benchmarking of cable condition monitoring methods ... · 6 different cables xlpe/cspe ......
TRANSCRIPT
![Page 1: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/1.jpg)
Benchmarking of cable condition monitoring methods – lessons learned from a recent IAEA
co-ordinated research programme
Dr Sue BurnayJohn Knott Associates Ltd, UK
![Page 2: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/2.jpg)
John Knott Associates Ltd 2
Objectives of IAEA programme
Identify practical condition monitoring methods for cable insulation & jacket materials of interest to NPPs Methods that can trend degradation Suitability for ageing assessment
Assess the reproducibility of these methods Benchmarking using different test laboratories Identify ways to improve test methods
![Page 3: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/3.jpg)
John Knott Associates Ltd 3
Participants & materials
17 active participants from 13 different countries, plus several observers
6 different cables XLPE/CSPE (Rockbestos, USA) EPR/EVA (Eupen, Belgium) PEEK/XLPO (Habia, Sweden) SiR/SiR (Hew, Germany) XLPO/XLPO (Shanghai Special Cable, China) EPR/EPR (Changzhou Bayi Cable Co., China)
![Page 4: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/4.jpg)
John Knott Associates Ltd 4
CM methods included in programme
Mechanical Elongation at break (EAB) Indenter (IM) Recovery time
Chemical/thermal Oxidation induction time
(OIT) Oxidation induction
temperature (OITP) Thermogravimetric analysis
(TGA) Fourier transform infrared
spectroscopy (FTIR)
Electrical Dielectric loss (Tan delta) Dielectric spectroscopy Insulation resistance (IR) Time domain reflectometry
(TDR) Frequency domain
reflectometry (FDR)
Other Ultrasonic velocity Density
![Page 5: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/5.jpg)
John Knott Associates Ltd 5
Overall plan
Each lab decided which materials and CM methods to test
Labs prepared their own samples from whole cable
Test method agreed at start of programme Comparison of unaged data initially, method revised if
necessary Labs did thermal ageing at agreed temperature Limited no. of radiation aged samples Data collated in standard format
![Page 6: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/6.jpg)
John Knott Associates Ltd 6
Main concern – variability between labs
Initial results on unaged material – update to test methods
On aged materials, variability was considerable for some CM methods
Important to understand reasons for variability
![Page 7: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/7.jpg)
John Knott Associates Ltd 7
Variability between labs – EAB (1)
CSPE jacket (Rockbestos)
0
100
200
300
400
500
600
0 500 1000 1500 2000 2500
Ageing time at 120˚C (hr)
EAB
(%)
AMSLABOR-SCKVEIKIVUJENIST
![Page 8: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/8.jpg)
John Knott Associates Ltd 8
Variability between labs – EAB (2)
EPR Insulation (Eupen) - Black
0
100
200
300
400
500
600
700
0 2000 4000 6000 8000 10000
Ageing Time at 135˚C (hr)
EAB
(%)
AMS
CNEA
LABOR-SCK
SECRI
VEIKI
VUJE
UJV
NIST
![Page 9: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/9.jpg)
John Knott Associates Ltd 9
Variability between labs – EAB (3)
EVA jacket (Eupen)
0
100
200
300
400
500
600
0 4000 8000 12000
Ageing time at 120 C (hr)
EAB
(%)
AMSLABORSECRIVEIKIVUJECNEANISTUJV
![Page 10: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/10.jpg)
John Knott Associates Ltd 10
Cable types
CSPE jacket – reasonably uniform thickness, easy to cut dumb-bell samples for EAB
EVA jacket – non-uniform thickness, moulded around insulated wires
EPR insulation on solid tinned Cu – very difficult to strip
![Page 11: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/11.jpg)
John Knott Associates Ltd 11
Reasons for variability - EAB
Test method – type of extensometer used Optical extensometer Clip-on extensometer Cross-head movement
Sample preparation – biggest source of variability Dumb-bell samples – removal of surface irregularities,
dual layers, wrap materials Tubular samples – method used to extract wire from
insulation Difficult to prepare specimens from aged cable
![Page 12: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/12.jpg)
John Knott Associates Ltd 12
