A simple test for quickly measuring shielding effectiveness of cable and connectors.

Size: px

Start display at page:

Download "A simple test for quickly measuring shielding effectiveness of cable and connectors."

Gilbert Goodwin
5 years ago
Views:

1 A simple test for quickly measuring shielding effectiveness of cable and connectors. Servilio Sebastiani ELASS (Sistema ricerca FAT nel Mezzogiorno) Via E. Mattei Chieti Scala - TALY. ABSTRACT A method to measure the shielding effectiveness of cable and connectors is proposed. The testing methodology has been chosen, based on a double TEM cell, in order to obtain data comparable with the method described in EC-96-1, normally used for the measurement of surface transfer impedance. The surface transfer impedance Zt, can be related to shielding effectiveness as follow: Let shielding effectiveness be SE = (Pout / PL) Pout= where PL= Power to end of line = (VL~ / ZO) NTRODUCTON Surface transfer impedance testing was developed as a means to predict the shielding effectiveness of coaxial cables with a single inner conductor and one or more cable shields. This testing technique calls for an overall shield (or two) to placed over the cable and for measurements to be taken. t is designed to measure near field effects. n the 1930 s, Schelkunoff showed that the surface transfer impedance was the intrinsic shielding property of cable, connectors/backshells and cable assemblies. nitially, he treated only solid cylindrical shields, however, his work has been extended to include imperfections in the shield such as apertures, porpoising, etc. n Schelkunoff 5 work he defhled the surface transfer impedance of a shield as follows: f a current is caused to flow along the conductor of a cable, with its return path on the sutiace of the shield, then the longitudinal voltage along an incremental length which results on the inside surface of the shield is related to that current by the surface transfer impedance, and has units of impedance per unit length. This thought can be expressed mathematically as: The following equation for transfer impedance can be derived from known results: Zt = v 202 -ZOZ V (l/se) (F/L) where Zol = Characteristic impedance of the drive line. Zo2 = Characteistic impedance of the pickup line. L = length of the cable sample. F=(l 4 ( l- e 12 e 22) (1-e 21 ed + ed ( l+ e22> ( l- el2> ( l- e21> ell = Reflection coefficient of the drive line at the source. e21= Reflection coefficient of the drive line at the termination. e12 = Reflection coefficient of the pickup line at the termination e22 = Reflection coefficient of the pickup line at the detector. Zt = ( V shield / shield ) /97/$

2 fwe now solve for shielding eflectiveness, then: SE = Zol Zo2 / Zt ( F/L)2 Which when rewritten in logarithmic fonn to express the result in db gives: SE = 10 loglo (Z,,) + 10 log,, (Z,,) + 20 loglo (F) - 2Olog10 (zt) dl3 97 The various terms of this formula are related to cable construction. The first term ( 10 loglo Z,,) characterizes the relationship of the cable shield to all other conductors in the vicinity including the shield itself. The second term ( 10log10 Zo2) characterizes the dependence of the shield s effectiveness on the internal construction of the cable. The third tenn (2010g10 F) characterizes the dependence of the shielding effectiveness of the way in which the conductors and the cable shield are terminated. The fourth term (2010g10 L) indicates that the shielding effectiveness for short cables (less than one-quarter of a wavelength) goes down as the square of the cable length decreases as surface transfer impedance decreases. Of the five terms characterizing shieldding effectiveness, only the final term describes the shield leakage, so only that term is affected by the actual design of the shield. f it were possible to completely control all of the other terms, the surface transfer impedance would be sufficient to totally specify shielding. The interrelationship of screening effectiveness measured in decibels and the surface transfer impedance in Ohm presents the designer with a conversion difficulty. External hames circuits vary, as do the coupling characteristics and it is therefore only possible to give an approximate conversion. The normal conversion from decibels to Ohm is to a reasonable approximation: Shielding Effectiveness (db)se = log lo(zt) The constant term is developed from the expression for the characteristic impedance of the line formed by the harness screen and the ground plane and the internal characteristic impedance of the inside of the harness. The relationships from screening effectiveness to surface transfer impedance is shown in figurel. O.OOO O Frequency (MX) Fig. 1 - Shielding effectiveness and surface transfer impedance relationships (Schelknoff Model) Measurement techniques for detennining shield transfer impedance are many and varied. The usual method for the measurement of surface transfer impedance and admittance of coaxial cables uses two propagation lines: a disturbing line and the cable under test. n most cases, the disturbing line is made up of a hollow tube which acts as an outer concentric conductor, the inner conductor being the cable shield itself. This disturbing line is driven at on end and the transfer parameters are deduced from the voltages appearing at each end of the coaxial cable. At low frequency, thus when the cable length is much smaller than the wavelength, the transfer parameters are easily determined fi-om the measurements. However, if the f?equency increases, the propagation along the line cannot be neglected. We shall see on an example that corrections based on a mathematical analysis does not allow to get results with a high degree of confidence for frequencies above about 100 MHZ. The only solution is to decrease the cable length but putting a cable few centimeters long in a usual bench is not realistic. This type of test fixture is called a triaxial test fixture because it has three electrodes; the core of the sample under test, the shield of the sample under test, and, the outer shield. Figure 2 showes the triaxial test fixture specified in EC The dielectric betwen the shield and the test and the outer shield has a dielectric constant and a relative phase velocity of 1. The test fixture is shorted at one and; therefore, it looks like a shorted transmission line 1 meter long with air dielectric. when the test fixture is a quarter wave length long (about 75 MHz for 1 meter test fixture) the input impedance of the test fixture will be a minimum and the current flowing in the test fixture will be maximum. 115

