Eective circuit modelling and experimental realization of an ultra-compact self-rectier ux pump B.P.P. MallettS. Venuturumilli J. Clarke B. Leuw J.H.P. Rice D.A. Moseley C.W. Bumby and R.A. Badcock

2025-05-03 0 0 6.77MB 9 页 10玖币

侵权投诉

Eﬀective circuit modelling and experimental realization of an ultra-compact

self-rectiﬁer ﬂux pump

B.P.P. Mallett,∗S. Venuturumilli, J. Clarke, B. Leuw, J.H.P. Rice, D.A. Moseley, C.W. Bumby, and R.A. Badcock

Paihau Robinson Research Institute, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand

J. Geng

Wuhan National High Magnetic Field Center, Huazhong University of

Science and Technology, 430074 Wuhan, People’s Republic of China

(Dated: October 25, 2022)

This paper presents experimental and modelling results of an ultra-compact self-rectiﬁer ﬂux pump

energizing a superconducting coil. The device ﬁts inside a volume of 65x65x50 mm and generates up

to 320 A dc through the coil and a peak output voltage up to 60 mV. We also develop and present

a full electromagnetic eﬀective circuit model of the ﬂux pump and compare its predictions to the

experimental results. We show that our model can reproduce accurately the charging of the load coil

and that it reproduces the systematic dependence of the maximum load coil current on the input

current waveform. The experiments and modelling together show also the importance of dc-ﬂux

oﬀsets in the transformer core on the ﬁnal achievable current through the coil. The miniaturization

possible for this class of ﬂux pump and their minimal heat-leak into the cryogenic environment from

thermal conduction make them attractive for applications with demanding size, weight and power

limitations. Our eﬀective circuit model is a useful tool in the understanding, design and optimization

of such ﬂux pumps which will accelerate their progression from research devices to their application.

I. INTRODUCTION

Flux pumps (FP) are superconducting devices which

output a large dc current to a superconducting coil or

magnet [1]. Self rectifying FPs are a subset of such de-

vices that rectify an asymmetric, alternating input cur-

rent to a transformer to generate the dc current through

a superconducting load [2–7]. Rectiﬁcation is achieved

by driving current larger than the critical current, Ic,

through part of the superconducting circuit (the ‘bridge’)

for part of the input waveform cycle, as originally de-

scribed for high-temperature superconductors by Vysot-

sky et al. [8]. This self-rectiﬁcation method can be com-

pared with rectiﬁers that employ various active switching

mechanisms to vary the Icof the bridge [1, 9–12]. Self-

rectifying FPs are capable of delivering large currents

at modest voltages [3] without a requirement for direct

electrical and thermal connection between the supercon-

ducting load and an ambient temperature power supply.

Together with the absence of moving parts and relatively

few components, they may be an attractive alternative to

conventional power supplies, or other types of ﬂux pumps

[1, 13, 14], for applications with demanding size, weight

and power requirements.

Previous experimental work on this class of ﬂux pump

has focused on demonstration of principle [2] or maximiz-

ing the delivered dc current [3] rather than its potential

advantages, comprising a simple device architecture and

physical compactness. This previous work successfully

described and developed what we believe to be the es-

sential operating principles of the FP [3, 6]. However,

∗ben.mallett@vuw.ac.nz

quantitative and accurate prediction of the performance

of these FPs involves numerically solving the coupled

equations describing them. Recent numerical modelling

work utilizing an eﬀective-circuit (or ‘lumped parame-

ters’) methodology has been limited to exploring the in-

ﬂuence of sub-components of the FP, such as the bridge

[15] or the transformer [16]. Combining these subsys-

tems into the same model has only just been reported

by Zhai et al. [17] with some explanatory success for an

air-core transformer self-rectiﬁer. What such modelling

should enable however is; (i) quantitative performance

prediction to facilitate model validation, and (ii) an un-

derstanding of the interplay between all subcomponents

of the FP such as the input waveform, transformer and

circuit impedances. Together these would allow for use-

ful full-system design and optimization, which has been

identiﬁed as an important issue to be addressed for ﬂux

pumps [14]. The utility of such an eﬀective-circuit ap-

proach for system-level modelling and has recently been

demonstrated in other areas of superconducting power

electronics [18–21].

In this work, we construct a simple prototype ‘ultra-

compact’ self-rectiﬁer FP and demonstrate its function-

ality. We then develop a system-level eﬀective circuit

model of the FP and compare its predictions to experi-

mental results. Our ﬁrst ﬁnding is that the FP does not

require a superconducting secondary-winding and can de-

liver more than 300 A dc to a superconducting coil with

up to 60 mV peak output voltage. The eﬀect of modify-

ing the input current waveform is explored and the im-

plications on performance discussed. Our second ﬁnding

is that our eﬀective circuit model matches experimen-

tal behaviour and is capable of describing the maximum

load current and its dependence on the input current to

the transformer as well as the magnitude and time depen-

arXiv:2210.12999v1 [cond-mat.supr-con] 24 Oct 2022

0 . 0 0 0 . 0 5 0 . 1 0 0 . 1 5

- 5

1 0

1 5

2 0

P r i m a r y c u r r e n t , i1 ( A )

T i m e ( s )

f = 1 0 H z

{ 1 5 , - 4 . 5 }

{ 1 2 , - 4 . 5 }

{ 2 0 , - 2 . 7 }

( b )

FIG. 1. (a) An annotated picture of the ‘ultra-compact’ self-

rectiﬁer and test load coil. (b) Examples of the current wave-

form supplied to the primary windings of the transformer.

dence of the voltage across the load. The paper concludes

with a discussion of insights from the model regarding the

operation of self-rectifying ﬂux pumps, particularly the

eﬀect of dc oﬀsets of ﬂux in the transformer core, and the

limitations of our current model.

