Dataweek Electronics & Communications Technology | 25 July 2012
Transcript of Dataweek Electronics & Communications Technology | 25 July 2012
![Page 1: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/1.jpg)
R16.50 (incl VAT)
25 July 2012
FEATURING: Test & Measurement, Analysis, Calibration & Compliance • Audio & Video Electronics
![Page 3: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/3.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 125 July 2012 Vol. 35 No. 14
News
Events
Literature showcase
Systems, components, design
General
Hot chips
QuickFind index
Test & measurement, analysis, calibration & compliance
Audio & video electronics
regulars
All rights reserved. No part of this publication may be reproduced, adapted, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording or otherwise, without the prior written permission of Technews Publishing (PTY) Ltd. Reg No. 2005/034598/07 ISSN 0256-8934
features
Published by:Technews Publishing (Pty) Ltd
1st Floor Stabilitas,
265 Kent Avenue, Randburg
Tel: (011) 543 5800
Fax: (011) 787 8052
Printed by: Intrepid Printers (Pty) Ltd
9 Grix Road, Willowton
Pietermaritzburg
GautengTracy Wolter
Tel: 011 543 5800
Fax: 011 787 8052
KwaZulu-NatalJane van der Spuy
Tel: 031 764 0593
Fax: 031 764 0386
AdvertisingPrint and OnlineSales Manager: Malckey Tehini
EditorBrett van den Bosch
B Eng (Electrical & Electronic)
Subscription servicesFor address changes,
subscriptions, renewal status
or missing issues call 011 543
5800 or fax 011 787 8052 or
WRITE TO:Technews Publishing (Pty) Ltd,
Box 385, Pinegowrie 2123
Subscribe onlinewww.technews.co.za
Western CapeContact Durban or
Jhb numbers for details
2
3
4
6
28
32
33
contents
10
18
highlights . . .
www.dataweek.co.za
On the cover
Microchip’s article on page 18 details
how designers of high-quality audio
systems can benefi t from the digital audio
features built into the PIC32 family of
microcontrollers.
Fluke designs tools for profes-
sionals who take pride in their
work. The process begins with
observing and listening to its
customers, with the result that,
in this year alone, through its
distributor, Comtest, Fluke
has launched six new instru-
ments, with more to come. The
companies bring expertise in
power and indoor air quality
monitoring, process and predic-
tive maintenance with thermal
imagers, infrared thermometers
and vibration screening, help-
ing customers to accurately
troubleshoot problems, predict
failure and prevent costly
repairs.
For more information contact
Comtest, +27 (0)11 608 8520.
OnOnOnOOO ttttthhhhhe cover
With the buzz surrounding wireless charging
and the push to get the technology into
mainstream use, IDT has developed a high-
power transmitter/receiver combination
(page 8).
![Page 4: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/4.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News2 Electronics news digest
DisclaimerEditorial contributions relevant to the focus of this publication are welcome but the publishers cannot accept responsibility for unsolicited material. While every eff ort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily refl ect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication.
Advertisements, inserts and company contact details are printed as provided by the advertiser. Technews Publishing (Pty) Ltd can not be held responsible for the accuracy or veracity of supplied material.
• Cirtech has been awarded its ISO 9001: 2008
accreditation after passing the fi nal audit in
March this year. The set of standards dictates
specifi c levels that need to be maintained for
quality, environmental friendliness, safety, reli-
ability, effi ciency and interchangeability. This
news follows the recent awarding to Cirtech of a
BBBEE level one contributor certifi cate.
• Neil Kayton, former Chief Financial Offi cer
of Power Technologies (Powertech) has been
appointed Chief Executive Offi cer (CEO) of the
company eff ective from 1 July 2012. Kayton
succeeds Norbert Claussen, who recently
stepped down from the position after 16 years
with the group.
• Mantech has bolstered its stockholding by
purchasing Avnet Kopp’s locally warehoused
stock. This deal came about as a result of Avnet’s
centralisation of its warehouse in Europe. The
stock in question includes passives, integrated
circuits, microprocessors, enclosures, transform-
ers, switches and more. Mantech expects the
deal to increase its already wide product off er-
ing which now stands at close to half a million
product lines. The products will be available
through its three South African distribution cen-
tres, and could benefi t existing users of these
components in industry where off -the-shelf
stock is required.
• The long-standing distribution partnership
between TTI and TRX has been made offi cial,
with the companies having signed an agree-
ment that gives TRX access to TTI’s entire range
south africa
overseas
of products, including passive components,
discretes, connectors, relays and switches.
Business
• XP Power announced the opening of its new
manufacturing facility in Ho Chi Minh City,
Vietnam. The factory, believed to be the world’s
most environmentally advanced power supply
manufacturing facility, incorporates the use of
sustainable power and energy effi ciency; the
building’s highly insulated construction includes
the use of a solar PV array, eco-friendly glass
to prevent solar heat gain, low-energy lighting
and ultra-effi cient air conditioning. In addition,
rainwater is captured for use in the building and
for irrigation systems. All the environmental fea-
tures are managed by a fully integrated building
management system.
• Xilinx announced its intention to invest
$50 million in an expansion of its electron-
ics engineering operations at the company’s
EMEA headquarters in Dublin and Cork, Ireland.
The company will recruit 45 senior silicon and
electronics engineering staff for its regional
headquarters in Dublin and for its engineering
centre in Cork. In addition, a further 15 staff
will be hired across a broad range of disciplines
supporting the continued growth of Xilinx in
Ireland.
Companies
• Aeroflex / Metelics announced a strategic
alliance conveying a licence for proprietary
information with Picosecond Pulse Labs for
the manufacture and design of a family of
non-linear-transmission-line based comb gen-
erators. Under the terms of the agreement,
Aeroflex / Metelics will produce the prod-
ucts in its Sunnyvale, California facility
and supply parts both to Picosecond for its
internal use, and to serve the test and measure-
ment industry.
• NXP Semiconductors announced the
appointment of Sean Hunkler as executive vice
president and co-general manager of opera-
tions. The company also announced that
René Penning de Vries has stepped down as
member of the NXP management team and
as CTO. Hans Rijns, senior vice president of
research and development, will take over as
interim CTO.
• Sierra Wireless has entered into an exclusivity
agreement for the acquisition of Sagemcom’s
machine-to-machine (M2M) business. Sierra
Wireless has submitted a binding off er to
acquire substantially all of the assets of the busi-
ness for 44,9 million Euros in cash consideration
plus assumed liabilities.
• u-blox announced the acquisition of Cognovo,
a company specialising in software defi ned
modem chip development technology. The
deal gives u-blox access to 4G chip technol-
ogy and enables it to address new markets for
high-speed LTE modems. u-blox acquired 100%
of the shares of Cognovo at a price of $16,5
million.
• Ramtron’s board of directors has rejected
Cypress Semiconductor’s off er to acquire all
outstanding stock of Ramtron at a price of $2,68
per share in cash, stating that, in consultation
with its fi nancial and legal advisors, the board
“unanimously concluded that the tender off er is
inadequate, does not refl ect the intrinsic value
of the company, and is not in the best interest of
Ramtron and its stockholders.”
• Cadence Design Systems has acquired
Sigrity, a signal and power integrity technology
provider. Sigrity provides gigabit signal and
power network analysis technologies, including
a unique power-aware signal integrity analysis
capability for system, printed circuit board (PCB)
and IC package designs. The acquisition was
made at a purchase price of approximately $80
million.
• Micron Technology and Elpida Memory have
signed a defi nitive sponsor agreement for
Micron to acquire and support Elpida. Under
the agreement, approximately $2,5 billion, less
certain reorganisation proceeding expenses,
will be used to satisfy the reorganisation claims
of Elpida’s secured and unsecured creditors.
Micron will acquire 100% of the equity of Elpida
for roughly $750 million.
Industry
• The Semiconductor Industry Association (SIA)
announced that worldwide sales of semicon-
ductors reached $24,4 billion for the month
of May 2012, a 1,4% increase over the prior
![Page 5: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/5.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 3
EVENTSmonth’s fi gure of $24,1 billion. Combined global
sales for the March-April-May period increased
over December-January-February by the high-
est rate (6,4%) since June 2010. Additionally,
May marked the third consecutive month that
sales have grown over the previous month – the
longest streak of sequential monthly growth
since September 2010. However, sales from May
2012 were 3,4% lower than the May 2011 total
of $25,2 billion, and 2012 year-to-date sales
were lower across all regions than at this time
last year.
• Following a spectacular recovery in 2010, the
power semiconductor discrete and module
market grew by a more modest 9% in 2011,
according to a new report from IMS Research.
(NYSE: IHS). Total power semiconductor reve-
nues grew to just under $18 billion in 2011. The
market for power modules grew by 32%, much
The DM330014 kit enables
designers to quickly leverage
the capabilities and perform-
ances of the dsPIC33F GS
series of digital signal control-
lers (DSCs) to develop LED
lighting products.
The DSC and kit allow
developers to create a 100%
digitally controlled ballast
function, while including advanced features
such as dimming and colour hue control.
The kit includes an LED baseboard with an
onboard dsPIc33FJ16GS504, a DC-DC buck
daughter board and a DC-DC boost daughter
board.
LED lighting applications supported by
the development kit include dimmable LCD
backlighting, signage, LED replacement of
LED lighting development kit up for grabs
PIC24 hands-on workshop
3 August 2012Military Museum, Johannesburg
A hands-on training course based on
Microchip’s 16-bit PIC24 platform; suitable for
students and professionals alike.
Email [email protected] to register
Microchip MASTERS 2012 Conference
6 – 11 August 2012Arizona, USA
Technical training for embedded systems
design engineers, with numerous topics
taught by Microchip product developers and
technical experts.
Register at www.microchip.com
NEPCON South China
28 – 30 August 2012Shenzhen, China
Exhibition showcasing electronics manu-
facturing technologies; co-located with ATE
Register at www.nepconsouthchina.com
Electronica India & Productronica India
11 – 13 September 2012Bangalore, India
The leading trade fair for the electronics
industry in India and South Asia, covering
components, manufacturing technologies
and much more.
Register at www.electronica-india.com
European Microwave Week
28 October – 2 November 2012Amsterdam, The Netherlands
Incorporates three major conferences cover-
ing microwave ICs and technology, wireless
technology and radar. Includes workshops
and seminars.
Register at www.eumweek.com
faster than that for discrete power semiconduc-
tors, which grew only 3%. Infi neon remained
the leading supplier of power discretes and
modules, while other manufacturers whose
share of the market grew included power
module specialists Mitsubishi Electric, Fuji
Electric and Semikron.
Technology
• Audi is using Sierra Wireless AirPrime embed-
ded wireless modules to help develop and test
an LTE-connected infotainment system. Still in
the development and testing stage, the system
has been showcased at several industry events
this year. Sierra Wireless AirPrime modules are
available with varying combinations of LTE
frequency bands and fallback technologies, and
were the fi rst LTE modules available, allowing
Audi to work on the leading edge of wireless
technology as LTE networks roll out.
Dataweek is off ering
readers the chance to
win a Microchip LED
lighting development kit.
fl uorescent tubes and incandescent bulbs,
architectural lighting and automotive lighting
applications. Automotive lighting prod-
ucts include exterior applications, such as
headlights, daytime running lights and signal
lights.
For more information visit
www.microchip-comps.com/dw-led
![Page 6: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/6.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News4
LITERATURE SHOWCASE
Contact Technews to place your literature
in the Literature Showcase
IP&E Component Guide
Avnet Kopp has released its 2012 IP&E components
guide containing an overview of passive, electro-
mechanical, interconnect and enclosure products
available from the company’s suppliers. The guide
includes colour photos and performance sum-
maries of all products, as well as line cards for
the various product types. An electronic ver-
sion can be downloaded at www.avnet.co.za/
Download/2012_IPE_Components_Guide.pdf.
Avnet Kopp, +27 (0)11 809 6100
u-blox has released its Product Catalogue 13,
containing the latest updates to all its products
and services, including the latest generation
of low-power GNSS receiver products based
on the new u-blox 6 platform, plus compact,
cutting-edge wireless communication
modules and reference designs implement-
ing 2G and 3G mobile communication
technologies.
RF Design +27 (0)21 555 8400
u-blox Product Catalogue
ts
Interconnect Solutions Catalogue
Samtec’s new 2012 Interconnect Solutions
Catalogue includes a number of new products,
new high-speed test data and updated specifi ca-
tions on existing products, as well as an overview
of Samtec’s ever-expanding RF product line. In
addition, the catalogue highlights expanded
product features such as lower-cost plating
options, new rugged screw-down options for
board-to-board and discrete wire systems,
and latching options for ultrafi ne-pitch
connectors.
Otto Marketing, +27 (0)11 791 1033
Skiff y Catalogue No. 65
The latest version of Skiff y’s catalogue includes
information on the wide range of plastic and
metal parts supplied by the company, such as
nuts, washers, spacers, fasteners, caps and plugs.
Among the new products featured are PCB spac-
ers, knobs and handles, shrink tubes, grommets,
cable glands and clamps.
Skiff y, +27 (0)11 314 8750
Samte
Catalo
new
tion
of S
add
pr
o
b
a
O
Skiff
The l
info
me
nu
Am
e
c
IP&E Co
Avnet K
guide
mech
avail
incl
ma
th
s
D
Avnet Kopp, +
N 65
Schroff Innovations Brochure
This brochure details some of Schroff ’s new
developments in the areas of subracks,
card locks, systems and cabinets. Colour
photographs and product specifi cations
are provided.
Actum Electronics, +27 (0)11 608 3001
Small-scale renewable energy solutions, produced from zero- or low-carbon
source technologies and derived from naturally regenerative or practically
inexhaustible resources, are the latest addition to the list of energy-effi cient
technologies funded through Eskom’s Standard Off er Programme.
Introduced in consultation with NERSA as a pilot project with total capacity
achieved during this phase limited to 10 MW, Eskom views small-scale renew-
able energy solutions as a way to further broaden the scope of optimal energy
usage in South Africa through its range of energy-effi ciency rebate programmes.
For this purpose, renewable energy sources include biomass waste, wind
energy, and geothermal, solar, thermal gradient and ground source heat.
Solar thermal liquid heat is excluded from the programme but municipal solid
waste will be considered a source. Small-scale renewable energy solutions will
be limited to an installed capacity range of between 10 kW and 1 MW.
Customers or project developers wishing to apply for funding must
submit electrical and structural designs certifi ed by qualifi ed engineers as
part of the application, and a letter from the electricity supplier (Eskom or a
municipality) confi rming knowledge and acceptance of the proposed
network connection, and registration of the project as per NERSA requirements.
To qualify for funding, equipment should be installed on the energy user’s
side of the meter; systems should be new and in compliance with perform-
ance and safety standards applicable to all components of the installation.
In addition, standalone systems not tied to the grid must replace existing
national electricity connections to substitute energy that would have been
drawn from the grid. Grid-tied systems must comply with all regulatory
and embedded generation interconnection requirements; a letter from the
relevant local authority or electricity utility will be required as part of the
application to confi rm the applicant’s knowledge and acceptance of the
proposed connection onto the network.
Initially, funding will only be available for systems that do not feed elec-
tricity onto the grid at a rate of R1,20 per kWh, and be paid as per the usual
Standard Off er Programme milestone payments based on the agreed rate,
and the total savings delivered over the contract period of three years.
For more information visit www.eskomidm.co.za/industrial
Eskom promotes small-scale renewable energy
Ember is the latest manufacturer to board the ZigBee Light Link band-
wagon. The company’s ZigBee platform – chips, software and tools – has
achieved Golden Unit certifi cation from the ZigBee Alliance for the newly
ratifi ed Light Link standard.
ZigBee Light Link is a wireless interoperability standard for consumer
LED lighting and control products. It allows consumers to wirelessly gain
control over their LED fi xtures, light bulbs, timers, remotes and switches.
The standard ensures that ZigBee certifi ed products from diff erent vendors
will easily connect into a single network, without the need for additional
devices to coordinate the network.
The standard will also give consumers the ability to control lighting
remotely over the Internet, providing yet another avenue for managing
energy use and making their homes greener. In addition, the standard
enables lighting manufacturers to target the broader ZigBee home automation
market through the capability to integrate with other ZigBee standards.
As one of the fi rst fi ve Golden Units certifi ed by the ZigBee Alliance,
Ember’s ZigBee platform, including EM300 series wireless chips, EmberZNet
PRO networking software and development tools, will serve as a preferred
development platform to build and test future ZigBee Light Link products.
For more information contact Michelle Parreira, CST Electronics,
+27 (0)11 608 0070.
More ZigBee Light Link ‘golden units’
![Page 7: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/7.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 5
Local electronic research and development house
Keystone Electronic Solutions has been awarded
the contract to provide its Remote Site Monitor
solution for all DFA monitoring facilities in order
to provide proactive monitoring of the cable on
DFA’s new generation networks (NGN).
Jacques van Loggerenberg, DFA technical execu-
tive, comments: “DFA has a build, own and maintain
model whereby it is responsible for deploying the
fi bre infrastructure, for our clients to light up. The
initial build was focused around metro build in
Johannesburg, Pretoria, Cape Town and Durban.
“We then completed two long haul links
from Mtunzini to Pretoria and Cape Town to
Yzerfontein. We now cover all three international
cable landing stations through Mtunzini (EASSy,
Seacom and SAFE), Melkbos Strand (SAT2, SAT3
and SAFE) and Yzerfontein (WACS).
“Last year we initiated an NGN project where
we evaluated all the existing building blocks we
were deploying on the network. The project was
part of the DFA Total Quality Management Pro-
gramme (TQM) whereby products get evaluated
on a continuous basis to ensure that the best prod-
uct is used for immediate and future requirements.
“One of the new building blocks identifi ed
in this process was the need for a Primary Cross
Connect (PCC) facility. The purpose of the primary
cross connect is to partition and provide fl exible
connectivity between the metro core network
and the long haul fi bre network.”
He continues: “These PCC nodes will be housed
in containers or building locations, and will cater
for up to 4320 fi bre connections at a single node.
These facilities will also be used to house the
remote testing units (RTUs) which are used for
proactive monitoring of the DFA fi bre cables.
“Because monitoring and security are of the
utmost importance to DFA it was decided that we
required a remote site monitoring solution for the
PCC sites. Although we had used monitoring equip-
ment before, we needed to integrate the best prod-
uct for our specifi c requirements on this project.
“We developed a list of requirements, and
together with an external consultant for objectiv-
DFA chooses Keystone’s remote site monitor
ity, we eventually shortlisted three products, one
of which was the Remote Site Monitor developed
locally by Keystone. We installed the products
in a test environment in one of the facilities, and
then began stringent testing,” he explains.
One of the criteria was fast and effi cient
local support, and so the consultant requested
changes and additions to the functionality on a
regular basis, in order to assess how the compa-
nies would handle rapid and diverse changes.
Keystone was able to turn around most of the
changes within 6-8 hours.
The fi ndings of the testing were documented
and evaluated according to the predetermined
criteria. Keystone scored the highest and was
selected to provide its monitoring solution for the
DFA PCC monitoring facilities rollout. In fact, the
Keystone test site was able to go live onto the net-
work immediately as the fi rst monitored facility.
John Eigelaar, Keystone technical director,
explains the solution: “The Remote Site Monitor
consists of modular rack-mounted hardware as
well as a server-based management platform that
can be accessed remotely through a web-based
interface. The geo-referenced interface plots each
facility’s location and status on a map of South
Africa, and immediately refl ects any critical change
in the status of the facility on the icon on the map.”
For the DFA installations, the RSM monitors
access, power and environmental factors such
as air temperature, movement and humidity.
This enables DFA to have full monitoring capa-
bility to all its PCC facilities.
Says van Loggerenberg: “One of the advan-
tages of the Keystone solution was that it easily
integrated with our existing network software.
We had used Keystone previously to provide
other network management solutions, so it was
really just a case of expanding the functionality
of the existing network software.
“Another bonus was that unlike other prod-
ucts, Keystone was able to provide us with only
those modules that we required. Other products
came as a bundled package, so would have cost
a lot more, and come with a range of features
we didn’t require. And additionally, we have
confi dence that Keystone can develop modules
for any future requirements that we may have,
making it totally scalable,” he concludes.
For more information contact Keystone
Electronic Solutions, +27 (0)12 460 4135.
Jacques van Loggerenberg (DFA), John Eigelaar (Keystone) and Derek Koekemoer (DFA).
Dark Fibre Africa (DFA), founded in 2007,
is an open access, carrier neutral dark
fi bre provider to all licensed entities. It
currently has more than 6000 km of duct
route installed across South Africa.
![Page 8: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/8.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News6 regular Systems, components, design
Premier Farnell has extended its relation-
ship with ARM through the launch of
the latest series of ARM Cortex-M3 and
Cortex–M4 processor-based STM32F2xx
and STM32F4xx development kits.
The deal will see the kits shipped with
Keil MDK-Lite development tools and the
ULINK-ME debug adaptor. Developers will
benefi t from support by Premier Farnell’s
ARM trained engineers, the element14
ARM development tools group and rich
technical content, video support materi-
als, and design examples on the Knode
on element14.
STM32 development kits
The MCBSTM32F200 board features
an STM32F207IG ARM Cortex-M3 based
microcontroller which off ers 1 MB Flash
and 128 KB RAM on-chip memory, and
the MCBSTM32F400 features an STM-
32F407IG Cortex-M4 based microcon-
troller with 1 MB Flash and 192 KB RAM
on-chip memory.
In addition, both evaluation boards
are populated with 8 MB NOR Flash,
512 MB NAND Flash, 2 MB SRAM, and
8 KB I2C EEPROM with an NFC interface.
They provide a colour QVGA LCD and
interfaces for SD card, CAN, 10/100 Ether-
net and Full/High-speed USB. The boards
also feature a 3-axis digital accelerometer,
a 3-axis digital gyroscope, a digital micro-
phone and a digital VGA camera.
Also featured are a dedicated audio PLL
and two full-duplex I2S. The kits have up
to 15 communication interfaces (including
six USARTs running up to 10,5 Mbps,
three SPI running up to 42 Mbps, three
I2C, two CAN, SDIO). Analog functions
include two 12-bit DACs, three 12-bit
ADCs reaching 2,4 MSps or 7,2 MSps in
interleaved mode. The boards have up to
17 timers, 16-bit and 32-bit running up to
120/168 MHz.
For more information contact
Electrocomp Express, 0860 10 20 20.
It off ers high linearity with
25,9 dBm IIP3 (input third-order
intercept), high conversion gain
of 7,4 dB owing to its integrated
IF amplifi er, and a noise fi gure of
11,3 dB.
These performance character-
istics enable excellent dynamic
range performance for both main
receivers and digital pre-distortion
receivers. Moreover, the IC’s RF
input can withstand strong interfer-
ence signals without signifi cantly
degrading its noise fi gure, thus
preserving receiver sensitivity and
enhancing robustness.
The LTC5544’s 6 GHz reach is
ideally suited for wireless systems
such as point-to-point broadband
microwave links, 5 GHz licence-free band
WiMAX radios, satellite receivers, radar systems,
avionics, public safety radios and RF test
systems.
The device incorporates an on-chip, high-
gain IF amplifi er capable of supporting IF
frequencies up to 1 GHz. Also integrated are an
LO buff er along with two RF balun transformers.
They enable the RF and LO inputs to operate
single-ended and with 50 Ω matching. The LO
Down-converting RF mixerLinear Technology announced the LTC5544, a 4 GHz to 6 GHz, high dynamic
range down-converting mixer.
input requires only +2 dBm drive level.
The chip operates on a single 3,3 V supply,
drawing 194 mA of current. It has a power-down
feature with 0,6 μs turn-on and turn-off time,
supporting Burst Mode operation or TDD (time
division duplexed) systems. When disabled, the
device consumes a maximum of 500 μA.
For more information contact Arrow Altech
Distribution, +27 (0)11 923 9600.
Texas Instruments has
expanded its award-
winning WEBENCH suite
of tools by introducing
System Power Architect,
an online tool to speed
design of multi-output,
high-performance
DC-DC power supply
systems.
The tool enables the
user to build and opti-
mise a complete power
system from the edge of
the card to the smallest
point-of-load. In addi-
tion to the design and
simulation of individual
point-of-load converters,
it adds the ability to accommodate multiple
intermediate bus architectures and safety isola-
tion in conjunction with TI’s hot-swap control-
lers and isolated power products.
With the System Power Architect, TI claims
designers can create isolated power systems
with hot swap capability in just about 60
seconds. When driving multiple loads, the
designer can customise output voltages and
Power supply design tool
load currents, and also select FPGAs and pro-
cessors, which are preconfi gured with critical
requirements, load grouping and sequencing.
Other options to select include ripple fi lter,
low dropout regulators (LDO), external sync or
separate supply.
For more information visit www.ti.com/
webenchspa-pr-eu
on element14.
![Page 9: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/9.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 7Systems, components, design
Nordic Semiconductor announced the fi rst members of its new nRF51
series of ultra-low power (ULP) RF ICs that feature a new multi-protocol
2,4 GHz radio and a 32-bit ARM Cortex-M0 based processor.
Nordic’s novel software architecture for Bluetooth low energy and
ANT system-on-chip (SoC) solutions frees designers from the integra-
tion eff ort, complexities and restrictions of chip vendor-supplied
software frameworks and instead allows customers to develop their
designs quickly and easily using the highly popular ARM Cortex pro-
gramming environment.
This is achieved by using a new software architecture featuring
a unique and powerful separation between protocol stack and user
application code. This separation provides developers a clean bound-
ary between application and protocol stack, and removes the need
to struggle with integration of application code as part of a vendor-
imposed application development framework.
Multi-protocol 2,4 GHz SoCs
These GaAs pHEMT MAPS-010145 and MAPS-010165 are 4- and 6-bit
packaged devices for use in C-band applications from 3,5 to 6,0 GHz.
They maintain low phase error and minimal attenuation variation over
the full 360 degrees and frequency range.
The MAPS-010145 has a built-in CMOS driver that allows for serial
or parallel control and low power consumption. The 360 degree phase
shift is broken down to steps of 22,5 degrees. It off ers insertion loss of
5,0 dB with a ±1,0 dB attenuation variation and high phase accuracy
over the operating band.
The MAPS-010165 also has a built-in CMOS driver and is capable of
either serial or parallel operation, while the 360 degrees of phase shift
is divided into 5,6 degree step sizes. It has 5,5 dB of insertion loss with
a low ±0,8 dB attenuation variation and high phase accuracy over the
entire band.
The phase shifters operate using both positive and negative voltage
supply. They are packaged in industry standard 4 mm, 24-lead PQFN
style packages. They are intended for use in commercial weather radar
and military phased array radar applications.
For more information contact M/A-COM Technology Solutions,
+27 (0)11 908 6679.
Digital phase shiftersM/A-COM Technology Solutions has added two new
members to its digital phase shifter product family.
The fi rst two Flash-based ICs to début in the new series are the
nRF51822 multi-protocol Bluetooth low energy / 2,4 GHz proprietary RF
SoC, and the nRF51422 ANT SoC.
The new chips provide -92,5 dB receive sensitivity in Bluetooth low
energy mode, and up to +4 dBm output power in all modes. They
feature sub-10 mA peak currents running off a 3 V coin cell battery.
For more information contact Andrew Hutton, RF Design, +27 (0)21 555 8400.
![Page 10: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/10.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News8 regular Systems, components, design
Hittite has expanded its clocks and
timing product portfolio to include the
HMC1031MS8E, a low-power integer-N PLL
clock generator. This new device is ideal as
a building block for jitter attenuation, OCXO
and TCXO frequency multiplication and low-
power VCXO-based clock generators driving
data converters (ADCs/DACs), FPGAs and
SERDES applications.
The chip includes a phase/frequency de-
tector with a charge pump current output
and a low-power integer-N divider resulting
in reference frequency multiplication ratios
of x1, x5 and x10. Together with an external
loop fi lter and a VCXO, VC-TCXO or OCXO,
the HMC1031MS8E forms a complete clock
generator solution targeted for jitter clean-
up and reference clock multiplication /
generation applications.
The IC may be used with a low phase
noise VCXO to multiply 10 MHz TCXO or
OCXO sources to 50 MHz or 100 MHz with
excellent long-term stability. Such a topol-
ogy reduces system cost and improves
control over the design because high-
frequency OCXOs and TCXOs may be more
expensive, have higher ageing rates and
degraded phase noise compared to their
low-frequency 10 MHz counterparts.
PLL clock generator
Additional applications of the HMC-
1031MS8E may include low-frequency clock
jitter attenuation and clock frequency trans-
lation or synchronisation for 25 MHz and
125 MHz Ethernet clocks and 27,000 MHz,
74,250 MHz and 74,1758 MHz video clocks.
The integrated phase detector and
charge pump are capable of operating up
to 140 MHz which means the designer can
eliminate noise-generating divider circuitry,
further reducing cost; additionally, VCXOs as
high as 500 MHz can be used.
The low-power architecture of the de-
vice consumes less than 2 mA under typical
conditions, making the device well suited
for power-sensitive battery powered and
portable applications. Additional features
include integrated lock detect indicator on
a dedicated hardware pin, and a built-in
power-down mode for further power
saving.
For more information contact Andrew
Hutton, RF Design, +27 (0)21 555 8400.
The ADN469xE series
includes eight transceivers,
each of which is capable of
connecting 32 data/clock nodes using a single
diff erential cable pair at data rates of 100 Mbps
or 200 Mbps. Traditional LVDS communication
links, by comparison, must use 32 separate
point-to-point nodes, which signifi cantly
increases power, connector size, cabling costs
and total board space.
The devices deliver 8 kV IEC ESD protection
to improve the reliability of insertable/
removable boards and cards used in wireless
base station and networking infrastructure, data
acquisition, automated test equipment and
other high-speed, highly networked backplane
and cable applications.
Available in half- and full-duplex versions,
the transceivers are fully compliant with the
Multipoint LVDS transceivers
TIA/EIA-899 M-LVDS standard. The ADN4691E/
ADN4693E are Type 1 receivers with 200 Mbps
data rates that include symmetrical input
threshold voltage and are designed to improve
clock timing performance.
The ADN4696E/ADN4697E Type 2 receivers
with 200 Mbps data rates are designed for
data applications and feature a receiver-input
threshold voltage off set of -100 mV. This ensures
a known receiver output state when the bus is
idle (bus-idle failsafe) or the bus inputs are open
(open-circuit fail-safe).
For more information contact
Arrow Altech Distribution,
+27 (0)11 923 9600.
The IDTP9030 and IDTP9020 comprise a
wireless power transmitter and receiver solu-
tion designed to meet the Wireless Power
Consortium (WPC) Qi standard, which ensures
interoperability with any other device meeting
the standard.
Both transmitter and receiver are capable
of multi-mode operation, supporting both the
Qi standard as well as proprietary formats for
added features, improved safety and increased
power output capability. Built-in protocol
detection enables dynamic switching between
Qi and proprietary modes, creating seamless
transitions and a trouble-free user experience.
These devices can be used in a myriad of mobile
applications for convenient battery charging.
The IDTP9020 is a high-effi ciency wireless
power receiver delivering up to 5 W in Qi mode.
When used with the IDTP9030 transmitter
in a proprietary confi guration, it can deliver
up to 7,5 W. The devices also feature propri-
etary multi-layered foreign object detection
Wireless charging chipsetIDT has developed a single-chip wire-
less power transmitter accompanied
by a high-output-power single-chip
receiver solution.
(FOD), utilising sophisticated multi-parameter
algorithms to assure a high level of safety while
avoiding FOD false alarms.
The solution features over-temperature,
over-voltage and over-current protection. In
addition, the receiver includes support for
charging the battery via USB cable when a wire-
less charging station is not available, eliminating
the need for external USB adaptor switches in
mobile devices.
For more information contact Marian
Ledgerwood, Future Electronics,
+27 (0)21 421 8292.
Recently introduced
by Analog Devices is
a series of multipoint,
low-voltage, diff erential
signalling (M-LVDS)
transceivers with
the highest ESD
(electrostatic discharge)
protection of any
multipoint LVDS
transceivers.
![Page 11: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/11.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 9regularSystems, components & design
The MCUs are system-on-chip devices based on
the Cortex-M4 core with FPU, and are optimised for
effi cient handling and processing of mixed signals in
circuits such as three-phase motor controls, biometrics
and industrial sensor outputs or audio fi lters. They are
intended to simplify design, cut power consumption
and reduce PCB size in consumer, medical, portable fi t-
ness, system monitoring and metering applications.
The new F3 series extends the scope of ST’s STM32
family by allowing designers to tackle mixed-signal
control applications. Featuring versatile analog
circuitry, the series provides low to medium memory
sizes. The Cortex-M4 with FPU core augments the
proven Cortex-M3 CPU with additional digital signal
processing (DSP) capability, optimised single-cycle and
saturating arithmetic instructions, and fl oating-point
unit (FPU), to boost performance over ST’s STM32 F1
Cortex-M3 devices.
Its features position the new series between the
STM32 F1 and the STM32 F4, which features the
Cortex-M4 with FPU core and large memory capacity
combined with high-speed 168 MHz maximum CPU
frequency for complex applications.
With the new F3 microcontrollers, ST’s STM32
family now off ers more than 350 variants addressing
applications from entry-level, price-sensitive designs
through to those demanding high performance and
rich on-chip features. The F3 family comprises four
product lines, diff erentiated according to on-chip
memory density and integrated peripherals.
STM32 F30x devices have seven fast comparators
with 50 ns response time, four programmable gain
amplifi ers supporting four diff erent gain settings with
1% accuracy, two 12-bit DACs, and four 12-bit, 5 MSps
ADCs. ADC performance can extend up to 18 MSps in
interleaved mode.
The devices also have two advanced motor-control
timers running at up to 144 MHz. These allow one
microcontroller to manage two motors plus power
factor correction (PFC) via its on-chip analog peripher-
32-bit ARM MCUsSTMicroelectronics is now delivering the fi rst samples of its new STM32 F3 series of micro-
controllers, enabling early evaluation of this new line in ST’s product family featuring the ARM
Cortex-M core embedded.
als, in equipment such as home appliances. Among
other capabilities of the advanced timers, timing
resolution better than 7 ns allows applications such as
digital power supplies in telecom infrastructure or data
servers, micro-inverters in solar installation and LED
lighting.
To help developers optimise code execution time,
the F30x devices have 8 KB of SRAM mapped to the
instruction bus as Core Coupled Memory (CCM-SRAM).
Critical routines loaded in the 8 KB CCM at startup
can be completed at full speed with zero wait states,
achieving 94 Dhrystone MIPS at 72 MHz. This compares
with 62 DMIPS when executing from Flash or SRAM;
equivalent to a 52% performance increase for critical
routines.
The F30x series also has up to 40 KB of SRAM
mapped on the data bus. The 8 KB CCM-SRAM can
also be used for data storage with no performance
trade-off . Devices are pin-compatible with the STM32
F1 series, and also share the same API, which simplifi es
migration to higher-performing devices.
The F37x devices deliver a diff erent blend of
peripheral functions. They are the fi rst microcontrollers
from ST to integrate a 16-bit Sigma-Delta ADC on-chip,
allowing the STM32 family to solve a wider range of
precision-sensing applications and providing a single-
chip solution capable of replacing a discrete general-
purpose processor and separate ADC.
Up to three 16-bit Sigma-Delta ADCs are integrated
on-chip, with separate analog supply from 2,2 V to
3,6 V, up to 21 single or 11 diff erential channels, and
seven programmable gains per channel.
Other STM32 F37x analog peripherals include two
fast comparators, three 12-bit DACs and one 12-bit,
1 MSps ADC. There is also an enhanced CEC (Consumer
Electronics Control) unit, making these devices relevant
for consumer multimedia products.
For more information contact Avnet Kopp,
+27 (0)11 809 6100.
![Page 12: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/12.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News10 feature Test & measurement, analysis, calibration & compliance
Using a particle counter is relatively simple;
however, understanding the features
that distinguish counters can sometimes
be a challenge. The following terms are
commonly used to describe the accuracy,
effi ciency and other attributes of an air-
borne particle counter.
Count mode: The count mode defi nes
how the particle counter displays data to
the user. Concentration and raw count are
two typical sampling modes, and the Fluke
985 adds an audio mode. Concentration
mode samples a small volume of air then
calculates the value based upon the volume
setting (cm3, ft3 or litre) in the counter while
raw count mode allows the user to collect
actual particle counts as they accumulate
per sample time.
It can be set to cumulative (the total
number of particles > each channel size)
or diff erential (the number of particles fall
between the channel sizes). Audio mode is
useful when searching for areas that exceed
predefi ned particle levels. Once exceeded,
the counter notifi es the user in audio alarm.
The count mode defi nes how the particle
counter displays data to the user.
Zero count: Zero count is a measure of
the particle counter’s accuracy, and should
be taken prior to use and periodically there-
after, or when sampling error is suspected.
The zero count fi lter is attached to the par-
ticle counter per the manufacturer’s instruc-
tions, then the counter is run for 15 minutes.
The counter should not have detected more
than one particle greater than 0,3 μm in a
fi ve-minute period.
Coincidence loss: Coincidence loss occurs
when two particles cross the counter’s light
beam simultaneously, creating a single
pulse and resulting in a single particle
count. This type of error occurs more fre-
quently as the concentration of particles
increases within the sample. Per 21501-4,
coincidence loss must be ≤10% by the maxi-
mum particle number concentration. Fluke
985 is 10% at 4 000 000 particles per ft3.
Counting effi ciency: The probability that
the counter will sense and count a par-
ticle passing through the sample volume.
Counting effi ciency is a function of size up
to a minimum sensitivity threshold, above
which all particles are sensed and counted.
A counting effi ciency of 50% at the most
sensitive threshold is typically considered
optimal, and facilitates consistent compari-
sons between counts from OPCs and those
of higher-resolution instruments.
Sensitivity: A device’s ability to detect
small particle sizes at a certain counting
effi ciency. The Fluke 985 detects 0,3 μm at
50% counting effi ciency.
Calibration: A set of operations or ac-
tions taken to establish the relationship
between the measured values obtained
through a device and the values of the
corresponding parameters as defi ned in a
standard. The Fluke 985 is calibrated using
PSL (polystyrene latex) spheres, widely
used due to their uniform size and light-re-
fraction properties and meets ISO 21501-4:
Light scattering airborne particle counter
for clean spaces.
NIST traceable: Traceability is a charac-
teristic of a measurement or standard and
its relationship to stated references, which
are often national or international stan-
dards. The PSL spheres used in the Fluke 985
calibration process can be traced to NIST (US
National Institute of Standards and Technol-
ogy) standards.
Making sense of particle counter features
Leaving the system off until occupants arrive
in the morning might save money, but allows
pollutants to build up in the air. If an HVAC
system is not properly maintained and the air
fi lter checked, it could lead to IAQ problems. The
eff ects of IAQ have been highlighted in numer-
ous environmental risk reports and joint agency
eff orts by the US Environmental Protection
Agency (EPA). Further studies indicate that
indoor air in some commercial buildings is up to
fi ve times more polluted than outdoor air.
Airborne particulates come in a variety of
forms ranging from animal dander, plant pollen
and airborne bacteria, to fi bre-glass, asbestos
and combustion particles that are generated
from part of the production equipment or
process. To properly identify and troubleshoot
IAQ problems, the technician needs a tool that
not only reads particle concentrations for spot
check, but also provides continuous monitoring
for process control.
Why particle counts matter
Diff erent locations have varying levels of accept-
able particulate concentrations. In a residential
and commercial environment (i.e. homes, offi ces,
Tips for monitoring indoor air qualityInformation from Fluke.
Energy savings are top of facilities owners’ agendas, and HVAC systems tend to be the largest contributor to the energy bill. Tax
rebates are often based on energy savings per square metre, but savings on HVAC energy costs can cause signifi cant indoor air
quality (IAQ) issues due to improper ventilation, temperature and humidity control.
hotels), health and comfort concerns and fear of
litigation often drive IAQ investigations.
In industrial and institutional environment
(hospitals, food and beverage plant, electronic
and precision manufacturing), energy cost,
contamination control and production yields
are the primary concern. Excessive levels can
result in medical conditions such as Sick Building
Syndrome, lower productivity, contaminated
product or all of the above.
Maintaining acceptable air quality levels not
only lowers the costs associated with downtime,
but also reduces or removes costs associated
with expensive fi xes in the future. The fi rst step
in establishing an IAQ maintenance programme
is to determine if a problem currently exists.
IAQ investigation
An IAQ investigation is the fi rst step in an ongo-
ing maintenance programme or in responding
to complaints potentially associated with air
quality. No matter if it is a commercial/residen-
tial or industrial/ institutional environment, the
methodology is similar:
1. Conduct a survey of credible staff at the facil-
ity. Who has fi led complaints, and what are
the symptoms? Are those who complained
centrally located, or dispersed throughout
the facility? The purpose is to gauge the level
of toxicity or concentration as evidenced by
allergic reactions or irritation in human com-
fort or increased contamination in product.
2. Research the building’s history. When was the
building built and/or remodelled? Have there
been any instances of excessive damage,
and how were the repairs carried out? What
![Page 13: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/13.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 11Test & measurement, analysis, calibration & compliance feature
are the maintenance practices
within the facility? For exam-
ple, roof or plumbing leaks
may have been repaired, but
the underlying water damage
may not have been addressed.
3. Perform a physical inspec-
tion. Technicians need to be
aware of their test environ-
ments by considering harmful
particulate sources. Within a
given location, areas contain-
ing exhaust vents, furnaces,
cleaning supplies, as well as
areas with fresh paint and/or
carpeting must be accounted
for, especially if present within
a complaint area. Are there any
odours or visible sources (i.e.
mould)?
4. Take air quality measurements.
When conducting a complete
IAQ investigation, tempera-
ture, humidity, CO and CO2
readings should also be taken
to identify problems related to
inadequate and/or contami-
nated ventilation, potentially
creating a particulate problem.
For example, temperature and humidity
readings play a key role in identifying mould
and bacteria. A location with high relative hu-
midity and higher concentrations of particles
3,0 μm or larger may indicate the presence
of mould spores, which should be remedied
once identified. High concentration of 0,3 μm
to 10 μm particles may imply the presence of
bacteria which may endanger patients in a
surgery room or special infirmary.
The most efficient procedure
for assessing indoor air quality is to
obtain several outdoor air read-
ings as baselines, noting where the
readings are taken in relation to the
facility. At least one of the readings
should be from near the building’s
fresh air intake vent.
Note, however, the location of
the intake vent to ensure baseline
readings are not skewed by pollut-
ant sources, for example, locations
near a loading dock. An indoor
air particulate ‘target’ is then cal-
culated by modifying the baseline
readings by the efficiency of indoor
filtering.
In the case of cleanroom applica-
tions, ‘as built’, ‘at rest’ or ‘opera-
tional’ (the three cleanroom stages)
can be used as baseline. But once
the baseline is established, data col-
lected should always be compared
against the same baseline.
Particles tend to diffuse very
quickly into the surrounding air,
making source identification a
challenging task. One method is to
take multiple indoor readings, starting with
the complaint area first, then move outwards.
Get a map of the HVAC systems, as installed
and use that to create an inspection route.
Within each zone, take a measurement in the
![Page 14: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/14.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News12 feature Test & measurement, analysis, calibration & compliance
middle of the space as well as near the air
intakes, outtakes and any other HVAC system
elements.
Make particular note to measure both
upstream and downstream at any HEPA
filters. As data is collected, take note of any
unusual increases in particulate quantity
and size. One can make use of the Fluke 985
location naming and data storage capability
to distinguish particle concentration in dif-
ferent locations for comparison.
Compare the particle readings against the
baseline to get a feel for the relative severity
of the particulate concentration, and identify
hotspots and pathways that may lead to the
particulate source. Continue to follow the
path of higher concentrations until
the source is identified. Once the source
is remedied, the area is reassessed to
ensure the corrective action addressed the
problem.
Combined with the Fluke 971 tempera-
ture and humidity meter, the technician can
also gauge the temperature and humidity
readings against the accepted parameters
(see ASHRAE Standards 55 and 62) for a basic
IAQ investigation, and take appropriate steps
to treat the problem.
Interpreting the data
A correct interpretation of the data requires
an understanding of the test area. Is the
area residential or commercial? Is it a clean-
room or precision manufacturing area? Is
the cleanroom at rest or operational? Is the
location exposed to tobacco smoke or com-
bustible particles? Is there construction at
or near the location? A proper assessment of
the environment can help narrow down the
list of problem particulates.
Concentration limits vary widely accord-
ing to the size and type of facility, among
other variables. However, a high-level assess-
ment can provide direction on whether or
not a problem exists. The outside air read-
ings in Figure 1 provide a high-level point of
reference for the technician.
Scenario 1: The particulate levels dis-
played in Figure 2 are from a new residence
(<5 years), and do not indicate any concen-
trations outside of the norm. In a residential
setting, particle levels are sometimes higher
than outside readings due to more potential
particle sources (i.e. pet dander), smaller
diffusion area, and often less sophisticated
filtration.
Scenario 2: The particulate levels dis-
played in Figure 3 are representative of
an average office workspace, and do not
indicate any concentrations outside of the
norm. In a commercial setting, particle levels
should be significantly less than outside
readings due to better filtration and better
dilution with outside air.
Scenario 3: The particulate levels in Figure
4 are from an older residential location with
visible mould. The readings are significantly
higher, and steps should be taken to remedy
the mould and address the root cause of the
problem.
Scenario 4: If the particle source in Scen-
ario 3 is not visible, use particle size tables such
as Figure 5 to identify possible sources. Obtain
a sample of the particles and submit to a lab for
further analysis.
A cleanroom exercise (certification
and monitoring)
Cleanrooms are an excellent application for a
particle counter. Certification for cleanrooms
is usually conducted at ‘as built’ stage. For
illustration purposes, let’s put the Fluke 985
to the test in evaluating an ISO Class 5 (per
ISO 14644) cleanroom. To qualify as a Class
5 cleanroom, levels cannot exceed the limits
for the class in each particle size stated in
Table 1 (Page 14).
Our test is concerned with the concen-
tration of 0,3 μm particulates in the room.
Several 2 litre samples are taken from six
different locations over six different times
inside the cleanroom, with the results as
shown in Table 2 (Page 14).
The individual readings are well within
the limitations for the cleanroom; however,
we can take the following steps to determine
the statistical validity of the readings:
SD = ( (AC1 – M)2 + ... + (AC
6 – M)2)/(L –1)
116
= ( (674–995)2 + (1154–995)2 + (1097–995)2 + (841–995)2 + (828–995)2 + (1376–995)2)/(6–1)
SE = SD/ L
47,36 = 116/ 6
Step 1: Calculate the mean average particulate concentration.
M = (AC1 + AC
2 + AC
3+ AC
4+ AC
5+ AC
6) / L
995 = (674 + 1154 + 1097 + 841 + 828 + 1376) / 6
Step 2: Calculate the standard deviation of the averages.
Step 3: Calculate the standard error of the mean of the averages.
Step 4: Establish the upper confi dence limit (UCL) – see Table 3 (Page 14).
![Page 15: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/15.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 13Test & measurement, analysis, calibration & compliance feature
Figure 1. Reference reading.
Figure 2. Typical particulate levels in a new residence.
Figure 3. Typical particulate levels in an offi ce space.
Figure 4. Typical particulate levels in an older
residence with visible mould.
Figure 5. Typical particle sizes in μm.
The resulting mean count for all locations is within the
requirements of a Class 5 cleanroom.
The Fluke 985 provides particulate data over six chan-
nels on a single display, allowing the technician to view
all readings at a glance. Though the cleanroom exercise
focused on 0,3 μm particulates, the single display would
immediately alert the technician to anomalies in other
particle size concentrations.
With the Fluke 985, the technician can confi gure the
number of channels displayed for diff erent particle sizes
and also set an alarm alert via audio or highlighted text for
count over-concentration for the particular particle size of
concern.
Once the cleanroom is certifi ed, the technician can
also use the trend graph function of the instrument to
monitor particle concentration over specifi c time periods
they can defi ne. This can be used at the operational stage
of the cleanroom. This allows the technician to monitor
the manufacturing process if there is any unusual spike
of particle concentration and track down the event that
introduces contaminates to the process.
This also allows the technician to view the trend imme-
diately on the unit before data is exported, saving time on
data download. With the Fluke 985 USB/ Ethernet cradle,
the technician can quickly transfer the data to computer
via USB or over the network for immediate data analysis so
that corrective actions can be quickly deployed.
