Available online at www.sciencedirect.com

Nuclear Energy and Technology 2 (2016) 251–255 www.elsevier.com/locate/nucet

In-service change in the flexural rigidity of the VVER-1000 fuel assemblies

S.V. Pavlov

Sosny R&D Company, 4a Dimitrova Ave., Dimitrovgrad, Ulyanovsk Reg. 433507, Russia

Available online 1 December 2016

Abstract

In-service dimensional stability of the VVER-1000 reactor fuel assemblies (FA) depends to a great extent on their flexural rigidity. A

decrease in the flexural rigidity in the process of the FA operation in the reactor may lead to such FA bowing as will make it difficult for the absorber rods of the reactor control and protection system to move in the FA guide channels. This is not admissible from the point of view of the reactor operation safety.

This paper describes a method and a facility for the hot cell testing of the irradiated VVER-1000 FA flexural rigidity. The method is based on measurements of the FA bowing induced by cross-sectional loading. The load applied to the spacer grids is perpendicular to the grid face, and the FA bowing is measured optically using a TV camera. The facility can also be used to test the flexural rigidity of the FA

skeleton after all of the fuel rods are removed. Several tens of VVER-1000 FAs with a burnup of ∼4 to ∼65 МW day/kg U were tested by Dimitrovgrad Research Institute of Atomic Reactors. The generalization and an analysis of the test results have made it possible to identify the major factors that contribute to the in-service change in the flexural rigidity of the VVER-1000 FAs and to determine the experimental dependence of its change on burnup.

It has been shown that an increase in the burnup causes the flexural rigidity of TVSA and TVS-2 FAs with a rigid skeleton to decrease to the minimum value of ∼5 kg f/mm, the burnup being 45–50 МW day/kg U, and then to start growing again. It has been found out that it is the fuel bundle, specifically the change in the force of the fuel rod compression in the spacer grid, which is responsible for the change in the FA flexural rigidity. The maximum TVSA and TVS-2 FA bowing is in the range of 8–11 mm whereas the burnup is 48–63 МW day/kg U. The newly adopted TVSA and TVS-2 FA designs have contributed to the safe operation of the VVER-1000 control and protection system’s absorber rods. Copyright © 2016, National Research Nuclear University MEPhI (Moscow Engineering Physics Institute). Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0/ ).

Keywords: Fuel assembly; Fuel rod; VVER; Flexural rigidity; Fuel burnup; Spacer grid; FA skeleton; FA bowing.

r

o

t

t

c

s

o

M

0

r

r

c

c

t

a

t

h2B

Author's Personal Copy

Safe operation of nuclear power units with VVER-1000eactors depends to a great extent on the geometrical stabilityf the fuel assembly (FA) structure (bowing degree and shape,wist angle, length, lateral dimension, warping of spacer grids)hroughout the operation period. The robustness of the reactorontrol and protection system’s (CPS) absorber rods and theafety of the FA handling in the course of refueling dependn changes in the FA geometrical parameters.

E-mail address: pavlov@sosny.ru . Peer-review under responsibility of National Research Nuclear University

EPhI (Moscow Engineering Physics Institute). Russian text published: Izvestiya vuzov. Yadernaya Energetika (ISSN

204-3327), 2016, n.3, pp. 42-52.

t1

t

b

a

ttp://dx.doi.org/10.1016/j.nucet.2016.11.005 452-3038/Copyright © 2016, National Research Nuclear University MEPhI (Mos.V. This is an open access article under the CC BY-NC-ND license ( http://creati

To improve the neutronic performance of the VVER-1000eactor cores, the commercial FA with a steel skeleton waseplaced for an improved FA (IFA), in which the steel guidehannels (GC) and spacer grids (SG) were substituted for zir-onium ones. The spacer grids are attached to the FA centralube, and the GCs are attached in the FA tail support gridt the bottom and in the FA head at the top and run freelyhrough the SG [1,2] .

The results of studies on six IFAs at NIIAR have shownhat it is as soon as after one fuel cycle (a burnup of 10–6 MW day/kg U) that the IFA bowing reaches ∼15 mm andhat it has a zigzag shape. The potential causes for the IFAowing are axial loads from the reactor’s safety tube blocknd lateral loads from the adjoining FAs, as well as the

cow Engineering Physics Institute). Production and hosting by Elsevier vecommons.org/licenses/by-nc-nd/4.0/ ).