3 This results in a resonance in the surface transfer impedance. The frequency of this resonance depends on the frequency at which the test fixture is a quarter wave length long. f the dielectric constant within the test fixture is air, such as in the case of the EC-96-1 fixture, the resonance will be about 75 MHz. Figure 3 showes the surface transfer impedace of a double braid cable according the EC-96-1 method. DUAL TEM CELL MEASUREMENTS A TEM cell is a wide band linear transducer both in phase and in amplitude. t is used to convert radiofiequency signals into electromagnetic fields. Since the TEM fkctions in bidirectional mode, it can be used to measure electromagnetic fields when it is coupled through a common opening with another cell. A double TEM cell therefore consists of a cell with an aperture on the top side and whose generated electromagnetic field is coupled with similar cell located above. (Fig.4) The measurement is of substitutive type and is made in two phases: a) measurement of the power of the signal transduced with an unshielded cable in place of the septum imler conductor (~1); b) measurement of the power of the signal transducer with a shielded cable (~2) ; shielding effectiveness is defined as follow: S.E. = 10 log (WP2) Fig. 2 - Surface transfer impedance test fixture (CE-96-l) SOfl, load,-, Dual EMCell Control Unit i JEFE-488 f _.._...,-.. c _ % PlOttW Fig. 3 - Transfer inpedance plot of duble braid cable Fig. 4 - Dual TEM cell measurement system 98

4 ---~ ~t -.-_-. ~.,- i - _..~-.,~-. ( ~ _._.,_. -i j G--;----j.. AM _ 80 &j i START 400 km STOP MHZ RES BW 30 khz BW 38 CHZ SWP 20 rnsec Fig. 5 - Measured attenuation in dual TEM cell 4ooKHz-2MHz Fig. 8 - Measured attenuation in dual TEM cell 3oMHz-3ooMHz Fig. 6 - Measured attenuation in dual TEM cell 2MHz-1oMHz Fig. 9 - Measured attenuatiou in dual TEM cell GHZ it-ii i i - Ait -8ddB i-'-?-l 99 Fig. 7 - Measured attenuation in dual TEM cell lomhz-3omhz Frequency m) Fig Attenuation (ansl~elded.kl~elded cable) 3OOKHz- 1 GHz wo 3cco

5 Reference - S.A. Schelkunoff, The Electromagnetic Theory of Coaxial Trasmission lines and Cylindrical Shield. Bell Systems Tecnical Publication, Monograph B-816, nternational Electrotechnical Commission, EC Standard Publication 96-lA, Radio Frequency Cab/es, Part 1: General Requirements and Measuring Methods. Fig Comparison between the two type of measurement in the range MHz CONCLUSON Dual TEM cell measurements offer an alternative that is not as dependent upon the total design of the cable and its shield. Measurements of an equivalent system without a shield are made; then measurements of the same system with a shield are made. Dual TEM measurement by their very nature provide a direct result of shielding effectiveness. The technique calls for one set of measurements to be made with an unshielded cable of exactly the same construction as the shielded cable. The shielded cable then replaces the unshielded cable in exacctly the same location. The measurements are then made with the shielded cable in place. The shielding effectiveness of the shielded cable is the represented by; - Vance, E. F. : Shielding effectiveness of braided-wire shields. EEE. Trans.on EMC, vol. 17 (1975) - L.O. Hoeft Measurement of surface transfer impedance of cable and connectors EMC Expo B. Demoulin,P. Duvinage, P. Degauque Measurement of transfer paramters of shielded cables at frequencies above 100 MHz 6th EMC Symposium Zuric Darrell Fernald Cable assembly testing for,fa47 TEM A. B. Martin and M.Mendenhall, A fasf accurate, and sensitive methodfor measuring surface transfer impedance. EEE Trasaction on EMC, Vol. EMC-26, no 2, May SE = voltage (db) unshielded cable - voltage (db) shielded cable Since this thechnique relies on an empirical solution and is only as good as the repeatability of the measurements, the following possible sources of error must be considered: - Differences between the shielded cable core and the unshielded cable. - Variances in the positioning of the two cable. - Variation in the output power level of the power amplifier during the two runs. This method is typically of greatest value when used in determining absolute level emitted by a system in a test to some standard such as FCC part 15, subpart j. Even in this measurement, extreme care is taken to ascertain that the dual TEM cell is calibrated and all resonant frequencies are accounted for before proceeding. -loo

Conduit measured transfer impedance and shielding effectiveness (typically achieved in the RS103 and CS114 tests)

Conduit measured transfer impedance and shielding effectiveness (typically achieved in the RS3 and CS4 tests) D. A. Weston K. McDougall conduitse.doc 5-2-27 The data and information contained within this