CA1287890C - Automatically preionized double-sided lc inversion circuit - Google Patents
Automatically preionized double-sided lc inversion circuitInfo
- Publication number
- CA1287890C CA1287890C CA 565436 CA565436A CA1287890C CA 1287890 C CA1287890 C CA 1287890C CA 565436 CA565436 CA 565436 CA 565436 A CA565436 A CA 565436A CA 1287890 C CA1287890 C CA 1287890C
- Authority
- CA
- Canada
- Prior art keywords
- capacitors
- load
- amplifier
- series
- port
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired
Links
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03K—PULSE TECHNIQUE
- H03K3/00—Circuits for generating electric pulses; Monostable, bistable or multistable circuits
- H03K3/02—Generators characterised by the type of circuit or by the means used for producing pulses
- H03K3/53—Generators characterised by the type of circuit or by the means used for producing pulses by the use of an energy-accumulating element discharged through the load by a switching device controlled by an external signal and not incorporating positive feedback
- H03K3/537—Generators characterised by the type of circuit or by the means used for producing pulses by the use of an energy-accumulating element discharged through the load by a switching device controlled by an external signal and not incorporating positive feedback the switching device being a spark gap
Landscapes
- Lasers (AREA)
Abstract
AUTOMATICALLY PREIONIZED DOUBLE-SIDED
LC INVERSION CIRCUIT
Abstract of the Disclosure A high voltage, high current pulse amplifier. First and second series-connected capacitors are connected in parallel across third and fourth series-connected capacitors. A voltage source is connected in parallel across the first and second capacitors. A first port of a load is connected between the first and second capacitors, and a second port of the load is connected between the third and fourth capacitors. The first and third capacitors are positioned adjacent one another on a first side of the load; and, the second and fourth capacitors are positioned adjacent one another on a second side of the load, opposite the first side. The load is preionized via a spark gap connected in series between the voltage source and the first port of the load. First and second spark arrays are connected between the spark gap and the first port of the load. The spark arrays are positioned, respectively, on the first and second sides of the load. The circuit is configured to minimize inductance.
LC INVERSION CIRCUIT
Abstract of the Disclosure A high voltage, high current pulse amplifier. First and second series-connected capacitors are connected in parallel across third and fourth series-connected capacitors. A voltage source is connected in parallel across the first and second capacitors. A first port of a load is connected between the first and second capacitors, and a second port of the load is connected between the third and fourth capacitors. The first and third capacitors are positioned adjacent one another on a first side of the load; and, the second and fourth capacitors are positioned adjacent one another on a second side of the load, opposite the first side. The load is preionized via a spark gap connected in series between the voltage source and the first port of the load. First and second spark arrays are connected between the spark gap and the first port of the load. The spark arrays are positioned, respectively, on the first and second sides of the load. The circuit is configured to minimize inductance.
Description
o AUTOMA'rICALLY PREIONIZED DOUBLE sID~n LC INVERSION CIRCUIT
Field of the Invention This application pertains to an amplifier for delivering high voltage, high current pulses to a dual port load. The invention has particular application in the amplification of ultrashort CO2 laser pulses.
It is relatively easy to obtain a self-sustained glow discharge from a laser if the laser is operated at a comparatively low pressure (i.e. about 1 atmosphere). However, it is preferable to operate CO2 lasers at higher pressures approaching 10 atmospheres in order to broaden the wavelength over which laser trans-mission occurs. Only high pressure amplifiers, with their large gain bandwidths due to pronounced collis-ional broadening and significant line overlap, can be used to amplify picosecond CO2 laser pulses oE the sort which are useul in photochemistry, spectroscopy, laser ranging and other applications.
` 25 Conventional high pressure TE CO2 ampliEiers use an arrangement of switches, resistors and capacitors known as a l'Marx bank" to produce the high voltages needed to initiate the glow discharge. The capacitors are connected in parallel, with resistors connected between the plates of each adjacent capacitor; and with switches (i.e. spark gaps) connected between opposed plates of each adjacent capacitor. The capacitors are slowly trickle charged through the resistors and then ; 35 discharged by closing the switches in synchronization.
.
..
~. :
.
,. , . . : .
~ :' ~ , ' ' ' .;' .. , : ' ' ' ' rrhe spark gap switches have inherent inductance, ~hich reduces the operational speed of the circuit (the cir-cuit rise time is proportional to (LC)-5 where "L"
represents the total circuit inductance, and "C" repre-sents the total circuit capacitance). To maximize oper-ational speed, both inductance and capacitance must be reduced. However, the circuit capacitance represents the energy stored by the circuit, which is pre~erably maximized. To store high energies in small capaci-tances, a very high voltage per capacitor is necessary,and immersion in an oil bath or a pressurized gas cham-ber is usually required to contain the multi spark gap device. Often, the compromise between energy and speed results in low input energy densities and/or small dis-charge volumes, and the resultant devices are bettersuited to use as oscillators rather than amplifiers.
Moreover, at higher pressures, the glow of a sel~-sustained discharge tends to constrict over time, even-tually forming streamers and then sparks~ To obtain a clean glow discharge with a Marx bank, one must minimize circuit capacitance and maximize circult inductance by increasing the number of stages, so as to increase the speed of the discharge; both of which are undesirable trade-o~fs, for the reasons aforesaid.
The principal of vector inversion (i.e. the generation of high voltage pulses by the systematic reversal of alternate interconnected capacitors which have been charged such that their charges add vec~or-ially to zero) has proved successful in a variety ofsmall high pressure laser devices, both in its contin-uous form, as with Blumelin circuits, and in its dis-crete form, with single-sided and double-sided LC inver-sion circuits. These single-stage LC inversion circuits are not fast enough to be applied to the type of high ., .
' ' ' . .
: .
: ', ' . ~ : .
.
. -, , . . ~ :
37~s~3~3 energyt high pressure, larye aperture amplifier required to amplif~ ultrashort CO2 laser pulses, because of the inherent instability associated with glow discharges at high pressures. However, the basic principles of opera-tion can still be utilized. The discharge circuit ofthe amplifier described below is designed to be signifi-cantly faster than simple capacitor transEer circuits, LC inversion circuits, or ~arx banks, for comparable bank energies and glow discharge volumes.
Summary_of the Invention In accordance with a first embodiment, the invention provides an amplifier for delivering high voltage, high current pulses to a dual port load. The amplifier comprises first and second series-connected capacitors; third and fourth series-connected capaci-tors; and, a voltage source which is connected in parallel across the first and second capacitors. The capacitors each have substantially identical capaci-tance. The first and second capacitors are connected in parallel across the third and fourth capacitors. A
first port of the load is connected between the first and second capacitors; and, a second port of the load is connected between the third and fourth capacitors. The first and third capacitors are positioned adjacent one another on a first side of the load; and, the second and fourth capacitors are positioned adjacent one another on a second side of the load, opposite the first side.
A preionization means is provided for preion izing the load. The preionization means comprises a spark triggering means and first and second spark - arrays. The spark triggering means is connected in :, , . : ., : . .
: : .. - - . - . :
. ~ . .. . . . .
, .
,:
,- '~
series between the voltage source and the Eirs~ ~or~ oE
the load. ~he first spark array is connected between the spark triggering -means and the first port of the load: and is positioned on the first side o the load.
The second spark array is also connected between the spark triggering means and the Eirst port of the load but is positioned on the second side of the load.
The capacitors and the load are advantageously connected and configured to minimize circuit inductance.
In the preferred embodiment, this is accomplished by employing equal pluralities of discrete capacitors to form each of the first, second, third and fourth capaci-tors mentioned above. The first capacitors, the third - 15 capaci~ors and the load define a first electrical cir-cuit loop about a first axis. The second capacitors, the fourth capacitors, and the load define a second electrical circuit loop about a second axis. The cross- -sectional area of the first and second circuit loops is minimized in planes which are respectively perpendicular to the first and second axis. The lenyth of the first and second circuit loops is maximized in the direction of the first and second axes respectively.
, In accordance with a second embodiment, the inventi.on provides an alternative amplifier for deliver-ing high voltaye, high current pulses to a dual port load. This amplifier comprises first and second series-connected capacitors; third and fourth series-connected capacitors; fifth and sixth series-connected capacitors;
: and, seventh and eighth series-connected capacitors. A
first voltage source is connected in parallel across the first and second capacitors; and, a second voltage source is connected in parallel across the third and fourth capacitors. The capacitors each have substan-_ 4 :
. :..
: ' . , ~ , ' . :.. . ' . ' ' .
.3~8~
tially identical capacitance. The first and second capacitors are connected in parallel across the third and fourth capacitors. Similarly7 the Eifth and sixth capacitors are connected in parallel across the seventh and eighth capacitors. The fifth and sixth capacitors are further connected in series between the second and fourth capacitors and, the seventh and eighth capacitors are further connected in series between the first and third capacitors. A first port of the load is connected between the Eirst and second capacitorst and, a second port of the load is connected between the third and fourth capacitors. The first, third, seventh and eighth capacitors are positioned adjacent one another on a first side of the load; and, the second, fourth, fifth and sixth capacitors are positioned adjacent one another on a second side of the load, opposite the first side.