252 S.V. Pavlov / Nuclear Energy and Technology 2 (2016) 251–255

Fig. 1. Layout of the FA flexural rigidity measurement facility: 1 – FA head attachment; 2 – TV camera; 3 – measuring bar; 4 – force loading ring; 5 –FA; 6 – FA tail attachment; 7 – cable hoist; 8 – dynamometer; 9 – cables; 10 – TV camera rack; 11 – crane sling.

Fig. 2. FA axis shape during tests.

t

t

s

f

s

t

c

t

r

d

w

i

T

i

w

r

T

t

r

t

l

a

k

f

l

k

d

Author's Personal Copy

in-service decrease in the IFA flexural rigidity [3,4] . The bow-ing reaches 18 mm at a burnup of ∼40 MW day/kg U. Thesetwo factors, the degree of bowing and its shape, have beenfound to be primarily responsible for the increase in the timeof the CPS absorber rods drop in the FA GCs and even forthe potential absorber rods sticking in the event of the reactorscram, which is not admissible in terms of the reactor safeoperation [5,6] .

A procedure and an hotcell facility have been developed atNIIAR to study the flexural rigidity of irradiated FAs basedon determination of the FA flexural rigidity from the degreeof bowing in conditions of lateral FA loads [6–8] .

The facility ( Fig. 1 ) consists of three major components:a force loading system, the FA attachments and an FA bow-ing measurement system with a TV camera moving along

he FAs. The FA attachment inside the facility reproduceshe attachment conditions in the reactor. The FA bowing isimulated through a load being perpendicularly applied to theaces at the level of the eighth or the twelfth spacer grid (re-pectively SG8 and SG12). The load is transferred throughhe force loading ring fit tightly onto the SG. A cable isonnected to the ring which is led, with the use of rollers,hrough a shielded hotcell horizontal port into the operatoroom where it is connected to the force loading hoist via aynamometer.

The load is applied to the SG in steps of 10 to 15 kg fith the load removal after each step, whereas a TV camera

s used to measure the displacement of each SG in the FA.he results are processed and smoothed and the FA axis shape

s restored during the loading and after the load is removed. As an example, Fig. 2 presents the shape of the FA axis

ith SG8 being loaded in two opposite directions: in the di-ection from the FA’s face 3 towards face 6 and vice versa.he figure also shows the shape of the initial bowing after

he FA operation and of the residual bowing after the loademoval. The residual bowing is caused by the interaction ofhe fuel cladding and the GCs with the spacer grids duringoaded FA bowing. It is enough to suspend the assembly onto crane and pull it slightly up for the bowing to disappear.

The FA flexural rigidity is characterized by the coefficient FA

which is determined with an error of not more than 4%rom the dependence of the bowing degree B on the appliedoad F ( Fig. 3 ) [7] :

FA

= d F /d B. (1)

Normally, the coefficient k FA

is determined for the FA mid-le at the level of SG8 for the FA designs with fifteen grids.

S.V. Pavlov / Nuclear Energy and Technology 2 (2016) 251–255 253

Fig. 3. FA bowing at the level of SG8 as a function of the force applied.

Fig. 4. VVER-1000 FA flexural rigidity change as a function of burnup.

fl

t

w

t

I

g

b

T

f

e

t

I

3

t

t

a

1

f

t

Fig. 5. VVER-1000 fuel cladding diameter change as a function of fuel burnup.

d

T

s

u

o

i

d

t

t

t

T

1

t

i

t

n

t

4

e

t

l

t

i

(

c

t

a

c

c

t

f

a

F

s

o

T

m

Author's Personal Copy

The results of the IFA studies ( Fig. 4 ) have shown that theexural rigidity decreases by more than a double, from 8.3

o 3 to 4 kg f/mm, at a burnup of about 38–48 MW day/kg U,hich leads to a heavily bowed FA [9, 10] . Two designs, TVSA and TVS-2, both with a rigid skele-

on, have been developed for increasing the flexural rigidity.n the TVSA, the skeleton rigidity is achieved via six an-les from the E635 alloy, which are connected to the SGy resistance spot welding. The angles are attached to theVSA tail at the bottom. The central tube and the GCs run

reely through the SGs and are not attached to them [11] . Inarly designs, the angle covered three fuel rods (together withhe corner one) on the TVSA face [12] . As compared withFAs, the flexural rigidity was increased twofold at burnups of7–55 MW day/kg U (see Fig. 4 ). Further, the angle width inhe TVSAs was reduced, which naturally led to a decrease inhe assembly structure’s flexural rigidity.

