A flow disturbance can result from either a loss of power to the pump motors causing the pump to coast down or a mechanical failure of the pump. The former coast-downs can occur in one or more pumps simul­taneously depending only on how the pump motor electrical supply lines are interconnected. However, a mechanical failure of a pump is expected to occur in only one pump at any time since it is a very unlikely fault.

The loss of electrical supply may occur in two or three primary pumps or in one primary and a secondary pump, just what happens to the core flow depends on a detailed flow balance in the system. Flow conservation equations are used together with assumed pump characteristics to derive core flow as a function of time t. Such flow coast-down curves take the form:

relative flow = 1 — [t/(t + ?j)] (2.3)

In a typical pump failure in an LMFBR due to a loss of electrical power the time constant is of the order of 2 sec, while a pump seizure would be somewhat faster with a tx of the order of 1 sec.

The core flow due to a mechanical failure of a pump in one loop of a two-loop plant might typically take the form:

relative flow = a + b tan(c — dt) (2.4)

where a, b, c, and d are constants.

This core flow behavior is input to a calculational model to predict changes in coolant and fuel temperatures with and without a reactor trip in the event of either of these flow perturbations. Figure 2.4a shows a typical flow rundown due to a loss of all electrical power to the pumps of a LMFBR, while Fig. 2.4b shows the resulting core temperatures.

It will be noted in Fig. 2.4b that the initial temperature rise is cut back as the reactor is scrammed, but it rises again as the flow drops rapidly to 5% or less than the decay power level. However the temperatures again decrease as the power decays still further and becomes less than the flow level relative value.

Notice that the monitored temperatures are the coolant and cladding temperatures as the former first changes due to the flow decrease and the failure of fuel pins would be the result of excessive cladding tem­peratures.

2.2.2.1 Trips

In the event of a flow perturbation the following trip signals would be available: (a) primary cause signals, electrical supply board power loss, or pump speed indication; (b) flow (say 85%) and power-to-flow rqtio; and (c) outlet temperature. These trips are in order of occurrence. Figure 2.4b assumes a flow trip.

When the fuel center temperature attains 6500°F the vapor pressure starts to rise rapidly from 10 to 100 atm at 7250°F. Under these conditions the high pressures could force molten fuel through cracks into contact with cladding and cause cladding failure.

On the other hand, if there is a large amount of fuel melting the excessive fission gas release will cause excessive pressures on the cladding. The amount of molten fuel judged to be excessive depends, of course, on the condition of the cladding and the burn-up of the fuel.

Thus at start-, middle — and end-of-life, a fuel temperature of 6500°F may be judged to portend failure. However, 60 mil of fuel melting (25% areal extent) represents failure at the start-of-life while 20 mil may represent a similar failure at the end-of-life due to the weakening of the cladding and the rise of fission gas pressures.

The position of the failure depends on the pattern of fuel melting (5).

The modeling and the analysis is borne out by experimental results of failures and by radiographs of fuel structure and cladding cracking at various burn-up levels. This latter information is unfortunately only obtained when the fuel has cooled off, of course.

If a bubble of sodium vapor is surrounded by sodium liquid at the same pressure, thus providing a condensing medium, the bubble will then collapse. Initially the rate of collapse is slow and the rate of condensation keeps pace with it; however as the collapse rate increases the rate of condensation de­creases as the heat transfer area is also decreasing. Thus the rate of con­densation no longer matches the rate of collapse, so the internal bubble pressure will increase very rapidly indeed. The bubble then grows once more and collapses again and finally is fully condensed after several collapse and rebound cycles (6). The whole collapse and rebound cycle can occur in 10 msec and peak pressures of up to 8000 atm have been calculated for a 10-cm bubble. Such collapse pressures have been observed experimentally, although the energy possible in the pressure pulse is very limited. In the case cited it is about 4000 J. However this effect has not been observed in any reactor installation, although the phenomenon should be accounted for in accident analysis.

