superior performance.

powerful technology.

Transmission Level HTS Fault Current

Limiter
Chuck Weber
8th Annual EPRI Superconductivity Conference
Oak Ridge, TN
November 12, 2008
SuperPower, Inc. is a subsidiary of Royal Philips Electronics N.V.

SFCL program overview

138 kV, 650 kV BIL
Bushings

Partners

Pressure Vessel
Vacuum Vessel

"
2
6
.3
7

"
8
4
.2
5
1

"
2
6
.3
7

"
8
4
.2
5
1

Inner
Height

Specifications

YBCO based, resistive type FCL

138 kV class device
Fault Current 13.8 kA
Load Current 1,200 Arms
Design fault current 37 kA
Design device response Recover
to superconducting state after a fault
carrying full load current

HTS
Assembly
Height

Matrix Assembly

8th Annual EPRI Superconductivity Conference November 12, 2008

Assembly diameter
Inner diameter

Generalized SFCL Specification Development

Baseline Design for Program was the AEP SPORN substation site
This is a niche application site, operating at 400Arms, 138 kV
Prospective fault current 26 kArms (~90 kA peak) and 13.8 kArms
(~ 37 kA peak)
Working with AEP, we have identified a site with broader general
application
TIDD substation
1,200 Arms, 138 kV
Prospective fault current is 13.8 kArms (~37 kApeak)

8th Annual EPRI Superconductivity Conference November 12, 2008

TIDD Substation (Partial) One-Line Diagram

Proposed SFCL
Installation
Location

8th Annual EPRI Superconductivity Conference November 12, 2008

Prior accomplishments
Proof-of-Concept demonstrated
MCP 2212 (2004)
2G YBCO (2006)
Beta device testing specifications
established
Completed design and testing of HV
bushings (SEI)
Investigated several engineered 2G
architectures for improved RUL

Thermal simulation of RUL process

Weibull plots of standard 2G failures
Conceptual CRS & vessel design
Investigated LN2 dielectric properties

Probability of failure [%]

Design and laboratory testing of shunt

coils to withstand high fault transient loads

2G FCL - Probability of failure for 2G tapes as function of energy

input
100
10
1
0.1
0.01
20

25

30

35

40

45

Energy [J/cm/tape]
Probability of Failure - Test data
Probability of Failure Calculated using Weibull Distributuon

8th Annual EPRI Superconductivity Conference November 12, 2008

50

Improvements to shunt coil and contact design

Shunt coil improvements:
Manufacturing improvements
(easier assembly, more robust
coil)
Mechanical strength
Multi-Layer winding (size
reduction)
Connector improvements:
Shape optimization to avoid
contact hotspots
Improvement in RUL Time
Improvement in RUL Current
Improvement in consistency of
contact resistance
8th Annual EPRI Superconductivity Conference November 12, 2008

Tape heating near contact during fault impacts

RUL

8th Annual EPRI Superconductivity Conference November 12, 2008

Correlation between different contact geometries

Total Current (80A peak)

Superconductors Current

Total Current (80A peak)

Straight Thick Contacts
(M3-460 Tape):
I load = 80 A
RUL = 82 sec.

Superconductors
Current

Recovery Voltage
Recovery Voltage

Straight -Tapered Contacts

(M3-460 Tape):
I load = 80 A
RUL = 3.5 sec.

Total Current (80A peak)

Superconductors
Current

Straight -Tapered Contacts

(M3-460 Tape):
I load = 80 A
RUL = 2.8 sec.

Recovery Voltage

8th Annual EPRI Superconductivity Conference November 12, 2008

Recent KEMA tests

Recent rounds of KEMA testing focused on critical AEP reclosure sequence on
an HTS element
5 Cycles
Fault
13kA/7kA

5 Cycles
Fault
13kA/7kA

18 Cycles
Load Current

5 Cycles
Fault
13kA/7kA

15 sec
Load Current

5 Cycles
Fault
13kA/7kA

5 Cycles
Fault
13kA/7kA

135 sec
Load Current

160 sec
Load Current

Straight elements were

used
Improved connector
designs were used
Standard, pre-qualified
tapes were used
Test Dates: May 2008,
July 2008
8th Annual EPRI Superconductivity Conference November 12, 2008

Breaker opens
and locks-out

Recovery under
NO Load Current

2G RUL capabilities tested at KEMA

Standard SF12100 2G wire used
Total R ecovered Pow er, 2x5 cycles Faults at 37kA w ith 10
mOhm

Test conditions
- 37 kA fault

250000
200000

- follows AEP sequence

150000
Loa d P ow e r (VA)

- Shunt impedance
- Number of parallel tapes
- System voltage (v/cm/tape)
- Load Current

8th Annual EPRI Superconductivity Conference November 12, 2008

50000

300V
250V

Vo
lta
ge

Test variables

100000

200V

16 Tapes

Para8 Tapes
lle
Tape l
s

100V
4 Tapes

Achieving RUL is a difficult task

Without load current recovery is very fast

w/o Load

w/ Load

3 x load
Base-Line
Voltage
Adding
current
makes
recovery much more difficult
8th Annual EPRI Superconductivity Conference November 12, 2008

