Electromagnetic Pulse Research (1970 – 1993)
Electro-Magnetic Pulse or EMP is a very intense impulse of electromagnetic energy that can occur during a nuclear explosion. In 1962, EMP was discovered after a nuclear test at 400 km altitude, about 800 km from Hawaii. The effects on July 9, 1962, in Honolulu were described in a magazine: “The quiet predawn in Honolulu in 1961 was shattered by the simultaneous pealing of hundreds of burglar alarms. At the same instant, circuit breakers on the power lines started blowing like popcorn“.
Only sometime later, when no cause was found, a correlation was made with the nuclear test at a large distance. The defects and failures that occurred were caused by EMP and were comparable effects of a short-range lightning strike. EMP is, therefore, a phenomenon in which the strengths of the electric and magnetic fields increase to very high values in less than one hundred-millionths of a second. As a result, for example, a peak voltage of 25,000 V occurs in a rod antenna of a portable radio, as a result of which transistors in the radio are fried. It is likely that electrical systems and installations with long wires and cables will be disrupted or are destroyed, such as electrical power stations, refineries with process controls, communication centres, computers and the like. The treacherous aspect is that EMP effects may occur in an area with a radius of more than 1000 km when a nuclear weapon explodes high in or just outside the atmosphere. Even when this happens far outside the Dutch territory, it still can be dangerous.
Since 1970, TNO Waalsdorp has therefore been using EMP simulators and computer programs to investigate exactly what the effects of EMP are on electronic systems and what countermeasures can be taken. In summary, an EMP simulator requires a high electrical voltage, very short rise times and very high frequencies. Therefore, such a simulator consists of a high-voltage pulse generator connected to a so-called flat-plate transmission line, which is terminated at the end with an ohmic load R = Z0 where Z0 is the characteristic impedance of the transmission line.
EMIS-1, EMIS-2 and EMIS-3
At first, a small EMP-simulator was built called EMIS-l with the nickname “Piggy bank”. It size was about 0.8 x 0.8 x 1 meter. Only small devices up to 50 x 50 x 50 cm could be tested in EMIS-1 for their EMP sensitivity up to 50 kV/m.
Later, a large 60-meter simulator was built behind the laboratory: EMIS-2. EMIS-2 allowed testing of entire systems with dimensions up to 3 x 8 x 25 meters. To simulate an EMP in EMIS-2, a capacitor battery was charged up to 500,000 V and then discharged into the transmission line using a very fast switch (spark gap). In less than 5 nanoseconds, a current of about 5,000 Ampères would flow into the transmission line. Between the “plates” of the transmission line, the generated impulse-shaped electromagnetic field then corresponds to a nuclear EMP. The strength of the generated field could be set because it was necessary to know at which EMP level system defects would occur. The highest field strength that could be generated in the test volume of EMIS-2 was 80 kV/m.
At the start of an EMP test, the power flows in cables, antennas and circuits were first measured using special measuring equipment. The possibly vulnerable parts could then be identified from the measurements and from studying the schematics of the equipment. If the EMP-test showed that defects occurred, TNO investigated what kind of protection measures could be taken.
EMIS-3 was an even larger, mobile EMP simulator which was installed at the military airfield Ypenburg in 1982. EMIS-3 allowed testing of aircraft, ships, and complete communication centres for their sensitivity to EMP. In 1983, a horizontal emitter for EMIS-3 was installed and a transmission line that could be used to test accommodations and aircraft at threat level was completed. In the next years, the system was continuously in use to test equipment of the Armed Forces. In 1987 experiments started with the current induction generator with which, at EMP threat level, currents could be induced in cables of communication systems.
In the period 1979 – 1980, research in EMP modelling was carried out. Topics were the penetration of EMP into water, the transfer of EMP to cables, and the shielding effect of different type of metal. TNO’s model-based calculations of shielding were internationally recognised as achieving a high degree of accuracy.
In 1981, the US Defense Nuclear Agency helped to TNO get hold of the PRESTO software package which was developed by Boeing Aerospace Company for EMP-calculations on the Minuteman missile. Three Boeing employees were flown in from the US to install the package and get it running on TNO’s Control Data Cyber 74 with its limited memory. At Boeing, the systems had a so-called extended memory where large blocks of data (arrays) could be temporarily stored; a sort of self-organised virtual memory. If there was no extended memory, the same subroutine was used to write/read the blocks of data from/to disk, which of course was slower, especially because standard I/O library routines were used.
At that time, TNO’s system programming had developed a small Compass (assembler) module which requested PP CIO to write a block of data of n * 64 words to disk while the CPU continued calculations. Only when a subsequent write action was required, the routine ‘tested’ whether the previous write action had finished. In the same way, a read action was ‘thrown over the fence’ until the data was really needed for calculations. With only the adjustment of a couple of Fortran statements, we replaced the I/O subroutine which dealt with the swap-in/out of large chunks of data with TNO’s special input-output routine, a routine with only a handful of assembler instructions. As a result, the I/O-bound PRESTO package suddenly became CPU-bound. Within a few days, instead of the planned three weeks, we completed all the standard tests of the package. Moreover, we found some errors in the package as our system was configured in a way that we could detect uninitialised variables.
In July 1992, a completely renewed pulse generator for the EMIS-3 was transported to the EMIS installation at Ypenburg. This simulator met the newest, 1990, NATO test requirements. On 14 October 1992, all acceptance tests were successfully completed and the installation became operational. Aside from EMIS-3 at Ypenburg, only an EMP test installation existed in the USA that was capable of generating a comparable pulse, albeit with some limitations.
It is worth noting that the EMIS-2 antenna became inoperative when all systems of an Anti-Aircraft tank, type CA-1 (CAESAR), were undergoing an EMP test. When the Royal Netherlands Navy flew along the coast with a Breguet Atlantique, the AA-system automatically tracked that aircraft as a target. The gun barrels followed the Navy plane and ruptured the EMIS-2 antenna system.
Directed Energy Systems
In 1989 an exploratory study was conducted into the field of non-nuclear EMP and, more general, Directed Energy Systems. In later years work was performed on the effects of and protection against High Power Microwave (HPM).