The vessel plug had been designed to take several positions depending on a servosystem related to temperature and pressure measurements. This proved far too complex and moved when it was not required to do so. It was altered to a fixed position for simplicity.

The high-speed argon blowers were changed for conventional low-pressure

blowers with oil bearings. The more advanced components had previously been sticking.

Some of the previously inerted vaults below the operating floor were changed to air for convenience, although nitrogen was still retained within the reactor vault and around the primary system.

The air conditioning system within the containment began giving rise to a water collection problem that had originally not been recognized as a possibility. This problem was solved before start-up with the sodium — cooled reactor.

In order to standardize siting requirements for the purposes of long term 1000 MWe LMFBR reactor assessment by the industry, the AEC selected a hypothetical site (Middletown, USA) on which to place these reactors. The properties of this site give a good idea of the characteristics of any site with which a plant designer might have to contend. Many of these character­istics are safety-oriented, and others have safety overtones. The description is reprinted from the literature (14a). The monetary values are as of 1961 and therefore, should oney be regarded as indicating trends.

General

The Atomic Energy Commission has established Ground Rules for use in the prepara­tion of design studies, cost normalization studies and other types of studies relating to the economic factors associated with power generation. They relate to site conditions, certain design guide data and a basis for estimating fixed charges. In the absence of specific instructions, these rules, or the applicable portions thereof, shall be used when like factual data for a specific proposal are not available. The utilization of the Ground Rules together with the AEC Classification of Construction Accounts (Section 105) will ensure to the maximum extent possible that for study and proposal evaluation purposes:

(1) The site conditions for the plant, which is designed on the basis of the Hypothetical Site, will be as uniform as possible.

(2) The estimate is prepared on a uniform Classification of Construction Accounts.

(3) The estimated indirect costs are applied to the estimated direct costs on a uniform basis.

(4) The computation of production costs is on a uniform basis.

Hypothetical Site Conditions

When location or site conditions are not specified, a selected Hypothetical Site shall be used and the plant designs and costs shall be based on the Hypothetical Site condi­tions described herein. The site layout and plant arrangement shall be essentially as shown in Fig. 5.4. This layout is typical for a 300 MWe nuclear power plant. Adjustments should be made as required for the proposed nuclear reactor concept.

The ultimate objective of the safety engineer is to be able to show that the system which he has nurtured is indeed safe, and to be able to show this with such conviction as to obtain licenses to build and to operate the plant. To this end he provides a safety evaluation of the plant as a report in support of the license application.

Before 1800 (Table 1.1), all of the world’s power needs were satisfied by wood, wind, and water, all of which were energy sources that could be renewed.

Simultaneously with the growth of industry, indeed as an initiator for much of the growth, energy from coal deposits began to supply an ever — increasing proportion of the world’s requirements at the end of the 19th

century. This growth in coal use continued until it supplied 90% of the world’s requirements by the beginning of this century.

However, the growth of the population in recent times has a doubling time of about 40 years, and the energy demand rises even faster with a doubling time of about 25 years. Thus the supply of energy by coal could not keep pace with the demand.

A partial solution was found around 1930 when the problem of the transportation of naturally occurring oil and gas was solved, so that these two new sources of energy began to supply an ever increasing proportion of the amount required. In particular, the growth of modern transport absorbed a large proportion of the energy produced from oil. Thus by 1970 coal was supplying about a fifth of the demand with coal and oil each sup­plying in the neighborhood of one-third to two-fifths. The amount of energy supplied by wood, wind, and water had undoubtedly increased with the advent of new hydroelectric plants but nevertheless it supplied a very small fraction of the total (Fig. 1.1).

Unfortunately coal, gas, and oil are all exhaustible sources of energy. With the power demand rising on an increasing scale it behooves us to look to the future. The United States has the highest energy demand in

the world, but the growth of that demand is relatively small compared to that of other nations. Figure 1.2 shows that the growth rate is larger for countries with smaller gross national products, as one might expect. This means that as the world becomes more and more affluent, the power re­quirements will rise to extremely high values.

