Digital determination of TR-1 control rod worths and time behaviour of neutron flux

T.A.E.C.

CEKMECE NUCLEAR RESEARCH AND TRAINING CENTER

I

stanbul

-

turkey

gNAEM - R - 164

D I G I T A L D E T E R M I N A T I O N O F

T R -I C O N T R O L R O D W O R T H S A N D

T I M E B E H A V I O U R O F N E U T R O N F L U X

by

g . E r t e k

P.K. 1, Hava Alam, Istanbul, Turkey

1976

Digital determination of TR-I control rod worths and time behaviour of neutron flux

By Qetin Eriek

TAFC, Cekmeco ISuoso Rehear :h and Training Centre,. Istanbul, Turkey

Abstract

In this work, the control rod reactivity worth of the swimming­ pool type reactor (TR-I) in £NAEM, £ekmece Nuclear Re­ search and Training Centre has been measured dig itally and the mean neufron lifetime has been estimated by a special

miniature fission chamber with 60 nanosecond resolving time. The time behaviour of thermal neutrons is also compared with reactor control console results.

Zusammenfassung

D igitale Bestimmung des Reaktivitätsäquivalents des TR-I-Sleuersfabes und Zeitverhalten des Neutronenflusses

Das Reaktivitätsäquivalent des Steuerstabes des TR-l-Swimming-pool-Reak- tors im £N A E M , Cekmece Kernforschungszentrum, wurde d ig ita l gemessen. Die m ittlere Neutronenlebensdauer wurde m it einer spezieller. M in ia tu r­ spaltkammer mit 60 ns Zeifaufiösung abgeschätzt. Das Zeitverhalten ther­ mischer Neutronen w ird m it Meßergebnissen am R eaktorkontrollpult v e r­ glichen.

INIS DESCRIPTORS

REACTIVITY INSERTIONS CONTROL ROD WORTHS TR-1 REACTOR

THERMAL NEUTRONS

1. Introduction

Reactivity measurements by the aid of specially developed fast miniature fission chambers have been very successful. In this work an investigation has been made to find the d iffer­ ence (if there is any) of the control rod worths measured at different power levels in the TR-I reactor and the mean neutron lifetime is estimated. In Section 2, we introduce the control rod configuration and the value of control rod No. 1 reactivity worth which is determined digitally. In Section 3, this result is compared with reactor control console results, mainly big fission chamber and lo g N ionization chamber results. In Section 4, we discuss the gamma-ray compensation procedure. Finally, in Section 5, we have estimated the mean neutron lifetime fo r this system.

2. Digital reactivity worth of control rod

The control rod No. 1 of the swimming-pool type reactor (Fig. 1) was initia lly 81.68% out from the reactor core and then immediately inserted into the core with scram at the power level of 1 kW.

The miniature fission chamber was made at Grenoble, France, and has 60 ns resolving time. It was installed 30 cm from the bottom of the TR-I core in the middle of the MTR fuel elements. A 400-channel analyser has been used in multi-scaler mode. The time between channels can be adjusted from 10 \i$ up to orders o f magnitude of a second. W e have obtained results fo r 2, 10, 40, 200, 400 ms time intervals between channels. The total time fo r each data analysis were respectively 0.8; 4; 16; 80; 160 seconds. Figs. 2 and 3 show linear and logarithmic drop curves, respectively.

One of the main aims of the experiment was to check whether the control rod worths determined by conventional rod-drop or doubling-time measurements agree with the present digital experimental technique at low and high reactor power levels. The reason why we have chosen the control rod position 81.68% (out from the reactor core) was simply because this

TIME DEPENDENCE LIFETIME

CALIBRATION FISSION CHAMBERS

©

)

(T)

Fig. 1: Core configuration (No. 92) o f the TR-I MOO 3000 1300

s

Fig. 2: M iniatu re fission chamber result (linear representation) 400 ms per channel, control rod No. 1 scrammed from 81.68 Vo, reactor power 1 kW , 30 cm from the bottom o f the core

position was calibrated before at a power level of 10 W att and estimated to have a reactivity value of — 1 % by d iffe r­ ential doubling-time measurements. This reactivity value by the presented digital technique is also found to be — 1 %. The statistical uncertainty in counting rates was changing in the range of 1 to 6% . After obtaining d ig itally the drop of fission rates follow ing the scram of the No. 1 control rod a well-known treatment has been applied to find the reactivity value [1].

3. Reactor console results

The digital technique using the miniature fission chamber at 1 kW power level described in Section 2, gave the same reactivity value as found by the reactor console instrumen tation at 10 W att power level. Since the information is registered by reactor console instruments (mainly fission chamber and log N compensated ionization chamber channels)

F:g. 3: M iniature fission chamber result (logarithmic representation} 400 ms per channel, control rod No. 1 scrammed from 81.68%, reactor power 1 kW, 30 cm from the bottom of the core

during the scram of control rod No. 1 we have also measured the reactivity worth with this two channels. The big fission chamber belonging to TR-S reactor control console gave the reactivity worth for control rod N o.l (scrammed from 81.68% at 1 kW working power) to be — 0.5 %.