~ preionizing means comparable to that des-cribed above in relation to the Eirst embodiment of the invention is preferably utilized with the second embodi-ment of the invention. The circuit inductance of the second embodiment is also preferably minimized by connecting and configuring the capacitors and the load in a manner comparable to that described above in relation to the first embodiment of the invention.
Brief Description of the Drawinqs Figure 1 is an electronic circuit schematic diagram of an amplifier according to the first embodi-ment of the invention.
-~ Figure 2 is a cros~-sectional pictorial illus-tration of an amplifier constructed in accordance with the first embodiment oE the invention.
'' ' : . ' ' ', :
~, . . .
-.
7~
Figure 3 is a pictorial illustratlon of a spark array preionizer rod.
~ iyure 4 is an electronic circuit schematic diagram of an ampli~ier according to the second embodi-ment of the invention.
Figure 5 is a cross-sectional pictorial illus-tration of an amplifier constructed in accordance with the second embodiment of the invention.
Detailed Description of the Preferred Embodiment The first embodiment of the invention to be described is an automatically preionized, double-sided LC inversion circuit. The second embodiment of the invention, described subsequently, is an automatically preionized, double-sided four fold ~C inversion circuit.
The first embodiment of the invention is appropriate for use in a laser oscillator; the second embodiment is use-ful as an amplifier of pulses produced by a laser oscillatorO
The first embodiment of the invention will now be described with reference to Figures 1 through 3~
As depicted in Figure 1, amplifier 10 com-prises first and second series-connected capacitors 12, 14; and, third and fourth series-connected capacitors 16, 18. A voltage source VO is connected in parallel across capacitors 12, 14. Capacitors 12, 14, 16 and 18 each have substantially identical capacitance, which may conveniently be designated as "C/4", such that the bank o~ capacitors, connected as described, has a total capacitance of "C".
.
.
- : ~ . . : . :
. ~ , . '. .
~' . , : . ' ' ' . ~: . . . :.. , AmpliEier 10 is conEigured to de].iver high voltage, hiyh current pulses to dual port load 20, which may be the main discharge electrodes of a CO2 laser.
Capacitors 12, 14 are connected in parallel across capacitors 16, 18. A first port 22 of load 20 is connected between capacitors 12, 14; and a second port 24 of load 20 is connected between capacitors 16, 18.
As may be seen in Figure 2, capacitors 12~ 16 are positioned adjacent one another on a first side of load 20; and capacitors 14, 18 are positioned adjacent one another on a second side of load 20, opposite the first side.
In a conventional LC inversion circuit, the two parts of the capacitor bank are mounted on one side of the load. For a comparable capacitor bank of total capacitance C and single-side inductan~e Ll the double-sided LC inversion circuit has a rise time of about (LC/8)-5, as compared to about (LC/4)-5 for the conventional LC inversion circuit. The main loop in either of these circuits contains no switching elements, so the inductance can be made to be much lower than that in a Marx bank, which is limited by the inductance of the required switches. Double-sided LC inversion circuits constructed in accordance with the first embodiment of the invention are therefore faster than the prior art discrete capacitor circuits, so glow dis-charges in nitrogen, excimer, and CO2 lasexs at higher operating pressures are possible.
The first embodiment of the invention also ~ pxovides a preionization means for automatically pre-- ionizing load 20. ~ "spark triggering means"; namely, spark gap 26, is connected in series between voltage source VO and first port 22 of load 20, through charging resistor 29. First and second spark arrays 28, :
, ~.:, ~ , ,. . '; ' .
-- ~
-........ . . ' : : .-37~g~
30 are connected across resistor 29 (i.e. between spark yap 26 and first port 22 of load 20). First spark array 28 is positioned on the first side of load 20, and second spark array 30 is positioned on the opposed, second side of load 20. The inversion eurrent is accor dingly routed through spark arrays 28, 30 in parallel, so preionization UV light is automatically produced before and during the main discharge.
Capaeitors 12, 14, 16 and 18 are preferably connected and configured, relative to load 20, so as to minimize eircuit inductance. This may be accomplished by utilizing a plurality of discrete capacitors to eonstruct each of first, seeond, third and fourth eapaeitors 12, 14, 16 and 18 respectively. The first capacitors, the third capaeitors, and load ~0 together define a first electrieal eireuit loop about a first ; axis whieh i5 perpendieular to the plane of the paper on which Figure 1 appears. The first eireuit loop is illustrated diagramatieally in Figure 1 by arrow 31.
Similarly, the seeond eapaeitors, the fourth capaeitors and load 20 define a second eleetrical eireuit loop (illustrated diagramatieally in Figure 1 by arrow 33) about a seeond axis whieh is also perpendieula~ to the plane of the paper on whieh Figure 1 appears. The eross-seetional area "A" of the first and seeond elreuit loops is minimized in planes whieh are respeetively per-pendieular to the first and seeond axes aforesaid. rrhe length "1" of the first and seeond eireuit loops is pre-Eerably maximized in the direetion of the first andseeond axes respeetively. The main eireuit loop has an effeetive induetanee of about L/2 where "L" is the singleside induetanee of about uoA/l.
. ~ . . . ..
- : . ~ . :, ' ' ' ' - - , . . ..
, . :, :
t3~7~9(~
An amplifier was constructed in accorda~ce with the first embodiment of the invention by employing 24 cm. long brass electrodes as load 20 and by using four 2.7 nF capacitors as each of the first, second, third and fourth capacitors. The electrodes were flat over a 5.8 x 234 mm2 area, with rounded corners, and were glued into slots in a Lucite pressure chamber, with a separation of 6 mm. The preionizer spark arrays each consisted of an array of twelve 400 um sparks, spaced every 21 mm. Each array was connected into the inver-sion loop using two 30 cm. links of 50 ohm coaxial cable, with Belden~#8868 high voltage leads as the center conductor. Aluminum plates 32, 34, and 36 were used to interconnect the capacitors in a low inductance configuration. To avoid corona production on these plates and elsewhere, all exposed edges were machined round - no oil bath was required.
The actual current in both the spark-gap loop "Ip", and in the main loop IIImll was measured using an integraph to integrate the dI/dt signals of subnanc-second rise-time pick-up coils. The main current had a 0.1 - 0.9 rise time of 13.5 ns, a FWHM (iOe. full width at half maxi~um) of 38 ns, and a peak value of 9.6 kA.
Although the glow resistance is a function of time, the main current shape was found to be quite close to that of a critically damped circuit. The total preionizer current Ip reached a maximum value of 7.5 kA at t =
106 ns, or 114 ns before the start of the main current pulse IM. The main electrode voltage was measured using a Tektronix~ A12 dual plug-in unit, arranged to take the difference between two identical, 8-ns rise-time, resistor-divider probes connected to the main electrodes. It took 220 ns, with a jitter of less than -~ 3 ns, to raise the voltage to Vmax = 48 kV
. .
- , ~ , .:
37?3~.30 required for initiation of the glow discharge curre~t I M ~
The free-ringing behaviour of the Cp = 22nF
primary circuit was determined by pressurizing the cham-ber to 11.5 atm, so that no main breakdown could occur~
The corresponding series Lp-Rp-Cp equivalent circuit had a ringing frequency of ~ = 2~r/(470 ns) an~
a l/e decay time of ~= S00 ns. It follows that Lp =
C ~ 2+~-2) 1=250nH, and Rp= 2Lp/1~
=1.0 ohm. The preionizer current Ip and the main electrode voltage VM during the voltage inversion are given by:
Ip(t) = VO ~~lLp le t/~ sin ~t VM(t)=VO[l-e~t/~(cos~t+~ 1~ lsin ~t)]
In the free-ringlng case, VM reaches a maxi-mum of VO [1 + exp(-~ ~ 1 ~ 1)] = 50 kV, at t = 235 ns.
Integrating the differential equation with these parameters, we find that the energy delivered to 2S the preionizers, before the main current pulse, is 6.1 J, out of the initial energy of CVo2/2 = 19.8 J
stored in the bank at 30 kV. A small voltage of 11.5 kV
remains on the capacitors on the side opposite to the primary loop, after the main current pulse has subsided.
This represents an energy of 1.4 J, leaving 12.3 J for the glow discharge. This value agrees well with 12.9 J
found by integrating the power VMIM over the current pulse. The specific energy deposited by the main dis-charge is therefore 150 Jl~l atm -1. The above , . ' . , . . .' ~.2~h37 890 estimates and calculations are based on the norninal, un-charyed capacitance of 2.7 nF per capacitoe. When the dependence of capacitance on voltage is taken into account we Eind an actual deposited energy denslty of 110 Jl~l atm -1 a~ 10 atm, at VO = 30 kV.
The 16 ceramic capacitors used were Murata Corp. no. DHS60~5V272Z-40, each having a capacitance of 2.7 nF and a 40-kV dc voltage rating. The bank is cap-able, thereEore of supplying specific energies in theoptimum range of (100-200) Jl-l atm -1, so as optical losses in the resonator change, the gain may be varied as compensation. A 10 atm mixture of 6% CO2, 6% N2, and 88% He was passed over the surface of liquid tripropylamine in an auxiliary pressure cell, after which it was directed longitudinally through the amplifier section, at a flow rate of 7.5 cm3 s-l, determined ~sing a curved-tube Elowmeter. ~lmost complete gas replacement for the 135-cm3 chamber volume occurs within about 20s, and the laser is fired at a repetition frequency of up to 0.05 Hz. Reliable glow discharges with this mixture were observed up to about 11 atm, while using a mixture of 3% CO3, 3%
N2, 94% He, and tripropylamine. Reliable operation is expected to occur at pressures up to about 16 atm.
