227.

00-13

Nuclear Theory - Course 227

REACTIVITY CONTROL

Reactivity mechanisms represent the final control

elements which cause changes in the neutron multiplication
constant k (or reactivity 6k) hence, reactor power. There
are two separate requirements of the reactivity mechanisms
which are preferably fulfilled by two independent systems.
These requirements are:

1. Reactor Regulation. The three basis functions of the

reactor regulation systems are:

a) Maintain k = 1 for steady power operation.

b) Provide small changes +ve or -ve in 6k to change

reactor power.

c) Prevent the development of flux oscillations.

2. Reactor Protection. The principal purpose of the

prutective system is to rapidly insert a large amount
of negative reactivity to shutdown the reactor (TRIP).

From a reactor safety viewpoint it is desirable to

have reactor regulation and protection performed by
separate systems. From a practical viewpoint no single system
can adequately fulfill all the requirements for reactor
regulation let alone regulation and protection together.

Requirements of Reactivity Mechanisms

As well as independence between (1) and (2) the complex

physical and nuclear changes occurring in core during reactor
operation mean that an effective regulating system will have
to consist of more than one type of reactivity mechanism.
A convenient breakdown of the various in core reactivity
changes which require compensating/regulating controls is
listed in Table 1 and grouped in terms of the most important
parameters of any reactivity mechanism namely:

(i) reactivity worth (or depth) 6k (mk).

This must be somewhat larger than the reactivity change for

which the mechanism must compensate or control, and

(ii) operational time interval.

- 1 -

August 1980
 227.00-13

This is the time period during which the mechanism has to be

able to supply or remove reactivity and this will hence
determine the reactivity insertion rate (sometimes called
the ramp reactivity rate), .6k per unit time (mk/s).

Each of the tabulated reactivity changes is now briefly

described and typical ~k worths necessary to adequately
control these changes as they occur in our stations are shown
for comparison in Table 2. Where these values change from
fresh fuel to equilibrium fuel load conditions then the diff-
erence is noted.

In Core Reactivity Changes

(a) Power Changes (Ref. Lesson 227.00-12)

Because the temperatures of the fuel and coolant increase

as power increases from a hot shutdown condition to a hot full
power condition, reactivity changes. Under normal (ie, non
excursion) type conditions there will be a negative reactivity
worth change called the power coefficient of reactivity. These
are tabulated in Table 2. In order to maintain criticulity an
equal but opposite reactivity worth must be supplied by some
other means, (eg, by removing an equivalent reactivity worth
from the Zone Control System) .

(b) Fuel and Coolant Temperature Change~ (Ref. Lesson 227.00-12)

As the fuel and coolant are heated from a cold shutdown

condition (-25° C) to a hot shutdown condition (-276°C)
reactivity decreases, Table 2.

(c) Moderator Temperature Changes (Ref. Lesson 227.00-12)

Normally moderator temperature is kept fairly constant

(typically 70°C maximum in the calandria and 40°C at the heat
exchanger outlets) but variation could be obtained by changing
the rate of heat removal from the heat exchangers. The
accompanying reactivity change is usually negative with in-
creasing temperature for a freshly loaded core but changes to a
small positive value at equilibrium fuel burn up as shown in
Table 2.

(d) Fresh Fuel Burn Up (Ref. Lesson 227.00-7)

From an initial fresh fuel charge to equilibrium fuel

burn up there is a large increase in negative reactivity load
over a period of 6 - 7 months as a result of build up of long
lived neutron absorbing fission products (not including Xe 135 )
and depletion of fissile material. Figures for our reactors
are quoted in Table 2. This is a slow but continuous reactivity
change.

- 2 -
 227.00-13

(e) Equilibrium Fuel Burn Up

At equilibrium fuel burn up, when the operating target

excess reactivity has been reached, fission products continue
to be built up and fissile material continues to be depleted.
Continuous on power refuelling is of course the most import-
ant method of compensating for this continual depletion of
fissile material at equilibrium burn up. The rate of
reactivity loss for our reactors without refuelling is shown
in Table 2 and for comparison the reactivity increases due
to the refuelling of a single typical central channel are
also [Link].

(f) Equilibrium Xe Load Build Up (Ref. Lesson 227.00-11)

Following a long reactor shutdown (>2 - 3 days) an

equilibrium reactivitv load (up to 28 mk see Table 2) will
be built up due to Xe 135 accumulating in the fuel after
start up.