In the TVS-2, the assembly skeleton rigidity increase ischieved in another way: the guide channels are attached to all5 SGs by resistance spot welding, the SG height is increasedrom 20 to 30 mm, and the SG subchannel width is increasedo 0.3 mm [13] . And the TVS-2 flexural rigidity does not

iffer greatly from the flexural rigidity of the narrow-angleVSA (see Fig. 4 ).

The procedure and the facility developed were used totudy the fuel rod contribution to the VVER-1000 FA flex-ral rigidity. After the FA study (examination, measurementf geometrical parameters and determination of flexural rigid-ty), the FA head was dismantled and all fuel rods were with-rawn from the skeleton. The force of the fuel rod withdrawalhrough all FA spacer grids was measured in parallel. Thenhe FA head was reinstalled onto the skeleton and the skele-on’s flexural rigidity without fuel rods was measured.

The studies have shown that, after one fuel cycle of theVSA and TVS-2 operation, starting with the fuel burnups of5–18 MW day/kg U and up to a burnup of ∼54 MW day/kg U,he skeleton rigidity in these FA designs does not change ands in a range of 4–4.5 kg f/mm (see Fig. 4 ). At the same time,he TVSA and TVS-2 flexural rigidity decreases with bur-up from approximately 10 kg f/mm for non-irradiated FAso the minimum values of about 5 kg f/mm at burnups of5–50 MW day/kg U, and then starts to grow (see Fig. 4 ).

Such flexural rigidity behavior is explained by the fuel rodffects. Under the action of an excessive coolant pressure, inhe process of operation, the fuel rod diameter decreases to theevel of the cladding’s tight contact with the fuel pellets. Forhe VVER-1000 fuel rods, this contact takes place at the max-mum fuel rod burnup value in a range of 50–60 MW day/kg U Fig. 5 ) [14–17] . Meanwhile, the fuel cladding diameter de-reases by approximately 0.08 mm. The cladding’s initial dis-ortion takes place then under the action of the fuel swelling,nd the diameter starts to increase.

A decrease in the fuel cladding diameter leads to a de-rease in the fuel rod compression force in the spacer gridells, and, as a consequence, to a decrease in the FA struc-ure flexural rigidity. The decrease in the fuel rod compressionorce inside the spacer grid cells is indirectly confirmed by change in the force of the fuel rod withdrawal from theA skeleton from burnup. Fig. 6 presents the averaging re-ults for all of the 312 fuel rods in the FA and the intervalsf changes in the force of the fuel rod withdrawal from theVSA and TVS-2 skeletons as a function of burnup. Theinimum of the withdrawal force falls on the average FA

254 S.V. Pavlov / Nuclear Energy and Technology 2 (2016) 251–255

Fig. 6. Force of the fuel rod withdrawal from the VVER-1000 FA skeleton.

Fig. 7. Fuel bundle contribution to the FA flexural rigidity as a function of burnup.

Fig. 8. FA bowing change as a function of burnup.

p

p

3

n

t

o

f

2

s

t

p

C

r

i

b

s

t

p

t

n

g

r

d

a

c

e

R

Author's Personal Copy

burnup of ∼44 MW day/kg U [10] . With the “physical” dif-ference in the FA average burnup values and the maximumfuel rod burnup value taken into account, the minimums inFigs. 5 and 6 fall approximately on the identical burnup in-terval in terms of the average burnup in the FA.

Numerically, the fuel rod contribution to the FA flexuralrigidity, as a function of burnup, can be obtained from datain Fig. 4 , using a simple formula

k fr / k FA

= ( k FA

− k ske ) / k FA

, (2)

where k fr is the flexural rigidity introduced by the fuel bundle;and k ske is the flexural rigidity of the FA skeleton.

Since the skeleton flexural rigidity was studied not for allof the FAs shown in Fig. 4 , it was assumed that the flexuralrigidity of the TVSA and TVS-2 skeleton did not practicallychange in a burnup range of 0–70 MW day/kg U. In accor-dance with this, the FAs whose skeleton flexural rigidity wasnot measured had its degree assumed to be equal to the aver-age value of k av

ske = 4.23 kg f/mm obtained from experimen-tal data for five FAs (see Fig. 4 ) with a burnup of ∼15 to∼54 MW day/kg U.