Charging resistor 29 was a 10 Meg ohm tiger resistor, while the other resistors were 1 Meg ohrn tiger re~slstors.
The second embodiment of the invention will now be described with reference to Figures 4 and 5.
As depicted in Figure 4, amplifier 110 com-prises first and second series-connected capacitors 112, 114; third and ~ourth series-connected capacitors 116, ' . ' ' .' ' . .''' , :
- . . .
~3~
118; fifth and sixth series-connected capacitors 120~
122; and, seventh and eighth series-connected capacitors 124, 126. A first voltage source ~Vc is connected in parallel across capacitors 112, 114. A second voltage source -Vc is connected in parallel across capacitors 116, 118. The capacitors each have substantially iden-tical capacitance, which may conveniently be designated as "C/8", such that the bank of capacitors, connected as described, has a total capacitance of "Cl'.
1~
Amplifier 110 is configured to deliver high voltage, high current pulses to dual port load 128 which may, as in the case of the first embodiment of the invention, be the main discharge electrodes of a CO2 laser. First and second capacitors 112, 114 are connec-ted in parallel across third and fourth capacitors 116, 118. Similarly, fifth and sixth capacitors 120, 122 are connected in parallel across seventh and eighth capaci-tors 124, 126. Eifth and sixth capacitors 12C, 122 are further connected in series between second and fourth capacitors 114, 118. Seventh and eighth capacitors 124, 126 are similarly further connected in series between first and third capacitors 112, 116. A first port 130 of load 128 is connected between first and second capacitors 112, 114; and a second port 132 oE load 128 is connected between third and fourth capacitors 116, 118. As may be seen in Fiyure 5, first, third, seventh and eighth capacitors 112, 116, 124 and 126 are positioned adjacent one another on a first side of load 128 (i.e. to the right of load 128, as viewed in Figure 5); and second, fourth, fifth and sixth capacitors 114, 118, 120 and 122 are positioned adjacent one another on a second side of load 128, opposite the first side (i.e.
to the left of load 128, as viewed in Figure 5).
.
: ~
: . . . . . ~ : .
.
.
.
' . ' .
~ 3~ 3 The second embodiment of the invention also provides a preionization tneans for automaticall~ preion-izing load 128. A "spark triggering means": namely, spark gap 134, is connected in series between the first and second voltage sources as shown in Figure 4. Firsk and second spark arrays 136 and 138 are connected in parallel across the first and second voltage sources.
First spark array 136 is positioned on the first side of load 128, and second spark array 138 is positioned on the opposed, second side of load 128.
The first through eighth capacitors and the load are preferably connected and configured so as to minimize circuit inductance. This is achieved, in the second embodiment of the invention, by employing equal pluralities of discrete capacitors for each of first, second, third, fourth, fifth, sixth, seventh and eighth capacitors 112, 114, 116r 118, 120, 122, 124 and 126 respectlvely. The first, third, seventh and eighth capacitors, together with load 128, define a flrst elec-trical circuit loop about a first axis which is perpen-dicular to the plane of the paper on which Figure 4 appearsO The first circuit loop aforesaid is illus-trated diagramatically in Figure 4 by arrow 140. Simi-larly, the second, fourth, fifth and sixth capacitors,together ~ith load 128, define a second electrical circuit loop about a second axis which is also perpen-dicular to the plane of the paper on which Figure 4 appears. The second circuit loop is illustrated dia-3~ gramatically in Figure 4 by means of arrow 142. Thecross-sectional area of the first and second circuit loops is preferably minimized in planes which are respectively perpendicular to the first and second axes aforesaid. The length of the first and second circuit .
, ~
, .
:. ~. ,.
3~ 7~
loops is preferably maximized in the direction of the first and second axes respectively.
An ampli~ier was constructed in accordance with the second embodiment of the invention by employing eight banks, of 11 capacitors each, about the main dis-charge electrodes. These were high-voltage ceramic - capacitors (Murata Corp. #DHS60Z5V272Z-40) rated at 2.7 ; nF, 40 kV dc. In addition, eight capacitors o~ the same type (labelled 150 through 164 in Figure 4) were used for the preionization bank. A PVC main body tube 144 was machined to ha~e an inside diameter of 57.2 mm and an outside diameter of 76.2 mm. The two aluminum main discharge electrodes 128, shaped to fit on the inside wall of main body 144, were spaced 15 mm apart. The surfaces are essentially flat over a 10 x 350-mm2 area~ with edges machined round using a 1/4 inch radius cutter. Connections between each electrode and its corresponding aluminum plate (130,132~ were made with 12 small brass fittings with individual o-ring seals. The aluminum plates used for interconnecting the capacitors ` were machined round at each exposed edge, to prevent corona discharges. Immersion in oil was not required.
UV preionization was accomplished using two preionizer rods 136, 138 located along the walls of main body 144. Each preionizer rod, consisting of an array of 15 gaps 1 mm wide, spaced 25.4 mm apart, was con-structed using angled sections o~ stainless-steel tubing glued onto a 42-cm length of glass tubing of 5 mm o.d., with a coaxial cable center conductor (Belden #8868 high voltage lead) passing inside the glass tubing. The length of the shielded part of each preionizer lead was 1 m. The 7 atm pressurized spark gap and preionization .~ - -: : . . .
~ ~ .
. . ' ', ' ' . ' . , ' ' , . , . , : ~ -.
. .
~ ~7~0 capacitors were supported above the main body by l/4inch diameter copper rods 146, 148, which also served as electrical connections. Por triggering the spark gap, a trigger pin (not shown) was provided inside the hollow spark gap cathode. Optical access to the amplifier was through 5-mm thick KCl flats located at the ends o~ main body 144. The flats were mounted at 10 from normal to the body axis, to eliminate the possibility of self-lasing due to Fresnel reflections. The Lucite~nd flanges and the salt flats were sealed using O-rings, leaving a 22-mm clear aperture. The entire amplifier was supported within a Plexiglas box, which was covered with fine brass mesh to provide shielding against elec-trical noise generated by the amplifier circuit.
The charging and grounding resistors were TRW
tiger resistors. The total main discharge capacitance was c = 240 nF, while the total preionization capaci-tance C' = 22nF. The charging voltages of +Vc and -Vc against ground were supplied by a dual power supply. With 50-Meg ohm charging resistances, the amp-lifier requires 30 seconds to reach full charge.
Upon spark gap breakdown, preioni~ation commences and the four main discharge capacitor banks across which the spark gap is connected undergo voltage reversal, while the four banks connected to the main electrodes maintain their original relative voltage. A
potential is developed between the main electrodes, with the upper electrode becoming the cathode and the lower one becoming the anode. If the inversion is allowed to reach its peak, and if spark gap losses are neglected, the voltage across the main discharge gap would reach a theoretical maximum of 4Vc. The amplifier is normally ': ,', . " : ~.
- . , . . :
..
~2~t30 operated such that breakdown oE the main gap occurs just before the inversion peak.
The inductance "L" of one side of the dis-charge clrcuit can be approximately evaluated by consid-ering the configuration to be that of a sinyleturn solenoid. A current sheet, passing through the dis-charge electrodes and the capacitors of one side of the amplifier, encloses an effective area A - 131 cm2.
With a 1 = 35 cm long discharge, the inductance is L =
uOA/l = 47 nH. Using the total capacitance of C = 240 nF, the rise time of the discharge current pulse can be estimated. If the circuit was connected to an impedance matched load of 4(2L/C)-5 = 2.5 ohms/ the critically damped current would reach a maximum of vmaxe-l[c/(8/L)] 5 = 31 kA with a rise time of (LC/32)-5 = 19 ns.
The electrical and gain characteristics of the amplifier, operating at 10 atm with a gas mix composed of 6~ CO2, 6~ N2, and 88% He and a trace amount of tripropylamine were measured. The charging voltage was Vc = + 27.5 kV as determined by the two power supply meters, while the transient main-electrode voltage was 2S obtained using two identical high-voltage probes, connected to the aluminum plates attached to the elec-trodes. The probes were (7 k ohm: 50 ohm) resistor dividers, with measured rise times of 30 ns. By taking the difference o~ the attenuated voltage signals on an oscilloscope, the total electrode potential was dis-played. A Tektronix 7344 dual-beam oscilloscope, with two 100-MHz 7A12 plug-in units, was used for synchron-ized measurements of the voltage, currentt and small signal gain factor. The main discharge and preioniza-tion currents were measured using subnanosecond risetime . . .
.
,' ' , ' ' ,' ~' ' . ', : ~ ' ' ' Rogowski pick-up coils. Integration of the dI/dt siy-nals from these coils gave the currents, with a Eurther integration used for calibration. By adjusting condi-tions for observing free ringing oE the primary (inver-sion) loop of capacitance Cp = 30 nF, it was deter-mined that the spark-gap resistance was Rp a 210 Meg ohms and the primary loop inductance was Lp = 390 nH.