(g) Xe Transient Build Up (Ref. Lesson 227.00-11)

Within 12 hours of a reactor shutdown (or large derating

due to operational problems, or a load following situation)
there is a very large transient rise in Xe poison concentra-
tion (up to -80 mk above the equilibrium level at Pickering,
Table 1). To enable us to restart the unit, Xe OVERRIDE or
BOOSTING CAPABILITY is provided to compensate for this
reactivity loading providing an override time, measured after
shutdown, which gives reactivity capability of restarting
a unit within this time. Actual reactivities available and
the override times thus obtained are listed in Table 2 for
all our stations.

(h) Flux Oscillations (Ref. Lesson 227.00-11)

As localized flux/power changes occur in the core

(from, for example, refuelling part of a channel or movement
of a localized control rod) these can result in quite large
undamped power swings (Xenon oscillations) being set up
with periods between 15 - 30 hours.
To counterbalance these oscillating unbalanced
reactivity loads in various regions (called ZONES) of the
core, the ZONE CONTROL system is used. Total reactivity
worth of these systems are shown in Table 2, and are actually
larger than required to control only the flux oscillations as
these systems are also used for bulk power control.

- 3 -
 227.00-13

(i) Plutonium and Samarium Build Up (Ref. Lessons 227.00-7&11)

After shutdown plutonium builds up from the decay of

neptunium adding positive reactivity and samarium builds up
from the decay of Promethium adding negative reactivity.
The overall effect is positive as shown in Table 1.

TABLE 1

In core reactivity changes

Source of in-core reactivity lIk depth time

changes. interval

(a) Power changes, hut shutdown medium seconds,

to hot full power. (+ve, -ve) minutes

(b) Fuel and Coolant temperature medium seconds,

changes. (+ve, -ve) minutes

(c) Moderator temperature small minutes

change. (+ve, -ve)

(d) Fresh fuel burn up. large 6 - 7

(-ve) months

(e) Equilibrium Xe load build large 40

up. (-ve) hours

(f) Xe transient build up. large <12

(-ve) hours

(g) Flux Oscillations. medium 15 - 30

(+ve, -ve) hours

(h) Equilibrium fuel burn up. small days

(-ve) (continuous)

(i) Plutonium and Samarium medium 300

build up. (+ve) hours

- 4 -
 'rABLE 2: COMPARISON OF STATION REACTIVITY LOADS

REACTIVITY WORTH CHANGE NPD DOUGLAS PICKERING BRUCE

POINT A & B A & B
-
(a) Power Coefficient fresh fuel -3.3 mk -6 mk -7 mk -9 mk
- -- ~._-----_ ..- ---- .~

hot shutdown - equilibrium

hot full power fuel -1.2 mk -5 mk -3 mk -3.5 mk
0- ____._._ ___._ i--_...-.--.--- .._-..- - - . -
!-- . ____ o _ _ • •_ _ _ _ • _ _ _ _ _ _ _. . . .

-----
I (b) Fuel and Coolant fresh fuel -·3 mk .. 6 Ink -8 mk -9 mk
II
temperature 25°C
_equilibriwn
_-------- 1---------.---.----.------.--- f---------.-.
.... ~~,._---
- _._-.~

! -1 mk -4.5 mk. -2.5 mk -3 mk

to 275°C fuel
-----._--------- --
(c) Moderator Tempera tun.:: fresh fuel ~O.O8 mk/ C -0.06 mk/ C -0.06 mk/ C -0.07 mk/ C
Coefficient equilibrium
+0.01 mk/ C +0.03 mk/ C +0.08 mk/ C +0.09 mk/ C
fuel

(d) Fresh Fuel Burn Up -9 mk -20 mk -26 mk -22 mk

(e) Xe Equilibrium Load -24 mk -28 mk -28 mk -28 mk

(f) Xe Peak Load -46 mk -107 mk -98 mk -105 mk

Xe Override Capability* +2.4 mk +10 mk +18 mk +15 mk
Xe Override Time 35 min 30 min 45 min 40 min

( g) Zone Control Reactivity Worth NONE 3 mk 5.4 mk 6 mk

(h) Reactivity Loss (Equilibrium Fuel) -0.15 mk/day -0.3 mk/day -0.3 mk/day -0.5 mk/day

Reactivity Gain/Refuelled +0.1 mk +0.2 mk +0.2 mk +0.5 mk

Central Channel
o
(i) Plutonium and Samarium Build Up +2.5 mk +6 mk +6 mk +6 mk o
I
f-'
W
* New elements only, will decrease by -30% at end of life burn up.
tIl
 227.00-13

As you can see the range of reactivity depths and

insertion rates make it imnractical to try to design
a single control mechanism:

Methods of Reactivity Control

Before we can discuss actual control mechanisms we must

look at the theoretical methods of reactivity control. Re-
calling that:

we will examine which of the six factors we can use to change/

control reactivity (remember 6k = k - 1 )
k

First neither the fast fission factor (E) nor the

resonance escape probability (p) are easily varied. They
depend on the amount of U-238 present and the lattice
spacing in the reactor. Therefore, we make no attempt to
control reactivity by controlling E or p.

