25th International Conference on Electricity Distribution Madrid, 3-6 June 2019

Paper n° 1606

COMPARATIVE RESEARCH BETWEEN XLPE AND P-LASER MV-CABLE

Piet SOEPBOER Blandine HENNUY Jos VAN ROSSUM

Enexis Netbeheer – the Netherlands ENGIE-Laborelec - Belgium Prysmian Netherlands – the Netherlands
[email protected] [email protected] [email protected]

Tjeerd BROERSMA Robin SIMAL Sander LAUWERS

Enexis Netbeheer – the Netherlands ENGIE-Laborelec - Belgium Prysmian Netherlands – the Netherlands
[email protected] [email protected] [email protected]

ABSTRACT VERIFICATION TESTS

A few years ago, Cable manufacturer Prysmian developed During the verification test it was concluded that both
a new insulation material for MV-cables called “P-laser” cables for the comparative research fulfil the require-
which is based on PP instead of PE [1]. Dutch DSO ments of standard NEN-HD 620 S2 and that valid type test
Enexis, research institute ENGIE-Laborelec and Prysmian certificates and test reports are available. Dimension
Netherlands together set up and conducted a research measurements according to IEC 60811-1-1 [4] were
project to investigate the behaviour of a P-laser cable in performed on straight cable samples. The reference
comparison with an identical cable with XLPE-insulation. thicknesses are given in table 1.
Based on the outcome of this study Enexis decided to
introduce P-laser as one of their standard cable types. Requirement XLPE P-Laser
Conductor screen ≥ 0.5 mm 0.623 mm 0.637 mm
INTRODUCTION Insulation (nom 3.4 mm) ≥ 2.96 mm 3.530 mm 3.620 mm
The cable tender of Enexis in 2016 was partially awarded Insulation screen ≥ 0.5 mm 0.629 mm 0.627 mm
to Prysmian Netherlands with the possibility to purchase
cable with P-laser insulation. The advantages of this “High Table 1: Reference thicknesses.
Performance Thermoplastic Elastomer” insulation are e.g.
its recyclability and lower CO2-emissions during The average insulation thickness of the XLPE cable under
production. Enexis however learned from its experiences test appeared to be 3.53 mm and the average insulation
after purchasing the 1st generation XLPE cable in the ‘70’s. thickness of P-laser 3.62 mm where the standard requires
To prevent a similar scenario, a research project was set up a nominal thickness of 3.4 mm. A theoretical verification
by Enexis in collaboration with ENGIE-Laborelec and of this small overdimensioning showed an insignificant
Prysmian in order to investigate the behaviour of a influence of less than 1% on the cable ampacity.
1x630Al P-laser 6/10kV cable in comparison with an
identical XLPE cable.
The thermal conductivity was determined by the “Hot Disc
Based on earlier studies on P-laser and based on how MV- TPS” Method according to ISO 22007-2 [5] and out of
cables are installed and operated in the network of Enexis, these results it was concluded that the measured thermal
the following tests were determined and performed: resistivity of P-laser (4.55 K.m/W) is in line with the value
in the available literature (4.5 K.m/W).
1. Verification with standard NEN-HD 620 S2 [2] and
setting reference values for cable comparison. The DSC analysis showed that the observed melting
2. Ampacity-in-air test (comparison of temperatures and temperatures are those that can be expected for XLPE and
cable loads of XLPE and P-laser).
3. Overbending test (cable behaviour under a too small HPTE and that these are sufficiently higher than the
bend radius at -15 °C, 90 °C and 110 ° C). nominal operating temperatures of XLPE and P-laser.
4. Short-circuit test (short-circuit behaviour up to 250 °C
on the minimum permissible bend radius). The detected glass transition temperatures (-31°C to -
49°C) of P-Laser and XLPE cable during the DMA
After these electrical and mechanical tests, DSC, DMA, analysis are significant lower than the lowest temperature
dimension measurements and stereomicroscopy analysis ever measured in the Netherlands (-27.4°C on 27 January
were performed by ENGIE-Laborelec to study the material 1942 in Winterswijk). From that it can be concluded that
properties.
there is no risk for both materials in relation to the usual
An investigation on watertree sensitivity was left out storage and processing temperatures of cables at Enexis.
because this was already investigated by Enexis and
ENGIE-Laborelec in a preliminary phase showing not a AMPACITY-IN-AIR TEST
single watertree in P-laser insulation [3].
A piece of P-Laser cable and a piece of XLPE cable with
lengths of 8 meters were installed in a test loop according

