PROGRESS IN THE DEVELOPMENT OF SUPERCONDUCTING TURBOGENERATORS AT KWU/SIEMENS

HAL Id: jpa-00223617

https://hal.archives-ouvertes.fr/jpa-00223617

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 pub- lished 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.

PROGRESS IN THE DEVELOPMENT OF

SUPERCONDUCTING TURBOGENERATORS AT KWU/SIEMENS

D. Lambrecht, Grünewald, Liese, Weghaupt, L. Intichar, H. Neumüller, C.

Schnapper

To cite this version:

D. Lambrecht, Grünewald, Liese, Weghaupt, L. Intichar, et al.. PROGRESS IN THE DEVEL-

OPMENT OF SUPERCONDUCTING TURBOGENERATORS AT KWU/SIEMENS. Journal de

Physique Colloques, 1984, 45 (C1), pp.C1-713-C1-716. �10.1051/jphyscol:19841145�. �jpa-00223617�

JOURNAL DE PHYSIQUE

Colloque CI, supplkment au no 1, Tome 45, janvier 1984 page C1-713

PROGRESS IN THE DEVELOPMENT OF SUPERCONDUCTING TURBOGENERATORS AT K W U / S I E M E N S

D. Lambrecht, Griinewald, L i e s e , Weghaupt, L. I n t i c h a r * , H.W. Neumiiller* and C . schnapperr

Kraftwerk Union AG, Mlheim-Ruhr, F.R. G

*siemens AG, Research Laboratory, ErZangen,

F.R.G.

Resume - Le developpement d e s c r y o a l t e r n a t e u r s

2

m / S i e m e n s e s t d e c r i t . A b s t r a c t - The development of superconducting g e n e r a t o r s a t KWU/Siemens i s d e s c r i b e d .

1 KRAFTWERK UNION (KWU) PROGRAM OF ADVANCED GENERATOR TECHNOLOGIES

The multi-step development program for advanced generator technologies combines the evolution of conventional genera- tor design with the development of superconducting generators and is presented in Fig. l [ l t o 41. I t IS characterized by

--- -- - - - -

700 MVA Advanced Conventional Generator

-

an independent generator development and test facility at the KWU Mulheim Factory which exclusively serves this program (Fig. 2).

-

a large monitoring and data acquisition system t o measure and process data

1 from up t o 1,000 measuring points

simultaneously to meet the outstand- ina obiect~ves of this Droaram.

--

^{- ,}

. -

- a qenerator frame w ~ t h endsh~elds.

v- 6 I201400 MVA

SC 1-3, Rotor

bearings, coolers and all necessary aux-

-

B R o t a t l p

1201400 MVA iliarles (oil, gas, water) t o be combined

sc G-I~~o, with replaceable rotors and inner-cage

Stators t o form various test generators of different des~gn (steps 1 t o 6 ) ,

C Commeroal Sole - three different rotors (A, Band C) and

- 3

850 MVA 850 MVA SC Gsnsralor two different inner-cage stators (A/AW

SC Rotor

Superconducting Generator Development i and

B)

as indicated in Fig. 1,

mwer man,

oparatwn

-

a liquid helium refrigeration system

w ~ t h a capacity o f about 200 I/h of Itq- Fig. 1 KWU development program for advanced generator technologies. uid helium as indicated in Fig. 3.

As outlined in Fig. I , steps 1 t o 3 contribute t o further improvement of turbogenerators of conventional design with water and gas cooling, while steps 4 t o 6 are part of the program for the development of superconducting generators for utility operation. The basic design and major parameters of the various test rotors and stators were presented in [4].

I n step 4 of the program, the superconducting test rotor will run in the conventional slot-type stator AW to undergo the fundamental cryogenic tests at various speeds and to prove the proper performance of the rotor and the cryogenic system in conjunction with the refrigerator during startup and shutdown and at transient field current loading. Due t o temperature limitations in the stator teeth, the continuous field current has to be reduced at fu:l speed. However, during short periods and at lower speeds, the rotor loading can be increased up t o the critical field current. I n this way, important test data and operational experience of a fullsize machine can be gained at an early stage of superconducting generator development, allowing corrective measures t o be taken for the following phases of the debelopment program.

In step 5 of the program, the superconducting test rotor B and the slotless stator B provided with an air-gap winding will form a superconducting prototype generator with a nominal rating of 120 M V A with the test rotor partially wound and of 400 MVA i f the rotor were fully wound. A full prototype test program will be performed with this unit, including open- circuit and short-circuit tests, unbalanced load tests and sudden three-phase short-circuits at the generator terminals with voltage successively increased up t o the rated value.

With the experience gained during manufacturing, testing and operation of rotor B, the fullsize rotor C will be built t o replace test rotor B in step 6, thus transforming the prototype generator into the first commercial-size superconducting generator of about 850 MVA, equivalent to the output of modern coal fired power plants in Germany. This unit is designed for continuous operation in a power plant after comprehensive factory tests have been successfully performed.