Next is the reproduction factor (n).

Recall that:
fuel
n = V _L.=.,f_ _
L afuel

f
ue
If we increase the amount of fissile material present
(I ) we will increase n. That is, more neutrons will be
f
produced per neutron absorbed by the fuel.

Thermal utilization ([J is the fraction of neutrons

absorbed by the fuel to those absorbed in the whole core:
fuel
L:
a
f = + L:non-fuel
a

If we increase or decrease the amount of non-fuel ab-

sorption, we vary f, hence reactivity. variation of neutron
absorption is by far the most common method of control.

Finally we have the fast and thermal non-leaka~e

probabilities (~f & At ). If we vary the leakage of neutrons
from the reactor we will vary reactivity.

- 6 -
 227.00-13

Reactivity Mechanisms

In order to discuss the reactivity mechanisms presently

in use we shall divide them into five groups based on their
basic function in the reactor. The five functional groups are:

1) Automatic Reactor Regulation - (includes bulk

power and zone control)

2) Xenon Override.

3) Long Term Reactivity Control - (includes fresh fuel

t:urn lJ.P r the build up of equilibrium xenon and the
build up of plutonium and samarium after shutdown).

4) Equili b rilun Fuel Burn up.

5) Shutdown Systems

For each of these catagories we will discuss the methods

used and the significant advantages and disadvantages of those
methods (See 433.50-1 for a discussion of the mechanics of
the systems). Table 3 indicates which systems are used at
each station and the reactivity depth of each system.

Automatic Reactor Regulation

a) Moderator Level Control.

Small changes in moderator level change the thick-
ness of the reflector on top of the reactor thus
varying leakage (A & At ).
f
Advantages:

1) Easily incorporated into a system using moderator

dump for protection.

Disadvantages:

1) Zone control is not possible.

2) Lowering the moderator level distorts the over-

all flux distribution.

b) Control Absorbers.
Solid rods of a mildly absorbing material (typically
stainless steel) which can be operated vertically
in the core. Because they are parasitic absorbers
the control absorbers change the thermal ulitization
(f) •

- 7 -
 227.00-13

Advantages:

1) Provide additional reactivity at minimal cost.

Disadvantages:

1) In core guide tubes represent, permanent,

reactivity loss (fuel burn up loss).

c) Liquid ~one Control (LZC)

Zone Control Compartments inside reactor which contain a
variable amount of light water (a mild neutron
absorber). Varying the amount of light water in
the LZC, varies parasitic absorption hence thermal
utilization (f).

Advantages:

1) Individual zone levels can be independently

varried for zone control.

2) Operating equipment is mainly outside contain-

ment, therefore, accessible during reactor
operation.

3) Cooling easily accomplished.

4) Only slight distortion of the overall flux

pattern.

Disadvantages:

1) Requires special design to insure that the

zones fail safe (ie, fill).

2) In core structure represents a reactivity (or

fuel burn up) loss.

Xenon Override

a) Booster Rods.
Solid rods of highly enriched (-90%) U-235. In-
sertion of booster rods increases the amount of
fissile material in the reactor hence the repro-
duction factor (n). It also increases f.

Advantages:

1) Can provide large override capability.

- 8 -
 227.00-13

Disadvantages:

1) Enriched Uranium is a very expensive, non-Canadian

product.

2) Require highly reliable source of cooling (loss

of cooling to an inserted booster at high
power could cause the rod to melt down in
about 5 seconds).

3) Because of cooling requirements additional trips

are required thus complicating the reactor pro-
tection systems.

4) Limited lifetime as the reactivity worth decreas-

es with each use.

5) A criticality hazard exists in the storage of

both new and irradiated booster rods.

6) Because of all of the above reasons, the AECB

requires special licenses, which, at this writ-
ing (June 1979) BNGS A does not have.

b) Adjuster Rods

Solid rods of a neutron absorbing material (Cobalt

or Stainless Steel). Normally fully inserted in
the reactor thus increasing parasitic absorption
(decreasing f). Positive reactivity is provided
by withdrawing the adjuster rods.