The main-gap voltage reached a peak of Vmax = 106 kV at 310 ns after spark gap breakdown, which represents 96% o~ the ideal maximum of 4 Vc.
~he time of the initial drop in voltage upon breakdown was observed to correspond closely to that of the initi-ation of the main discharge current, which occurred with a jitter of less than 6 ns. After the main discharge, a ringing potential, with an average potential of VF =
23 kV was observed across the main gap. The discharge current pulse reached its maximum value of 17.5 kA with a 27-ns 0.1-0.9 rise time and the pulse had a 70-ns FWHM. The preionization current started along with the inversion, after a small ormation lag time of about 50 ns, and reached a peak of 7.3 kA at t = lO0 ns, or 260 ns before the main current peak.
The product of the voltage and current curves yields the power pulse during the main discharye, while the discharye resistance R(t) may be found by taking the ratio. At the time of the peak discharge current, the resistance was 3.4 ohms, which indicates that the condi-tions are quite similar to criti.cal damping. The rise time, the peak value, and the overall form of the current pulse are comparable to those of the critically damped approximation above.
{~) The single-pass small-signal yain ~actor r, of the amplifier was measure~ on the 9.29 um, 9R(16) rotational-vibrational transition. The small-signal gain coefficient ~ is given by r = exp (~1), where 1 = 35 cm is the discha~rge length. The amplifier may be used to amplify pico-second laser pulses centered at ~his wavelength, as part of a system based on an injection-locked, Q-switched, mode-locked and cavity-dumped 10-atm CO2 oscillator. The gain signal r was measured together with the current signal using a l-W probe beam from a TEMoo mode, grating tuned cw C2 laser. Using a beam splitter, the cw laser output was simultaneously monitored on an Optical Engineering~
C~2 spectrum analyzer. The 9-mm beam, after making a single pass down the center of the amplifier discharge volume, was focused, using a 10-cm focal length KCl lens, onto a Labimex P005 HgCdTe room temperature detector. The detector and oscilloscope were located within an electrically screened room.
The gain signal r exhibited an initial dip during the time of the current pulse, which is attribu table to inverse bremstrahlung absorption by the dis--charge~ The signal had a 10%-90% rise time of 120 ns and a FWHM of 750 ns, both measured relative to the r=
1 line. The signal reached a maximum of rmax = 7 3~
corresponding to a small-signal gain coef~icient on the 9.294-um, 9R~16) line of ~ ax = 5.7~ cm~l.
Photographs of the discharge volume showed an approximately 10.5-mm-wide discharge, and the nominal specific energy input for the (15 x 10.5 x 350)~mm3 discharge was calculated to be 156 Jl~l atm~l.
The energy deposited in the discharge was taken to be the initial energy stored in the main bank capacitors, .
".
.. ~ , ' 37~
minus the energy of CVF2/32 remaining in the bank after the main discharge. The 156 Jl~l atm -1 figure was calculated on the basis of a nominal capaci-tance of 2.7 nF per capacitor. At high voltages, such capacitors exhibit reduced capacitances, and taking this ef-fect into account results in an actual deposited energy density of 120 Jl~l atm -1. By allowing suffici~nt time for complete gas replacement between shots, the amplifier performed ~ith excellent shot-to-shot reliability and reproducibility. The peak small-signal gain coefficient varied by only + 0.1% cm~l over five consecutive shots, taken at 60-s intervals.
For the 750-cm3 chamber volume, and an axial flowrate of 20 cm3s~l, measured using a curved tube flow-meter, essentially complete gas replacement occurs within this time. Reducing the concentration of C02 and N2 to 3~ each, with higher charging voltages, the amp-lifier is expected to operate reliably to about 15 atm.
With the concentration of CO2 and N2 at 6~ each, as used for the present study, 10 atm is close to the limit for reliable operation. The glow discharges contain, as is usual, several small streamers which are observed to extend, from the cathode, about 2 mm into the glow.
The-effectiveness of this amplifier has been clearly demonstrated by the measured discharge speed and observed relability. The amplifier is highly suitable for its role as a large-aperture, high input energ~, compact, and efficient device. The 5.7% cm~l small-3~ signal gain coefficient is, to the inventors' knowledge, the highest reported for any large aperture U~ preion-ized TE CO2 laser.
;
As in the case of the double-sided LC inver-sion circuit of the first embodiment, as many stayes as , .
'; ' ' ' ~ . ' . ' - . , ',' : ~,' .: ', , ' .
.
., :, . .. . .. .
desieed may be added to the double-sided four fold LC
inversion circuit of the second embodiment, using several of the eight-part circuits of Figure 4 connected in series. Such an arrangement would be appropriate Eor larger electrode separations and/or higher pressures.
Although the rise time of (LC/32) 5 depends inversely on the number of stages n, for laser excita-tion at optimum energy density, one should expect the required C to increase with pressure and discharge volume. The rise time would end up to be quite similar to the 27 ns of the n = 1 circuit described here, show-ing that such circuits would indeed be successful. For excitation of many types of gas ].asers, this circuit provides a realistic alternative to Marx bank or conven-tional LC inversion methods.
As will be apparent to those skilled in theart in light of the foregoing disclosure, many altera-tions and modifications are possible in the practice of this invention without departing from the spirit or scope thereof. Accordingly, the scope of the invention is to be constLued in accordance with the substance defined by the following claims.
.: . ..... , , .
- , . . : ". . .
. . .
Field of the Invention This application pertains to an amplifier for delivering high voltage, high current pulses to a dual port load. The invention has particular application in the amplification of ultrashort CO2 laser pulses.
It is relatively easy to obtain a self-sustained glow discharge from a laser if the laser is operated at a comparatively low pressure (i.e. about 1 atmosphere). However, it is preferable to operate CO2 lasers at higher pressures approaching 10 atmospheres in order to broaden the wavelength over which laser trans-mission occurs. Only high pressure amplifiers, with their large gain bandwidths due to pronounced collis-ional broadening and significant line overlap, can be used to amplify picosecond CO2 laser pulses oE the sort which are useul in photochemistry, spectroscopy, laser ranging and other applications.
` 25 Conventional high pressure TE CO2 ampliEiers use an arrangement of switches, resistors and capacitors known as a l'Marx bank" to produce the high voltages needed to initiate the glow discharge. The capacitors are connected in parallel, with resistors connected between the plates of each adjacent capacitor; and with switches (i.e. spark gaps) connected between opposed plates of each adjacent capacitor. The capacitors are slowly trickle charged through the resistors and then ; 35 discharged by closing the switches in synchronization.
.
..
~. :
.
,. , . . : .
~ :' ~ , ' ' ' .;' .. , : ' ' ' ' rrhe spark gap switches have inherent inductance, ~hich reduces the operational speed of the circuit (the cir-cuit rise time is proportional to (LC)-5 where "L"
represents the total circuit inductance, and "C" repre-sents the total circuit capacitance). To maximize oper-ational speed, both inductance and capacitance must be reduced. However, the circuit capacitance represents the energy stored by the circuit, which is pre~erably maximized. To store high energies in small capaci-tances, a very high voltage per capacitor is necessary,and immersion in an oil bath or a pressurized gas cham-ber is usually required to contain the multi spark gap device. Often, the compromise between energy and speed results in low input energy densities and/or small dis-charge volumes, and the resultant devices are bettersuited to use as oscillators rather than amplifiers.
Moreover, at higher pressures, the glow of a sel~-sustained discharge tends to constrict over time, even-tually forming streamers and then sparks~ To obtain a clean glow discharge with a Marx bank, one must minimize circuit capacitance and maximize circult inductance by increasing the number of stages, so as to increase the speed of the discharge; both of which are undesirable trade-o~fs, for the reasons aforesaid.
The principal of vector inversion (i.e. the generation of high voltage pulses by the systematic reversal of alternate interconnected capacitors which have been charged such that their charges add vec~or-ially to zero) has proved successful in a variety ofsmall high pressure laser devices, both in its contin-uous form, as with Blumelin circuits, and in its dis-crete form, with single-sided and double-sided LC inver-sion circuits. These single-stage LC inversion circuits are not fast enough to be applied to the type of high ., .
' ' ' . .
: .
: ', ' . ~ : .
.
. -, , . . ~ :
37~s~3~3 energyt high pressure, larye aperture amplifier required to amplif~ ultrashort CO2 laser pulses, because of the inherent instability associated with glow discharges at high pressures. However, the basic principles of opera-tion can still be utilized. The discharge circuit ofthe amplifier described below is designed to be signifi-cantly faster than simple capacitor transEer circuits, LC inversion circuits, or ~arx banks, for comparable bank energies and glow discharge volumes.
Summary_of the Invention In accordance with a first embodiment, the invention provides an amplifier for delivering high voltage, high current pulses to a dual port load. The amplifier comprises first and second series-connected capacitors; third and fourth series-connected capaci-tors; and, a voltage source which is connected in parallel across the first and second capacitors. The capacitors each have substantially identical capaci-tance. The first and second capacitors are connected in parallel across the third and fourth capacitors. A
first port of the load is connected between the first and second capacitors; and, a second port of the load is connected between the third and fourth capacitors. The first and third capacitors are positioned adjacent one another on a first side of the load; and, the second and fourth capacitors are positioned adjacent one another on a second side of the load, opposite the first side.
