Part 21 Report - 1996-210

 
 

ACCESSION #: 9601300209



NUMERICAL APPLICATIONS, INC.



Engineering Software

Development and Applications



825 GOETHALS DRIVE, SUITE A o RICHLAND, WA 99352 o

TELEPHONE (509) 943-0861 TELECOPY (509) 943-6617



                                             January 16, 1996



Director, Office of Inspection and Enforcement

U.S. Nuclear Regulatory Commission

Washington, D.C. 20555



Dear Sir:



This letter is to inform you of a reportable defect under 10CFR Part 21. 

The defect is in the early versions of the GOTHIC and FATHOMS computer

programs for containment and general thermal-hydraulic analysis and is

described in the attached error notification being sent to all users of

record with the affected versions.



The notification applies to all organizations with current or future use

of GOTHIC 3.1 or earlier and all versions of FATHOMS.



The defect was confirmed on January 15, 1996 and all affected users are

being notified.



Version 3.2 of GOTHIC (July 1990) eliminated the identified problem. 

Earlier code versions with the identified problem are not frequently

used.  Modifications are available from NAI to correct the problem.



If there are any questions, please contact me at the number shown above.



Sincerely,



Thomas L. George





                                                       January 15, 1996



                GOTHIC ERROR NOTIFICATION - GOTHIC.3.1.34



To: GOTHIC Users



An error has been discovered in GOTHIC_S 3.1 and earlier versions, and

all versions of the precursor code FATHOMS, that may effect the accuracy

of some calculations.  The error does not exist in version 3.2 or later

of GOTHIC.



The error may result in an overprediction of the vaporization of water in

superheated environments.  GOTHIC 3.1 and earlier versions, including

FATHOMS, use a single energy equation for the combined liquid and drop

phases and correspondingly a single source term for the total

vaporization rate of drops and liquid.  In certain analyses, e.g., a

steam line break, there should be a minimal amount of drops in the

atmosphere.  However, the code selected initial value for the drop

surface area and a small drop area source term associated with the break

in these early code versions, results in a substantial drop surface area

during the blow-down, even though the drop mass is negligable.  Because

the drop and liquid vaporization are combined, the large drop area

combined with the significant liquid from condensation on heat sinks,

results in a high estimate for the vaporization rate.  This vaporization

cools the atmosphere and results in lower peak containment temperatures

and pressures.  This effect does not cause a problem in steam/water

blowdowns because the significant drop injection quickly leads to

realistic values for the drop surface area and the total rate of phase

change.



Any safety analyses performed with these early code versions should be

reviewed with regard to the identified error.



Necessary corrections may be obtained from NAI.



NAI considers this error to be a reportable defect under 10CFR Part 21

and an error notification has been sent to the NRC.





*** END OF DOCUMENT ***