The delay function of the Fluke 985 is useful for
determining ‘at rest’ cleanroom particle concentrations
after facility occupants have exited the cleanroom and
the room’s ventilation system has fi ltered the air and the
environment has stabilised. Continued on page 14
![Page 16: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/16.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News14 feature Test & measurement, analysis, calibration & compliance
EXFO announced the addition of its iOLM
technology to its FTB series OTDR (opti-
cal time domain reflectometer) modules.
Originally developed for the challenging
FTTH/PON fibre test environment, iOLM
technology now extends its realm of fibre/
cable testing applications to more common
architectures, such as access, point-to-point,
FTTA/DAS, metro, CWDM and long-haul
networks.
Having iOLM on these OTDRs will enable
network operators to eliminate complex
OTDR trace interpretation and human errors
from their network testing cycles. Using an
automated multi-pulse acquisition approach
and advanced algorithms, the iOLM is an
OTDR-based software application that deliv-
ers detailed information on every element on
the link, in a single button operation.
Powered by Link-Aware technology, the
EXFO proprietary iOLM self-manages the
setting of all test parameters, requiring
virtually no training for technicians of any
skill level. Now available on new FTB-7300
and FTB-7400 OTDRs, the application can be
combined with various advanced test func-
tionalities – CD, PMD, WDM-Aware, Ether-
SAM, SONET/SDH, Ethernet, etc. – inside
EXFO’s FTB Ecosystem platforms, including
the popular FTB-200 Compact platform.
Since its first release about a year ago,
EXFO’s iOLM has been well received by
the market, as it is already being used in
60 countries around the world and cur-
rently undergoing numerous field trials with
major operators. EXFO intends to develop
additional targeted features for the iOLM to
address the needs of even more customers in
multiple applications.
For more information contact Chris Nel,
Lambda Test Equipment, +27 (0)12 349 1341.
EXFO enhances OTDR modules
Continued from page 13
Concentrations (C1) AVE. Concentration
Location (L) 1 2 3 4 5 6 (AC1)
A 750 560 655 730 674
B 1575 1250 750 950 1100 1300 1154
C 1300 850 980 1125 1350 975 1097
D 1150 775 450 825 845 1000 841
E 825 855 730 940 695 925 828
F 1700 1585 1135 900 1725 1210 1376
Upper control limit (UCL) factor for 95% confi dence
# Locations 1 2 3 4 5 6 7 8 8+
95% UCL factor 6,31 2,92 2,35 2,13 2,02 1,94 1,9 1,86 NA
UCL=M + (UCL Factor * SE)
1,087 = 995 + (1,94 * 47,36)
Table 1. ISO particulate limits.
Table 2. Example concentration levels.
Table 3. Calculating the upper confi dence limit.
ISO Class particulate limits
Classifi cation 0,1 μm 0 0,2 μm 0,3 μm 0,5 μm 1,0 μm 5,0 μm
m3 m3 m3 m3 m3 m3
1 10 2
2 100 24 10 4
3 1000 237 102 35 8
4 10000 2370 1020 352 83
5 100000 23700 10200 3520 832 29
6 1000000 237000 102000 35200 8320 293
7 352000 83200 2930
8 3520000 832000 29300
9 35200000 8320000 293000
Particle counting in perspective
The key to a successful IAQ investigation is
to be aware of the environment as a whole.
Location, building history, complaints, process
records, and measurable factors such as tem-
perature and humidity, pressure and particle
concentration all play a role in uncovering IAQ
problems.
When using a particle counter, be aware
that a particle’s source may only be a symp-
tom of a much larger issue looming under the
surface. Remediation of the source may not
address core problems of poor filtration, ven-
tilation or excessive moisture. Left unchecked,
these conditions will cause the same symp-
toms, or worse, to reoccur with certainty.
Particle counting as part of ongoing
preventive maintenance is critical to ensure
a healthy environment and good production
yield. The Fluke 985 is a powerful, rugged,
and easy to use tool to assist the techni-
cian in identifying particulate problems and
authenticating the efforts to address their
root causes.
For more information contact Comtest,
+27 (0)11 608 8520.
![Page 17: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/17.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 15Test & measurement, analysis, calibration & compliance feature
Newsteo recognises that some customers
are likely to have continuing and evolving
needs for product capability and adaptation
requirements. Newsteo has strong experi-
ence in radio transmission and sensors
interfacing, and is willing to supply exper-
tise as needed.
The Newsteo Logger LGS allows users
to connect any potentiometric sensor and
record the measurement or transmit it by
radio to the supervision centre. The installa-
tion and maintenance of the sensor are thus
simplified. The logger is available in two
configurations, one with a potentiometric
Potentiometric sensor loggersensor input and the second with three
inputs.
The units are made versatile thanks to
the capability to read a number of sensors.
These sensors include crack meters, pressure
sensors, photoresistance sensors and poten-
tiometer for angular measurement, to name
but a few. In areas where the supervision
centre is out of reach of the radio transmit-
ter, Newsteo also has a repeater with an
external antenna.
For more information contact Conical
Technologies, +27 (0)12 345 1545.
Unlike standard cor-
rugated test leads
that experience rapid
failures due to kinking
and connector/cable
interface breakage, the
SilverLine-LP is steel-
armoured and has a
massive, solid metal
backshell to protect
the cable-to-connec-
tor interface against
virtually all possibili-
ties for damage. The
robust design results
in much longer prod-
uct life and reduces
the occurrence of faulty test results.
SilverLine PIM is ideal for use with portable
PIM testers in fi eld test applications and for use
with benchtop PIM analysers in a lab or factory
environment.
In the fi eld this test cable represents a de-
pendable test lead, eliminating the need to re-
build or re-terminate a test lead on site or cancel
a test entirely. In the factory it saves labour by
providing more accurate and consistent results
over a far longer product life, thus reducing the
PIM testing cableWebb Industries has added Times Microwave’s SilverLine-LP, a test cable specifi cally
designed for fi eld and production passive intermodulation (PIM) testing, to its
product off ering.
possibility of rejects caused by faulty test leads.
Other features and benefi ts include: better
than -117 dBm (-160 dBc) performance; a set
of low PIM adaptors with connector saver (op-
tional); and a retractable, OneTurn connector for
fast-on, fast-off mating. The SilverLine-LP test
cable is easier to handle than corrugated cable,
is water-resistant and RoHS compliant.
For more information contact Lodie Potgieter,
Webb Industries, +27 (0)11 719 0000.
Model 7131A from Measurement Specialties is
a miniature triaxial IEPE accelerometer specifi -
cally designed for NVH, engine vibration, road
response, modal analysis, structural testing, ride
handling and fl ight testing.
Available in ±50 G to ±2000 G dynamic
ranges, the rugged device features a hermeti-
cally sealed titanium construction with a 4-pin
Microtech connector interface. It incorporates
stable piezoceramic crystals in annular shear
mode which provide a fl at frequency response
up to >10 kHz.
The accelerometer features stable operating
performance over an operating range of -55°C
through +125°C with linearity of ±1% (FSO).
Operating from 18 to 30 V d.c., it provides a
±5 V d.c. output with an 8 – 12 V d.c. bias volt-
age. Excitation current is 2-10 mA. The 7131A
is case grounded and adhesively mounted for
ease of installation.
Each unit is shipped with a NIST traceable
amplitude calibration certifi cate. A 4-pin 1/4-28
to 3x BNC cable assembly (315) is off ered as an
accessory, as is the Model 161A 4-channel PE &
IEPE signal conditioner.
For more information contact Accutronics,
+27 (0)11 781 2645.
Triaxial accelerometer
+27 (0)11 781 2645.
![Page 18: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/18.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature16 feature Test & measurement, analysis, calibration & compliance
The target of cost optimisation means that
standard commodity silicon processes with no
special manufacturing steps must be used. This
leads to a paradox: how to use an ASIC made
entirely of elements whose critical parameters
vary by several hundred ppm /°C to make mea-
surements with a precision of 1% to 2% over all
operating conditions.
The target is made even more challenging
when the need for a very fast response time is
added. This article describes a new
LEM ASIC which achieves this objec-
tive while at the same time adding
extra functionality.
Technology and quality
evolution
In many current measurement
applications, it is advantageous that
the measuring circuits are isolated
from the measured current, and in
these cases the magnetic fi eld due to
the current can be concentrated at a
sensor to which it is sensitive. In an
ASIC, a Hall cell may be used to trans-
form the magnetic fi eld into a voltage.