The compensated ionization chamber result (the log N channel) also gave — 0,5% for the - 1 °/o reactivity worth. There are two main reasons to explain this large difference:

a) The distance ct high power is large between the detector and the recctcr. Because of the shielding effect of water the reactivity effect seen by the detector is less than its actual value. The reference to this effect can be given by seed and blanket critical reactor experiments [2]. It is stated by S. H. Levine et al. in Ref. [2] p. 12 that “ the shielding effect of the water was so great that the counter showed a decrease in multiplication rate when, in actuality, the rate was in­ creasing” .

b) The second reason, perhaps the more dominant one is the effect of gamma-ray contribution to the detector counting rate at high reactor power levels.

4. Gamma ray compensation effects

It 3s an experimental fact that the big fission chamber belonging to the reactor control console was more sensitive to gamma rays than the miniature fission chamber. This is because of the larger volume of the former. The larger the volume of the fission chamber, the larger the probability of interaction with the detector walls hence the gamma-ray measuring sensitivity of the system increases. Since there is approximately 7*107 R/h gamma-ray radiation level in the system all the neutron measuring devices are sensitive to gamma-ray energy and gamma-ray field intensity.

The miniature fission chamber must not be considered completely insensitive to gamma rays or gamma-ray build-up effects, in fact, we tried to measure the reactivity value of the seme control rod worth at lOkW reactor power and have found such a large contribution from gamma rays that it was impossible to measure the reactivity worth. At higher power levels the detector was almost insensitive to neutrons due to the high gamma-ray build-up.

We have also measured control rod No. 1 worth at the 1 kW power level after the rod drop operation using the compen­ sated ionization chamber of reactor control console. The

reactivity value is found to be - 0.5% which is haif less than its actual — 1 % value which is well calibrated with different techniques at low-power levels (10 W). The explanation of this large, expected difference lies in the compensation proc­ ess itself. The ionization chamber compensation is done at low-power levels such as 10 W. The reversed bias voltage applied to ihe ionization chamber can not be changed after the adjustments at the 10W power level. At the higher power levels such as 1 kW, 10 kW and 1 MW the compensation becomes ineffective. O f course, the large ionization chambers are more sensitive to gamma rays and gamma-ray build-up effects than small ones. So, the negative reactivity value found by the reactor control console instruments at high level of power was less than its actual value. In the positive reactivity insertion case this causes the underestimation of reactivity worth which is important from the reactor safety point of view. But this type of reactivity measurements is not usually done at high power.

5. Estimation of the mean neutron lifetime

A preliminary estimation of mean neutron lifetime has been done finding the well-known semi-logarithmic drop curve inclination determination1 (by means of the digital values of Fig. 3) and is found to be

7

Since the TR-I reactor is long-lived, flux-hardened and over­ poisoned, many changes of mean neutron lifetime can be expected. This point may be investigated further at different burn-up intervals.

6. Conclusions

The following conclusions could be derived from this study. - The control rod reactivity worths are usually calibrated at

low-power levels (1 to 10 W) and the calibration values should not depend on the reactor power levels But if we try to measure the reactivity value of some change from Ihe reactor console at high-power level e. g. at 1 kW or 10 kW levels, it is found that this change is underestimated. - Neutron measuring devices are highly sensitive to gamma

rays and gamma-ray build-up effects and attention must be paid to the instant and unknown reactivity effects seen by the control console at especially high-power levels.

- Miniature fission chambers are found to be also sensitive to gamma rays and gamma-ray build-up effects for the reactor power levels of more than 10 kW. In this work it is also found that if we try to measure the reactivity value of a control rod at 1 kW power level and if we apply this power by lowering the reactor power from 1 MW to 1 kW, even the miniature fission chamber gives misleading results due to the high gamma radiation field and gamma build-up. - The digital method gave very good agreement with differ­ ential doubling-time measurements which were performed only Gt 10 W low-power level and the former method was giving reliable results even up to 1 kW power level. - Large contribution from gamma rays are experimentally

verified even for the so-called “ compensated” ionization chambers. (Received on 12. 8.1975) References

[1] Sturm, W. J.: ANL, Reactor Lab. Experiments. ANL 6410, p. 46 [2] Levine, S. H. et a l.: Critical experiments with a seed and blanked-s!ab

geometry assembly, Bettis Plant, WAPD TM 130. Contract A T-ll-l-G EN-14 [3] Report on the first start-up experiments of TR I. £NAEM (March 1961] [4] Operation and maintenance procedures of Turkey’s pool-type research

reactor. QNAEM (Jan. 1961) 1 — 1/r * r inclination