My second EE job, the Lawrence Livermore Lab, Laser Department, AVLIS Project

Almost all of my LLNL projects were unclassified. Only a few projects were classified and I am not including those projects in my blog. There were a few projects that were somewhat sensitive and those projects are also not included. The only projects that I am including are those that have been reported in the open literature such as unclassified papers in symposiums or conferences or in unclassified patents that are available in the Internet.

Laser AVLIS
After our LGF tests at the Hopkinton gravel pit, I started looking for a new project and contacted an old buddy in the Laser Electronics Division, Fred Strange, having heard that there was to be a big project coming up, AVLIS, the Atomic Vapor Laser Isotope Separation project to use lasers to refine Uranium for atomic power plants. Fred invited me over and I had interviews with Edward Moses, the head of the AVLIS project, Mary Spaeth, Earl Ault and Richard O'Neal. Moses told me that he wanted me to work on a very important project, the Wavelength Meter, but first he needed me to build up a new laser lab, the Dye Master Oscillator Development Lab (DMODL). After my interviews with the AVLIS leaders, Fred Strange offered me a job in his EE Group. I quickly accepted, moved to an office in B3724. After settling into my new digs, I got started working on the DMODL, a big year-long project. Later on, my new EE Group Leader Roger L. Peterson, another old buddy from my Chemistry days, came on board and assigned me to be the lead EE for the Laser Diagnostics for the Dye Corridor. Besides the Dye Laser diagnostics, I also co-developed our calorimeter system for high power copper vapor and dye laser measurements. I had many projects during my 8-year tenure in the Laser Electronics Division but it's been over twenty years ago and only a few of my projects are listed below.

Dye Master Oscillator Development Lab
I got started on design of the DMODL, working with a Mechanical Engineer from the Oak Ridge National Laboratory (ORNL) who had been assigned to Livermore to work on the AVLIS project. The DMODL project involved the design and installation of Copper Vapor Lasers (CVL's) which I had never before even seen or had any knowledge of, except that I knew that high voltage and high power would be required. Previously I had only worked on low voltage circuitry and so the CVL

Green light from a high-powered LLNL Copper Vapor Laser.

design was a new challenge. I learned that the lasers in the AVLIS system consisted of chains of CVL's providing pump power to energize Dye Master Oscillators and Dye Laser Amplifiers, a completely new technology to me. The Dye Laser Amplifiers were arranged in chains to provide sufficient power for the AVLIS isotope separation process.

A Dye Laser Amplifier Chain in LLNL's AVLIS System.

Fortunately I got a lot of help from others in the AVLIS project, explaining CVL operation and design, showing me installations and providing me with schematics of existing systems. And a surplus power supply was available to us, which we had installed on a concrete pad outside the laser room. Moses had insisted that eight CVL's be built into the DMODL, four lasers stacked on each side of a laser table stood on its edge. The ORNL Mechanical Engineer and I worked well together, producing a compact design. Each CVL was mounted on a steel shelf, accessible via an interlocked latching steel door. Each laser was totally enclosed in a cabinet. Each laser could be viewed through a Lexan polycarbonate window that had a wire mesh screen shield. The system design was complex including a computer-controlled helium gas system, a comprehensive safety interlock system and optics to combine the laser beams and deliver the beam to the Dye Master Oscillators (DMO's) mounted on an edge-mounted granite table. As mentioned above, I got help on the electronics circuitry. However, I was still to make a mistake in the high-voltage circuitry but fortunately it was easily fixable. I had designed the 10 kV circuitry to be laid out on 1/2" thick fiberglass boards approximately 12" x 12" in size and had the fiberglass boards hard mounted to the well-grounded steel shelves in the laser cabinets. When we first fired up the 10 kV power to one of the lasers, the circuits on the fiberglass board lit up like a neon sign. The problem was the proximity of the HV circuitry to the steel base-plate, resulting in ionization of the air surrounding the circuitry. The simple fix was to raise the fiberglass circuit boards off the steel base-plates with 1" tall ceramic standoffs and the ionization was eliminated. Then the system was quickly operational, each CVL producing approximately 30 Watts of green light to power the DMO's on the granite table. Following completion of the DMODL it was time for me to begin work on the Wavelength Meter.