Variability between labs – IM (1)
CSPE Jacket (Rockbestos)
0
5
10
15
20
25
30
35
0 500 1000 1500 2000
Ageing Time at 120⁰C (hr)
IM (N
/mm
)
AMS
KHNP
LABOR
AECL
INSS
![Page 13: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/13.jpg)
John Knott Associates Ltd 13
Variability between labs – IM (2)
XLPO Insulation (Shanghai)
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
0 1000 2000 3000 4000 5000 6000 7000
Ageing Time at 135⁰C (hr)
IM (N
/mm
)
AMSINSS (corrected)SECRIAECL
![Page 14: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/14.jpg)
John Knott Associates Ltd 14
Variability between labs – IM (3)
XLPO Jacket (Shanghai)
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Ageing Time at 135⁰C (hr)
IM (N
/mm
)
AMSSECRILABORINSS (corrected)AECL
![Page 15: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/15.jpg)
John Knott Associates Ltd 15
Reasons for variability - IM
Probe tip diameter (correction to standard) Sample temperature Force range used for analysis Clamping force Variation in sample thickness
![Page 16: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/16.jpg)
John Knott Associates Ltd 16
Recovery timeXLPO jacket (Shanghai)
0
1
2
3
4
5
6
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Ageing time at 135⁰C (hr)
Reco
very
tim
e (s
)
AECL
Only 1 lab testing this method Often shows larger change than for indenter e.g for this XLPO,
IM changes by factor of 2 and recovery time by factor of 5
![Page 17: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/17.jpg)
John Knott Associates Ltd 17
Variability between labs – OIT (1)
EPR insulation (Eupen)
0
10
20
30
40
50
60
70
80
90
100
0 1000 2000 3000 4000 5000 6000 7000
Ageing time at 135⁰C (hr)
OIT
at 2
40⁰C
(min
)
KHNP-green (250C)
SECRI-green
SECRI-black
SECRI-blue
SECRI-brown
NIST-black (230C)
NIST-green (230C)
AMS-yellow
![Page 18: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/18.jpg)
John Knott Associates Ltd 18
Variability between labs – OIT (2)
EPR insulation (Changzhou)
0
10
20
30
40
50
60
0 1000 2000 3000 4000 5000 6000 7000
Ageing time at 135˚C (hr)
OIT
at 2
40˚C
(min
)
SECRI
NIST
AMS
![Page 19: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/19.jpg)
John Knott Associates Ltd 19
Variability between labs – OITP
EPR insulation (Eupen) - yellow/green
150
170
190
210
230
250
270
290
310
0 1000 2000 3000 4000 5000 6000 7000
Ageing time at 135 C (hr)
OIT
P ( ⁰C
)
VUJE-blue
SECRI-Blue
SECRI-yellow/green
SECRI-brown
KHNP-yellow/green
NIST-black
NIST-yellow/green
SECRI-black
![Page 20: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/20.jpg)
John Knott Associates Ltd 20
Reasons for variability – OIT/OITP
Must use same set temperature throughout series of tests for OIT
Sample preparation for both OIT and OITP Particle size and packing Selection of sampled material (colour, dual layers) Analysis method – specified in IEC/IEEE standard but
may need to be done manually
![Page 21: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/21.jpg)
John Knott Associates Ltd 21
Variability between labs – FTIR
XLPE insulation (Rockbestos)
0.000
0.100
0.200
0.300
0.400
0.500
0.600
0.700
0.800
0 1000 2000 3000 4000 5000 6000 7000
Ageing time at 135C (hr)
FTIR
Pea
k ra
tio
KHNP 1735:1351NEL 1732:2917NEL 1507:2917NEL 3400:2917
![Page 22: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/22.jpg)
John Knott Associates Ltd 22
Reasons for variability – FTIR
Main problem is determining which peaks to compare –often not known before ageing programme starts
![Page 23: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/23.jpg)
John Knott Associates Ltd 23
Potentially useful CM methods
These have only been tested by 1 or 2 labs so there is no data on variability between labs
Methods showing trends with ageing (for at least some materials) Ultrasonic velocity Dielectric spectroscopy Frequency domain reflectometry (FDR)
![Page 24: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/24.jpg)
John Knott Associates Ltd 24
Ultrasonic velocityXLPO Jacket (Shanghai)
1400
1600
1800
2000
2200
2400
2600
0 1000 2000 3000 4000 5000 6000 7000
Ageing Time at 135°C (hr)
Ultr
ason
ic V
eloc
ity (m
/s) SNPSC
![Page 25: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/25.jpg)