Advantages:

1) Provide flux flattening which must be provided

by some other method if booster rods are used for
xenon override.

2) No significant decrease in reactivity worth

over normal lifetime.

Disadvantages:

1) Presence of adjusters results in a fuel burnup

penal ty of _ 8 %• (The adj usters reduce f, there-
fore, we must increase one of the other factors.
Thus n is increased by not allowing the fuel to
burn out as much.

- 9 -
 227.00-13

Long Term Reactivity Control

The method of long term reactivity control presently in

use is the addition of soluble poison to the moderator.
While solid rods could be used for this purpose, soluble
poison systems are cheaper and cause no flux distortions.
However, the addition of poison to the moderator does reduce
the flux reaching the ion chambers sufficiently to require
that the power reading from out of core ion chambers be
corrected for the presence of the poison. Boron in the form
of boric acid D3B03 or gadolinium in the form of gadolinium
nitrate Gd(N03) 3 ·6HzO are the poisons presently in use.
Gadolinium has the advantage over boron for Xe load simulation
because the neutron burn up rate of the neutron absorbing
gadolinium isotopes (Gd155 and Gd157) and the Xe build up are
sufficiently complementary that little adjustment of the
gadolinium concentration by IX control is necessary during
start up. The IX columns are, however, used to remove the
reactivity build up of low cross section gadolinium absorp-
tion products to limit their accumulation in the moderator.

Using boron to simulate Xe load needs a closely monitored

operation of the cleanup circuit to obtain the rapid reduct-
ion of boron required (3.5 ppm = 28 mk), boron removal being
essentially only dependent on the IX removal rate rather
than neutron burn up rate. Much more IX column capacity is
also needed for B removal than for the Gd system. Gadolin-
ium is not used at Pickering A as there is some concern that
it may lead to high deuterium gas levels in the cover gas
system due to increased radiolysis of the moderator.

Equilibrium Fuel Burn Up

On power refueling is used in all of Ontario Hydro's

reactors. This essentially keeps the amount of fissile
material constant by replacing irradiated fuel with fresh
fuel more or less continually. This system of refueling
has several distinct advantages:

1) No downtime for refueling

2) Better average fuel burnup

3) Better flux shaping.

4) Failed fuel can be removed easily without a shutdown.

- 10 -
 227.00-13

There are of course some disadvantages mainly the high

capital cost of the fueling machines and the maintenance
which is required for them.

If the fueling machines are unavailable for some reason,

there is a limited time the reactor can continue to operate.
A Bruce 'A' reactor normally consumes 0.5 mk/day. (That
is the reactivity worth of the fuel diminishes at that rate) .If
the LZC were at 50% at full power in an equilibrium fuel
condition, about 3 mk of excess positive reactivity would be
available. That gives approximately 6 days of operation
before we must start inserting the boosters (actually
undesirable for this purpose) or reduce the operating power
(called derating) or shutdown the reactor.

On the other hand, if you overfuel the reactor you may

have to derate the reactor due to the high flux in the area
of the new fuel, (called regional overpower).

Shutdown Systems

Early Candu designs had a single shutdown system. As

the design of the reactor became more sophisticated, the
requirement for extremely high reliability dictated that
two independent shutdown system be provided. There are
presently three types of shutdown systems in use.

1) Moderator Dump

As the moderator level decreases, the physical size of

the active portion of the core decreases. As the core gets
smaller, leakage increases (A and At go down).
f
Advantages:

1) Simple, fail safe with gravity system.

2) Absolute shutdown, with the moderator dumped the

core cannot be made critical.

[Link]:

1) Slow for a large reactor. The initial reactivity

insertion rate may not be adequate to protect the
reactor from certain types of accidents. Figure 1
shows reactivity vs time for moderator dump at PNGSA.
Note that in the first two seconds only -2mk of
reactivity has been inserted.

2) Time required to pump the moderator back into the

calandria is so long (-50 min. at PNGSA) in a larger
reactor that a poison out is quite possible.

- 11 -
 227.00-13

REACTIVITY VERSUS TIME FROM INITIAL MODERATOR DUMP SIGNAL

(100 PERCENT MODERATOR LEVEL, EQUILIBRIUM FUEL CONDITIONSl
H.01l00-H

FIGURE 1

12
 227.00-13

2) Shutoff Rods

Hollow cylinders of neutron absorbing material (normally

stainless steel sheathed cadmium) which can be gravity dropped
into the reactor. Their presence greatly increases parasitic
absorption thus reducing the thermal utilization (f).

