HEMP Filters & MIL-STD 188-100 Series
The effects of a high-altitude electromagnetic pulse, or HEMP, are well documented and researched. There could be serious problems for electronic devices and electrical grids which are not protected from the effects. And HEMP is not easy to handle with a single protection device. This is because the effects range from very fast, about 1 nanosecond risetimes, to very long duration lasting minutes, all from the same pulse.
With such a significant threat, it become important to protect the equipment from HEMP with equipment that has been properly designed and tested as compliant to specific requirements. To support this work, several standards are used to help test the electronics and the filtering system designed for this environment.
Standards which have been written to address the design of protection include the MIL-STD 188-100 series. The purpose of these standards is to establish the “minimum requirements and design objectives for high-altitude electromagnetic pulse (HEMP) hardening of fixed ground-based facilities that perform critical, time-urgent command, control, communications, computer, and intelligence (C4I) missions.”1 MIL-STD 188-125 has two sections: part 1 of the document addresses HEMP hardening for fixed facilities; part 2 addresses transportable systems. The early versions of MIL-STD 188-125-1 remain in the public domain while the latest revision of the military standard, MIL-STD 188-125-1A, is (CUI) Confidential Unclassified Information and (ITAR) International Traffic in Arms Regulations controlled. Similar information is provided in UK Ministry of Defence DEF STAN 59188 Part 1 Issue 3.
A variety of individuals, companies, and groups have taken on the task of modeling the HEMP waveform. This has resulted in establishing three specific regions of the impulse, designated E1, E2, and E3. From equations by C. E. Baum and an IEC-77C presentation2, the double exponential waveforms were solved and the E1 and E2 waveforms plotted as shown in Figure 1 and Figure 2. Note that the time scale between the two are radically different. What is lost in these normalized waveforms is the difference in amplitude. E1 peak amplitude is 500 times higher than the E2 amplitude, typically given as 50 kV/m for E1, and 100 V/m for E2.
However, this does not explain a second challenge with these waveforms. Energy is based on amplitude and time. With E1, the voltage is below 1% of the peak by 122 nS after start of the pulse. In contrast, the E2 waveform does not decay significantly until after 1 second has passed, a factor of 107 greater duration for this pulse. This results in a pulse with a great deal more total energy than E1.
Figure 1 – E1 portion of HEMP event
The above figure can also be found in IEC 61000-2-9 and is reproduced in MIL-STD 464D and MIL-STD 461G in the test requirements for RS105. This portion of the waveform is extremely short, but very intense in amplitude, thought to be about 50 kV/m or more. Due to the short duration, the wavelengths created by this pulse are also relatively short as compared to other portions. Thus, for this and much of the E2 portion, the power distribution network can become a good antenna for receiving this energy.
Due to the very fast risetime of this pulse, some transient suppression devices, especially those designed for high voltage and high current, such as varistors and gas discharge tubes, do not respond fast enough to clamp the impulse. As a result, equipment using such devices would not be protected from the E1 portion of HEMP.
The above figure can also be found in IEC 61000-2-9 and is reproduced in MIL-STD 464D and MIL-STD 461G in the test requirements for RS105. This portion of the waveform is extremely short, but very intense in amplitude, thought to be about 50 kV/m or more. Due to the short duration, the wavelengths created by this pulse are also relatively short as compared to other portions. Thus, for this and much of the E2 portion, the power distribution network can become a good antenna for receiving this energy.
Due to the very fast risetime of this pulse, some transient suppression devices, especially those designed for high voltage and high current, such as varistors and gas discharge tubes, do not respond fast enough to clamp the impulse. As a result, equipment using such devices would not be protected from the E1 portion of HEMP.
Figure 2 – E2 portion of HEMP event
To assure that the equipment, including filters, shielded enclosures and the like, meet the HEMP Hardness Assurance required by the standards, several tests have been designed to verify the energy induced does not propagate through the barrier, either filter, shield, or what have you. For a shielded enclosure, the techniques used to measure shielding effectiveness are those commonly employed by shielding companies. The requirements of a minimum of 80 dB of isolation up to 1 GHz does not seem difficult on the surface. However, once all the panels and seams are measured, all penetrations are verified, and every cable coupling is checked, the testing can be very time consuming and rather tedious. Even a small amount of debris or contamination on any mating surface can cause a significant reduction in shielding effectiveness.
Paths into shielded enclosures, the point of entry (POE) will be the main weakness in maintaining shielding effectiveness. Of course there is the need to allow personnel into the facility, which must be performed through two shielded entry doors, of which only one may be open at any time. But there is a need also for all electrical (power, audio, data, and signal lines), ventilation, water and sewage, and any thing that needs to cross into or out of the shielded enclosure. Each POE must be properly filtered, shields bonded, and assured it meets required shielding effectiveness.