For nonirradiated TVSAs and TVS-2s, the flexural rigidityof the fuel bundle ( Fig. 7 ) accounts for 60% of the assembly’sflexural rigidity. After the initial year of operation, the fuel rodcontribution to the FA flexural rigidity decreases by 20 % andreaches its lowest of 20% in the course of further operationup to a burnup of ∼48 MW day/kg U. This is followed bya growth in the fuel rod contribution to the flexural rigiditycaused by an increase in the fuel rod diameter due to the fuel

ellet swelling. With a burnup of ∼65 MW day/kg U, the fuelin contribution to the FA flexural rigidity increases to about4%.

The maximum recorded degree of the TVSA bowing doesot exceed 8 mm up to the burnups of 63 MW day/kg U, andhe TVS-2 has its bowing not exceeding 11 mm at a burnupf 48 MW day/kg U ( Fig. 8 ). And the bowing shape changesrom zigzag to arched.

The increase in the flexural rigidity of the TVSA and TVS- structures has made it possible to solve in principle theafety problem involved in the operation of the reactor con-rol and protection system’s absorber rods which is confirmedrimarily by a positive experience of their operation [18–20] .

onclusion

The development of new VVER-1000 FA designs with aigid skeleton (TVSA and TVS-2) has made it possible toncrease greatly the FA flexural rigidity and to reduce theirowing to the maximum values of 8–11 mm, which ensuresafe operation of the CPS absorber rods. It has been shownhat an in-service change in the FA flexural rigidity is definedredominantly fuel rods and by the change in the force ofheir compression inside the spacer grid cells. Introduction ofew solutions such as a decrease in the number of the spacerrids from 15 to 12 and then to 8; a redesign of the fuelod (a decrease in the cladding thickness, a decrease in theiametrical gap between the cladding and the fuel pellets, andn increase in the fuel rod length) are expected to lead to ahange in the VVER-1000 FA flexural rigidity and require anxperimental justification.

eferences

[1] A. Enin , Y. Bezborodov , D. Pluzhnikov , in: Proceedings of the VIIIthInternational Conference on WWER Fuel Performance, Modelling andExperimental Support, Helena Resort, Bulgaria, 26 September–4 Octo-ber 2009, 2009, pp. 201–209 .

[2] V.D. Shmelev , Aktivnye zony VVER dlya atomnykh stantsiy [VVERCores for Nuclear Power Plants], IKTs Akademkniga Publ., Moscow,2004, p. 220. (in Russian) .

[3] V. Troyanov , Yu. Likhachev , V. Folomeev , Izv. vuzov. Yad. Energ. (2)(2002) 44–54 (in Russian) .

[4] V. Troyanov , Yu. Likhachev , V. Folomeev , Izv. vuzov. Yad. Energ. (3)(2002) 14–19 (in Russian) .

S.V. Pavlov / Nuclear Energy and Technology 2 (2016) 251–255 255

[

[

[

[

[

[

[

[

[

[

[

Author's Personal Copy

[5] Materialy rabot mezhvedomstvennoy komissii po vyyasneniyu prichinzastrevaniya OR SUZ na energoblokakh s VVER-1000 [Proceedings ofthe Interdepartmental Committee for Identification of Causes for theControl Rod Sticking at VVER-1000 Units]. V. 1–8. Moscow. Rosen-ergoatom Concern Publ., 1996 (in Russian).

[6] V. Polenok , S. Pavlov , V. Smirnov , A. Smirnov , D. Markov , in: Sbornikdokladov pyatoy mezhotraslevoy konferentsii po reaktornomu materi-alovedeniyu [Proceedings of the Fifth Interindustry Conference on Re-actor Material Science], Dimitrovgrad, 8–12 September 1997, 1, JSCSSC RIAR Publ., Dimitrovgrad, 1998, pp. 47–58 .

[7] S. Pavlov, V. Polenok, V. Smirnov, Metodika opredeleniya progibovTVS VVER-1000 pripoperechnykh nagruzheniyakh [Procedures to mea-sure the VVER-1000 fuel assembly bending caused by transverse loads].Sbornik trudov OAO “GNTs NIIAR” [Collection of JSC SSC RIAR pa-pers]. 1998, v. 3, pp. 85–93 (in Russian).

[8] S. Pavlov , A. Sukhikh , S. Sagalov , Nerazrushayushchaya diagnostikasostoyaniya elementov aktivnykh zon yadernykh reaktorov [Nondestruc-tive Examination of Nuclear Reactor Internals], DITI MEPHI Publ.,Dimitrovgtrad, 2015, p. 320. (in Russian) .