CIRED 2019 1/5

 25th International Conference on Electricity Distribution Madrid, 3-6 June 2019

Paper n° 1606

to figure 1. Three thermocouples (Tc) were placed on each OVERBENDING TEST

cable with a spacing of 1 m in between. The current and
the conductor temperatures were recorded by means of a The performed overbending test is a self-defined test in
data logger. A current transformer was used to inject the order to investigate and compare the cables’ behaviour
current until the desired temperature was reached (tests under circumstances that might be present in the practise
were conducted at conductor temperatures of 30 °C, 60 °C, of cable installation e.g. the possible cable storage at low
90°C and 110°C). After heating up to the relevant temperatures and the impact of possible installation by
conductor temperature, the temperatures were kept stable means of a bend radius that is too small. For this purpose,
for 2 hours by "freezing" the current. the test set-up in figure 3 was chosen. Cable pieces of 6 m
long were installed in a bend with a diameter equal to 2 *
0.75 * the minimum permitted bend radius (2 * 0.75 * 0.9
= 1.35 m). The test loop was closed with an equal cable
equipped with a thermocouple to measure the temperature
on the conductor. During the overbending test, the same
cable piece was bent and straightened several times,
underwent several temperature changes and was subjected
to high voltage AC-tests and partial discharge tests
according to NEN-HD 620 S2. The complete overbending
test program is described in table 1. To achieve the high
temperatures, a current transformer has been used. To
reach the low temperature, a cold room has been used. The
ambient temperature was approximately 20 °C.

Figure 1: Test loop for the ampacity test.

Assuming a max. permissible equal conductor temperature

of 90 ° C, the ampacity of the tested P-laser cable is lower
than that of the XLPE cable (1046 A vs. 1156 A). This is
due to the higher thermal resistivity of P-laser. However,
considering the max. permissible conductor temperature
based on P-laser’s properties, the P-laser cable can be
loaded up to a nominal temperature of 110 °C where XLPE
is limited to 90 °C. Based on these nominal temperatures
the tested P laser cable has a higher ampacity than XLPE
(1190 A vs. 1156 A). Furthermore, it was found that the P-
Laser cable showed a 7°C higher core temperature than the
XLPE cable at its maximum permissible current i.e. at its
nominal temperature of 90 °C. A study on the impact of Figure 3: Test loop for the overbending test.
this higher temperature on accessories was not included in
the comparative research project because a relevant study
on three MV-joint types with satisfactory outcome was
already performed in 2007 by Prysmian and Dutch DSO
Liander [6, 7, 8].

Table 2: Overbending test programme.

Figure 2: Temperature as a function of current.

