Conductive Plastic Compounds
The performance of a shielding device, whether it is constructed of metal, conductive coated plastic or a conductive thermoplastic compound, will be determined by several interrelated design criteria. The shape and thickness of the housing, the relative position of the electronics, and the type of joints and holes are factors that determine the shielding performance of a device.1-12
Often these criteria are dictated by practical considerations, such as size (or available space), assembly, cooling, mechanical and cosmetic requirements. Shielding performance is almost an afterthought.
Many times, changes or add-ons have to be incorporated to boost the electromagnetic compatibility (EMC) performance of a device designed primarily to meet geometric, mechanical or cosmetic targets. The added time and expense of iterative testing and re-engineering can be prohibitive.
The first step in designing a conductive plastic device to counter electronic noise is to fully define and understand the EM attenuation needed. Ideally, the emission spectrum of the electronic circuitry is known or can be measured.
Subtracting the defined maximum allowable noise (agency-mandated such as FCC, VDE or EU; or performance-defined such as susceptibility limits) from the emission spectrum should indicate the frequencies needing attenuation and the degree to which they need to be attenuated (Figure 1).
Shielding effectiveness (SE) in decibels is defined in Equation 1,
SE = 10 log (Pi/Pt) [1]
where: Pi is the energy of the incident wave.
Pt is the energy of the transmitted wave.
In principle, conductive thermoplastic compounds can provide 65 to 70 dB of electromagnetic noise attenuation, or SE, over a broad range of frequencies. Practically, though, attaining SE greater than 45 dB with conductive thermoplastic compounds is difficult.
Stainless steel fibers are generally the preferred conductive additive for shielding applications. These fibers are typically 8-µ dia 302 stainless steel with an aspect ratio of approximately 750.
Unlike glass or carbon, stainless steel fibers are soft, which allow them to form an interconnected conductive network inside a molded part. Also, the low fiber loading required to achieve conductivity has minimal effect on mechanical performance.
The result is that stainless steel-filled materials behave more like unreinforced than reinforced resins. This allows for designs incorporating snap fits, living hinges and molded-in custom colors.
Part Design Criteria
SE is directly determined by the conductivity of the material and the thickness of the device. At a given conductivity and wall section, SE is relatively constant over a broad frequency range (kilohertz to gigahertz). This gives the designer freedom to balance wall section (mechanical performance, flow and weight) with material cost (higher conductivity higher cost per pound) given a defined SE target.
SE is also linked to the free flow of EM-induced eddy currents in the part. Restrictions to current flow denigrate the device’s shielding capability.13 Joints pose such a threat. Current flow through a joint is determined by either the joint resistance or joint capacitance. At frequencies less than 200 MHz, high resistivity at the joint, relative to the bulk material, inhibits current flow. At frequencies greater than 200 MHz, capacitance coupling across the mating surfaces makes the joint transparent to current flow despite high resistivity at the joint.5
Designing a shielded enclosure for assembly will result in some type of joint or seam where the conductive plastic parts mate. A typical clamshell enclosure, for example, will contain a butt or overlap seam between the two halves.
Since the surface resistivity of a molded part made with a stainless steel compound is typically higher than that of the bulk material, standard designs for plastic joints may not be sufficient to achieve adequate electrical conductivity across the mating surfaces. This is particularly important if the device needs to attenuate low-frequency EM radiation.
Design options to lower the contact resistivity across a joint or seam include self-tapping screws, sonic metal inserts, insert molded pins, and sonically (or vibrationally) welded seams. The principle behind these design solutions is simple: All form a conductive “bridge” between the stainless steel fiber networks in the two mating parts.
Another less effective, and often less practical, design option would simply increase the contact surface area between mating parts. This solution would maximize the probability of establishing electrical contact between stainless steel fibers at the surface of the two parts.
Slots or holes employed to ventilate or to provide access to a device can also dramatically impact the SE of a device. Ultimately, they limit the frequencies that can be contained.
Holes or slots can have two effects. They can be large enough for EM radiation to penetrate the shield unattenuated. They also can be oriented to inhibit the free flow of induced current in a shielding device.6
Design solutions for holes and slots are the same as those applied to metallic shielding devices. Holes should be small enough to prevent unwanted frequencies from passing through the shield unattenuated, approximately one-sixth the wavelength of the highest frequency to be shielded.
Slots can be oriented in the direction of current flow to minimize their impact. This solution applies only to situations where the induced current flow can be predicted; that is, when shielding an emitter of known polarity and orientation.
Summary
The effects on shielding of the various design options are illustrated in Figure 2. Figure 2a shows the SE of an ideal closed housing (no holes or seams) under worst-case conditions: near an EM field emitter.
