Evaluation of the XLPE MV lifetime Boris Dardel

WETS D’15 Workshop Organization: Jicable and Prospective 21OO

Palais des Congrès de Versailles, France Thursday, 25 June 2015

WETS D’15 2.3 Dardel

60kV cable with vapour crosslinking: 2 samples

Production Cortaillod 1973 Graphite outer semicon

Conductor Copper Tan δ @ 1.5U0 4.5 10-4 1.8 10-4

Tan δ @ 2.0U0 7.2 10-4 3.5 10-4

Breakdown at 90kV 5 min 2 min PD 10 pC 40 pC

Analysis of old cables

Analysis of old cables

60kV cable with vapour crosslinking:

Production Cortaillod 12/78

Production Cossonay 1978?

Conductor Copper Tinned copper Tan δ @ 1.5U0 4.9 10-4 7.9 10-4

Tan δ @ 2.0U0 6.2 10-4 5.7 10-4

Breakdown at 90kV 10 min 12 min PD 50pC 50pC

What about first synthetic cables?

0

2

4

6

8

Bre

akdo

wns

(#

/100

km/y

ear)

Year of manufacture

Vapour crosslinking Graphite outer SC

PVC sheath Extruded outer SC Strippable

Dry crosslinking Triple extrusion

Non-strippable SC Laminated sheath

Data in the litterature

Defects in insulation

3 types of defects can be present: 1. Point defects (inclusions/impurities) that lead to a

rapid breakdown and determine the lifetime of the insulation

2. Voids that reduce the stress ageing resistance of the material.

3. Irregularities on the SC/insulation interface

Selection criteria for routine test

15 kV/mm voltage test during 30 minutes

• No partial discharges • Breakdowns as well as PD are localized and analyzed for corrective

action

Elimination of point defects

A voltage test is done at a high stress and, if a breakdown occurs, the time to breakdown is recorded (routine test).

15kV/mm

1E-05

0.0001

0.001

0.01

0.1

1

0.01 0.1 1 10 100 1000

Cu

mu

lati

ve p

rob

abili

ty

Time [min]

Break-down statistics over 2000 tons of XLPE material

15kV/mm

Routine test at15 kV/mm (4 to 6 U0 depending on insulation thickness). The mean length per breakdown is 320km single phase (1 ton approx. corresponds to 1km - phase cable)

Determination of n

200 tons were tested at 20 kV/mm

15kV/mm

One year

1E-05

0,0001

0,001

0,01

0,1

1

0,01 0,1 1 10 100 1000 10000 100000 1000000

Cum

ulat

ive

prob

abili

ty

Time [min]

Break-down statistics with 2 different levels of proof test

15kV/mm 20kV/mm Equivalent 15kV/mm, n=14

In the present case n is estimated to be around 14

E.t1/n = constant

Defect distribution after the proof test

• The largest defects are removed by the proof-test • The residual defects could be detected if the voltage application was 30

minutes or more • These are quite rare and will define the cable lifetime.

One year

15kV/mm

1E-05

0.0001

0.001

0.01

0.1

1

0.01 0.1 1 10 100 1000 10000 100000 1000000

Cum

ulat

ive

prob

abili

ty

Time [min]

Troncature effect on Weibull Statistics

Before proof test

After proof test

Ageing of residual defects

3kV/mm

One year Fifty years

0,0001

0,001

0,01

0,1

1

10

1,E-01 1,E+00 1,E+01 1,E+02 1,E+03 1,E+04 1,E+05 1,E+06 1,E+07 1,E+08 1,E+09 1,E+10

Cum

ulat

ive

prob

abili

ty

Time [min]

Tronquated population evolution after ageing

• If we suppose an n value of 10 to take into account the service conditions (higher temperature), then the extrapolated lifetime at 3kV/mm of the routine tested cable is far above 50 years …

IEC 60502.2 3.5Uo - 5min

Conclusions

Since the 90’s, actions have been taken to warrant a lifetime of the cable over 50 years.

• Material selection • Processing • Testing

At the conditions that the laying and the accessories installation are done properly, that the accessories are of good quality and that the maintenance are done properly. For cables that are more than 30 years old, we will learn progressively about the existing problematic. A good evaluation of the characteristics (original values, service conditions, …) of the cables that are replaced is then necessary and will require a strong collaboration between the manufacturers and the operators.

What can be expected

0%

2%

4%

6%

8%

10%

12%

0 10 20 30 40 50 60Temps [années]

Pro

babi

lité

de c

laqu

age

1970

1980

2000

1970

19751980

19851990

1995200020050%

1%2%3%4%5%6%7%8%9%

10%11%12%13%14%15%16%17%18%19%20%

19701980

19902000

20102020

20302040

20502060

Prob

abili

té d

e cl

aqua

ge

Thank you for your attention