John Sethian Naval Research Laboratory April 4, 2002 Electra title page Electra Electra NRL J. Sethian M. Friedman M. Myers S. Obenschain R. Lehmberg J. Giuliani JAYCOR S. Swanekamp Commonwealth Tech F. Hegeler SAIC M. Wolford Titan PSD, Inc D. Weidenheimer Work sponsored by DOE//NNSA/DP The Electra Laser Facility 500 keV, 100 kA, 100 nsec @ 5 Hz (x 2 sides = 50 kW)
16
Embed
John Sethian Naval Research Laboratory April 4, 2002 Electra title pageElectra NRL J. Sethian M. Friedman M. Myers S. Obenschain R. Lehmberg J. Giuliani.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
John SethianNaval Research Laboratory
April 4, 2002
Electra title page
ElectraElectraNRL
J. Sethian M. Friedman
M. MyersS. ObenschainR. LehmbergJ. Giuliani
JAYCOR S. Swanekamp
Commonwealth TechF. Hegeler
SAIC M. Wolford
Titan PSD, IncD. Weidenheimer
Work sponsored by DOE//NNSA/DP
The Electra Laser Facility500 keV, 100 kA, 100 nsec @ 5 Hz (x 2 sides = 50 kW)
The Electra Laser Facility500 keV, 100 kA, 100 nsec @ 5 Hz (x 2 sides = 50 kW)
Laser GasRecirculator
Topics This Time
RearMirror
OutputOptics
Laser Cell(Kr + F2)
PulsedPowerSystem
ElectronBeam Foil
Support(Hibachi)
AmplifierWindow
Cathode
Bz
Front runner for Advanced Pulsed Power should meet IFE requirements (< $10.00/J, >80% eff)
Notes:Cost: $ / e-beam Joule, for 100 kJ systems in quantities, NOT Electra; Efficiency: Flat top e-beam/wall plug
TTIDIODE
Marx/PFN
Marx / Pulse Forming Network
too new to evaluatecost and efficiency
XFMRS1 DIODEPFL
Laser-gated switch stack/array
Transformer + PFL + HV laser output switch
$ 8.35/J 87% eff
TTIDIODE
MC-1
Fast Marx
Fast Marx w/ laser gated switches + 1 stage Magnetic Compressor
$ 7.15/J 85% eff
Burst mode @ 5pps – 104 shotsShots to failure of chip under test – 1.36 x 105
Max voltage – 3.2 kV (limited by external insulation)Max current density – 2.7 kA/cm2
(121% of IFE requirement)Max di/dt – 1.36 x 1010 A/sec/cm2
(154% of IFE requirement, limited by circuit) > 10 x higher than non laser triggered devices
Max charge transfer per pulse – 1.7 mCb/cm2
(152% of IFE requirement)Laser output ~40% of maximum (tested at >107 shots)Successful demo of thyristor/ laser diode integration
Advanced Pulse Power Switch Development-1
RESULTS TO DATE
LGPT (Laser Gated and Pumped Thyristor)
Flood entire switch volume &junction with laser light…Gives high turn-on (dI/dt)
*Photon 1105 nm = 1.36 eV I.E. Just above the band edge of the PN junction
DiodeLaser*
Si Switchp
n+n
n++
p++
CONCEPT
Prototype has demonstrated concept
Switch
Lasers (hidden)
Feed Electrodes
Pulse Sciences Division
Advanced Pulse Power Switch Development-2
Full required performance (16.7 kV)NEXT GENERATION
Generation 2 Forecasts
Maximum voltage > 16.7 kV (IFE requirements) Thicker device with two sided laser pumping Design allows use of advanced encapsulants (insulation) (soldered or alloyed interface of silicon and electrodes)
Action (I2t) in excess of IFE requirements
Integration of larger area reverse parallel co-planar diode to meet 100% of IFE reverse current requirement
Extrapolates to “rail gap” aspect ratio easily (full IFE scale)
EXPECT TO BE READY FOR TESTING WITHIN 6 MONTHS
Pulse Sciences Division