Advantages:

1) Rapid reactivity insertion as required for protection

in certain worst case accidents. Figure 2 shows
reactivity vs time for PNGSA shutoff rods. Note that
in 2 secondS-the rods have inserted -22 mk.

2) Rapid recovery from a trip is possible. (-3 minutes

to withdraw the rods).

Disadvantages:

1) Limited reactivity depth. As presently designed

shutoff rods do not provide enough reactivity for a
guaranteed long term shutdown.

2) Complex system (relative to dump) subject to mechani-

cal failure. Safety analysis normally assumes that
the two most reactive rods don't drop on a trip.

3) Poison Injection

Poison (Gadolinium) is injected into the moderator under

high pressure. This causes a large reduction in the thermal
utilization (f),

Advantages:

1) Rapid insertion of reactivity. Figure 3 shows reac-

tivity vs time for BNGSA poison injection system.
Note that -33 mk is inserted in 1.5 seconds. Total
worth is approximately -675 mk.
Disadva~tages:

1) Poison must be removed from the moderator by ion

exchange which is costly and slow (-12 hours).
If poison injection shuts down the reactor, a Xenon
poison out will occur before the moderator poison
can be removed.

- 13 -
 227.00-13

• -\+

TIME FHOM Th!1' SI(;"f< i

MOOERATOH LE vEL,
4401100-

14
 35 1
55 MK AT 2 .9 SECONDS

I ;::::---
~
30
---
V
//
25 /'

V
~
H
GJ
c::
~
::;<
E
>-
20

:/
/
~
~
........
~ 6TANKS-
----- ~
--r-J
~

/
w t:
>
t-
o
«
UJ 15
a:
/ TIME MEASURED FROM INSTANT WHEN
HELIUM VALVES START TO OPEN

10
/ I
1200 PSI DRIVING PRESSURE
I
RESULTS CALCULATED BY COMPUTER
I
I

/ CD SIMULATION OF HYDRAULIC SYSTEM

5
//

/
V I

I o
o
o I
I
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 I-'
LV
TIME IN SECONDS

Reactivity Vs Time, For Poison Injection.

 N
NPD Douglas Pt. Plckering A Bruce A Pickering B Bruce B N
-...J

o
Automatic Reactor Primary Moderator Level 4 Control 14 Liquid 14 Liquid 14 Liquid 14 Liquid o
i
Regulation (1) Control Absorbers Control Zones Control Zone~ Control Zones Control Zones i-'
(3mk) (S.4mk) (6mk) (6mk) (6mk)
~-------I-------------+-------+---------t----------I--------+-----.---

Secondary NONE Moderator Moderator 4 Control 4 Control 4 Control

Level Level Absorbers Absorbers Absorbers
Control Control (7mk) (lOmk) ([Link])

Xenon Override 1 Booster Rod 8 Booster 18 Adjuster 16 Booster 21 Adjuster 24 Adjuster

(2.4mk) Rods Rods Rods Rods Rods
(lOmk) (18mk) (lSmk) (18mk) (l8mk)

Long Term Reactivity (

Control Moderator Level Moderator Poison Addition (Variable reactivity depending on poison I
_. . l-_c_o_n_t_r_o_l L--. . ~=~~~a tion) - ----1
1

Equilibrium Fuel All Stations use on power refueling

Burn up

Shutdown Systems SDS 1

( 1) The primary stem is nGrma~,ly used for reactor regulatjoll~ If the primary system is ~na't,,'ailab.l or
has insufficient reactivit~7 iepth; the secondary systerD will act automatically.

(2) Operation of the dump system at Pickering A is not entirely independent of the shutoff rods.

TABLE 3

Reactivity Control Systems

 227.00-13

ASSIGNMENT

1. A Bruce reactor trips inserting - 40mk due to the shutoff

rods. Using the information in Table 2 and assuming an
equilibrium fuel condition, would you expect the reactor
to remain shutdown (subcritical) if the heat transport
system was kept at normal operating temperature? If
the heat transport system was cooled down? Justify
your answers.

2. Both methods of Xenon Override require derating when

used. Explain why.

3. Simple chemical analysis for boron or gadolinium is not

considered sufficient to determine the reactivity worth
of moderator poison, explain why.

J.E. Crist
A. Broughton

- 17 -