[9] D. Markov , S. Pavlov , A. Novoselov , in: Sbornik dokladov IXRossiyskoy konferentsii po reaktornomu materialovedeniyu [Proceed- ings of the Ninth Russian Conference on Reactor Material Science].Dimitrovgrad, 14–18 September 2009, Dimitrovgrad., JSC SSC RIARPubl., 2009, pp. 24–46 . (in Russian).

10] S. Pavlov , in: Proceedings of the Xth International Conference onWWER Fuel Performance, Modeling and Experimental Support, 7–14September 2013, Sandanski, Bulgaria, 2013, pp. 213–227 .

11] O. Samoylov , V. Kaydalov , A. Romanov , A. Falkov , A. Shishkin , in:Sbornik dokladov VIII Rossiyskoy konferentsii po reaktornomu materi-alovedeniyu [Proceedings of the Eighth Russian Conference on ReactorMaterial Science], JSC SSC RIAR Publ., Dimitrovgrad, 2008, pp. 4–12 .

12] V. Molchanov , A. Starikov , O. Samoylov , V. Kaydalov , A. Falkov ,A. Romanov , A. Shishkin , in: Proceedings of the VIIIth InternationalConference on WWER Fuel Performance, Modeling and ExperimentalSupport, Helena Resort, Bulgaria, 26 September–4 October 2009, 2009,pp. 191–200 .

13] S. Ryzhov , V. Mokhov , I. Vasilchenko , S. Kobelev , V. Vyalitsyn ,S. Kushmanov , S. Semiglazov , A. Turkin , V. Makhin , in: Sbornik dok-ladov IX Rossiyskoy konferentsii po reaktornomu materialovedeniyu[Proceedings of the Ninth Russian Conference on Reactor Material Sci-ence]. Dimitrovgrad, 14–18 September 2009, ., JSC SSC RIAR Publ.,Dimitrovgrad, 2009, pp. 46–52 .

14] A. Novoselov , S. Pavlov , V. Polenok , D. Markov , V. Zhytelev ,G. Kobylyanskiy , A. Kostyuchenko , I. Volkova , Fiz. Khim. Obrab. Ma-terialov (2) (2009) 24–32 (in Russian) .

15] V. Polenok , D. Markov , V. Zhytelev , S. Perepelkin , G. Kobylyanskiy ,I. Volkova , in: Sbornik dokladov VIII Rossiyskoy konferentsii po reak-tornomu materialovedeniyu [Proceedings of the Eighth Russian Confer-ence on Reactor Material Science], JSC SSC RIAR Publ., Dimitrovgrad,2008, pp. 138–147 .

16] V. Zhytelev , D. Markov , Ye. Zvir , V. Polenok , G. Shevlyakov , in:Sbornik dokladov IX Rossiyskoy konferentsii po reaktornomu materi-alovedeniyu [Proceedings of the Ninth Russian Conference on ReactorMaterial Science]. Dimitrovgrad, 14–18 September 2009, Dimitrovgrad.,JSC SSC RIAR Publ., 2009, pp. 121–129 . (in Russian).

17] D. Markov , S. Pavlov , A. Novoselov , V. Polenok , A. Zhytelev , Ye. Zvir ,V. Chesanov , G. Kobylyanskiy , in: Proceedings of 2010 LWR Fuel Per-formance/TopFuel/WRFPM, Orlando, Florida, USA, September 26–292010, 2010, pp. 504–512 .

18] V. Molchanov , A. Panyushkin , V. Zheleznyak , O. Samoylov , V. Kuul ,V. Kurylev , in: Sbornik dokladov VI Rossiyskoy konferentsii po reak-tornomu materialovedeniyu [Proceedings of the Sixth Russian Confer-ence on Reactor Material Science]. Dimitrovgrad, 11–15 September2000 (in 3 volumes: V. 2, Part 1), Dimitrovgrad, 2001, pp. 53–66 . (inRussian).

19] V. Tyukin , in: Proceedings of the VIIth International Conferenceon WWER Fuel Performance, Modelling and Experimental Support,Albena Congress Center, Bulgaria, 17–21 September 2007, 2007,pp. 139–145 .

20] V. Molchanov , A. Starikov , O. Samoylov , V. Kaydalov , S. Shishkin ,A. Falkov , in: Proceedings of the VIIth International Conferenceon WWER Fuel Performance, Modelling and Experimental Support,Albena Congress Center, Bulgaria, 17–21 September 2007, 2007,pp. 218–222 .