T. Tomlinson and R.A. Lillie:

User's Guide and Description of the Perturbation Code Modules DGRAD and TPERT

ORNL/CSD/TM-71 (September 1978)

- E.T. Tomlinson, R.L. Childs, and R.A. Lillie:

DOS Perturbation Modules DGRAD/VIP/TPERT, ORNL/CSD/TM-116 (May 1980)

- J. Pena:

DOT 3.5/E-JEN, Two Dimensional Discrete Ordinates Radiation Transport Code

- P. Barbucci and F. Di Pasquantonio:

Implementation of Exponential Supplementary Equations on DOT-III and DOT 3.5

Codes (Document received November 1977)

- P. Barbucci and F. Di Pasquantonio:

Exponential Supplementary Equations for Sn Methods: The Two-Dimensional Case

Reprint of Paper, Proceedings of Fifth International Conference on Reactor

Shielding, Knoxville, Tennessee (April 18-22, 1977)

CCC-0276/07, included references:

- W.A. Rhoades:Comments on the DOT 3.5 Version of DOT III (November 1975)

- Utility Network of America:

Program Additions to DOT 3 (July 1975)

- W.A. Rhoades and F.R. Mynatt:

The DOT-III Two-Dimensional Discrete Ordinates Transport Code

ORNL-TM-4280 (September 1973)

- P.A. Read, W.E. Selph and R.J. Cerbone:

DUCT Code Manual, Gulf-RT-10654

- J.T. West:

SORREL (November 1975)

- J.P. Jenal:

Common Symmetric Quadratures and the DOQDP Computer Code (August 1975)

- R.L. Childs:

VIP Input Instructions (November 1976)

- R.L. Childs, D.E. Bartine and W.W. Engle, Jr.:

Perturbation Theory and Sensitivity Analysis for Two-Dimensional Shielding

Calculations, ANS Transactions, Vol. 21, pp. 543-544 (June 1975)

- J.P. Jenal et al.:

The Generation of a Computer Library for Discrete Ordinates Quadrature Sets

ORNL/TM-6023 (September 1977)

- E.T. Tomlinson and R.A. Lillie:

User's Guide and Description of the Perturbation Code Modules DGRAD and TPERT

ORNL/CSD/TM-71 (September 1978)

- E.T. Tomlinson, R.L. Childs, and R.A. Lillie:

DOS Perturbation Modules DGRAD/VIP/TPERT, ORNL/CSD/TM-116 (May 1980)

- J. Pena:

DOT 3.5/E-JEN, Two Dimensional Discrete Ordinates Radiation Transport Code

- P. Barbucci and F. Di Pasquantonio:

Implementation of Exponential Supplementary Equations on DOT-III and DOT 3.5

Codes (Document received November 1977)

- P. Barbucci and F. Di Pasquantonio:

Exponential Supplementary Equations for Sn Methods: The Two-Dimensional Case

Reprint of Paper, Proceedings of Fifth International Conference on Reactor

Shielding, Knoxville, Tennessee (April 18-22, 1977)

[ top ]

11. MACHINE REQUIREMENTS

Card input, printed and punched output, and the 3 scratch data sets may be located on any external storage device, as may the 9 optional data sets. A small, but useful, problem can be run in as little as 256k bytes of fast memory. A clock, if available, provides timing data.

Card input, printed and punched output, and the 3 scratch data sets may be located on any external storage device, as may the 9 optional data sets. A small, but useful, problem can be run in as little as 256k bytes of fast memory. A clock, if available, provides timing data.

[ top ]

Package ID | Computer language |
---|---|

CCC-0276/02 | FORTRAN+ASSEMBLER |

CCC-0276/07 | FORTRAN-IV |

[ top ]

[ top ]

[ top ]

[ top ]