Electrical stress on the tapes can limit RUL

RUL time can affected by
increasing the V/cm on the
tape

RUL

Base-Line Voltage

Limits of the design

optimization are understood

RUL

RUL

1.5 x Base-Line Voltage

3 x Base-Line Voltage

8th Annual EPRI Superconductivity Conference November 12, 2008

Factors impacting RUL defined by test results

Total Recovered Power, 2x5 cycles Faults at 37kA with 4 Tapes

80000
70000
60000
50000
40000 Load Power (VA))
30000
20000

1.67 Sm-Ohm
hu
nt
Im
5pem-Ohm
da
nc
e

10000

100 V

tageV
ol250
200 VV

0
300 V

Sample surface plot of RUL conditions

8th Annual EPRI Superconductivity Conference November 12, 2008

Ability to predict RUL over wide design space

Maximum Load Current as a function of shunt impedance, operating voltage & number of
tapes

1000
900
800
700
600

Maxim un Recovered
Load Current

500
400
300

pes
a
T
,#

16Tapes, 100V

8Tapes, 100V

4Tapes, 250V

4Tapes, 100V

eda
nce

5 m-Ohm

Imp

1.67 m-Ohm

8Tapes, 250V

100

e
ltag
o
V

16Tapes, 250V

200

Recovered Current with 2 Asymmetrical 37kA Faults, 5 cycles each

8th Annual EPRI Superconductivity Conference November 12, 2008

Worst case conditions at Tidd can achieve RUL

RUL with 90% of the Power recovered within

the 2nd and the 3rd 37 kA Faults

Full recovery expected with optimal bath conditions

8th Annual EPRI Superconductivity Conference November 12, 2008

Bath Conditions Impact on Ability to Recover

During the fault transient, tape heats up to film boiling region.
Bath conditions (pressure, subcooling) shift boiling heat transfer curve
Bath conditions have an impact on the dielectric strength of LN2
No Recovery Due to Film Boiling

Boiling Heat Transfer for LN2

600

Heat Out
Heat In

100.0

500

Lower Zshunt,
Higher Ztape

10.0

Power (W)

q/A (W/cm )

400

300

200

1.0

100

0.1

1.0

10.0

100.0

T wall - T sat (K)

100

150

20 0

1000.0

250

300

350

400

450

500

Temperature (K)

Lowering the shunt coil value or

increasing the resistance of the
stabilizer layer will help with film boiling.

8th Annual EPRI Superconductivity Conference November 12, 2008

550

Bath Conditions Impact on Ability to Recover

Once film boiling threshold is crossed, nucleate boiling ensues
Bath conditions (pressure, subcooling) shift boiling heat transfer curve
Bath pressure shifts saturated boiling temperature, limiting nucleate boiling
recovery
No Recovery Due to Nucleate Boiling
600

Boiling Heat Transfer for LN2

Heat Out
Heat In

100.0

500

Lower pressure

Power (W)

q/A (W/cm )

400

10.0

300

200

1.0

100

0.1

1.0

10.0

100.0

1000.0

75

80

85

90

Temperature (K)

T wall - T sat (K)

Lowering the operating pressure will

help with nucleate boiling, but
decreases dielectric properties
8th Annual EPRI Superconductivity Conference November 12, 2008

95

100

Modeling indicates where operating conditions

for successful RUL exist
Recovery Under Load vs Number of Tapes
1300

Baseline
1200
1100

Pressure = 0.5 atm

Tbulk = 71.922 K

Recovery Load Current (arms)

1000
900

Pressure = 0.5 atm

Tbulk = 71.922 K
Stabilizer = 1% AgAu

800
700

Pressure = 0.75 atm

Stabilizer = 1% AgAu

600

Pressure = 0.75 atm

Stabilizer = 2.2% AgAu

500

Baseline Substrate = 4 mil

Stabilizer = Ag
Shunt Coil = 10 m/m
No dielectric coating
Ic @ 77 K = 250 amps
n-value = 20
Pressure = 1 atm
Tbulk = 72 K

400
300
200
100

Baseline
Shunt = 5 m/m

Baseline
Shunt = 7.5 m/m

0
0

10

12

14

Number of Tapes per Element

8th Annual EPRI Superconductivity Conference November 12, 2008

16

Introducing bubbles in LN lowers

breakdown strength: FCL recovery
Bubbles form thermally or electrically and can affect the breakdown strength

Two experiments
Open bath LN
Pressurized cryostat
Nitrogen gas provided
by fused silica capillary
tube
Varied flow rates
Parallel plane profiled
SS electrodes

2 mm gap
0.5 mm capillary tube

BD strength of LN is
~5x the gas at 1 bar
Important for FCL Recovery under Load

19 Managed by UT-Battelle
for the Department of Energy

DOE Peer Review 2008

Effect of externally provided bubbles

on LN Breakdown: AC breakdown
Effect of Bubbles

16

all data w and w/o bubbles

Cumulative Failure Probability (%)

Average Electric Field (kVrms/mm)

18

14
12
10
8
6
4
2

99.9
99.0
95.0
90.0
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
5.0

1.0

0
without bubbles

with bubbles

7 8 9 10

20

Breakdown Field (kVrms/mm)

Presence of bubbles

Liquid nitrogen at 1 bar

Bubbles in LN lowers breakdown strength

Change in slope at lower probability indicates change in BD
mechanism
20 Managed by UT-Battelle
for the Department of Energy

DOE Peer Review 2008

30

Summary
Significant progress in understanding and impacts of:
RUL
Variables impacting RUL studied and understood
Worst case conditions at TIDD can be met
Impact of device design and cost under evaluation

LN2 Dielectrics
Impact of bubbles on breakdown mechanism and dielectric
strength

Loss of cryogenic partner a setback, but not fatal

Next step: Alpha detailed design

8th Annual EPRI Superconductivity Conference November 12, 2008

Thank You for your attention!

For more information:

www.superpower-inc.com
or
cweber@superpower-inc.com

8th Annual EPRI Superconductivity Conference November 12, 2008