Three Surplus tank lasers are arranged side by side in a MOPA (Master Oscillator and Power Amplifier) configuration

Home Return to Main Page

Notes on a Ruby MOPA 2 Arrangement

The purpose for this arrangement is to generate holographic light.

reflect2.gif Diagram of single beam reflection hologram setup

Tank lasers were setup in a MOPA (Master Oscillator and Power Amplifier) configuration. This arrangement is used inorder to get a cleaner beam and also obtain power for good development of film. Although these lasers are used by some to burn things and with modifications can be setup to do just that although with limitations as you reach a point when you burn antireflection or surface pit the optics. This arrangement with the specified below power supply is capable of generating multimode output in excess of 500 mj. Instead the object is to reduce the power of the Master Oscillator to keep other spatial and longitudinal modes from developing. The desire is to have a good TEM00 spatial mode and a good coherence length. Result was a laser that produced good sharp holograms but dim. This also may be due to developer I was using at the time.

The tank laser has many good features for use in designing a Holographic laser:

Q switch for short pulse high peak power, resonant reflectors improving coherence length in addition to a folded resonator about 33.5 cm long (compact and a good length), good prism optics, Porro and rotating prism make easy alignment for a plane-parallel resonator. Half cavity for good heat dissipation.

MOPA Head Arrangement

Three surplus tank lasers are arranged side by side. See Picture and Diagram.

rw_mopa2.gif Diagram of mopa version 2

rw2a.jpg Picture of external case

rw2b.jpg Picture with front cover off

rw2c.jpg Picture with chassis out of case

rw2d.jpg Close up of mopa

Two 90 degree angle prisms are also used to fold the beam path to the power amplifiers. The Power amps have been modified by enlarging the 6mm bores for the flashtube to accommodate the 9 mm bore of the larger tubes. Each power tube is powered by 800 joule electrical input. The Master Oscillator is powered by the standard 4mm tube at 200 joule electrical input.

The Oscillator resonator has a variable aperture inserted for TEM00. Set at 1 mm the TEM00 mode is about 5 mj output. The delay to fire for the oscillator flashtube can be adjusted to fine tune the laser input just above threshold to improve transverse and axial modes. The Resonant reflector improves the longitudinal mode selection. But due to not keeping the temperature finely controlled, this also can still be problematic. Basically I mounted the lasers on a large aluminum slab ½ inch thick mounted on top of a large inverted heatsink with radiator fins. The combination of a large latent heat sink of the slab and radiation of the radiator helps keep the unit somewhat temperature stable using ambient environmental cooling and firing the laser once every 25 minutes. I did not want to use cooling fans due to dust contamination or other cooling/heating methods as I wanted to basically keep it simple. One of the thermister in the cavity is used in parallel with one mounted on the slab. This is plugged into a Radio Shack temperature module for reference. The power amplifiers due to their rod lengths basically double the output for each stage passed. By tuning the delay to fire for the flashtube I was able to get 2.3x output from each stage. Yielding a final output of 25-30mj, but eventually with the oscillator running at a higher level and a slightly larger aperture the oscillator yielded 8 to 10 mj and the amps brought the power to 54mj.

Alignments

Inorder to obtain the gain desired it is important to align the ruby’s C-Plane axis of both the oscillator and amplifiers. All should be in the same plane either vertical or horizontal. Since Porro prisms are used it is important to keep it either vertical or horizontal inorder to keep good polarization. The E field will be perpendicular to the C axis.. Due to this arrangement it will also create the most absorption when firing the oscillator through the unpowered amplifiers. I found that a 12mj beam will emerge from the first pass of the two amp stages at about 1mj power. That same test with both amps powered up will yield 63 mj. Inorder to align the prisms for the beam path, I had to operate one amp just to keep from losing the signal while aligning with alignment paper. Care must be taken to allow for 1/2 times the TEM00 beam size clearance on any optic as not to create additional diffraction (the 2x to 3x diameter beam size aperture rule). In our case I centered the beams within the rod.