CCC-0276/02

File name | File description | Records |
---|---|---|

CCC0276_02.002 | DESCRIPTION OF VARIABLES IN COMMON | 399 |

CCC0276_02.003 | SOURCE PROGRAM (ORNL FEB-77,F4,EBCDIC) | 5946 |

CCC0276_02.004 | HEADER,MESAGE,LOCO,SUBR (F4,EBCDIC) | 292 |

CCC0276_02.005 | CNNP SUBROUTINE (FEB-79,F4,EBCDIC) | 209 |

CCC0276_02.006 | GRIND SUBROUTINE (F4,EBCDIC) | 413 |

CCC0276_02.007 | WWESOL SUBROUTINE (F4,EBCDIC) | 124 |

CCC0276_02.008 | GRIND SUBROUTINE (ASSEMBLER,EBCDIC) | 1520 |

CCC0276_02.009 | WWESOL SUBROUTINE (ASSEMBLER,EBCDIC) | 400 |

CCC0276_02.010 | ALOCAT SUBROUTINE (ASSEMBLER,EBCDIC) | 44 |

CCC0276_02.011 | CMVBT SUBROUTINE (ASSEMBLER,EBCDIC) | 37 |

CCC0276_02.012 | JOBNUM,ITIME,IOR,IAND,ICLOCK (ASSEMBLER) | 236 |

CCC0276_02.013 | CMVBT,ALOCAT,ITIME,IFTIME (F4,EBCDIC) | 38 |

CCC0276_02.014 | ERROR MESSAGES | 50 |

CCC0276_02.015 | OVERLAY CARDS | 27 |

CCC0276_02.016 | SAMPLE PROBLEM INPUT DATA | 1027 |

CCC0276_02.017 | SAMPLE PROBLEM PRINTED OUTPUT | 7932 |

CCC0276_02.018 | JCL FOR RUNNING DOT-3.5 SAMPLE PROBLEM | 193 |

CCC0276_02.019 | UPSCATTER SAMPLE PROB. PRINTED OUTPUT | 2678 |

CCC0276_02.020 | VIP SOURCE PROGRAM (F4,EBCDIC) | 1086 |

CCC0276_02.021 | BSAM ROUTINES (ASSEMBLER,EBCDIC) | 421 |

CCC0276_02.022 | BSAM ROUTINES (F4,EBCDIC) | 269 |

CCC0276_02.023 | 'FORWARD' FOR VIP SAMPLE PROB. INPUT DATA | 386 |

CCC0276_02.024 | 'FORWARD' FOR VIP SAMPLE PROB. OUTPUT | 3649 |

CCC0276_02.025 | 'ADJOINT' FOR VIP SAMPLE PROB. INPUT DATA | 381 |

CCC0276_02.026 | 'ADJOINT' FOR VIP SAMPLE PROB. OUTPUT | 3296 |

CCC0276_02.027 | VIP SAMPLE PROBLEM INPUT DATA | 19 |

CCC0276_02.028 | VIP SAMPLE PROBLEM PRINTED OUTPUT | 899 |

CCC0276_02.029 | SWANLAKE S. PROB. INPUT(USING VIP OUTPUT) | 42 |

CCC0276_02.030 | SWANLAKE S. PROB. PRINTED OUTPUT | 304 |

CCC0276_02.031 | JCL FOR DOT-VIP-SWANLAKE RUN | 75 |

CCC0276_02.032 | SORREL SOURCE PROGRAM (F4,EBCDIC) | 1210 |

CCC0276_02.033 | JCL FOR SORREL SAMPLE PROBLEM | 32 |

CCC0276_02.034 | SORREL SAMPLE PROBLEM INPUT DATA | 221 |

CCC0276_02.035 | SORREL SAMPLE PROBLEM PRINTED OUTPUT | 3729 |

CCC-0276/07

File name | File description | Records |
---|---|---|

CCC0276_07.002 | DESCRIPTION OF VARIABLES IN COMMON | 399 |

CCC0276_07.003 | SOURCE PROGRAM (ORNL FEB-77,F4,EBCDIC) | 5946 |

CCC0276_07.004 | HEADER,MESAGE,LOCO (ENEL VERSION,F4,EBCDIC) | 312 |

CCC0276_07.005 | CNNP SUBROUTINE (FEB-79,F4,EBCDIC) | 209 |

CCC0276_07.006 | GRIND SUBROUTINE (ENEL VERSION,F4,EBCDIC) | 466 |

CCC0276_07.007 | WWESOL SUBROUTINE (F4,EBCDIC) | 124 |

CCC0276_07.008 | WWESOL SUBROUTINE (ASSEMBLER,EBCDIC) | 400 |

CCC0276_07.009 | ALOCAT SUBROUTINE (ASSEMBLER,EBCDIC) | 44 |

CCC0276_07.010 | CMVBT SUBROUTINE (ASSEMBLER,EBCDIC) | 37 |

CCC0276_07.011 | JOBNUM,ITIME,IOR,IAND,ICLOCK (ASSEMBLER) | 236 |

CCC0276_07.012 | CMVBT,ALOCAT,ITIME,IFTIME (F4,EBCDIC) | 38 |

CCC0276_07.013 | ERROR MESSAGES | 50 |

CCC0276_07.014 | OVERLAY CARDS | 27 |

CCC0276_07.015 | SAMPLE PROBLEM INPUT DATA | 167 |

CCC0276_07.016 | SAMPLE PROBLEM PRINTED OUTPUT | 3299 |

CCC0276_07.017 | JCL FOR RUNNING THE SAMPLE PROBLEM | 151 |

CCC0276_07.018 | 3-GROUP UPSCATTER PRINTED OUTPUT | 1202 |

Keywords: anisotropic scattering, discrete ordinate method, legendre polynomials, neutron transport theory, two-dimensional.