Home Return to Main Page

Version 7 Installs less amps and uses helical flashtubes for more uniform pumping and better beam quality.

Version 6 laser gave good power levels and made average quality holograms. These holograms were much brighter due to the 1 to 1.5 joule output power. Most of that power was used in object beam diffused lighting. Version 6's Dual linear flashlamps allowed for compact electrolytic capacitor power supplies. The beam divergence issue due to dual linears creating elliptical beams was solved by placing some amps with their flashlamps in the horizontal position and the other amps in the vertical position. This averaged the divergence differences to create a circular output beam.  Version 7 begins to address issues of diffraction limited qualities and spectral qualities of the beam.  Larger rods can reduce rod end fluence densities as beam size could be further expanded.  The new ruby rods are at a .03% cr doping level. It is anticipated, that the lower doping level and frosted barrel of the rod and use of helical flashlamps will allow a more even pump concentration in the rod. Amp 1 will use a 1/2" x 7.6" rod and Amp2 will use a 3/4" x 7.6" rod. The oscillator will use either a 7mm x115mm rod or a 1/4"x3" rod.

rwmopa7a.gif Diagram of upgraded version 6.

r7osc.htm Application notes on the oscillator upgrade.

r7h2.jpg Picture of version 7 during oscillator axial mode study. The resonator frame also was moved over to make way for the new amps that will be mounted separate from the resonator frame. Also the dye cell was moved behind the aperture so that the aperture shielded flashlamp light from hitting dye to improve the dye cell performance. This way the laser radiation is the light that is used for operating the dye q switch only.

r7h1.jpg Picture of R7 head looking toward the OC and output beam steering mirrors. Single longitudinal mode operation has been achieved by aligning the sapphire etalon with the OC and HR forming additional Fabry Perot resonant distances to both. The 3mm etalon is 23.5" from the HR and 7" from the OC.  Axial mode study was done with the 1/4"x3" oscillator. The output is 9 to 15 mj at 15 nanosec pulse duration TEM00 mode and single axial mode. Oscillator was ran at 900volt 520joule input. TEM00 mode assured by using an aperture set to the Fresnel number 0.6 (1.16mm). Later tests were able to confirm single mode operation even at single pulse levels of 24 millijoule output. Power supply was set at 1000volt 660joule input. Aperture was set at Fresnel number 0.97 (1.51mm). Dye concentration level adjusted to obtain the single pulse. For best spatial quality the holographic laser will be operated at Fresnel 0.6 and no more that 15mj output from the oscillator.

r7h3.jpg  Another Frame member was added and the 3mm etalon/KS1-t mount was mounted directly to the invar frame to minimize possible thermal variation of the Composite OC alignment.

r7p2.jpg Picture of version 7 power supply which has two power lines, three variacs, and three transformers: 1 microwave trans for osc, 2 4500vac trans for the amps. Amps are voltage doubled using doubler caps and rectifiers. The power supply also contains all low voltage transformers/rectifiers and OSC voltage digital meter. Front panel contains switches to switch off each power supply at the variacs and a switch to remove the 300v and 850v trigger voltages as a safety. Since all voltages can be turned off. The main switch can supply the low voltages for the HENE pwr supply and the logic circuits so that alignments to the optics can be made and timing circuits verified when needed. The power supply uses 2" Ceramic posts are used as terminals to connect HV cables which will carry upto 7.5kv.

r7p1.jpg Picture of version 7 amplifier caps to be used. Safety covers, shorting bars, voltage meters, inductors and additional circuits will be installed over these caps. Torque by mfg:  15 ft-lbs max.

r7p3.jpg Picture of version 7 amplifier power supply near completion.

r7p4.jpg Close-up Picture of completed version 7 amplifier power supply. Inductors are hand wound 10 gauge air coils dipped in resin and fiberglass wrapped. Meter voltage divider board  uses (13) 2watt resistors so that the voltage across each resistor did not exceed 553 volts. Ceramic Bleeder resistors for storage cap are the 100 and 200watt rating . Max voltage rating are 4000v each therefore two resistors are in series

r7sch.gif Schematic of Power Supply. Just amp circuit is shown. Physically the 5 to 8kv connections from the power supply to the laser head are made point to point with no plugs. The ends of each 12gauge 15kv cable is terminated  with a soldered ring lug terminal on a ceramic post 2" high. At the laser head these cables go directly to the pulse transformer connected by ring lug terminals. Corona points are minimized. Return ground wire is a 4 gauge wire from laser head to each power supply chassis.  .

