FIELD NONHOMOGENEITY MEASUREMENTS AT PULSED HEATING OF THE SUPERCONDUCTING MAGNET WINDING

(1)

HAL Id: jpa-00223648

https://hal.archives-ouvertes.fr/jpa-00223648

Submitted on 1 Jan 1984

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

FIELD NONHOMOGENEITY MEASUREMENTS AT PULSED HEATING OF THE SUPERCONDUCTING

MAGNET WINDING

P. Vassilev, A. Donyagin, I. Eliseeva, V. Lobanov, L. Makarov, G.

Reschetnikov, A. Smirnov, I. Schelaev

To cite this version:

P. Vassilev, A. Donyagin, I. Eliseeva, V. Lobanov, L. Makarov, et al.. FIELD NONHOMOGENEITY MEASUREMENTS AT PULSED HEATING OF THE SUPERCONDUCTING MAGNET WIND- ING. Journal de Physique Colloques, 1984, 45 (C1), pp.C1-845-C1-848. �10.1051/jphyscol:19841173�.

�jpa-00223648�

(2)

FIELD NONHOMOGENEITY MEASUREMENTS AT PULSED HEATING OF THE SUPERCONDUCTING MAGNET WINDING

P.G. Vassilev, A.M. Donyagin, I.A. Eliseeva, V.I. Lobanov, L.G. Makarov, G.P. Reschetnikov, A.A. Smirnov and I.A. Schelaev

J o i n t I n s t i i x t e f o r muclear Research, Dubna, U.S.S.R.

R6sum6 - Nous avons observe des changements de l'homogBn6it6 du champ dans l'ouverture d'un dipole sous l'action de courtes im- pulsions thermiques subcritiques dans la bobine supraconductrice, synchronisees avec le courant puls6 de l'enroulement.

Abstract

-

Changes of the homogeneity of the cyclic magnetic field in the aperture of a "window framevtype dipole have been observed and investigated at subcritical synchronous pulsed power supply to the winding.

A restricted normal zone(RNZ)can be initiated by means of external local pulsed power supply (total energy Q)to the winding of the su- perconduct~ng magnet.In the case of suitable transversal temperature gradient even if all wires have similar superconducting parameters, there exists some region of &&,(I) (:Q <&,=(I), where the RNZ does not spread over all cross-section of a composite conductor,which still remains capable to carry a nonzero transport current I at a zero voltage despite the fact that its load line in a self-magnetic field is more or less modified.Any development of such a RNZ leads to a respective current redistribution between the wires of the cable.

As this is equivalent to some change of the turn size and geometry, the magnetic field homogeneity in the aperture should be under in- fluence.The upper limit of admissible local overheatingpG(Q,I,r,t)=

= Tw(Q,I,r,t)

-

^To ⁽To is the bath temperature)of the winding for a given geometry,cooling environment and synchronization of the pulsed power supply and I(t) can be considered as a boundary condition of its dynamic stability (BCDS)beyond which a nonzero voltage appears.

In this paper we present sone observed changes of the fieldmnhomoge- neity ^,A ;n the aperture of a magnet dipole of the "window frame

type/'/ caused by synchronized subcritical pulsed power supply to the mnding in the vicinity of BCDS at linear increase of the feed

current.

I

-

EXPERIlVIENTAL PROCEDURE

The magnet winding was made of a multifilament HT50 composite conductor(each of 15 wires of the cable was 0.5 mm in diameter and consis- ted of 1045 Nb-Ti filaments in a copper matrix). pour electrical r u l . 3 Ohm heaters.with various geometry,arrangement and thermal environment were attached to the turns.Por example,Fig.l illustrates a part of the Iron yoke(1) and winding configuration together with the heater

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19841173

(3)

Cl-846 JOURNAL DE PHYSIQUE

H4 (6) made of electrically insulated (7) constantan wire 0.35 mm in diameter.In the figure is also shown a HT50 single wire ( 5 ) epoxy- glued(2)and feded indepently of the magnet.Using a thermally conduc- tive glue Bf-4 (lO),the carbon resistor thermometers AB4 (12) and TBO (9) were directly attached to the turn surface cleaned of turn- to-turn insulation /mylar ($) and fibre glass ribbon (3)/ and pro- tected against the liquid helium (8) by glasstextolite (11).

g. 1 :A general view of the winding with heater H4,

single wire HT50 and thermometers AB4 and TBO

The inductive field sensor(its

length exceeds that of the magnet in order to allow integral field measurements)consisted of several flat coi1s.It was fitted along the centre of the aperture so that the axis of its rotation coincided with the geo- metrical axis of the magnet.The sensor was rotated by steps of

h6=g0

(up to 360~) between any con- secutive pulses.The sensor coils ope- rated in a regime of depression of the dipole harmonics in order to im- prove a ~ d equalize the higher ones.

When changing the feed current,a constant step of 0.1 ,-1 was used for all magnetic measurements. The sig- nals of the sewor were integrated, amplified and coded by means of an analog-digital convertor (ADC).Si- milar-ADc-was used at the feeding of the ma et as we1l.A constant,as well as pulsed,triangle shape current

IR)

was applied to the magnet winding.Varying the delay ( d) of the local pulsed power supply relative to the rise of the magnetic field in the centre of the aperture Bo(t) = 1.028 I(t) (Bo in Tesla if I in kA) and/or changing the absolute value of Q,the synchronization conditions between overheating and the rise of B(t) could be varied in order to obtain maximum number of experimental points in vicinity of BCDS at zero voltage on the turns.It should be noted that despite a temperature diffeqence between the thermometers and winding,~T(Q,I,t)-measurements were very useful.In particular, it was possible to determine the BCDS in terms of maximum allowable overheating of the thermometers,AT ,under the action of the heater, using experimental data obtained at constant current and taking into account the corres onding delay of the pulaed power supply relative to the rise of I(t7 and B(t) as well as the dynamics of the temperature field for the given geometry.