A preionization means is provided for preion izing the load. The preionization means comprises a spark triggering means and first and second spark - arrays. The spark triggering means is connected in :, , . : ., : . .
: : .. - - . - . :
. ~ . .. . . . .
, .
,:
,- '~
series between the voltage source and the Eirs~ ~or~ oE
the load. ~he first spark array is connected between the spark triggering -means and the first port of the load: and is positioned on the first side o the load.
The second spark array is also connected between the spark triggering means and the Eirst port of the load but is positioned on the second side of the load.
The capacitors and the load are advantageously connected and configured to minimize circuit inductance.
In the preferred embodiment, this is accomplished by employing equal pluralities of discrete capacitors to form each of the first, second, third and fourth capaci-tors mentioned above. The first capacitors, the third - 15 capaci~ors and the load define a first electrical cir-cuit loop about a first axis. The second capacitors, the fourth capacitors, and the load define a second electrical circuit loop about a second axis. The cross- -sectional area of the first and second circuit loops is minimized in planes which are respectively perpendicular to the first and second axis. The lenyth of the first and second circuit loops is maximized in the direction of the first and second axes respectively.
, In accordance with a second embodiment, the inventi.on provides an alternative amplifier for deliver-ing high voltaye, high current pulses to a dual port load. This amplifier comprises first and second series-connected capacitors; third and fourth series-connected capacitors; fifth and sixth series-connected capacitors;
: and, seventh and eighth series-connected capacitors. A
first voltage source is connected in parallel across the first and second capacitors; and, a second voltage source is connected in parallel across the third and fourth capacitors. The capacitors each have substan-_ 4 :
. :..
: ' . , ~ , ' . :.. . ' . ' ' .
.3~8~
tially identical capacitance. The first and second capacitors are connected in parallel across the third and fourth capacitors. Similarly7 the Eifth and sixth capacitors are connected in parallel across the seventh and eighth capacitors. The fifth and sixth capacitors are further connected in series between the second and fourth capacitors and, the seventh and eighth capacitors are further connected in series between the first and third capacitors. A first port of the load is connected between the Eirst and second capacitorst and, a second port of the load is connected between the third and fourth capacitors. The first, third, seventh and eighth capacitors are positioned adjacent one another on a first side of the load; and, the second, fourth, fifth and sixth capacitors are positioned adjacent one another on a second side of the load, opposite the first side.
~ preionizing means comparable to that des-cribed above in relation to the Eirst embodiment of the invention is preferably utilized with the second embodi-ment of the invention. The circuit inductance of the second embodiment is also preferably minimized by connecting and configuring the capacitors and the load in a manner comparable to that described above in relation to the first embodiment of the invention.
Brief Description of the Drawinqs Figure 1 is an electronic circuit schematic diagram of an amplifier according to the first embodi-ment of the invention.
-~ Figure 2 is a cros~-sectional pictorial illus-tration of an amplifier constructed in accordance with the first embodiment oE the invention.
'' ' : . ' ' ', :
~, . . .
-.
7~
Figure 3 is a pictorial illustratlon of a spark array preionizer rod.
~ iyure 4 is an electronic circuit schematic diagram of an ampli~ier according to the second embodi-ment of the invention.
Figure 5 is a cross-sectional pictorial illus-tration of an amplifier constructed in accordance with the second embodiment of the invention.
Detailed Description of the Preferred Embodiment The first embodiment of the invention to be described is an automatically preionized, double-sided LC inversion circuit. The second embodiment of the invention, described subsequently, is an automatically preionized, double-sided four fold ~C inversion circuit.
The first embodiment of the invention is appropriate for use in a laser oscillator; the second embodiment is use-ful as an amplifier of pulses produced by a laser oscillatorO
The first embodiment of the invention will now be described with reference to Figures 1 through 3~
As depicted in Figure 1, amplifier 10 com-prises first and second series-connected capacitors 12, 14; and, third and fourth series-connected capacitors 16, 18. A voltage source VO is connected in parallel across capacitors 12, 14. Capacitors 12, 14, 16 and 18 each have substantially identical capacitance, which may conveniently be designated as "C/4", such that the bank o~ capacitors, connected as described, has a total capacitance of "C".
.
.
- : ~ . . : . :
. ~ , . '. .
~' . , : . ' ' ' . ~: . . . :.. , AmpliEier 10 is conEigured to de].iver high voltage, hiyh current pulses to dual port load 20, which may be the main discharge electrodes of a CO2 laser.
Capacitors 12, 14 are connected in parallel across capacitors 16, 18. A first port 22 of load 20 is connected between capacitors 12, 14; and a second port 24 of load 20 is connected between capacitors 16, 18.
As may be seen in Figure 2, capacitors 12~ 16 are positioned adjacent one another on a first side of load 20; and capacitors 14, 18 are positioned adjacent one another on a second side of load 20, opposite the first side.
In a conventional LC inversion circuit, the two parts of the capacitor bank are mounted on one side of the load. For a comparable capacitor bank of total capacitance C and single-side inductan~e Ll the double-sided LC inversion circuit has a rise time of about (LC/8)-5, as compared to about (LC/4)-5 for the conventional LC inversion circuit. The main loop in either of these circuits contains no switching elements, so the inductance can be made to be much lower than that in a Marx bank, which is limited by the inductance of the required switches. Double-sided LC inversion circuits constructed in accordance with the first embodiment of the invention are therefore faster than the prior art discrete capacitor circuits, so glow dis-charges in nitrogen, excimer, and CO2 lasexs at higher operating pressures are possible.
The first embodiment of the invention also ~ pxovides a preionization means for automatically pre-- ionizing load 20. ~ "spark triggering means"; namely, spark gap 26, is connected in series between voltage source VO and first port 22 of load 20, through charging resistor 29. First and second spark arrays 28, :
, ~.:, ~ , ,. . '; ' .
-- ~
-........ . . ' : : .-37~g~
30 are connected across resistor 29 (i.e. between spark yap 26 and first port 22 of load 20). First spark array 28 is positioned on the first side of load 20, and second spark array 30 is positioned on the opposed, second side of load 20. The inversion eurrent is accor dingly routed through spark arrays 28, 30 in parallel, so preionization UV light is automatically produced before and during the main discharge.
Capaeitors 12, 14, 16 and 18 are preferably connected and configured, relative to load 20, so as to minimize eircuit inductance. This may be accomplished by utilizing a plurality of discrete capacitors to eonstruct each of first, seeond, third and fourth eapaeitors 12, 14, 16 and 18 respectively. The first capacitors, the third capaeitors, and load ~0 together define a first electrieal eireuit loop about a first ; axis whieh i5 perpendieular to the plane of the paper on which Figure 1 appears. The first eireuit loop is illustrated diagramatieally in Figure 1 by arrow 31.
Similarly, the seeond eapaeitors, the fourth capaeitors and load 20 define a second eleetrical eireuit loop (illustrated diagramatieally in Figure 1 by arrow 33) about a seeond axis whieh is also perpendieula~ to the plane of the paper on whieh Figure 1 appears. The eross-seetional area "A" of the first and seeond elreuit loops is minimized in planes whieh are respeetively per-pendieular to the first and seeond axes aforesaid. rrhe length "1" of the first and seeond eireuit loops is pre-Eerably maximized in the direetion of the first andseeond axes respeetively. The main eireuit loop has an effeetive induetanee of about L/2 where "L" is the singleside induetanee of about uoA/l.
. ~ . . . ..
- : . ~ . :, ' ' ' ' - - , . . ..
, . :, :
t3~7~9(~
An amplifier was constructed in accorda~ce with the first embodiment of the invention by employing 24 cm. long brass electrodes as load 20 and by using four 2.7 nF capacitors as each of the first, second, third and fourth capacitors. The electrodes were flat over a 5.8 x 234 mm2 area, with rounded corners, and were glued into slots in a Lucite pressure chamber, with a separation of 6 mm. The preionizer spark arrays each consisted of an array of twelve 400 um sparks, spaced every 21 mm. Each array was connected into the inver-sion loop using two 30 cm. links of 50 ohm coaxial cable, with Belden~#8868 high voltage leads as the center conductor. Aluminum plates 32, 34, and 36 were used to interconnect the capacitors in a low inductance configuration. To avoid corona production on these plates and elsewhere, all exposed edges were machined round - no oil bath was required.
The actual current in both the spark-gap loop "Ip", and in the main loop IIImll was measured using an integraph to integrate the dI/dt signals of subnanc-second rise-time pick-up coils. The main current had a 0.1 - 0.9 rise time of 13.5 ns, a FWHM (iOe. full width at half maxi~um) of 38 ns, and a peak value of 9.6 kA.
Although the glow resistance is a function of time, the main current shape was found to be quite close to that of a critically damped circuit. The total preionizer current Ip reached a maximum value of 7.5 kA at t =
106 ns, or 114 ns before the start of the main current pulse IM. The main electrode voltage was measured using a Tektronix~ A12 dual plug-in unit, arranged to take the difference between two identical, 8-ns rise-time, resistor-divider probes connected to the main electrodes. It took 220 ns, with a jitter of less than -~ 3 ns, to raise the voltage to Vmax = 48 kV
. .