Of particular interest and challenge are the test requirements around the power and signal line filters used, especially for those on medium voltage distribution power lines. Standard techniques for determining filtering effectiveness are not adequate when dealing with current pulses, especially those generated by HEMP. These are the tests which are described and mandated by DEF STAN 59-188 and MIL-STD-188-125-1.
The power line current impulse tests are divided into three characteristics. The first is the Short Duration Pulse, which requires equipment having a minimum bandwidth of 750 MHz, and injection transients of at least 5,000 Amperes peak. Having a source impedance of 60 Ω which makes the required source voltage 55 kV. This pulse represents the E1 section of the transient.
The second is the Intermediate Pulse, representing the E2 portion of the HEMP event. This has a bandwidth from DC to 10 MHz and a peak current of 250 Amps. The source impedance is 10 Ω. Third is the Long Pulse which represents the E3 portion. The equipment requirement is for a bandwidth of 10 kHz, and a source capable of 1000 Amps with a 5 Ω source impedance.
The test is performed for the short and intermediate pulses by placing the filter into a shield room wall. The need for the shield room is first to isolate the cross coupled high frequency energy from the induced signal on the input side of the filter onto the output of the filter. Since the filter will be used in this manner, any cross coupled noise induced during testing will not allow accurate measurement of the performance of the filter. A well-designed filter typically has over 80 dB of filtration to CW signals from input to output, and cross coupled energy can mask this response, however impulse response can be less.
The pulse generator is placed on one side of the filter and coupled to the line connected to the filter. This line may be loaded or isolated from other connections, depending on requirements agreed to in the test plan. A current probe capable of measuring very high speed risetimes (~10 nS), such as a Rogowski Coil, must be used to accurately measure the source current.
On the output of the filter, inside the enclosure, the filter is loaded into an impedance to reference ground (shielded enclosure). The resistance historically has been 2 Ω however, much lower values may now be required. This can be a challenge since resistive loads capable of significant current are often rather inductive. Inductive loads will appear to have much higher impedance with higher frequency, such as in the case of fast risetime pulses induced on the filter.
The filter’s output line will have a second current measurement probe placed on it. When a pulsed current of up to 2,500 amps is induced on the input on a single line, the measured output current pulse must be less than 10 Amps, even in very low impedance loads. This can pose a significant challenge, since the transient path to the shielded enclosure in the filter must be much lower than the path through the load impedance.
When testing components, equipment, and subsystems for compliance with EMP standards, other test requirements become important. The International Electrotechnical Commission (IEC) has developed a number of tests and standards that are now adapted by MIL-STD 461G, MIL-STD 464D. IEC 61000-2-9 established a basis for impulsive fields created by the E1 portion of the HEMP event. The impulse, which is identical in shape to Figure 1, has an amplitude of 50 kV/m. This requirement has become the basis for the MIL-STD 461G RS105 test. This test is a radiated susceptibility test, performed using a parallel plate or radiating line antenna.
Other requirements in MIL-STD 461G which stem from a variety of source documents are CS115 and CS116. CS115 is an impulse current induced onto all lines of a subsystem or individual equipment. The pulse is 30 nS long, 5 amp current with a 2 nS rise and fall time. CS116 is a series of damped sinewave current pulses also induced onto all lines as CS115. The waveforms are at 10 kHz, 100 kHz, 1 MHz, 10 MHz, 30 MHz and 100 MHz, with a peak amplitude of 10 amperes depending on the frequency. Both CS115 and CS116 are performed using bulk current injection probes. These tests are bench-top test performed in a shield room laboratory in a carefully controlled test setup.
However, as described in MIL-STD 464D Appendix paragraph A.5.6, these tests only address the E1 portion of the HEMP waveform, although it may be argued that CS117, considered lightning testing, has some benefit for the E2 portion. To properly address and design equipment for the complete HEMP environment, it is necessary to review MIL-STD 2169, which discusses all threats, and is required for HEMP designed military systems. However, MIL-STD 2169, along with MIL-STD 188-125-1A and many other HEMP related requirements, are all classified and only available on a need-to-know basis.
Footnotes
- From MIL-STD 188-125-1 paragraph 1.1 – Purpose
- From Risetime Evolution in HEMP (High-Altitude Electromagnetic Pulse) E1 Waveforms – Technology and Standards, found in Sensor and Simulation Notes, Note 573, by D. V. Giri and W. D. Prather, December 25, 2015 This may be better addressed in Sandia National Laboratories Report SAND2020-1883, titled Double exponential approximation and inverse double exponential fit for Bell Labs and International Military Standard EMP waveforms, by S. Campione and L. K. Warne