John Knott Associates Ltd 25
Dielectric spectroscopy
Highlights the problem of electrical measurements on thermally aged materials – large step change in values as cable is dried
For irradiation at constant temperature, trends are more visible
![Page 26: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/26.jpg)
John Knott Associates Ltd 26
Frequency domain reflectometryEPR insulation (Eupen)
0%5%
10%15%20%25%30%35%40%45%
0 1000 2000 3000 4000 5000 6000
Ageing Time at 135 C (hr)
% C
hang
e fr
om B
asel
ine
Refle
ctio
n Co
effic
ient
AMS - blue (black)
![Page 27: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/27.jpg)
John Knott Associates Ltd 27
Summary of practical usefulness of CM methods for materials tested
Used a colour code to indicate practical usefulness of each CM method tested
This is specific to actual materials tested
![Page 28: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/28.jpg)
Example – very useful
John Knott Associates Ltd 28
![Page 29: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/29.jpg)
Example - useful
John Knott Associates Ltd 29
![Page 30: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/30.jpg)
Example – potentially useful
John Knott Associates Ltd 30
![Page 31: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/31.jpg)
Example – not useful
John Knott Associates Ltd 31
![Page 32: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/32.jpg)
John Knott Associates Ltd 32
Summary of practical usefulness of CM methods for materials tested
![Page 33: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/33.jpg)
John Knott Associates Ltd 33
Lessons learned
Three critical areas for practical use of CM methods in ageing assessment Specification of test method Sample preparation method Data analysis
![Page 34: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/34.jpg)
Specification of test method
All aspects must be tightly defined and consistent throughout monitoring programme Type of test equipment e.g. grips, extensometer for
EAB Test parameters e.g. OIT set temperature Identify those parameters that are critical to
reproducibility Not all parameters are critical e.g. cross-head speed in
EAB tests Ensure enough measurements are made to reduce
standard deviation of measurements
John Knott Associates Ltd 34
![Page 35: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/35.jpg)
Sample preparation
The most important factor in inter-lab variability For EAB need to define
Size of dumb-bells Surface treatment Removal of conductors from insulation Use of end-tabs for tubular insulation Type of grips
For OIT/OITP/TGA Selection of sample material e.g. dual layers Method of size reduction and particle size Packing of particles in test pan
John Knott Associates Ltd 35
![Page 36: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/36.jpg)
Data analysis
Check algorithms used in test equipment May need to analyse manually e.g. to define baseline
and threshold for OIT, OITP; force range used for IM In EAB, measurement of elongation relative to
start of test If using cross-head movement (not recommended)
define gauge length to be used Define how mean and standard deviation are
reported Do you discard outliers?
John Knott Associates Ltd 36
![Page 37: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/37.jpg)
Summary
Toolbox of useful CM methods available Several potentially useful methods identified Suitable CM methods are material dependent Test methods must be very well defined
Including sample preparation and data analysis Standards are being developed/revised for some CM
methods
John Knott Associates Ltd 37
![Page 38: Benchmarking of cable condition monitoring methods ... · 6 different cables XLPE/CSPE ... Dielectric loss (Tan delta) ... IM changes by factor of 2 and recovery time by factor of](https://reader033.vdocuments.us/reader033/viewer/2022060702/606f5f355f7bae7a636ec7d9/html5/thumbnails/38.jpg)
For more information
Summarised results are in IAEA-TECDOC-1825, can be downloaded from
http://www-pub.iaea.org/books/IAEABooks/11164/Benchmark-Analysis-for-Condition-Monitoring-Test-Techniques-of-Aged-Low-Voltage-Cables-in-Nuclear-Power-Plants
John Knott Associates Ltd 38