Beam Quality

I found it hard to eliminate the other TEM modes from the laser perhaps due to rotating mirror alignment during pulse buildup time. Using a 1mm aperture, I managed to get TEM00. Getting precise alignment centering of the Aperture helped due to the HR being a porro. Which by the way I had to replace by finding another one from another tank unit. I choose the best optic and motor bearing quality laser I had for my oscillator. But found that I was getting two beam modes that looked like rectangular tem01 mode. But instead it turned out to be a bad tem00 mode due to the roof line of the porro was poor. I salvaged the best one I could find out of my litter. Aligned up the unit and got good TEM00 or as best as it gets with these units. Beam pulse width was measured at about 40 nsec duration and in the single pulse mode. With the resonant reflector setup and a MOPA arrangement , I found I could get Holographic scene depth of around 4 inches by estimate using the single beam reflection method which implied an 8 inch coherence length. Again this is just a rough guess at it. I was impressed with the sharpness of the image, not impressed with the overall dimness of the hologram as I had to use sunlight to see it well. I assumed power levels was the issue and never investigated developers as an issue. Overall this setup made for a good holographic laser without much expense invested.

Q Switching and Firing Control

The Q switch motor was ran backwards to allow for the correct timings of firing the power amps and Master oscillator flashlamps due to a longer flash pulse to keep the flashtubes at a lower percent of explosion limit. Also the motor was voltage regulated to run at 30,000 rpms. Yielding 2msec between trigger pulses. The Osc was adjusted for a 1.1 msec delay. The power amp was adjusted for a 1.2 msec delay. This was accomplished by using pairs of 556 timers in their own cascaded arrangement. An additional 555 timer controls a relay for the motor run time of about 2 seconds. Firing protection is accomplished by using a manual safety switch. With the switch in the closed (Safety off) position it supplies the 300vdc for the primary trigger circuit. When the trigger fire momentary is closed this starts the 555 motor/relay. Allowing for motor surge to settle, 500 msec later a 556/relay in series with the safety manual switch completes the path to charge the trigger primary coil caps. Then another 556/relay closes the connection from the magnetic pickup (trigger timing pulses) 200 milliseconds later. The Timing pulses are rectified and amplified/inverted by a transistor. This then feeds the pair of 556 flashlamp timers. These 2 timers then fires into 3 SCR/trigger circuits to dump the trigger primary power for HV triggering at each flashlamp.

Oscilloscope of motor q switch operation and output rwmopao.htm

timcir.htm Timing circuit used to control Q switch timing. Power supply and SCR trigger circuit/Series Injection Trigger Transformer circuits. The only circuit not show is the motor control/ relay circuits etc that are based on the same NE556 delay circuit but the output drives a relay to control the motor, and make connections for the firing /safety system.

Rotating prism has three reflections during rotation. 1^st at the fire position, 2^nd at the 144 degree position and 3^rd at the 216 degree position. This should be kept in mind for firing timings.

Power Supply

Each amp is fed from a central 30kohm current limited fullwave diode bridge 850v transformer. Yield about 1150 volts fully charged. Each amp uses 3 450v 3900 ufd electrolytic capacitors in series with a resistor voltage divider. I used 2watt 500k carbon resistors across each cap to help keep the voltage within the limits and to provide a discharge path when the supply is turned off. The Osc uses 3 450v 1000 ufd caps. These 9 caps yielded the most compact arrangement. 3 HV diodes and 2 switches are used to provide three separate legs allowing to turn on/off which optical amp power supply and it‘s laser. The diodes are used to prevent one bank of caps from discharging into another laser through the common power supply. A 300 volt supply is created from a second transformer/Fwbridge. This dumps into a 400v cap to supply enough energy to fire the all the trigger SCRs. Each flash tube is arranged in a series injected circuit. A highvoltage trigger pulse from the secondary of the in series transformer/inductor provides the flash start. A SCR shorts a trigger cap into the trigger primary.