My second EE job, the Lawrence Livermore Lab, Liquefied Gaseous Fuels 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.

LGF Project
LLNL scientist Edward Teller raised concerns about the plans to transport liquefied natural gas by special LNG tanker ships to a terminal in Santa Barbara, saying that the energy in the shipments would be equivalent to a nuclear bomb and that the LNG tanker ships would be "floating targets." initiated the Liquefied Gaseous Fuels (LGF) project.

LNG Tanker Ship.

In order to collect data that would be used to help develop and fine-tune computer programs for the prediction of the effects of accidental LNG spills, we were to do spill tests of LNG into the pond at the U.S. Naval Weapons Center, China Lake California. William J. Hogan was the Project Leader, Ronald P. Koopman was the scientific leader and there would be a cast of tens of engineers, technicians, scientific folk and software developers to develop a control and data acquisition system to be deployed at the China Lake facility for the spill tests. For the data collection from the spill tests, an array of sensors and a data acquisition and recording system would be required to be installed at the China Lake site. Several different sensors such as gas sensors, humidity sensors, temperature sensors and anemometers would be required to be installed on towers arrayed on the site. Battery powered data acquisition and communication systems would be deployed at the sensor locations to retrieve analog signals from the sensors and pass the data by radio communication to a recording system located in a trailer, some 250 m to the south-west of the pond spill site. Initially I was named the EE Project Engineer on the Liquefied Gaseous Fuels (LGF) project. However, I wanted to concentrate on technical aspects of the project and consequently a new EE Project Engineer was named, Lyle M. Kamppinen, and a new Project Engineer from EG&G was named, Donald Baltz, to oversee the software development. Rodney D. Kiefer was the lead EE on the gas sensor subsystems. EG&G had a facility at the "Atlas" building on Losee Road in North Las Vegas. There EG&G engineering personnel would assemble the LGF data acquisition

Bare LSI-11 Microcomputer.

China Lake LGF Test Site, 1980.

Sketch of LLNL LGF Turbulence Station. 

system hardware and EG&G Computer Scientists would develop the software to operate the three LLNL's repackaged Digital Equipment (DEC) LSI-11/23 microcomputer systems, each with a 10 Mbyte hard disk. The LSI-11/23 systems were to be located in the Command, Control and Data Recording System (CCDRS) trailer and were the heart of our LGF system. We were to do spill tests of Liquefied Natural Gas at the U.S. Naval Weapons Center China Lake facility. The "lake" at China Lake had shrunk to be a pond with a diameter of about 50 m. The NWC folks enlarged the pond to about 58 m and it was a perfect location for the spill tests. However, the LGF schedule was incredibly aggressive with plans to do the first tests in a year's time. R.N. Maroney and D. E. Neff from the Fluid Mechanics and Wind Engineering Department of Colorado State University (CSU) had previously conducted spill tests at China Lake and reported their tests in a Journal of Wind Engineering and Industrial Aerodynamics report. The CSU tests were in the range of 4 to 5 cubic meters of LNG. Our tests were to be much larger at 40 cubic meters. Our LGF sensor and data acquisition system, the Instrument Array, comprised three subsystems, the Weather, Gas and Turbulence. There were 20 Weather stations, 25 Gas sensor stations and 5 Turbulence stations, each with a battery powered Data Acquisition Unit.

LLNL's LGF Site.

The Data Acquisition Unit consisted of a Pacific Cyber/Metrix CMOS microcomputer with a complement of data acquisition and data communication modules plus a UHF analog radio transmitter and receiver for receiving commands from and transmitting data to the LSI-11/23 systems. I was responsible for the data acquisition systems and radio communication, including the software development on the microcomputers. the Pacific Cyber/Metrix microcomputers that used the the same CMOS chip of the latest DEC PDP-8 computers. I wrote the code for the FDAS, Field Data Acquisition Systems, in DEC's MACREL assembler. I chose MACREL as it allowed me to write Structured Macros such as IF, THEN, ELSE and other C-like commands. I was very proud of the FDAS code as the resulting code was easy to read, develop and modify, subsequently allowing another LLNL programmer, Stanley A. Martins, to extend the code to add commands and data for an additional sensor. In order to enable the analog radios to transmit and receive digital data, I developed an interface board for the radios to communicate at 19,200 Baud using the so-called Bell Code which allowed data to be transmitted through an analog transmitter/receiver system without loss of the DC voltage level. Subsequently the radio company made and sold an interface board after my design to allow their analog radios to transmit/receive digital data. The FDAS systems are described in a LLNL report "FDAS Hardware and Firmware Description ..." with a follow-on report by Stan Martins, Hal Goldwire and Gary Johnson, "FDAS ... 1987." A site map is shown below. The Instrument Array was located from approximately 250 m to the SW of the pond and 900 m to the NE of the pond. Our Command, Control and Data Recording (CCDRS) trailer was located approximately 250 m to the SW of the pond spill facility. LSI-11/23 microcomputers located in the CCDRS trailer acquired and stored the data on hard disk. The data was later copied to magnetic tape for subsequent analysis at our Livermore offices.  A key subsystem was the Weather Array that provided a map of the wind direction. Before a test could be conducted, we viewed the Weather monitor to ensure that the wind was steadily from the south-west, assuring us that a spill test could be conducted safely. The screen copy below shows the wind direction.

Weather Array monitor screen copy.

Me installing a Weather Station at China Lake.

The Burro tests were very successful with only a few loss of data glitches due to failures in the microcomputer memory modules. We reported the initial 1980 tests, the "Burro" tests, in an LLNL technical report titled Burro Series Data Report ... Future tests, the Coyote Series, were even more successful as we replaced defective memory chips, tweaked the software a little and got more experience with the system. The Coyote test report is available on the NTIS.gov site at Coyote Series Data Report and also in a report by Donald L. Ermak, Heavy Gas Dispersion Test Summary Report. One of the surprising results of the tests was that the gas plume would split up into two plumes, separated when gravity pulled down the center of the plume. Another surprise is the non-chemical explosive Rapid Phase Transition (RPT) effect when a shell of water ice would encapsulate the liquefied gas, exploding when the gas warmed. Ron Koopman reported on the tests in his LNG Safety document. Also see the youtube video. Subsequently the system was relocated to Frenchman's Flat on the Nevada Test Site for spill tests with ammonia and nitrogen tetroxide. Several of us took an abbreviated system to Hopkinton, MA for spill tests in cooperation with a scientist from the Factory Mutual Research Company (FMRC). The FMRC scientist wanted to test his scheme for entraining the gas from an LNG spill by use of water fountains surrounding the spill area. Gary W. Johnson, Hal Goldwire, Jeff (an electronic technician) and I took the abbreviated system to a gravel pit in Hopkinton which also was a training site for fire fighters to extinguish gas and other fossil fuel fires. Below is a photo of some of the crew at the Hopkinton site eating Whoppers while the fire fighters are training in the background at our abbreviated LGF sensor array. The fire fighters would call us "Perfessers" because of the white labcoats we wore at the site.

Three "Perfessers" (Jeff, Baker and Goldwire) at the Hopkinton site with fire fighters training in the background.