1 cm2 silicon thyristorand co-planar diode
Laser array in electrode(anode and/or cathode)
(Single sided system shown)
Advanced Pulse Power Switch Development-3
Rail Gap GeometryFINAL CONFIGURATION
Pulse Sciences Division
Full-Stage Switch Assembly (16.4 kV)
11cm
108 cm total length
End View Side View
16 kVSi Switch
7 cm 1/2 Capacitor
1/2 Capacitor
1/2 Capacitor
1/2 Capacitor
Laser Drive Electronics
Si Switch
Laser
Application- Ultra Fast Marx (+/- 16.4 kV stage)
Oil Insulation
Test bed is evaluating lifetime of capacitor dielectric systems
Capacitor/Switch Test Bed Results· 2 x 108 shots on CSI capacitors (~0.03 J/cm3) – no failures· 1 x 108 shots on GA capacitors ( ~0.015 J/cm3) – no failures· 2 x 108 shots on ABB thyristor in service comparable to 3 stage MPC for IFE
Rep-rate: 55 Hz
Modified SLIA Rep Pre-Compression Stage
HV Transformer
Parallel Plate Test Volume
Coaxial Test Volume
Liquid Breakdown Dielectric Test facility nearing completion
Pulsed Power Systems use liquids to store electrical energy
V = V0
V =0
Oil or water
Electrodes, area A
d
E = V0.d
Electrical Breakdown strengthdetermines size of the system:
Oil (+): E = .48/(1/3 A.075)Oil (-): E = .72/(1/3A.075)Water (+): E = .23/(1/2 A.058)Water (-): E = .56/(1/3 A.070) ( is time voltage is above 63% of peak)
IFE driver..A(oil) = 108 cm2, A (water) = 3 x 108 cm2
Above formula derived for single shotevents, not long term repetitive operation.
E = Vo/d
d
Cathode(V =Vo)
Anode(V = 0)
Cathode development: Edge effect
0 20 40 60 80 current density (A/cm2)
Prediction
Elimination
Field-Shaper
Problems with this solution
1. Field shaper emits in rep-mode or longer pulse
3 cm x 30 cm“Strip” Cathode
2. Tough to do with strip cathode
0 0.5 1.0Cumulative charge (AU)
Verification
Radiachromicfilm
E = C-L
e-beam
Base
Emitter
“Floating Edge Removers” eliminate edge, fit strip geometry, and don’t emit (over 3000 shots)
(a)
(a)
(b)
(b)
Cathodebase
Cathode
Emitter (velvet)
Edge removers
Anode
0 50 100 current density (A/cm2)
(a)
(a)
(b)
(b)
Radiachromic Film
SeeFrank HegelerPoster
Expts & theory show beam subject to “transit time” instability. 2D PIC simulations predict resistive slots stabilize beam
dN
/dE
(A
rb.
un
its
) 1.0
0.5
0.00.0 0.5 1.0
E (MeV)
Not stoppedby gas
stopped by foils
0 2 4 6 8 10 12 14 160.0
0.2
0.4
0.6
0.8
FFT (B-dot signal)
2.5 GHz
Frequency (GHz)
0 5 10 15 20-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
B-dot signal (Arb. units)
Time (ns)
Cathode
Anode
e-
Before
0 50 100 150 200 Time (ns)
V (
MV
) 0.8
0.4
0.0
f=2.5 GHz
Experiment (Nike 60 cm AMPLIFIER_
Simulation
0 50 100 150 200 Time (ns)
0.4
0.2
0.0
1.0
0.5
0.0
-0.5
-1.0
X (
m)
/4
Slot withResistiveWire
-5 0 5 Z (cm)
Cathode
After
dN
/dE
(A
rb.
un
its
)
0.0 0.5 1.0E (MeV)
1.0
0.5
0.0
Not stoppedby gas
stopped by foils
Simulation
7
cu
rre
nt
de
ns
ity
A/c
m2
time (100 ns/div)
0
14
frequency (GHz)
am
pli
tud
e
0 1 2 3
0
.02
.04
-7
0
7
14
time (100 ns/div)
cu
rre
nt
de
ns
ity
A/c
m2
0 19
2 30
.02
.04
am
pli
tud
e
frequency (GHz)
e-
e-
The slotted cathode suppresses the transit-time instability on NIKE 60 cm Amplifier
Slotting cathode in other direction may completely eliminate instability