Wavelength Meter
Our Wavelength Meter, or Wavemeter, followed the invention of the Fizeau Wedge wavemeter by J.J. Snyder of the National Bureau of Standards and Snyder's several journal articles on his work. The accurate operation of the Wavelength Meter was necessary for wavelength control of the pulsed DMO's and therefore a necessary requirement for the success of the lasers and the AVLIS Project for enrichment of the uranium, so it was absolutely essential for me to get it working properly and reliably. I am very proud of the success of the Wavelength Meter and my work in its development.

The Fizeau Wedge used in Snyder's and our Wavelength Meter

A later article was in a NASA publication describing Snyder's Wavelength Meter development and computer programming algorithms, "A Laser Spectrometer and Wavemeter ..." The Wavelength Meter was to simultaneously measure and control the wavelengths of several pulsed Dye Laser Master Oscillators (DMO's) for the enrichment of Uranium in the Lab's AVLIS Project. Richard P. Hackel was the leader of our Wavelength Meter project, Mark Feldman was the optics designer and Robert Paris was the EE in charge of the DMO control system. According to Snyder's journal article and patent, the Fizeau Wedge produced an interferogram from the DMO laser light and the interferogram was imaged onto a photodiode array. The signals from the photodiode array were digitized with an Analog-To-Digital Converter (ADC) array and input to a computer for analysis of the interoferogram.

From http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19900019516.pdf

A temperature stabilized Helium-Neon Laser (HeNe) was used as a reference, also producing an interferogram from the Fizeau Wedge. Following Snyder's work, I purchased a 1024 element photodiode array module and interfaced it to a DEC LSI-11 microcomputer, then developed the software in Fortran IV for measuring the pulsed DMO wavelength according to Snyder's algorithms, also providing analog control signals to EG&G Lock-In Amplifiers, similar to the Princeton Model 5210, which provided a means to control the etalons or mirrors of the individual DMO's. Following Snyder's algorithm, my Fortran code computed the fringe spacing for the interferogram of each DMO and also for the interferogram of the temperature stabilized Helium-Neon laser. Then comparison of the fringe spacing of the DMO's interferogram to that of the HeNe produced the wavelength of the DMO as Snyder described in his original paper. I had purchased a single-mode fiber switching device to enable measuring the wavelengths of several DMO's and also interfaced and programmed the LSI-11 to control the fiber switching device to select the light from the several DMO's sequentially into our Wavemeter's optical system and to the Fizeau Wedge. Even with the essential information and guidance from Snyder's journal publications, development of our Wavelength Meter took me several months, largely because of a vexing problem that appeared to be temperature related. It appeared as though an additional interference pattern was being overlayed on the main Fizeau Wedge. It took me several weeks of testing to discover the cause of this vexing problem. However, luckily the solution was simple. It turned out that in our optical implementation that a portion of the DMO beam was being reflected from a parabolic mirror back to a ferrule which held the single-mode-fiber and the reflection from the ferrule resulted in the additional interference pattern on our 1024 element photodiode array. I had found that the laser light going to the Fizeau Wedge was only reflected and focused by the lower half of the parabolic reflecting mirror. Fortunately the part of the beam reflected to the ferrule was only from the upper half of the parabolic mirror and I simply covered that upper half with a piece of cardboard, thereby eliminating the additional interference pattern on the photodiode array. As a result of this simple cardboard fix, our Wavelength Meter now was operational to measure and control the wavelengths of the several DMO's. The error introduced by this additional interference pattern had been approximately a factor of 10 and after inclusion of the cardboard to block the laser beam from striking the ferrule holding the fiber optic cable, the measurement accuracy of our Wavelength Meter was approximately 1 x 10^-7, sufficient for accurate control of the DMO's in the isotope enrichment process. We patented our Wavelength Meter system, Patent #5189485.