Nevertheless our use of coal, oil and gas has been so rapid already that coal is likely to be exhausted in the world by about the year 2200 (2) with gas and oil not far behind. Thus the energy demand will not be satiable.

Certain solutions open to us are: the curtailment of population growth; the curtailment of energy demand per head of population; and a new source of energy which is renewable.

As I have said the final answer will be given by society, but I would suggest that the first two courses of action will only partially answer the problem. A new energy source will have to be made available.

Solar energy, energy derived from the fusion of light elements, tidal energy, and nuclear energy derived from fission are all candidates. Indeed Table 1.1 shows that nuclear energy is already supplying a significant pro­portion of the demand.

Research in the conversion of solar and fusion energy has not yet shown that these will ever be adequate sources of power for an industrial com-

0.50

Fig. 1.2. Per capita growth of energy usage as a function of the gross national product per capita (7). Legend: — 1955-1960; 1960-1965.

munity, while tidal power is limited in locality and availability. Nuclear power alone seems to provide the answer and by the year 2000 it is expected to provide nearly a fifth of the total needs.

Moreover, fast reactors are essentially a renewable form of energy. So while thermal reactor systems are undoubtedly short term solutions, it is only the fast breeder which will supply the long term needs in an inex­haustible form. We are thus led to the conclusion that fast reactor power plants will necessarily be built.

Using the coefficients of reactivity we have the final equation for the system representation, the feedback equation

dk = <5fc0 + £ aiiXi — Tia) + «v bV + aBS (1.59)

І

were temperature, voiding and burn-up feedbacks are included.

This equation, together with the kinetics equations and the thermal model equations, forms the complete representation of the reactor. The inter­dependence of these equations can be seen from closed-loop model diagrams.

These are treated in exactly the same way as primary pump coast-downs or seizures were, and the resulting flow changes are input to the model, which includes the secondary circuit details.

The resulting transients are, of course, much less serious than primary pump failures because of the attenuation by the secondary and cold leg primary sodium. They are more of interest in the operational sense than in the safety sense.

2.4.2 Tertiary Loop Failures

The main criteria are those which relate to the release of radioactivity to the environment, 10 CFR 20 and 10 CFR 100, which refer to sections of the Federal Register. They are detailed in Chapter 5.

The 1967 General Criterion 70 states:

Control of Releases of Radioactivity to the Environment. The facility design shall include those means necessary to maintain control over the plant radioactive effluents whether gaseous, liquid or solid. Appropriate holdup capacity shall be provided for retention of gaseous, liquid or solid effluents, particularly where unfavorable environ­mental conditions can be expected to require operational limitations upon the release of radioactive effluents to the environment. In all cases, the design for radioactivity control shall be justified (a) on the basis of 10 CFR 20 requirements for normal operations and for any transient situation that might reasonably be anticipated to occur, and (b) on the basis of 10 CFR 100 dosage level guidelines for potential reactor accidents of exceedingly low probability of occurrence.

This has been replaced by the 1971 General Criterion 60:

The nuclear power unit design shall include means to maintain suitable control over radioactive materials in gaseous and liquid effluents and in solid wastes produced during normal reactor operation, including anticipated operational occurrences. Sufficient holdup capacity shall be provided for retention of gaseous and liquid effluents containing radio­active materials, particularly where unfavorable site environmental conditions can be expected to impose unusual operational limitations upon their release to the environment.

Regulations 10 CFR 20 and 10 CFR 100 still apply by law although they have been omitted from the wording of the latest criterion. They are the minimum requirements for “suitable control.”

Other principal criteria deal with redundancy in protective systems and emergency safety features and these are dealt with specifically in Section 3.3. These criteria too are subject to debate.