- , ~ , .:
37?3~.30 required for initiation of the glow discharge curre~t I M ~
The free-ringing behaviour of the Cp = 22nF
primary circuit was determined by pressurizing the cham-ber to 11.5 atm, so that no main breakdown could occur~
The corresponding series Lp-Rp-Cp equivalent circuit had a ringing frequency of ~ = 2~r/(470 ns) an~
a l/e decay time of ~= S00 ns. It follows that Lp =
C ~ 2+~-2) 1=250nH, and Rp= 2Lp/1~
=1.0 ohm. The preionizer current Ip and the main electrode voltage VM during the voltage inversion are given by:
Ip(t) = VO ~~lLp le t/~ sin ~t VM(t)=VO[l-e~t/~(cos~t+~ 1~ lsin ~t)]
In the free-ringlng case, VM reaches a maxi-mum of VO [1 + exp(-~ ~ 1 ~ 1)] = 50 kV, at t = 235 ns.
Integrating the differential equation with these parameters, we find that the energy delivered to 2S the preionizers, before the main current pulse, is 6.1 J, out of the initial energy of CVo2/2 = 19.8 J
stored in the bank at 30 kV. A small voltage of 11.5 kV
remains on the capacitors on the side opposite to the primary loop, after the main current pulse has subsided.
This represents an energy of 1.4 J, leaving 12.3 J for the glow discharge. This value agrees well with 12.9 J
found by integrating the power VMIM over the current pulse. The specific energy deposited by the main dis-charge is therefore 150 Jl~l atm -1. The above , . ' . , . . .' ~.2~h37 890 estimates and calculations are based on the norninal, un-charyed capacitance of 2.7 nF per capacitoe. When the dependence of capacitance on voltage is taken into account we Eind an actual deposited energy denslty of 110 Jl~l atm -1 a~ 10 atm, at VO = 30 kV.
The 16 ceramic capacitors used were Murata Corp. no. DHS60~5V272Z-40, each having a capacitance of 2.7 nF and a 40-kV dc voltage rating. The bank is cap-able, thereEore of supplying specific energies in theoptimum range of (100-200) Jl-l atm -1, so as optical losses in the resonator change, the gain may be varied as compensation. A 10 atm mixture of 6% CO2, 6% N2, and 88% He was passed over the surface of liquid tripropylamine in an auxiliary pressure cell, after which it was directed longitudinally through the amplifier section, at a flow rate of 7.5 cm3 s-l, determined ~sing a curved-tube Elowmeter. ~lmost complete gas replacement for the 135-cm3 chamber volume occurs within about 20s, and the laser is fired at a repetition frequency of up to 0.05 Hz. Reliable glow discharges with this mixture were observed up to about 11 atm, while using a mixture of 3% CO3, 3%
N2, 94% He, and tripropylamine. Reliable operation is expected to occur at pressures up to about 16 atm.
Charging resistor 29 was a 10 Meg ohm tiger resistor, while the other resistors were 1 Meg ohrn tiger re~slstors.
The second embodiment of the invention will now be described with reference to Figures 4 and 5.
As depicted in Figure 4, amplifier 110 com-prises first and second series-connected capacitors 112, 114; third and ~ourth series-connected capacitors 116, ' . ' ' .' ' . .''' , :
- . . .
~3~
118; fifth and sixth series-connected capacitors 120~
122; and, seventh and eighth series-connected capacitors 124, 126. A first voltage source ~Vc is connected in parallel across capacitors 112, 114. A second voltage source -Vc is connected in parallel across capacitors 116, 118. The capacitors each have substantially iden-tical capacitance, which may conveniently be designated as "C/8", such that the bank of capacitors, connected as described, has a total capacitance of "Cl'.
1~
Amplifier 110 is configured to deliver high voltage, high current pulses to dual port load 128 which may, as in the case of the first embodiment of the invention, be the main discharge electrodes of a CO2 laser. First and second capacitors 112, 114 are connec-ted in parallel across third and fourth capacitors 116, 118. Similarly, fifth and sixth capacitors 120, 122 are connected in parallel across seventh and eighth capaci-tors 124, 126. Eifth and sixth capacitors 12C, 122 are further connected in series between second and fourth capacitors 114, 118. Seventh and eighth capacitors 124, 126 are similarly further connected in series between first and third capacitors 112, 116. A first port 130 of load 128 is connected between first and second capacitors 112, 114; and a second port 132 oE load 128 is connected between third and fourth capacitors 116, 118. As may be seen in Fiyure 5, first, third, seventh and eighth capacitors 112, 116, 124 and 126 are positioned adjacent one another on a first side of load 128 (i.e. to the right of load 128, as viewed in Figure 5); and second, fourth, fifth and sixth capacitors 114, 118, 120 and 122 are positioned adjacent one another on a second side of load 128, opposite the first side (i.e.
to the left of load 128, as viewed in Figure 5).
.
: ~
: . . . . . ~ : .
.
.
.
' . ' .
~ 3~ 3 The second embodiment of the invention also provides a preionization tneans for automaticall~ preion-izing load 128. A "spark triggering means": namely, spark gap 134, is connected in series between the first and second voltage sources as shown in Figure 4. Firsk and second spark arrays 136 and 138 are connected in parallel across the first and second voltage sources.
First spark array 136 is positioned on the first side of load 128, and second spark array 138 is positioned on the opposed, second side of load 128.
The first through eighth capacitors and the load are preferably connected and configured so as to minimize circuit inductance. This is achieved, in the second embodiment of the invention, by employing equal pluralities of discrete capacitors for each of first, second, third, fourth, fifth, sixth, seventh and eighth capacitors 112, 114, 116r 118, 120, 122, 124 and 126 respectlvely. The first, third, seventh and eighth capacitors, together with load 128, define a flrst elec-trical circuit loop about a first axis which is perpen-dicular to the plane of the paper on which Figure 4 appearsO The first circuit loop aforesaid is illus-trated diagramatically in Figure 4 by arrow 140. Simi-larly, the second, fourth, fifth and sixth capacitors,together ~ith load 128, define a second electrical circuit loop about a second axis which is also perpen-dicular to the plane of the paper on which Figure 4 appears. The second circuit loop is illustrated dia-3~ gramatically in Figure 4 by means of arrow 142. Thecross-sectional area of the first and second circuit loops is preferably minimized in planes which are respectively perpendicular to the first and second axes aforesaid. The length of the first and second circuit .
, ~
, .
:. ~. ,.
3~ 7~
loops is preferably maximized in the direction of the first and second axes respectively.
An ampli~ier was constructed in accordance with the second embodiment of the invention by employing eight banks, of 11 capacitors each, about the main dis-charge electrodes. These were high-voltage ceramic - capacitors (Murata Corp. #DHS60Z5V272Z-40) rated at 2.7 ; nF, 40 kV dc. In addition, eight capacitors o~ the same type (labelled 150 through 164 in Figure 4) were used for the preionization bank. A PVC main body tube 144 was machined to ha~e an inside diameter of 57.2 mm and an outside diameter of 76.2 mm. The two aluminum main discharge electrodes 128, shaped to fit on the inside wall of main body 144, were spaced 15 mm apart. The surfaces are essentially flat over a 10 x 350-mm2 area~ with edges machined round using a 1/4 inch radius cutter. Connections between each electrode and its corresponding aluminum plate (130,132~ were made with 12 small brass fittings with individual o-ring seals. The aluminum plates used for interconnecting the capacitors ` were machined round at each exposed edge, to prevent corona discharges. Immersion in oil was not required.
UV preionization was accomplished using two preionizer rods 136, 138 located along the walls of main body 144. Each preionizer rod, consisting of an array of 15 gaps 1 mm wide, spaced 25.4 mm apart, was con-structed using angled sections o~ stainless-steel tubing glued onto a 42-cm length of glass tubing of 5 mm o.d., with a coaxial cable center conductor (Belden #8868 high voltage lead) passing inside the glass tubing. The length of the shielded part of each preionizer lead was 1 m. The 7 atm pressurized spark gap and preionization .~ - -: : . . .
~ ~ .
. . ' ', ' ' . ' . , ' ' , . , . , : ~ -.
. .
~ ~7~0 capacitors were supported above the main body by l/4inch diameter copper rods 146, 148, which also served as electrical connections. Por triggering the spark gap, a trigger pin (not shown) was provided inside the hollow spark gap cathode. Optical access to the amplifier was through 5-mm thick KCl flats located at the ends o~ main body 144. The flats were mounted at 10 from normal to the body axis, to eliminate the possibility of self-lasing due to Fresnel reflections. The Lucite~nd flanges and the salt flats were sealed using O-rings, leaving a 22-mm clear aperture. The entire amplifier was supported within a Plexiglas box, which was covered with fine brass mesh to provide shielding against elec-trical noise generated by the amplifier circuit.
The charging and grounding resistors were TRW
tiger resistors. The total main discharge capacitance was c = 240 nF, while the total preionization capaci-tance C' = 22nF. The charging voltages of +Vc and -Vc against ground were supplied by a dual power supply. With 50-Meg ohm charging resistances, the amp-lifier requires 30 seconds to reach full charge.