helix.gif Diagram of helical cavity for amplifier stages. This is the redesign of 2/11/02. Earlier version attempted to enclose the entire lamp and terminals but had suffered thermal and arcing issues. 

r7amp.jpg     External view of helical amplifier cavity using the SG-2600 helical lamp. All aluminum design, small cavity with terminals exterior help eliminate arcing  and thermal issues. 

amp1.jpg   Internal view of helical amplifier cavity.

r7af.jpg   Test firing of osc and Amp. The Amp is producing a high intensity fluorescence light.

r7af2.jpg  Next video frame of test firing. Amp1 increased the 16.1mj osc pulse to 196.5mj.   Amp1 electrical input 9400 joules.

hel128.jpg 128 milljoule output from Amp1  as captured on Kentek Zap-it paper. Notice a very smooth output across the face of the beam and beam quality from the amp is more uniform than the dual linear amps of version 6.

r7mopa.jpg  Tested laser. No preamp used at this time. 24.3kj electrical input  (0.93kj osc, 11.1 kj amp1, 12.2kj amp2)  Output was 1.515 joules with 10mj oscillator.  Some thermal surface damage occurred on Amp1 and Amp2 due to SG-2600 lamps being too close to rod. SG-2800 lamp will be used on AMP2 with a 1.3 inch ID instead of SG-2600 ID of 0.9 inch. Amp1 will be ran with the SG-2600 lamp but a  lower power level of 8 to 9 kjoules to also help reduce thermal stress for the air cool setup.

amp2.jpg  view of amplifier 2 using the SG-2600 lamp.

r7resosc.htm    Above new oscillator cavity will now be used as a preamp and old oscillator was upgraded to a double resonant reflector setup.

r7fosc.jpg  New oscillator using dual linear flashlamps.  Cavity machined as double ellipse. Each flashlamp is set 0.66 inch from centerline of rod. Eccentricity 0.6 minor axis 0.8 inch. End plates use set screws to hold rod using a aluminum plug between screw and rod. Also a stainless steel set screw hold cathode terminal end of flashlamp. Anode terminals exits   the cavity onto clamp holders on ceramic post.  Each flashlamp  has separate power/trigger circuit supplying 900 joules each. New cavity setup provided 25 mj 15 nanosec single pulse and with amplifiers boosting 150x, the final output of the laser should be  3.7 joules.

r7osc.htm Application notes on the oscillator upgrade to help reduce axial modes and improve coherence length. Typical number of modes for version 6 was 2 to 6 modes or a coherence length around 0.5 meter on average. Version 7 operates in single longitudinal mode and coherence length increases to 2 to 3 meters instead. Limited by the linewidth of the transform limited pulsewidth of 12 nanosecs and the additional frequency chirp that ruby lasers generally undergo.

r7beam.htm Beam profiles of the version 7 oscillator  from the application notes on the oscillator upgrade. For Plano-Concave resonator, use weights.xls below to calculate the passive resonator geometry including the Fresnel number.

r7resosc.htm  R7 upgraded using double resonant reflectors installed in the oscillator resonator configuration.

r7mopa.jpg Assembled version 7 laser head. 24.3KJ electrical input. 1.5 joule TEM00 and single longitudinal mode output. 12 nanosec 10 mj pulse amplified 150x of helical pumped amplifiers: Amp1: 1/2 inch dia. x 7.6 inch long rod and Amp2:  3/4 inch dia. x 7.6 inch long rod. Osc ran at 384 electrical joules/cm³ of ruby rod.  Amp 1 ran at 453J/cm³ Amp 2 ran at 221J/cm³ . Output from osc was beam expanded by 25mmCC/150mmCX telescope (6x). Beam expander was adjusted to give a moderate beam divergence as it passed though the amps. Laser produced very smooth and round 12mm diameter beam at ouput.

r7fosc.htm New modification for version 7: Dual flashlamp oscillator.

HELICAL FLASHLAMP INFORMATION SECTION.

The helical flashlamps are Spaceglass SG-2600  for  both Amps   It is recommended not to run helicals over 40% of their max value and generally should be ran at 20% for longer life (50,000 flashes instead of 250).