CIRED 2019 2/5

 25th International Conference on Electricity Distribution Madrid, 3-6 June 2019

Paper n° 1606

The XLPE and P-laser cable samples both passed the couple to measure the temperature on the conductor. A
overbending test showing an equal behaviour on PD-level first short circuit test was conducted at a short-circuit
and PD-noise level. No irregularities were found in the current of 31.5kA during 2.7s. These values are prescribed
afterwards performed DSC and DMA analysis by ENGIE- by Enexis in specifications and can be seen as the worst-
Laborelec. Compared to the reference samples, both the case scenario occurring in the network, including a
selected XLPE and P-laser samples however showed some significant safety margin. For the second short-circuit test,
dimensional differences after the overbending test: values of 50kA/2.25s were chosen to achieve a short-
• The thickness of the conductor screen became larger for circuit end-temperature of 250 °C within 5 seconds. This
the XLPE sample after the overbending test. At the P- short-circuit temperature for XLPE was retrieved from
laser sample the thickness of the conductor screen standard NEN-HD 620 S2. Both P-laser and XLPE passed
became a bit smaller. the short-circuit tests based on the normative criterion of
• The insulation thickness of the P-Laser sample became a "no breakdown". Both samples also passed the included
bit larger after the overbending test. No significant high-voltage test and met the requirements regarding noise
difference was noted for the XLPE sample. level and PD level regarding the partial discharge
• The thickness of the insulation screen of the P-Laser measurement.
sample after the overbending test became a bit smaller.
No significant difference was noted for XLPE. No irregularities and no differences between P-laser and
XLPE were observed after the 31.5kA/2.7s short-circuit
These dimensional changes were probably caused by the test and successive DSC, DMA, dimension measurements
extreme bending of the test samples vs. unbended and stereomicroscopy analysis. After the 50kA/2.25s
reference samples, the cable clamps used for fixing the test short-circuit test, first it was observed that the insulation of
samples (reference samples never were clamped) or by the both XLPE and P-laser was shrunk with respect to the
high temperature and the thermoplastic character of P- conductor.
laser. In NEN-HD 620 S2 there are no dimension
requirements after carrying out type tests. The measured After the 50kA/2.25s short-circuit test, another difference
thicknesses after the overbending test of the conductor was visually determined between the shrinkage behaviour
screen, insulation and insulation screen of both the XLPE of P-laser with respect to XLPE (see figure 5). In P-laser,
and P-Laser samples however still fulfilled the minimum the conductor screen and the insulation bonded thereto had
"design requirements" of NEN-HD 620 S2. Based on the shrunk back approximately 12mm further than the
above, the measured thicknesses are considered as insulation screen and the insulation bonded thereto.
acceptable and the observed dimensional differences are
considered as insignificant.

SHORT-CIRCUIT TEST
Due to the thermoplastic nature of P-laser, the short-circuit
behaviour was investigated. No short-circuit tests are
defined in standard NEN-HD 620 S2. Therefore, it was
decided to perform a short-circuit test based on IEC 61442
[9] and HD 629-1 [10]. According to these standards for
cable accessories, 2 short-circuit tests were carried out. For
this test, the test set-up in figure 4 was chosen. Cable
pieces of 6 m long were installed in a bend with a diameter Figure 5: Observed shrink behaviour at 50kA/2.25s.
equal to 2 times the minimum permitted bend radius (2 *
0.9 = 1.8 m). The test loop was equipped with a thermo- This difference in "shrinkage shape” is probably caused by
the thermoplastic nature of P-laser and the high end-
temperature of 250 °C achieved during the test. This
temperature is above the "melting point" of P-laser. The
impact of the observed shrinkage shape of P-laser in
practice is however negligible as this phenomenon only
occurs at the end of the cable i.e. where the cable is
terminated in an accessory. In this case, the stress control
in this part is taken over by the accessory and this (peeled)
cable part is located within the stress controlling body of
the accessory at a position where this does not cause risks
regarding partial discharges or a disturbed electrical field.
Figure 4: Test loop for the short-circuit test.