II. METHODS

The self-rectifying FP studied in this work is shown

in Fig. 1(a). It is built around an iron core made from

0.3 mm laminations with cross section of 20 mm x 20 mm

and path length approximately 100 mm. The nominal

saturation ﬁeld at room temperature of the soft iron is

Bsat = 1.8 T. The primary windings are made from 45

turns of Cu tape. The secondary is made also from pure

Cu and comprises a single strip, 12 mm wide, 35 mm

long and 3 mm thick, bent in a U-shape around the

core. Soldered to the secondary winding is a 100 mm

long bridge made from commercial REBCO tape, which

completes the loop around the core and results in a trans-

former turns ratio of N= 45. The coated conductor

tape is SuNAM product code SCN12500-210222-01; 12

mm wide, stabilized by 20 µm of Cu and has a nominal

self-ﬁeld Icof 500 A at 77 K with n= 35 (see eq. 1 below

for the deﬁnition of n). A small load coil of eight turns of

the same REBCO tape with a total length of 1.9 m and

an inductance of 2.5µH, measured at room temperature,

was soldered to the bridge.

In all experiments, the current through the primary

windings was supplied by a Takasago BWS 40-15 bipo-

lar linear ampliﬁer controlled via a National Instruments

c-DAQ. Examples of the applied primary current wave-

form, i1, are shown in Fig. 1(b), where the maximum and

minimum of the current are speciﬁed by the two numbers

in braces in the legend. For example, i1{19.8,−4.45}

indicates an input waveform with a maximum current

peaking at i1,max = 19.8 A and a current in the reverse

direction of i1,min =−4.45 A. All waveforms used in this

work were driven at a frequency of 10 Hz, and all mea-

surements were carried out in liquid nitrogen (approxi-

mately 77 K). Current through the load-coil was mea-

sured via a home built open-loop hall sensor employing a

P15A sensor from Advanced Hall Systems Ltd. Current

supply to the primary was measured via the voltage de-

veloped across a thermally-sunk 1 Ω shunt resistor. The

voltage across the primary and load were also measured.

The operation of the self-rectifying ﬂux pump was

modelled using the eﬀective-circuit shown in Fig. 2. This

circuit involves coupled electrical and magnetic circuits

with non-linear componentry. It was implemented and

solved within the Simulink package in MatLab. The cur-

rent waveform input into the primary windings is speci-

ﬁed, along with a parameterization of the circuit compo-

nents.

The transformer is modelled as having a variable re-

luctance, R(H) = l.µ(H)−1.A−1, resulting from a mag-

netic material with non-linear permeability, B(H) =

µ(H)H=Bsat tanh(H/Hsat), and the physical dimen-

sions of the core length, l, and cross sectional area, A.

In addition, we include in the model a leakage reluctance

representing alternate paths outside of the transformer

core that magnetic ﬂux may take. This leakage reluc-

tance is signiﬁcant only as the ﬂux in the transformer

core approaches saturation. Ac-loss and hysteresis ef-

fects in the transformer core are not presently included

in the model.

The superconducting components are treated as cir-

cuit elements with a resistance, RSC , that is a non-linear

function of the current, I[22];

RSC =E0.l

IcI

Icn−1

(1)

Where E0= 1 µV.cm−1is the customary electric ﬁeld

criterion at which the critical current density of a su-

perconductor, Jc, is deﬁned. nis obtained from ﬁtting

characteristic experimental data [23] and captures the

steepness of the resistivity increase as Iexceeds Ic.lis

the length of the REBCO tape. We use Cu-stabilized

tape in which the Cu represents a parallel current path.

This is included as a parallel resistor in the model, such

that the full expression for the resistance of a component

of coated conductor (CC) tape in the model is:

R−1

CC =R−1

Cu +R−1

SC (2)

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

Eectivecircuitmodellingandexperimentalrealizationofanultra-compactself-rectieruxpumpB.P.P.Mallett,S.Venuturumilli,J.Clarke,B.Leuw,J.H.P.Rice,D.A.Moseley,C.W.Bumby,andR.A.BadcockPaihauRobinsonResearchInstitute,VictoriaUniversityofWellington,P.O.Box600,Wellington,NewZealandJ.GengWuhanNationalHighMa...

展开>> 收起<<

Eective circuit modelling and experimental realization of an ultra-compact self-rectier ux pump B.P.P. MallettS. Venuturumilli J. Clarke B. Leuw J.H.P. Rice D.A. Moseley C.W. Bumby and R.A. Badcock.pdf

共9页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Eective circuit modelling and experimental realization of an ultra-compact self-rectier ux pump B.P.P. MallettS. Venuturumilli J. Clarke B. Leuw J.H.P. Rice D.A. Moseley C.W. Bumby and R.A. Badcock

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: