Thesis (PhD)--Stellenbosch University, 2014. / ENGLISH ABSTRACT: Electromagnetic shielding plays a significant role in the protection of electronic equipment.
Its application is essential for mitigating radio-frequency interference for the Karoo Array
Telescope (MeerKAT) project in the Karoo region of the Northern Cape. In this context,
time-domain (TD) methodology for small enclosure shielding effectiveness (SE) is developed
using a reverberated environment technique. Interest revolves around measurement time
speed-up and an extended SE response which covers the under-moded condition of small
enclosures. Recommended IEC standard 61000-4-21 [1] e ciency of 0.75, for log-periodic
dipole array (LPDA) antenna, is also validated from a reverberation chamber (RC)
characterisation of a printed circuit-board (PCB) LPDA e ciency.
A built-in pulse generator and the Square Kilometre Array (SKA) receiver RATTY form
the main elements of the TD metrology. For validation purpose, a reference coaxial airline
is built. The cable is characterised with computational codes (CST and FEKO) and is
also modelled with Vance and Kley's analytical expressions. The results are compared
with TD transfer impedance (Zt) measurement within the RC. The study shows that the
cable fixture within an RC shapes the cable under-test (CUT) Zt. The airline itself is
also introducing an oscillating component within Zt. The resonance is proportional to the
CUT length and it is visible within the measurement data and the simulations. It is not,
however, taken into account by the theoretical models.
The consequence of an incorrect antenna efficiency on RC applications is also addressed
using a PCB LPDA antenna efficiency investigation. The unknown LPDA is simulated
with CST for the study. The result is compared to an RC measurement validating the IEC
61000-4-21 standard efficiency recommendation of 0.75 [1]. This methodology characterised
the unknown antenna parameter from a reference dipole antenna efficiency we investigated
with FEKO. Simulated Wheeler-cap techniques permitted the reference antenna validation.
We found that an inaccurate LPDA efficiency has little effect at higher frequency if the
IEC efficiency is adopted. However, a difference of more than 7 dB can arise at low
frequency if the real efficiency differs by more than 0.3 with respect to the IEC value.
The study highlights the importance of a correct antenna efficiency for accurate RC
applications. The nested-enclosure technique is regarded as the conventional method of investigating
small enclosures SE [2]. The technique is in general time-consuming and works for a frequency range higher than three times the enclosure under test (EUT) lowest cut-o
value. Our TD metrology covers a frequency band up to 1.4 GHz which coincides with
our enclosure under-moded region. The SE characterisation is not well-documented
within this particular region. The dissertation contributes to this field using a non-stirred
nested-enclosure configuration. In contrast to the conventional use of the nested-enclosure
methodology [2], the source is placed here within the EUT and the enclosure is treated as
a normal radiator. The SE definition according to the IEEE standard in [2] is followed
and the enclosure total transferred-power is computed from the port's reflection coeficient.
The approach does not require a stirrer for the EUT. Our measurement shows an SE
agreement between the modfied and the appropriate nested-enclosure technique from 390
MHz up to 4 GHz. The investigation is faster, but in addition the TD spectrum gives a
more detailed SE response than the FD approach. / AFRIKAANSE OPSOMMING: Elektromagnetiese skerming speel n belangrike rol in die beskerming van elektroniese
komponente. Die toepassing daarvan is noodsaaklik vir die versagting van radiofrekwensie
steurings in die Karoo Array Telescope (MeerKAT) projek wat tans ontwikkel word in die
Karoo omgewing van die Noord-Kaap. In hierdie konteks is daar op n tydsgebied metode
vir klein-omhulsel beskermingse ektiwiteit (BE) gefokus, deur gebruik te maak van n
weerkatingsomgewing tegniek. Die belangstelling handel om metingstye te verkort en n
uitgebreide BE reaksie wat die lae-modus toestand van klein omhulsels dek. Aanbeveelde
International Electrotechnical Commission (IEC) standaard 61000-4-21 [2] e ektiwiteit
van 0.75, vir log-periodiese dipool-skikking (LPDA) antennas, is ook bekragtig deur n
weerkaatsingskamer karakterisering van n gedrukte-stroombaanbord (PCB) LPDS antenna
e ektiwiteit.
n Ingeboude impulsgenerator en die Square kilometre Array (SKA) ontvanger, RATTY,
vorm deel van die hoof elemente van die tydsgebiedmetings. Vir bekragtigings doeleindes
is n koaksiale lugtransmissielyn gebou. Hierdie kabel is gekarakteriseer deur numeriese
sagteware (CST en FEKO) en is ook gemoduleer met behulp van Vance en Kley se
analitiese uitdrukkings. Die resultate is vergelyk met tydsgebied-oordragsimpedansie (Zt)
metings, wat in die weerkaatsingskamer gedoen is. Die studie wys dat kabel posisie binne
in die weerkaatsingskamer die Zt van die kabel-onder-toets vervorm. Die lugstreep voeg
ook n ossillerende component by tot die gemete Zt. Die resonansie is eweredig aan die
lengte van die kabel en is duidelik sigbaar binne die meting- en simulasiedata. Dit is egter
nie in-berekening gebring in die analitiese modelle nie. Die gevolge van n verkeerde antenna e ektiwiteit in die weerkaatsingskamer toepassing is
ook aangespreek deur ondersoek na die e ektiwiteit van n PCB LPDA. Die onbekende
LPDA is gesimuleer met CST vir hierdie studie. Die resultate is vergelyk met n weerkaatsingskamer
meting, wat die IEC 61000-4-21 standaard e ektiwiteit van 0.75 [1] bekragtig.
Die metode karakteriseer die onbekende antenna veranderlike deur n verwysings dipool
antenna e ektiwiteit wat ge-ondersoek is in FEKO. Gesimuleerde Wheeler-cap tegnieke
het die veri kasie van die verwysings antenna resultate toegelaat. Daar is gevind dat
n onakkurate LPDA e ektiwiteit n klein e ek op die ho er frekwensies het, as die IEC
e ektiwiteit aangeneem is. Daar kan egter n verskil van 7dB voorkom by laer frekwensies,
as die werklike e ektiwiteit met meer as 0.3 van die IEC waarde verskil. Hierdie studie lig die belangrikheid van n korrekte antenna e ektiwiteit uit vir akkurate weerkaatsingskamer
toepassings.
Die geneste-omhulsel tegniek word beskou as die konvensionele metode vir die bestudering
van beskermingse ektiwiteit vir klein omhulsels. Die tegniek is gewoontlik tyd-rowend
en werk net vir frekwensies wat drie maal ho er is as die af-sny waarde van die omhulselonder-
toets. Ons tydgebiedmeting dek net n frekwensieband tot 1.4GHz, wat ooreenstem
met ons omhulsel lae-modus gebied. Die beskermingse ektiwiteit karakterisering, in
hierdie spesi eke veld, is nie goed gedokumenteer nie. Hierdie verhandeling dra by tot
hierdie veld deur gebruik te maak van n onversteurde geneste-omhulsel kon gurasie. In
teenstelling met die konvensionele gebruik van die geneste-omhulsel metode, is die bron
geplaas binne die omhulsel-onder-toets en word dit gebruik as n gewone uitstraler. Die
de nisie van beskermingse ektiwiteit volgens die IEEE standaard in [2], is gevolg en die
totale oordragskrag van die omhulsel is bereken deur gebruik te maak van n poort se
re
eksie ko e si ent. Die benadering benodig nie n steurder vir die omhulsel-onder-toets
nie. Ons metings se beskermingse ektiwiteit het ooreengestem met die veranderde - en
die geskikte geneste-omhulsel tegniek, van 390MHz tot 4GHz in die spektrum. Hierdie
ondersoek is vinniger en lewer n meer gedetailleerde beskermingse ektiwiteit reaksie as
die frekwensiegebied benadering.
Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:sun/oai:scholar.sun.ac.za:10019.1/96037 |
Date | 12 1900 |
Creators | Andriambeloson, Joely Andrianina |
Contributors | Reader, Howard Charles, Stellenbosch University. Faculty of Engineering. Department of Electrical and Electronic Engineering. |
Publisher | Stellenbosch : Stellenbosch University |
Source Sets | South African National ETD Portal |
Language | en_ZA |
Detected Language | Unknown |
Type | Thesis |
Format | x, 119 p. : ill. |
Rights | Stellenbosch University |
Page generated in 0.0019 seconds