The SG-2600 flashlamp are powered by a 460 ufd 8kv Maxwell oil capacitor for each amp  Inaddition each circuit contains a LM640 trigger transformer and inductors to get to 400 uh for a current pulse of 1.2 millisecs. Each amp will be pumped to a 200-380j/cm3 electrical input level. Amp1 is  pumped at 9400 joules and amp2 13000 joules Which should make an energy storage of 2.5 to 3.3 j/cm3 for the amplifiers. At this level a 15mj single pulse from the oscillator should be amplified to a 2 joule pulse and due to the pulse having been beam expanded by a telescope the energy density should be around 3 j/cm2 and therefore below the damage threshold for the rod end and AR coat. The beam size exiting the amp rods will be around 10 to 12 mm in size. This should yield a 1.4x rod aperture factor over the beam size allowing for lower diffraction effects. All other optical components have a tilt to minimize a back reflection from entering a previous amplifier. In these amp rods the rod itself has had both rod ends ground and polished parallel with each other and with a 2 degree tilt with respect to the rod's axis to accomplish the same thing. Construction is under way and will report actual performance once completed.

Helicals are often filled with lower torr levels of around 300 to 350 for easy triggering and therefore will also depending on arc length have a min and max voltage range for the supply. For example for SG-2600 with an arc length of 113.9cm the min supply voltage is 3k and below this then it will not fire even if the correct trigger voltage is used. Above 10K for a supply voltage the tube may self trigger. Linears at 16cm and higher torr of 450 may only have a min of 600v and a max of 3kv. Triggering long helicals like the 6" to 8" variety generally require 25 to 35kv. It appears for the SG-26000 and the K6, the LM640 trigger transformer, a 25 to 1 voltage ratio, a 21kv trigger would work. Which means a primary supply of 819 volts upgrading from the 600v I had used before to trigger the 5" long linear flashlamps. The triggering SCRs will be upgraded to the 1200v 35amp type (I.R. 40TPS12) and going from the 0.2ufd currently used to 2 ufd for more current dumping into the primary. Peak current is estimated at 600 amps which is at the max value for the 40TPS12 SCR (circuit is 3uh primary, 2 ufd cap 0.65 joules).

Another issue is helicals should be supported in the middle of the coils or have a reflector that will do this. Inaddition due to shock and thermal issues the coils will expand and the unit will uncoil slightly during firing so the connections have to take this movement into account. Companies like Apollo and Korad used a flexible connection with a braided cable connected at one end to allow this movement. Since I will fire this only once per 30mins or more then it can be air cooled instead of water cooled. The main issue is the thermal time for the laser rod to cool as an air cooled head will absorb more of the long IR wavelengths from the hot flashlamp. Yet since it air cools by convection at a slower rate the thermal stress on the crystal is less than a water cooled rod. Ruby has a thermal shock value of 100w/cm compared to yag's 7.9w/cm and glass 1w/cm ( Walter Koechner. Solid State Laser Engineering).

Technical data on the  helicals:  helical.htm

In regard to the K6 lamps:

I would suggest a 1 msec pulse length instead and limit the joules to 8000 or less limiting to 28% of explosion or less. I found that to get the unit to flash I had to set the supply voltage at 4400 volts minimal and the trigger circuit had to use a 2 ufd cap at 819 volt fired into a LM640 (25 to 1) trigger transformer as an inline series injection setup which produced a high enough current spark to get the flashlamp to fire. For this lamp the DI/DT or the rate of current has to be great enough for the trigger to fire the lamp, even though the trigger voltage is enough. I fired the lamp at 5200 volts at 468 ufd into 360uh circuit to yield a 6400j pulse at 1.5 msec pulse width. I think the proper circuit would be around 200 ufd and 8000v to get the damping factor closer to 0.8.