The effect of adding a seam or joint to the housing is shown in Figure 2b. The magnitude of the reduction in shielding efficiency is determined by the type of seam and design solutions employed to reduce the contact resistivity. Careful attention to seam design and conductivity across the joint can minimize this shift.
Figure 2c shows the influence of holes on SE. Note that there is no impact on shielding efficiency until the wavelength of the EM radiation is small enough–frequency is high enough–to allow it to pass through the hole. At frequencies higher than that point, the shield is ineffective. Consequently, hole size determines the upper frequency limit that can be attenuated by the device.
The sum effects of all the design choices on shielding efficiency are shown in Figure 2d: compound conductivity, wall thickness, seam design and hole design. An optimally designed shielding device is depicted in Figure 2e. Ideally, there is just enough SE to attenuate the desired frequencies to the necessary level required.
Design recommendations for EMI shielding with conductive plastic compounds are:
Know the nature of the EMI to be attenuated, frequency, proximity, polarization and strength.
Define the size and location limits of the device.
Define cooling and/or other in/out openings required.
Select filler technology to suit the purposes.
Select material conductivity to minimize filler loading.
Design the assembly to maximize effectiveness of shielding.
References
1. White, R.J., Shielding Design-Methodology and Procedures, Interference Control Technologies, Gainesville, VA, 1986.
2. “Interference Reduction Guide for Design Engineers,” Report to the U.S. Army Electronics Labs, Volume 1 and 2, Accession AD 619666 and AD 61967, Filtron Co. Inc., Fort Monmouth, NJ.
3. Bannister, P.R., “New Theoretical Expressions for the Plane Shield Case,” IEEE Trans. on EMC, Volume EMC-10, No. 1, 1968.
4. Bannister, P.R., “Further Notes for Predicting Shielding Effectiveness for the Plane Shield Case,” Trans. on EMC, Volume EMC-11, No. 2, 1968.
5. Steenbakkers, L.W.; de Goeje, M.P.; Catrysse, J.A.; “Study of the Influence of Conductive Joints on the Shielding Efficiency of a Conductive Plastic Housing,” Proceedings on the 9th International Conference on EMC, Zurich, 1991.
6. Steenbakkers, L.W.; de Goeje, M.P.; Catrysse, J.A.; “The Influence of Slots and Holes on the Shielding Efficiency of a Housing,” Proceedings on the 10th International Conference on EMC, Zurich, 1993.
7. Borgmans, C.P.J.H.; Anaf, L.J.; Catrysse, J.A.; Comparative Testing on the Shielding Effectiveness of non-Conductive Joints for Conductive Plastics, submitted.
8. Bethe, H.A., “Theory of Diffraction by Small Holes,” Phys. Rev., Volume 66, 1944.
9. Butler, C.M.; Rahmat-Samii, Y.; Mittra, R.; “Electromagnetic Penetration Through Apertures in Conductive Surfaces,” IEEE Trans. on Antennas and Propagation, Volume 26 (1), 1978.
10. Shelkunoff, S.A., “Transmission Theory of Plane Electromagnetic Waves,” Proceedings of IRE, Volume 25, 1937.
11. Shelkunoff, S.A., Electromagnetic Waves, Van Nostrand, Princeton, NJ, 1943.
12. Gerbig, S.R., “Dealing With EMI/RFI,” Evaluation Engineering, May 1985.
13. Miller, D.A.; Bridges, J.E.; “Review of Circuit Approach to Calculate Shielding Effectiveness,” IEEE Trans. on EMC, Volume EMC-10, No. 1, 1968.
About the Authors
Chris P. J. H. Borgmans received a B.S. degree and an M.S. degree in chemical engineering from the University of Technology, Eindhoven, The Netherlands. He has been with DSM Polymers for the past six years. Mr. Borgmans has written 15 technical papers in the field of EMC and ESD, and is a member of the American ESD Association as well as the Dutch ESD and EMC Associations. DSM Polymers-Research and Technology, P.O. Box 604, 6160 AP Geleen, The Netherlands, 31-46-760-843.
Raymond H. Glaser, Ph.D., is the Industry Manager for Electrical and Electronic Programs at DSM. Before joining the company two years ago, he was affiliated with GE Plastics. Mr. Glaser received a B.S. degree from Marquette University and an M.S. degree from Florida State University, both in chemistry; and a Ph.D. from Virginia Polytechnic Institute and State University in Materials Engineering Science. He has authored 12 polymer publications and has one patent issued with 12 patents pending in the area of high-performance thermoplastics. DSM Engineering Plastics, 2267 W. Mill Road; Evansville, IN 47732-3333, (800) 333-4237.
Copyright 1995 Nelson Publishing Inc.
July 1995