The effects of stress relief annealing on the magnetic properties of cut laminations and assembled cores produced from nonoriented electrical steel

The effects of stress relief annealing on the magnetic properties of cut laminations and assembled cores produced from nonoriented electrical steel

Journal of Magnetism and Magnetic Materials 19 (1980) 65-68 © North-Holland Publishing Company THE EFFECTS OF STRESS RELIEF ANNEALING ON THE MAGNETIC...

215KB Sizes 0 Downloads 6 Views

Journal of Magnetism and Magnetic Materials 19 (1980) 65-68 © North-Holland Publishing Company

THE EFFECTS OF STRESS RELIEF ANNEALING ON THE MAGNETIC PROPERTIES OF CUT LAMINATIONS AND ASSEMBLED CORES PRODUCED FROM NONORIENTED ELECTRICAL STEEL R.G. HANCOCK GKN Sankey Ltd, Bankfield Division, Greenway Road, Bilston, West Midlands WV14 OTJ, UK

The strain was caused by a number of longitudinal cuts and expressed empirically

1. Introduction Fully processed electrical steel is sold on the basis of losses measured by the "Epstein Test". The creation of strain by mechanical working, blanking or welding, can result in an overall deterioration of the magnetic properties. The use of both submerged arc and TIG welding of stator cores is an established production process and provides an economical, practical, good quality product. There are, however, certain applications where the strains cannot be tolerated. The object of this presentation is to quantify the effect of such strain and thus provide assistance to machine designers. Methods of overcoming these problems are shown, which is of growing importance when energy conservation is considered.

Strain c~ =

cut length (cm) surface area (cm 2)

-

CL SA'

-

-

in the case shown in fig. 2: (25 + 25 + 25 + 25) =0.571. 7X25 Deterioration in magnetic performance is expressed as % increase on base value (no cuts), see figs. 5 and 6. 3.2. Conclusions

1. Deterioration in magnetic properties is caused by cutting strains. In areas of high cutting strains, i.e. stator teeth, deterioration is very significant. 2. Reannealing very much reduces the deterioration but some permanent effect exists, particularly at high strains.

2. Comparison between sheared and blanked edges Samples used in the exercise on cutting strains were sheared, whereas laminations in practice are blanked, including those in the welding tests. Photographs show the amount of "cut" and "break" to be approximately the same for both methods, hence, the effects can be assumed to be the same. 3. Cutting strains

Stator pack

3.1. Test method

~rlngle strip u n i t I - RM$ Ammeter

Samples were measured for total specific loss and magnetising force in a single strip test yoke, see figs. 1,2 and 3.

f

I

- Frequency counter

V - H i g h impedance voltmeter

Fig. 1. Diagram of the measuring circuit. 65

66

R. G. Hancock / Stress relief annealing

%W/kg increase 5O

:°2)I

40

(

30 20 lO 0 %w/k9 increase 5O

cuts

Fig. 2. Cutting strains.

!

l

I

l

1

2

3

4

CL

SA

4O ~

4OOHz b

30 20

2o.. )

10 0

4OOHz ~_ | 1

| 2

| 3

| 4

CL

SA %W/kg in'crease 50 40 3O 20 10

Fig. 3. Test equipment.

0

| I

! ~_

|

3

i 4

Ck SA

Fig. 5. Deterioration in magnetic performance, expressed as percentage increase on base value (no cuts), specific total loss 1.0 T, fully processed; (a) 1% silicon steel, 0.50 mm thick, (b) 2.3% silicon steel, 0.50 mm thick, (c) 2.8% silicon steel, 0.35 mm thick. M89netiling wlndinQ

;.m°,. C"m"

'o,.

w,n°,o,

• 2Bcm)..D

" "

"

[]

3. Practical m a c h i n e tests s u b s t a n t i a t e t h e value o f stress relief annealing.

,.,.

4. W e l d i n g s t r a i n s

4.1. Test m e t h o d {

\\\\\\\7

Idlminllled nickel/iron yoke

Fig. 4. Single strip yoke, cross section through yoke.

The packs were c o n s i d e r e d as ring samples a n d "allowance m a d e in c a l c u l a t i o n s o n losses for mass o f non fluxed teeth.

R.G. Hancock / Stress relief annealing

67

%H A i m increase 250

200

1

150

100

50

0

I 1

I 2

I 3

l 4 SA

%H A / m increase 250 --

200 --

Fig. 7. Test equipment, submerged arc welded stator pack

b

and TIG welded stator pack.

150

100

50

0

-}i2

I

3

4

Specific total loss W/k o 30

CL SA

a

25 %J'1A/m increase 20

250

15

200

c 10

150

Cond~tmns 3, 5 & 6 5--

100

0

50

O.6

I

|

I

I

0.7

0.8

0.9

1.0

induction (Tesla) 1

~-

3

4

CL SA

Fig. 6. See fig. 5, magnetising force 1.0 T.

Speclfic total loss W/kg 1.4

,.2 Tests were carried out on stator packs as detailed below, see also fig. 7.

b

~

1.0

°.8

4. 2. Submerged arc welding

O.6

J

CO~iltOfl5 3 g & 6

Material: 2.8% silicon steel; condition 1: as blanked laminations, condition 2: as blanked laminations, welded, condition 3: laminations annealed, condition 4: laminations annealed, welded, condition 5: as 2, annealed, condition 6: as 4, annealed.

0.4

0.2

0.0 0.6

I

I

I



0.7

0.8

0.9

1.0

InducliOn (Tesla)

Fig. 8. Specific total loss for the submerged arc welded stator pack at (a) 400 Hz and (b) 50 Hz,

R.G. Hancock / Stress relief annealing

68

Table 1 TIG welded stator packs, specific total loss 1.5 T at 50 Hz

Table 2 TIG welded stator packs, magnetising force 1.5 T at 50 Hz

Pack Loose Welded packs Annealed packs no. laminations (W/kg) (W/kg) (% INC) (W/kg) (% INC)

Pack Loose Welded p a c k s no. laminations (HRMS) ( H R M S ) (%INC)

Annealed packs

A B C D

A B C D

1067 1266 1151 1386

7.56 7.49 7.39 8.01

9.05 9.13 8.62 9.23

20 22 17 15

35.4 40.9 38.6 46.6

370 445 422 482

449 390 406 412

589 539 548 557

31 38 35 35

(HRMS) (%INC) 138 225 183 236

N.B. Pack A after unsticking gave a loss of 7.96 W/kg.

4.3. T.L G. welding Material: silicon free-semi processed steel laminations annealed and steam blued. Tests on four stator packs: A, B, C and D.

4.4. Conclusions 1. Submerged arc and T.I.G. welding both result in deterioration o f magnetic properties, see fig. 8 and tables 1 and 2. 2. Reannealing welded packs where laminations have interlaminar insulation which will withstand the heat treatment is a very satisfactory method of virtually eliminating the welding effects and at the

same time dealing with cutting strains. This has been substantiated on production machines operating at 400 Hz. 3. Reannealing welded packs where interlaminar resistance will not withstand the heat treatment is not recommended unless remedial processes are used to unstick laminations, but these can be unreliable and costly.

References [1] K.H. Schmidt, J. Magn. Magn. Mat. 2 (1975/1976) 136. [2] John Lysaght (Aus.) Ltd., Technol. Rept. no. 2702.