In regard to the SG-2600 lamps:

It turned out that in this setup, the main energy supply voltage had to be above 6000volts inorder for the flashlamps to fire and inaddition,  3ufd on trigger primary was required to get reliable firings. Most of this is due to the new cavity design for the flashlamp where the aluminum cavity is isolated from ground by 1/2 inch ceramic posts. Therefore no grounding planes are brought near the anode to help in triggering which would allow lower bank voltage and less trigger spark current inorder to fire. The use of a grounding plane was not used to prevent potential arc-overs. In the K6 lamp experiments the lamp was held over ground plane which allowed easlier triggering and lower bank voltage but cavity arc overs had also occurred when using aluminum enclosures. In the SG-2600 setup, the terminals are moved to the outside of the cavity to reduce uv ionization and the cavities are isolated from ground.  

amp1cur.gif    Actual performance of SG-2600 with circuit. 458ufd,  360 uh, 6400volts.   Based on lamp rise time of 200usec and 3000 amp peak. Damping factor was near 1.5,   Rt is calculated to be 1.28 ohms, Zo = 0.86 ohms. and Ko is around 112 ohm-amps^1/2     

Updates:

8/14/02.  Well almost finished, instead decided to upgrade osc to dual lamp design for more power output by using a longer rod (115mm). This will ease amplifier requirements. Amp2 lamp broke due to too loose support and it banged into 3/4 ruby rod. This also caused some surface damage from the hot lamp. Amplifier 2 will be rebuilt with larger diameter helical lamp SG-2800. Amp 1 will be ran at lower power levels.

7/28/02 Laser finished!!! Ready to checkout holographic quality.

7/14/02. Double Resonant reflectors abandoned for Plano concave arrangement again. 4 Plate etalon used for OC. 16mj single axial mode output. 1.1mm aperture (0.8 fresnel number) for TEM00. 19inch resonator length.

6/14/02. Resonator mirror separation was reduced to 20.5 inch and the 3mm etalon was placed at 17.5 inch. Since the laser was single axial mode then the coherence length was long enough even after this change. The change was made to allow a pre amp to be used in the future if higher energies was needed. The beam expander was increased to a 6x for better fill of the amplifiers. New oscillator cavity with 7mm x 115mm rod was removed for use as future pre amp and 1/4"x3" modified tank cavity was reinstalled as the oscillator.

6/03/02 at 8300 joules the helical lamp melted the glass wool used to shock absorb the lamp. Instead heavy duty aluminum foil was used. This also melted so braided stainless steel mesh used instead. This was obtained from braided water supply lines. Will make amp 2   larger than 2.3inch x 2.3 inch x 7.8 inch. Instead 2.3"x2.6"x 7.9 inch. 2.6" will allow more room for the helical glass terminals so lamp is symetrically centered over rod. 8" long will increase cavity inside length from 7.1" to 7.2" to minimize rod shielding by holders.

5/29/02 A smaller rectangular cavity was designed: 2.3"x2.3"x7.8" to contain the custom designed spaceglass SG-2600 Flashlamp (helical flashlamp with flexible leads). This cavity was mounted on 1/2 inch ceramic posts to isolate the cavity from ground. The flashlamp anode and cathode ends extended from two holes on top the  all aluminum cavity. 4" ceramic post are positioned beside the cavity for attaching the power supply cables and the flashlamp flexible leads. Flashlamp has a metallic reflector 1/16" sheet wrapped around the flashlamp. The flashlamp was installed in the cavity with refractory wool on outer edge of each flashlamp terminal to reduce the lamp shock within the cavity.The flashlamp fired at 5800 volts. 458ufd capacitor yielded 7800 joules during first test run without ruby rod installed.

1/16/02 - Used Teflon, PVC with overlapping dove tail joints to provide electrical isolation of the anode terminal. This allowed the cavity to be used but the plastics still suffered from heat and electrical stress. So will be abandoned as insulation material for the cavity. Instead will look toward ceramic sheets.

1/2/02 - Update: Tested the new helical cavity. Lamp and rod in large aluminum enclosure and immediately had flashover to cavity box ground. This HV short to ground also caused a voltage reversal and blew the charging diodes. This reversal was caused by having the inductors in the flashlamp path and not directly in the capacitor charging path. So the caps will be wired where they get charged and discharged with the cap and inductor in series. This will prevent a voltage reversal setup by oscillatory currents due to too low inductance vs resistance in the circuit if a flashover ever occurs again. The flashover was not caused by the 25kv trigger pulse as originally thought to have happened but by the flashlamp firing and additionally creating a UV ionized path to ground for the lamp anode. So currently seeking to find better insulation for the anode that can take UV and increase the path length even more as to prevent arc overs from occurring. Also better methods of securing the flashlamp and removing any strain on the terminals so the flashlamp can fire, expand (and slight uncoiling) and absorb the shock without putting too much stress on the lamp terminals where cracking and lamp failure has been occurring. This method needs to be able to also offer springiness for shock, high heat capability and uv resistance.