Annual monitoring and control of the effects of oxida­tion has produced information which is included in basic ‘ground rules’ covering the following topics:

• Oxide growth predictions at varying temperatures and gas compositions for a range of steels.

• The effects of coatings and surface treatments.

• Methods of assessing silicon content of materials, where this is unknown.

• Calculation methods for oxide growth in assemblies.

• Calculation methods for strain in assemblies.

• Calculation of bolt failure probability, thread disengagement and weld cracking.

Using the ground rules, calculations of the effects of oxidation on individual components are made, typical examples being:

TRANSITION STAGE Local breakdown of protective oxide film Number ano size of sites of excrescences grows. Duration of stage affected much as m protective stage Breakdown o* protective film accompanied by growth of underlying porous oxide which breaks through

PROTECTIVE STAGE

Oxide is a dense, even, dull gtey in colour. Very adherent, almost pure magnetite (FEjO.) Oxide growth roughly follows parabolic law, i e

w’ ■ к twhere n>2

Rate constant k) not closely related to gas pressure CO content, water content, or silicon content of steel Shot blasted surface has slightly higher rate than a poiisneo surface

Rale constant roughly increases with temperature Duration of this stage is much more sensitive to material and environment than rate constant.

Duration is inversely related to temperature, motslure. CO and surface roughness

Application of etch primer eg PA 21 can reduce length of protec live stage

BREAKAWAY STAGE

Further oxides grow through ano above protective

layer Excrescences which started at transition stage

are linked up over whole surface

Rough granular porous oxide. Cracks form m oxide ai

edges especially on lew silicon steel Films generally

vey aahereni until weight gams very high —

kOmgfcmr

Growth virtually linear

W = W,* kit-t. l

Rate constant increased Dy temperature, moisture, pressure and reduced by silicon content of steel Surface iinish has no effect on breakaway oxioatton rate.

Oxide contains about б^о by weight of caroon ie an approximate equi-molar mixture

Fig. 3.70 Stages of oxidation

• Bolted joints of charge pan structure or support systems within the reactor.

• Joints and linkages of core and charge pan support

structure.

• BCD and thermocouple supports throughout the reactor.

• Support structure of boiler units.

• Clearances of moving surfaces and assessing their effectiveness.

• Weld assessment.

For all these items, the material analysis and dimen­sional data are stored in a computer program. With the addition of the predicted oxide thickness, a strain and iailure probability is calculated for every compo­
nent. Annually, all the results are presented to the Nil in a document known as The Part 1 Oxidation Assess­ment which covers all reactors on each site.

This document is followed by the Part 2 Oxidation Assessment. In it, the practical consequences of the failure of any item shown to be suspect in the Part I Assessment are evaluated, i. e., the debris hazard, the loss of integrity of loss of movement are all considered as necessary (Figs 3.75 and 3.76). It is normally possi­ble to demonstrate that even where failure probabilities are quite high, there is either sufficient redundancy in the structure to cope, or the effects of failure are negligible.