Meanwhile, the test generator has almost completed the tests in step 1. The design work and the fundamental tests on

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19841145

C1-714 JOURNAL DE PHYSIQUE

Fig. 2 Generator development and test facility (GEVA) in the Mulheim factory of Kraftwerk Union.

~ i q ~ i d 6 1 1 u r n T a n k HTE

Fig. 3 Liquid helium refrigeration system f o ~ SC generator

superconductors, coils, insulation, rotor material, helium-cooling system and on key components of the rotor, were completed prior to the manufacture of the SC test rotor. The material procurement for the test rotor B is well within schedule. The major parts have been delivered and are at various stages of manufacture. The inner (cold) part of the rotor, carrying the superconducting coils, has been completely machined and is presently being wound. Fig. 4 shows the rotor body with the continuous slots for fully embedded coils and the numerous cooling ducts and holes for the helium cooling system.

For the slotless stator B with air-gap winding, the basic design work and model testing on various components have been successfully accomplished. As the next step, full-scale components of the air-gap winding are being built and tested t o pave the way for a smooth production of the stator B in 1984 and 1985. The liquid helium refrigeration system according t o Fig. 3 is under manufacture by Linde AG and will be installed at KWU's Miilheim Factory i n 1984. The helium transfer equipment (HTE) has been Installed separately in a test rig and is presently undergoing thorough tests to verify reliable performance.

As part of the quality control, the wound SC test rotor will be subjected t o a stationary I, test t o check the quench performance. The cryostat for t h ~ s test, capable of accommodating the large and heavy rotor part o f 10,000 kg, is under manufacture and will be available for the I,test at the end of 1983.

2 MODEL AND COMPONENT TESTING

Comprehensive tests have been conducted on numerous parts and subsystems of the superconducting rotor and air-gap winding, not only to verify reliable performance but also with regard t o economical production and effective quality control.

which is mandatory in the phase of industrialization. Only a selection of these tests is briefly outlined i n the following.

Extensive measurements were carried out on superconductors and coil models at the Siemens Research Laboratories i n order t o select the most suitable superconductor [ I , 61 and t o determine the current load capacity with a surrounding magnetic f ~ e l d under steady-state and transient conditions. Finally, a full-scale superconducting coil, as used for the SC test rotor but of reduced coil length, was manufactured under various conditions: w ~ t h and without vacuum-impregnation with epoxy resin. The coil was wound and tested in a specially designed rig that permits t o simulate the winding process i n the rotor, t o perform the/, tests and t o study the coil structure after the tests by disassembly of the rig. The SC test coil with the dismantled device can be seen in Fig. 5. The results of the I,test are given in Fig. 6. The test coil attained 96

+

^{2.5 %} in case 1

Fig. 4 Inner part of SC 4001120 MVA test rotor Fig. 5 SC test coil wlth aevlce disassembled after test

0 1 , , , , , 5 t , , c

1 2 3 4 5 6 7 8 9 1 0

Number of Quench U 70-

.-

0

6 0 -

Y

Z 5 0 -

3 r

^40-

@

30- 0

%

20-

Fig. 6 lc test results of the SC test coil

Case 1 ( )puench Initiated

x Case 2 a (1.Cool D O W ~ )

gJFf d","rL,""s2

^Coil

+ Case 2 b @.Cool Down)

0 Case 3 []Rapid Excitation

Fig. 7 Flux densities calculated and measured for superconducting test coil

(without impregnation), 94.5

+

^{4 %, i n case} 2 (with impregnation), 9 5

+

2.5 % in case 3 (with modified impregnation) without training. The percentages refer t o the short-sample 1, value.

This result cleary indicates that for the SC test rotor €3 and the follow-up large SC rotor C equally favorable behavior can be expected during operation, i n particular as the loading characteristics are similar and the heat removal capacity of the helium is much better during operation than during stationary tests.

The superconducting test coil n o t only served t o check the quench performance and the winding process but also t o verify the results of the three-dimensional magnetic field calculation. For this purpose, field measuring sensors were ~nstalled at different locations t o determine the components o f the magnetic flux densities i n the main directions. Fig. 7 depicts the comparison o f calculated and measured values and demonstrates the capability o f the ELMAG program employed based o n the extended scalar potential method as outlined i n

[5].

Moreover, the importance of a true three-dimensional field calcula- tion becomes evident i n the remarkable differences i n local magnetic flux densities between the curved part (sensor 4) and the straight part (sensor 3) o f the coil.

The test facility and test results of the helium cooling system for the SC test rotor are presented i n a separate paper. As already mentioned, the helium transfer equipment (HTE) has been installed and is being operated i n an independent test rig.

Testing will be carried out under various conditions t o simulate the actual operation i n a power station and w i l l comprise endurance tests over long periods of time. The endurance test w i t h a rotatlng vacuum seal, one of the most sensitive components. is running since 1980 and has meanwhile exceeded 20,000 hours o f successful operation.