CIRED 2019 3/5

 25th International Conference on Electricity Distribution Madrid, 3-6 June 2019

Paper n° 1606

During the dimensional comparison of the cable samples, national and international standards committees for
located at a distance from the cable clamps, no significant medium-voltage cables.
differences were observed between the reference values
and the values after performing the short-circuit test. This Based on the obtained test results, additional calculations
was valid for the conductor screen, the insulation, the
insulation screen, and the outer sheath. All measured on the conductor end-temperatures and occurring forces in
values were within the limits of the design criteria imposed various short-circuit situations have been performed as a
in NEN-HD 620 S2. This is valid for both the XLPE and risk analysis regarding the shrinkage and deformation
the P-Laser cable as well as the samples of the 31.5kA/2.7s behavior of P-laser. These calculations showed that the
and the 50 kA/2.25s short-circuit test. short-circuit force, using the Enexis specification, is
minimum 3.88 times lower than the force that occurred
Regarding the cable samples located beneath and directly during the 50 kA/2.25s short-circuit test. Furthermore, the
next to the cable clamps, both the XLPE and the P-Laser calculations showed that the conductor end temperature
cable appeared strongly distorted after the 50kA/2.25s during the short circuit at the Enexis specification, ranges
short-circuit test (figure 6). For P-laser both the internal
semiconductive screens and the insulation no longer between 53°C and 145°C where at the comparative 50
complied with the values in the design requirements of kA/2.25s short-circuit test an end temperature of 250°C
NEN-HD 620 S2, where this was also the case for the was used. Taken this into account, it can be concluded that,
insulation of the XLPE cable. This may be due to the even assuming the worst-case scenario in the network of
thermoplastic character of P-Laser and the high Enexis, the chance of deformation of the insulation and
temperature of 250 °C at the short-circuit test. This makes semiconductive layers of P-Laser and XLPE is negligible.
the different layers highly deformable in combination with
the high mechanical stress caused by the short-circuit CONCLUSION
forces. In this respect it should also be noted that IEC
60986 recommends that the end-temperature of Under practical conditions, P-laser and XLPE show
thermoplastic insulation materials should be reduced by 10 similar performance. Based on the recyclability of P-laser
°C when the cable is clamped. This was not taken into cable and the results of this comparative research and
account in the tests carried out on P-laser for the purpose
former studies, Enexis has decided to standardize P-laser
of a fair comparison with XLPE. After the 31.5kA/2.7s
short-circuit test, both the XLPE and P-laser samples insulated cable for utilization in its MV-network next to
beneath and directly next to the clamps showed no the already existing standard XLPE-cable.
dimensional changes with respect to the reference values.
REFERENCES

[1] Prysmian & Liander, CIRED 2011, Paper 0345,

Introducing High-Performance polypropylene Ther-
moplastic Elastomer insulation for MV-cables in the
Netherlands.

[2] NEN-HD 620 S2: 2016; Distribution cables with

extruded insulation for rated voltages from 3,6/6 (7,2)
kV up to and including 20,8/36 (42) kV (Netherlands
Figure 6: Example of distorted shape at P-laser and XLPE implementation from HD 620 S2 of part 1 and section
after the 50 kA/2.25s short-circuit test. 10J (revision).
Regarding the thickness of the outer-sheath the measured [3] Laborelec, 2015, “Enexis: Smartlife - Quality check
values at both the P-laser and the XLPE cable were within
of 5 MV cables”.
the limits of the design criteria imposed in NEN-HD 620
S2. Compared to the DSC and DMA analysis on reference
samples and samples after the overbending test, DSC and [4] NEN-EN-IEC 60811-1-1: 1996 +A1: 2001; Insulating
DMA analysis on samples after the short circuit test did and sheathing materials of electric cables - Common
not lead to new insights and conclusions. test methods - Part 1-1: General application –
Measurement of thickness and overall dimensions -
CONTEMPLATION Tests for determining the mechanical properties.”
During visual inspections carried out after the conducted [5] ISO 22007-2: 2015; Plastics - Determination of
tests, a deviation of material thicknesses was observed thermal conductivity and thermal diffusivity - Part 2:
several times compared to the "design requirements" in Transient plane heat source (hot disc) method.
NEN-HD 620 S2. This raised the question whether the
dimension requirements in the "design requirements"
should also be applicable after the execution of (type) tests. [6] Prysmian & Liander, 2007, “Compatibility Type Test
We would like to address this issue to the members of the of ABB Kabeldon joint on XLPE and P-laser

CIRED 2019 4/5

 25th International Conference on Electricity Distribution Madrid, 3-6 June 2019

Paper n° 1606

“Kudika” MV cable, Prysmian TVE07-17.1.

[7] Prysmian & Liander, 2007, “Compatibility Type Test

of Raychem joint on XLPE and P-laser “Kudika” MV
cable, Prysmian TVE07-18.1.

[8] Prysmian & Liander, 2007, “Compatibility Type Test

of Prysmian joint on XLPE and P-laser “Kudika” MV
cable, Prysmian TVE07-19.1.

[9] NEN-EN-IEC 61442: 2005; Test methods for acces-

sories for power cables with rated voltages from 6 kV
(Um = 7,2 kV) up to 30 kV (Um = 36 kV).

[10] NEN-HD 629-1, 2007; Test requirements on acces-

sories for use on power cables of rated voltage
3,6/6(7,2) kV up to and including 20,8/36(42) kV -
Part 1: Cables with extruded insulation.

CIRED 2019 5/5