The harmonic analysis of the data obtained during the experiment and

, A ,

all information processing'd'hsve been performed on line with a MERA 60-30 computer.The relative changes of the vertical

~ ~ y ( ~ ) and horizontal A~X(X) components of the field in the

Bo ⁰

mediane plane of the a p e r z ~ e may be expressed as follows:

+-

~B~(x,o, ZMZ '- /B~(O,O,Z)~Z AB~(X)

-00 i +Jx(x,0, .)dz

*'?

⁼ +?

(4)

I1

-

^RESULTS

The stability boundary of the magnet at constant feed current and pulsed power supply to the winding can be determined eimilarly to the BCDS,i.e.by the maximum allowable value of Q and the corresponding overheating of the thermometer,A Tma,beyond which at given

= I (I being the critical current of the winding in the self- m gneticCfiefd at I,=4.2K and Q = O)a nonzero voltage on the turns apPears.For the heater H4 in

Fig.2 it ia shown as a func

-

tion of the Q

( 2

⁾type for the composite cable of the winding(so1id curve)and for the single H150 wire having a critical current 15 times lower than

dashed

curve).Empty

2 f

\

\ \

'-*-7h --,-7-,7-y A

0 a~ a2 p, a4 as as ai a8 a9 r o t 0

Pig.2:&(?)

-

dependences for Fig.3:Tirne dependenoes of the over- the winding(so1id curve) heatingAT(t) TBO under the and the HT50 single wire action of H4( a:curves 1

+

⁵

(dashed curve). are for zero, curves 1

'+

⁵'- for nonzero voltage on the circles correspond to zero,points turns)at several values of -to nonzero voltage.The time axis = I/Ic and ~~DS(b:curve 6) for a given cycle of the feed

current is alao shown. for the chosen cycle of I(*) and B(lj).Curves 7,8,9 are for Pig. 3a presents some typical zero,8

-

for nonzero voltage P T(t)-de~endences at several on the turns at Q=0.81 J.

values o f & for H4 and TBO. Solid

curves cjorrespond to zero volta;ge,$8;9bed

-

to nonzero (curve fr&-;

2.02JY1 :Q=2.12J;cu$ve 2:Q=1.075,2 :Q=I. lOJ;cu$ve 3:Q=0.97J,3:Q=1.00 J;curve 4:&=0.79J,4 rQ=0.81 J;curve 5:&=0.61 J,5 :Q=0,62~).For pulsed feed current with I-=I500 A and a1 = 400 A/s the BCDS,determined upon the overheating-TBO due to a puised power supply by H4,is shown as curve 6 in Fig. 3b. Curves 7,8,9 represent overheating A !I?( t )of the thermometer TB0 at Q=0.81 J and t

-

⁰^,0.5 and 1.2 s ,respectively.

The curve 8'is for Q=l.lOJ and -0.5 s,at which the real A T(t)- dependence crosses the BCDS and in d-such a way there appears a nonzero voltage on the turns together with a Joule heating.

(5)

C1-888 JOURNAL DE PHYSIQUE

A By (x) L\B~(X) The Bo and

B 0

dependences,when Bo becomes equal to 1.34

Fig.4:Badial distribution of wires of the-composite cable of the winding since its overheating is no A By(d 4BX(x) m r e sufficient to form RNZ.

Tesla,are shown in Fig.$.Solid curves are for Q=O,dashed

-

^for

Q=0.81J.Other conditions are the fol1owing:heater H4, td=1.21 s

,

'ma =I460 A,dI/dt =380 A/s.It should be noted that the observed changes of the field nonhomogeneity reduce to zero when Q =

B 0 BO For magnets,the field of which is when the feed current I(t) mainly formed by a superconducting becomes equal to 1308 A winding,the appearence of such a at the chosen cycle for RNZ,when only a part of the cross- Q-0 (solid burve) and section of the composite cable goes Q-0.81 J(dashed curve). into normal state,larger changes of

%

,

A B - (x)

can be expected.This effect should be taken into account when magnet systems for the accelerator -0

if q=~onst.and Q decreases,i. e.

when the experimental conditions go out of the vicinity of BCDS.

For example,the transition from point C (Q=0,81J,

rd=l

.21 s )to points D (Q-0.565, td=0.5 s;

I I A I ) I , Q=0.8J9 td=0.3 S; Qz1.04 J,

-2.5 -2.0 -1.5 -10 -0.5 0 0.5 $0 1.5 20 25

rd=0.25S)in Fig.2 results in no difference between the radial dist- ributions of the magnetic field,

ABY(x) dB,(x) characterised by and---

Bo BO

with and without pulsed power supply

-3 by'H4.fChis is likely to be explained if under conditions,corresponding to

-2.5 -10 - \ 5 - 1 0 -0.5 0 0.5 (0

**x*cm**

1.5 2.0 1.5 points D,there is no redistribution of the transport current between the Id

are designed and constructed /4/

.

Const.and

td

decreases as well as

The authors are indepted to Dr,L.N.Zaitsev for the idea to perform this experiment.

I11

-

REFERENCES

1. Schelaev I. A.

,

^{Youdin I.}^P,

,

Preprint JINR R9-80-333, Dubna, 1 983.

2.Wiss C.,Proceed,5th Intern.Conf.on Rlagnet Technology,Roma,l975,p.231.

3.Vassilev P. G. et al, Preprint JINR R9-82-486 ,Dubna, 1982.

4. Vassilev P. G. et al, JINR R9-83-394,Dubna, 1983.