Calorimetric Laser Power Meter
A major problem in the AVLIS system was the accurate measurement of the laser power up to at least 1 kW. We had been using so-called calorimeters which were nothing but metal objects that were placed in the path of the laser beam for a certain number of seconds, the temperature of the object measured with a thermocouple and then a table of values was consulted to give a measure of the laser power level. These of course, were not true calorimeters and the measurements they provided were iffy at best. Accordingly, a real calorimeter was needed for accurate power measurement and we began work to develop one. Our team was Laser Scientist Leland F. Collins, Mechanical Engineer James V. Micali, Mechanical Engineer Thomas C. Kuklo and myself. Our plan was to measure the flow of Low Conductivity Water (LCW) through devices which would capture the energy of the laser beam. We used LCW for cooling of the lasers and also for the calorimeters. We would measure the rise in temperature of the LCW and compute the energy absorbed in the calorimeter devices from the water flow rate and the temperature rise. Micali and Kuklo designed the housings for the two different calorimeters that would capture the laser beams, a flat one for lower power levels and a cup-like device for the higher power levels. The flat housing was designed with a corrugated surface and the cup-like housing was designed to reflect the laser beam and absorb it within the cup. Both devices were "flame sprayed" with a flat black ceramic coating to absorb the laser beam. Particular care was taken in the mechanical design to minimize temperature measurement errors from loss of heat on the water couplings. The design of the cup-like device is shown below.

Calorimeter cutaway view from the patent.

Laser calorimter system, high-power calorimeter housing at left, flowmeter at rear, microcomputer at right front.

My part of the system design was the electronics and software development including selection of the flowmeters, selection of the thermistors for temperature measurement, selection of the microcontroller, a Zilog Super 8, design of the microcomputer board and software development of the system. I had previously developed some Structured Macros in my LGF project and used the same idea for the software in the microcontroller, modifying the macros for use with the Zilog Super 8 microcontroller in addition to adding to the macros and including some macros for floating point calculations. Some Structured Macros are described in IBM literature, "Using structured programming macros." I had decided to develop the floating point macros because of the wide range of values in the calculations and to correct for nonlinearities in the thermistor curves. I used an LED numeric display for the measured flow rate, temperature difference and the computed power level. The laser calorimeter was highly successful, providing highly accurate power measurements over a wide range of power levels for both the CVL's and the Dye Lasers. We patented our "Radiation Beam Calorimetric Power Measurement System" in 1992, granted to the Department of Energy and the University of California. A schematic from the patent application is shown below.

Schematic of our Laser Calorimetric Power Meter from our patent application.

Amplified Spontaneous Emission System
An additional system that I developed was that to measure the Amplified Spontaneous Emission (ASE) in a Dye Laser Amplifier. For maximum efficiency of a Dye Laser Amplifier, it is essential to minimize the unwanted ASE. For the ASE system, we had purchased an optical system that included an interferometer to produce interferograms of the Dye Laser Amplifier light and I added a photodiode array, similar to that I used on the Wavelength Meter, and also a microcontroller for analysis of the interferograms. For acquisition and analysis of the interferograms, I used a Zilog evaluation board for one of their microcontrollers and developed code to acquire the interferograms and display them on a dedicated oscilloscope. Also my code would measure the percentage of ASE and present that information digitally for the laser technicians to use in tuning the Dye Laser Amplifier. I used the Structured Macros that I had developed for my Calorimetric Laser Power Meter project and so got double-use of that software. Our ASE system was a very important tool in tuning the Dye Laser Amplifiers for maximum efficiency.

A typical spectrum showing the unwanted ASE.

My final task in the AVLIS project was to design and develop a stepper motor controller for steering the dye laser beams, installing several stepper motors and the controller boards. After completing the installation of the beam steering system, I decided it was time to leave the Laser Electronics Division and move on to Z-Division in the department to be later named the Global Security Department. The AVLIS Project was a great learning experience for me and I enjoyed working for and with so many highly talented scientists, engineers and technicians. It was a terrific experience and I cannot possibly name all those folks that I had so enjoyed working with and from whom I learned so much.