Intensive investigations have been conducted w i t h the nonmagnetic steel employed f o r the major rotor parts: inner rotor body, torque tube and outer cylinder. As outlined in [ I ] , the nonmagnetic steel X3CrNiMnMoNbN19 16 5 i s used as the rotor material. Large trial pieces o f several tons i n weight have been forged and tested t o prove strength, toughness and

Rp0.2.Rm in Wmm2

Rupture Strength

-

Radial Trepan

I

mi I" ---

Singk Specimen

hgential/Axbl)

'\

¹ Yield Stress

I

Elongation

Fig. 8 Distribution o f material properties i n a 4.7 Mg Fig. 9 Welding o f long cylinders test forging

CL-716 JOURNAL DE PHYSIQUE

homogeneity and t o optimize the production process. Fig. 8 shows the distribution of material properties over the cross section of a large forging. Thanks t o these investigations, the quality of the even larger forgings for the SC test rotor has been improved significantly. Machinability and weldability of the extremely tough nonmagnetic steel have been thoroughly investigated. Fig. 9 gives an impression of the welding process which has been developed t o connect thick cylinders t o the long cylindrical support structure of the SC test rotor.

The vibrational behavior of the multi-shell rotor, which IS the typical structure o f superconducting rotors, has been extensively tested with rotor models [ I ] . The results clearly Indicate the importance o f large-scale experiments which can reveal design imperfections not detectable b y analytical methods. I t is intended t o continue the investigations for the analysis of vibrational stability problems and balancing procedures b o t h w i t h special models and w i t h the SC test rotor i n different phases o f manufacture.

3 QUENCH DETECTION A N D PROTECTION

The conductor used for the superconducting field winding has a l o w matrix-to-superconductor ratio. It is therefore essential t o detect a quench within a very short time t o prevent the conductor from excessive heating. A time o f 100 ms should n o t be exceeded. Assuming normal quench propagation velocities, this limit is associated w i t h a normal zone of about 1 m and a voltage of about 100 m V across this length. Therefore quench detection has t o be quick and highly sensitive.

On the other hand, there are several interference voltages at the winding which make quench detection rather difficult. To highlight the problems involved In measuring 100 m V as a threshold for quench protection, the order of magnitude of various voltages occuring during operation has t o be considered:

DC voltage (inductive) during excitation and discharge 1000 V AC voltage induced by harmonics of the DC power supply 1 0 V A C voltage induced by the armature winding during transients 1 O V DC voltage due t o heat losses in the conductor durlng excitation and de-excitation 10 V DC voltage drop at soldered joints between coils and coil sections 0.1 V

Balance Resistors

Field Preamplifier

Rectifier

The protection circuit chosen is depicted i n Fig. 10. Any interference voltages are compensated b y a bridge connection o f the t w o halves of the winding. This method o f detecting quenches has been successfully applied t o stationary super- conducting coils. The signal can be transmitted t o the station- ary part o f the protection device as a preamplified analog signal or as a binary signal. This signal is used t o trip the unit, t o block the converter and t o de-excite the field winding rapid- ly. The magnetic energy will be removed from the rotor and dissipated t o a large extent i n the field discharge resistor.

- - - -

^-

-

-

-

-

Fast Deexcitation Threshold

Comparator 4 ACKNOWLEDGMENT

- - -

-

-

-

- -

The developmen: work has been sponsored by the Bundes- ministerium fiir Forschung und Technologle (BMFT).

Converter of Exciter System

& - -

- -

- - -

- -

Generator Trip

Fig. 10 Quench protection circuit for superconducting feld winding

REFERENCES:

[ I ] Lambrecht D. Status o f Development of Superconducting A C Generators, IEEE Trans. Vol. M A G 17 (1981) 1551

[2] Lambrecht D. Die Entwicklung supraleitender Generatoren fljr den Kraftwerksbetrieb, Paper presented at the A.I.M..

International Meeting o n Modern Electric Power Stations, LiBge, Belgium, Oct 26/30,1981. .

L3] Lambrecht D. et al, A n Advanced 700 M V A Turbine Generator i n its Own Independent Test Field for Permanent Testing and Development

-

A New Way t o Improve Consistently Turbine Generator Capability, Economy and Reliability, ClGRE Session 1982, Paper 11-03, Paris, 1-9 Sept., 1982.

[4] lntichar L., Lambrecht D. Technical Overview of the German Program t o Develop Superconducting A C Generators, Paper presented at the Applied SuperconductivitY Conference, Knoxville, Tennessee, December 3,1982.

[5] Liese M. et al, Comparison of Vector Potential and Extended Scalar Potential Methods and True Three-Dimensional Magnetostatic Field Calculation, Paper presented at the IEEE/PES Summer Meeting, Los Angeles, 1983.

[6] Neumiilfer H.W. Losses i n Superconducting Multi-Filamentary Roebel Bars, IEEE Trans. Vol. MAG 17 (1981) 2274