Upon spark gap breakdown, preioni~ation commences and the four main discharge capacitor banks across which the spark gap is connected undergo voltage reversal, while the four banks connected to the main electrodes maintain their original relative voltage. A
potential is developed between the main electrodes, with the upper electrode becoming the cathode and the lower one becoming the anode. If the inversion is allowed to reach its peak, and if spark gap losses are neglected, the voltage across the main discharge gap would reach a theoretical maximum of 4Vc. The amplifier is normally ': ,', . " : ~.
- . , . . :
..
~2~t30 operated such that breakdown oE the main gap occurs just before the inversion peak.
The inductance "L" of one side of the dis-charge clrcuit can be approximately evaluated by consid-ering the configuration to be that of a sinyleturn solenoid. A current sheet, passing through the dis-charge electrodes and the capacitors of one side of the amplifier, encloses an effective area A - 131 cm2.
With a 1 = 35 cm long discharge, the inductance is L =
uOA/l = 47 nH. Using the total capacitance of C = 240 nF, the rise time of the discharge current pulse can be estimated. If the circuit was connected to an impedance matched load of 4(2L/C)-5 = 2.5 ohms/ the critically damped current would reach a maximum of vmaxe-l[c/(8/L)] 5 = 31 kA with a rise time of (LC/32)-5 = 19 ns.
The electrical and gain characteristics of the amplifier, operating at 10 atm with a gas mix composed of 6~ CO2, 6~ N2, and 88% He and a trace amount of tripropylamine were measured. The charging voltage was Vc = + 27.5 kV as determined by the two power supply meters, while the transient main-electrode voltage was 2S obtained using two identical high-voltage probes, connected to the aluminum plates attached to the elec-trodes. The probes were (7 k ohm: 50 ohm) resistor dividers, with measured rise times of 30 ns. By taking the difference o~ the attenuated voltage signals on an oscilloscope, the total electrode potential was dis-played. A Tektronix 7344 dual-beam oscilloscope, with two 100-MHz 7A12 plug-in units, was used for synchron-ized measurements of the voltage, currentt and small signal gain factor. The main discharge and preioniza-tion currents were measured using subnanosecond risetime . . .
.
,' ' , ' ' ,' ~' ' . ', : ~ ' ' ' Rogowski pick-up coils. Integration of the dI/dt siy-nals from these coils gave the currents, with a Eurther integration used for calibration. By adjusting condi-tions for observing free ringing oE the primary (inver-sion) loop of capacitance Cp = 30 nF, it was deter-mined that the spark-gap resistance was Rp a 210 Meg ohms and the primary loop inductance was Lp = 390 nH.
The main-gap voltage reached a peak of Vmax = 106 kV at 310 ns after spark gap breakdown, which represents 96% o~ the ideal maximum of 4 Vc.
~he time of the initial drop in voltage upon breakdown was observed to correspond closely to that of the initi-ation of the main discharge current, which occurred with a jitter of less than 6 ns. After the main discharge, a ringing potential, with an average potential of VF =
23 kV was observed across the main gap. The discharge current pulse reached its maximum value of 17.5 kA with a 27-ns 0.1-0.9 rise time and the pulse had a 70-ns FWHM. The preionization current started along with the inversion, after a small ormation lag time of about 50 ns, and reached a peak of 7.3 kA at t = lO0 ns, or 260 ns before the main current peak.
The product of the voltage and current curves yields the power pulse during the main discharye, while the discharye resistance R(t) may be found by taking the ratio. At the time of the peak discharge current, the resistance was 3.4 ohms, which indicates that the condi-tions are quite similar to criti.cal damping. The rise time, the peak value, and the overall form of the current pulse are comparable to those of the critically damped approximation above.
{~) The single-pass small-signal yain ~actor r, of the amplifier was measure~ on the 9.29 um, 9R(16) rotational-vibrational transition. The small-signal gain coefficient ~ is given by r = exp (~1), where 1 = 35 cm is the discha~rge length. The amplifier may be used to amplify pico-second laser pulses centered at ~his wavelength, as part of a system based on an injection-locked, Q-switched, mode-locked and cavity-dumped 10-atm CO2 oscillator. The gain signal r was measured together with the current signal using a l-W probe beam from a TEMoo mode, grating tuned cw C2 laser. Using a beam splitter, the cw laser output was simultaneously monitored on an Optical Engineering~
C~2 spectrum analyzer. The 9-mm beam, after making a single pass down the center of the amplifier discharge volume, was focused, using a 10-cm focal length KCl lens, onto a Labimex P005 HgCdTe room temperature detector. The detector and oscilloscope were located within an electrically screened room.
The gain signal r exhibited an initial dip during the time of the current pulse, which is attribu table to inverse bremstrahlung absorption by the dis--charge~ The signal had a 10%-90% rise time of 120 ns and a FWHM of 750 ns, both measured relative to the r=
1 line. The signal reached a maximum of rmax = 7 3~
corresponding to a small-signal gain coef~icient on the 9.294-um, 9R~16) line of ~ ax = 5.7~ cm~l.
Photographs of the discharge volume showed an approximately 10.5-mm-wide discharge, and the nominal specific energy input for the (15 x 10.5 x 350)~mm3 discharge was calculated to be 156 Jl~l atm~l.
The energy deposited in the discharge was taken to be the initial energy stored in the main bank capacitors, .
".
.. ~ , ' 37~
minus the energy of CVF2/32 remaining in the bank after the main discharge. The 156 Jl~l atm -1 figure was calculated on the basis of a nominal capaci-tance of 2.7 nF per capacitor. At high voltages, such capacitors exhibit reduced capacitances, and taking this ef-fect into account results in an actual deposited energy density of 120 Jl~l atm -1. By allowing suffici~nt time for complete gas replacement between shots, the amplifier performed ~ith excellent shot-to-shot reliability and reproducibility. The peak small-signal gain coefficient varied by only + 0.1% cm~l over five consecutive shots, taken at 60-s intervals.
For the 750-cm3 chamber volume, and an axial flowrate of 20 cm3s~l, measured using a curved tube flow-meter, essentially complete gas replacement occurs within this time. Reducing the concentration of C02 and N2 to 3~ each, with higher charging voltages, the amp-lifier is expected to operate reliably to about 15 atm.
With the concentration of CO2 and N2 at 6~ each, as used for the present study, 10 atm is close to the limit for reliable operation. The glow discharges contain, as is usual, several small streamers which are observed to extend, from the cathode, about 2 mm into the glow.
The-effectiveness of this amplifier has been clearly demonstrated by the measured discharge speed and observed relability. The amplifier is highly suitable for its role as a large-aperture, high input energ~, compact, and efficient device. The 5.7% cm~l small-3~ signal gain coefficient is, to the inventors' knowledge, the highest reported for any large aperture U~ preion-ized TE CO2 laser.
;
As in the case of the double-sided LC inver-sion circuit of the first embodiment, as many stayes as , .
'; ' ' ' ~ . ' . ' - . , ',' : ~,' .: ', , ' .
.
., :, . .. . .. .
desieed may be added to the double-sided four fold LC
inversion circuit of the second embodiment, using several of the eight-part circuits of Figure 4 connected in series. Such an arrangement would be appropriate Eor larger electrode separations and/or higher pressures.
Although the rise time of (LC/32) 5 depends inversely on the number of stages n, for laser excita-tion at optimum energy density, one should expect the required C to increase with pressure and discharge volume. The rise time would end up to be quite similar to the 27 ns of the n = 1 circuit described here, show-ing that such circuits would indeed be successful. For excitation of many types of gas ].asers, this circuit provides a realistic alternative to Marx bank or conven-tional LC inversion methods.
As will be apparent to those skilled in theart in light of the foregoing disclosure, many altera-tions and modifications are possible in the practice of this invention without departing from the spirit or scope thereof. Accordingly, the scope of the invention is to be constLued in accordance with the substance defined by the following claims.
.: . ..... , , .
- , . . : ". . .
. . .
Claims (8)
1. An amplifier for delivering high voltage, high current pulses to a dual port load, said amplifier comprising:
(a) first and second series-connected capacitors;
(b) third and fourth series-connected capacitors; and, (c) a voltage source connected in parallel across said first and second capacitors;
wherein:
(d) said capacitors each have substantially identi-cal capacitance;
(e) said first and second capacitors are connected in parallel across said third and fourth capaci-tors;
(f) a first port of said load is connected between said first and second capacitors;
(g) a second port of said load is connected between said third and fourth capacitors;
(h) said first and third capacitors are positioned adjacent one another on a first side of said load; and, (i) said second and fourth capacitors are positioned adjacent one another on a second side of said load, opposite said first side of said load.
(a) first and second series-connected capacitors;
(b) third and fourth series-connected capacitors; and, (c) a voltage source connected in parallel across said first and second capacitors;
wherein:
(d) said capacitors each have substantially identi-cal capacitance;
(e) said first and second capacitors are connected in parallel across said third and fourth capaci-tors;
(f) a first port of said load is connected between said first and second capacitors;
(g) a second port of said load is connected between said third and fourth capacitors;
(h) said first and third capacitors are positioned adjacent one another on a first side of said load; and, (i) said second and fourth capacitors are positioned adjacent one another on a second side of said load, opposite said first side of said load.
2. An amplifier as defined in claim 1, further comprising preionization means for preionizing said load, said preionization means comprising:
(a) spark triggering means connected in series between said voltage source and said first port of said load;
(b) a first spark array connected between said spark triggering means and said first port of said load, and positioned on said first side of said load; and, (c) a second spark array connected between said spark triggering means and said first port of said load, and positioned on said second side of said load.
(a) spark triggering means connected in series between said voltage source and said first port of said load;
(b) a first spark array connected between said spark triggering means and said first port of said load, and positioned on said first side of said load; and, (c) a second spark array connected between said spark triggering means and said first port of said load, and positioned on said second side of said load.
3. An amplifier as defined in claim 1 or 2, wherein said capacitors and said load are connected and configured to minimize circuit inductance.
4. An amplifier as defined in claim 1 or 2, wherein:
(a) said first, second, third and fourth capacitors each comprise a plurality of discrete capaci-tors;
(b) said first capacitors, said third capacitors, and said load define a first electrical circuit loop about a first axis;
(c) said second capacitors, said fourth capacitors, and said load define a second electrical circuit loop about a second axis;
(d) the cross-sectional area of said first and second circuit loops is minimized in planes which are respectively perpendicular to said first and second axes; and, (e) the length of said first and second circuit loops is maximized in the direction of said first and second axes respectively.
(a) said first, second, third and fourth capacitors each comprise a plurality of discrete capaci-tors;
(b) said first capacitors, said third capacitors, and said load define a first electrical circuit loop about a first axis;
(c) said second capacitors, said fourth capacitors, and said load define a second electrical circuit loop about a second axis;
(d) the cross-sectional area of said first and second circuit loops is minimized in planes which are respectively perpendicular to said first and second axes; and, (e) the length of said first and second circuit loops is maximized in the direction of said first and second axes respectively.
5. An amplifier for delivering high voltage, high current pulses to a dual port load, said amplifier comprising:
(a) first and second series-connected capacitors;
(b) third and fourth series-connected capacitors;
(c) fifth and sixth series-connected capacitors;
(d) seventh and eighth series-connected capacitors;
(e) a first voltage source connected in parallel across said first and second capacitors; and, (f) a second voltage source connected in parallel across said third and fourth capacitors;
wherein:
(g) said capacitors each have substantially identi-cal capacitance;
(h) said first and second capacitors are connected in parallel across said third and fourth capaci-tors;
(i) said fifth and sixth capacitors are connected in parallel across said seventh and eighth capaci-tors;
(j) said fifth and sixth capacitors are further connected in series between said second and fourth capacitors;
(k) said seventh and eighth capacitors are further connected in series between said first and third capacitors;
(l) a first port of said load is connected between said first and second capacitors;
(m) a second port of said load is connected between said third and fourth capacitors;
(n) said first, third, seventh and eighth capacitors are positioned adjacent one another on a first side of said load; and, (o) said second, fourth, fifth and sixth capacitors are positioned adjacent one another on a second side of said load, opposite said first side.
(a) first and second series-connected capacitors;
(b) third and fourth series-connected capacitors;
(c) fifth and sixth series-connected capacitors;
(d) seventh and eighth series-connected capacitors;
(e) a first voltage source connected in parallel across said first and second capacitors; and, (f) a second voltage source connected in parallel across said third and fourth capacitors;
wherein:
(g) said capacitors each have substantially identi-cal capacitance;
(h) said first and second capacitors are connected in parallel across said third and fourth capaci-tors;
(i) said fifth and sixth capacitors are connected in parallel across said seventh and eighth capaci-tors;
(j) said fifth and sixth capacitors are further connected in series between said second and fourth capacitors;
(k) said seventh and eighth capacitors are further connected in series between said first and third capacitors;
(l) a first port of said load is connected between said first and second capacitors;
(m) a second port of said load is connected between said third and fourth capacitors;
(n) said first, third, seventh and eighth capacitors are positioned adjacent one another on a first side of said load; and, (o) said second, fourth, fifth and sixth capacitors are positioned adjacent one another on a second side of said load, opposite said first side.
6. An amplifier as defined in claim 5, further comprising preionization means for preionizing said load, said preionization means comprising:
(a) spark triggering means connected in series between said first and second voltage sources;
(b) a first spark array connected in parallel across said first and second voltage sources, and positioned on said first side of said load; and, (c) a second spark array connected in parallel across said first and second voltage sources, and positioned on said second side of said load.
(a) spark triggering means connected in series between said first and second voltage sources;
(b) a first spark array connected in parallel across said first and second voltage sources, and positioned on said first side of said load; and, (c) a second spark array connected in parallel across said first and second voltage sources, and positioned on said second side of said load.
7. An amplifier as defined in claim 5 or 6, wherein said capacitors and said load are connected and configured to minimize circuit inductance.
8. An amplifier as defined in claim 5 or 6, wherein:
(a) said first, second, third, fourth, fifth, sixth, seventh and eighth capacitors each comprise a plurality of discrete capacitors;
(b) said first, third, seventh and eighth capaci-tors, and said load, define a first electrical circuit loop about a first axis;
(c) said second, fourth, fifth and sixth capacitors, and said load, define a second electrical circuit loop about a second axis;
(d) the cross-sectional area of said first and second circuit loops is minimized in planes which are respectively perpendicular to said first and second axes; and, (e) the length of said first and second circuit loops is maximized in the direction of said first and second axes respectively.
(a) said first, second, third, fourth, fifth, sixth, seventh and eighth capacitors each comprise a plurality of discrete capacitors;
(b) said first, third, seventh and eighth capaci-tors, and said load, define a first electrical circuit loop about a first axis;
(c) said second, fourth, fifth and sixth capacitors, and said load, define a second electrical circuit loop about a second axis;
(d) the cross-sectional area of said first and second circuit loops is minimized in planes which are respectively perpendicular to said first and second axes; and, (e) the length of said first and second circuit loops is maximized in the direction of said first and second axes respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA 565436 CA1287890C (en) | 1988-04-28 | 1988-04-28 | Automatically preionized double-sided lc inversion circuit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA 565436 CA1287890C (en) | 1988-04-28 | 1988-04-28 | Automatically preionized double-sided lc inversion circuit |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CA1287890C true CA1287890C (en) | 1991-08-20 |
Family
ID=4137927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA 565436 Expired CA1287890C (en) | 1988-04-28 | 1988-04-28 | Automatically preionized double-sided lc inversion circuit |
Country Status (1)
| Country | Link |
|---|---|
| CA (1) | CA1287890C (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011081516A2 (en) | 2009-12-31 | 2011-07-07 | Specscan Sdn. Bhd. | Low inductance integral capacitor assembly |
-
1988
- 1988-04-28 CA CA 565436 patent/CA1287890C/en not_active Expired
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011081516A2 (en) | 2009-12-31 | 2011-07-07 | Specscan Sdn. Bhd. | Low inductance integral capacitor assembly |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US4455658A (en) | Coupling circuit for use with a transversely excited gas laser | |
| US4534035A (en) | Tandem electric discharges for exciting lasers | |
| US4876693A (en) | Integrated laser head and low inductance pulse forming circuit for pulsed gas lasers | |
| Kearsley et al. | A novel pre-ionization technique for discharge excited rare gas halide lasers | |
| RU2155443C2 (en) | High-voltage impulse generator | |
| US4542529A (en) | Preionizing arrangement for transversely excited lasers | |
| Martinez et al. | High-efficiency low-pressure Blumlein nitrogen laser | |
| CA1287890C (en) | Automatically preionized double-sided lc inversion circuit | |
| Taylor et al. | Magnetic-spiker excitation of gas-discharge lasers | |
| US4797888A (en) | Circuit for the preionization and main discharge of a pulsed gas laser | |
| Baker et al. | An efficient laser pulser using ferrite magnetic switches | |
| US4292600A (en) | Pulsed gas laser emitting high-power beam of short wavelength | |
| RU2368047C1 (en) | Device for generating volumetric discharge | |
| Sarjeant et al. | Parametric study of a constant E/N pumped high-power KrF* laser | |
| Stuart et al. | High gain, multiatmosphere CO2 laser amplifier | |
| Targ et al. | High-repetition-rate xenon laser with transverse excitation | |
| Houtman et al. | High‐speed circuits for TE discharge lasers and high‐voltage applications | |
| Marotta et al. | A simmered pre-pulsed flashlamp dye laser | |
| US4178563A (en) | System for generating a high-energy electrical signal through a brief time, and a laser comprising such a system | |
| US4612643A (en) | Electric glow discharge using prepulse and charge storage device | |
| Luches et al. | Output characteristics of an excimer laser with delayed double preionisation | |
| US4450566A (en) | Pre-ionizing trigger system for a gas laser | |
| Persephonis et al. | Capacitance allocation and its role in the performance of doubling-circuit pulsed gas lasers: its application to the N/sub 2/laser | |
| Sanz et al. | Peaking capacitor in an incomplete corona surface discharge preionized TEA CO/sub 2/laser | |
| CA2066804A1 (en) | Discharge-pumped gas laser with independent preionizing circuit |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| MKLA | Lapsed |