A Hall cell, however, is not only a particularly
unstable element, but additionally its properties
are usually not guaranteed in an ASIC manufac-
turing process.
Current transducer architectures have been
developed which are insensitive to the stabil-
ity and linearity of the magnetic sensor, for
example the closed-loop architecture shown in
Figure 1a.
However, such an architecture is still sensi-
tive to off sets in the sensor, and in any case
the ultimate goal for cost and reliability is to
develop an open-loop transducer containing
Current transducers benefi t from ASIC advances David Jobling, LEM.
Optimisation of cost and reduction of size are driving the need to integrate the components in current transducers using
application-specifi c integrated circuits (ASICs).
Figure 1a. Closed-loop current transducer architecture. Figure 1b. Open-loop current transducer architecture.
no electrical components except the ASIC – see
Figure 1b – while having the performance of a
closed-loop architecture.
Fully integrated ASICs for open-loop current
transducers such as LEM’s HXS series have been
developed in the past. The performance is sat-
isfactory, with typical specifi cation parameters
as follows:
• Gain drift: 350 ppm/°C;
• Input related off set drift: 2 μT/°C;
• Response time: 5 μs;
• Output noise density: 7 μV/rtHz with a
sensitivity of 40 V/T in a fi xed bandwidth;
• Current consumption: about 15 mA.
These ‘fi rst generation’ ASICs have a rigid
architecture where only gain (sensitivity) is
adjustable and is stored in a memory which can
only be written once, which limits the fl exibility
of their use.
Advances in silicon manufacturing tech-
nology and also in quality standards give a
serendipitous combination of process changes
that permit the design of a ‘second generation’
family of ASICs. These advances are:
• Analog elements which are fast and dense;
• Digital elements which are dense enough to
add useful functionality within a limited cost
budget;
• EEPROM memory which can be rewritten as
many times as needed;
• The three above features are available in
processes such as the 180 nm and 350 nm
nodes which are somewhat behind
the state of the art, and so are at a
reasonable cost;
• There is also an apparently unre
lated development, which is that
quality standards such as the auto
motive AEC-Q100 require testing
at 2 or 3 temperatures.
Of course these factors are
accompanied by some disadvan-
tages. The need for both high speed
and low noise increases power
consumption. The best transistors
are 3,3 V devices, but the market
requires operation in a 5 V environ-
ment.
However, putting these develop-
ments together has allowed LEM to design an
ASIC in which the company targeted at least
a twofold improvement in the critical param-
eters of the fi rst generation listed above. At
the same time the company has been able to
add features such as an over-current detection
(OCD) output, and the fi nal circuit confi guration
may be chosen at any time in the process, even
by the end user of the current transducer using
the ASIC.
The result is a cost-eff ective Hall-cell based
ASIC with fast response time and low tempera-
![Page 19: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/19.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
feature 17featureTest & measurement, analysis, calibration & compliance
Figure 3a. Gain error validation measurements. Figure 3b. Input related off set validation measurements.
Figure 4. ASIC die showing eight Hall cells
in the centre.
Figure 2. Block diagram of LEM’s new ASIC.
ture drifts. Figure 2 shows a block diagram of
the ASIC. To minimise the eff ect of off sets in the
Hall cells, conventional spinning techniques are
used with an output at 1,5 MHz and the eight
cells are placed in an optimised symmetrical
arrangement .
Using eight cells improves the
signal-to-noise ratio of the signal at
the input to the electrical chain to a
level similar to that of dedicated Hall
cells made with III-V compounds
such as Gallium Arsenide, though
the low signal level at this point
means that the following amplifi er
has to have a low noise level. It is
physically large and consumes a
high current: about 8 mA for the Hall
cell and amplifi er combination.
The signal is amplifi ed with high
gain-bandwidth amplifi ers and is
fi ltered before demodulation to
limit the noise which is folded back
into baseband. The fi lter uses the
same elements as the oscillator which sets
the spinning frequency, and so can have quite
a narrow bandwidth despite the process
variations.
The OCD block input is taken before the
demodulator and fi nal output amplifi er / fi lter
which has two advantages: it accommodates
signals outside the linear operating range of
the output, and it has a fast response which is
independent of the output fi lter bandwidth.
The latter may be chosen for the best
compromise between the response time of the
signal channel and the noise level at the output.
A timer block at the OCD output allows a hold
time of up to 1 ms, if needed, to ensure that
a short overload may still be detected by an
external microcontroller.
In order to meet a quality commitment at
the ppm level, testing of the packaged ASIC is
done at cold and/or hot before the fi nal test at
ambient temperature. This allows drifts of gain
and off set to be measured and an opposite cor-
rection to be stored in EEPROM.
These data are combined with the output of
an on-chip thermometer and digital-to-analog
converters to generate analog corrections of
the unavoidable gain and off set drifts in a stan-
dard silicon process. Indeed, several corrections
may be stored for diff erent ASIC confi gurations
and correction automatically applied for the
confi guration chosen.
If the Vref pin of the ASIC is forced to the
supply voltage, the output pin becomes the I/O
port of a single-wire bus interface. This means
that the fi nal ASIC confi guration may be de-
cided at any time, even in the fi nal application
of the LEM current transducer.
For example, two users may both want to
measure currents in the same range. For one a
fast response time may be the most important
parameter while the other prefers low output
noise. Each may choose the output fi lter band-
width by confi guring the ASIC according to their
preference.
The magnetic sensitivity may be
chosen between 5 and 200 V/T with
0,5% precision by sharing the 12 bits of
data between exponential steps of 15%
in the coarse gain adjustment and 32
steps of 0,5% in the linear adjustment.
The real challenge in design is to
improve all the critical parameters,
since generally they impose opposite
constraints. The measured specifi ca-
tion parameters of the new LEM ASIC
are given below:
• Gain drift: typical 100, maximum
200 ppm/°C;
• Input related off set drift: typical 0,3,
maximum 1,0 uT/°C;
• Response time: < 2 μs;
• Output noise density: 6 μV/rtHz with a sensi-
tivity of 40 V/T in a selectable bandwidth;
• Current consumption: 19 mA;
• Die area: slightly smaller than the fi rst-genera-
tion equivalent.
Measured parameters from LEM’s standard
validation setup which checks 9 circuits at a
time are shown in Figures 3a and 3b. A die
photograph is shown in Figure 4 – the eight Hall
cells can be seen at the centre of the die.
Conclusion.
The new LEM ASIC described meets the target
of improved precision for open-loop, Hall-eff ect
based current transducers. Looking forward,
the company believes that more improvements
are possible, both by adding more features and
by using new design techniques to achieve
the same level of precision while reducing the
calibrations needed.
For more information contact
Denver Technical Products,
+27 (0)11 626 2023
![Page 20: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/20.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature18 feature Audio & video electronics
High-quality audio applications using the PIC32 Jayanth Murthy Madapura, Microchip Technology.
In a typical docking station and device acces-
sory, a digital audio source that plugs into the
unit sends a serial stereo audio data stream
over the dock’s data transfer interface, such as
USB. The dock captures the data stream while
performing other crucial tasks, and routes the
stream to an audio codec or digital-to-analog
converter (DAC) at a specifi c sample rate, as
shown in Figure 1.
The captured stereo audio stream then
fl ows through a serial interface to the codec
or DAC. Since there
are many possible
sources of digitised
audio, and not all of
the sources use the
same sampling rate,
this serial interface
typically adapts the
sampling frequency to
the source, or converts
the sampled data
stream into a common
data rate.
Therefore, one of
the challenges in the
design of the docking
system or device acces-
sory is to perform the
sample-rate conversion
without degrading the
audio quality, and at
the lowest cost pos-
sible.
To deal with these
challenges, designers
have typically used a
dedicated sample-rate
conversion circuit and/
or a high-end audio
DAC that incorporates
Sample
rate
(kHz) Codec master clock/Time base with diff erent oversampling (MHz) Bit clock (MHz)
fs 128 fs 192 fs 256 fs 384 fs 512 fs 768 fs 1152 fs 32 fs 64 fs
32,0 — — 8,1920 12,2880 16,3840 24,5760 36,8640 1,024 2,0480
44,1 — — 11,2896 16,9344 22,5792 33,8688 — 1,4112 2,8224
48,0 — — 12,2880 18,4320 24,5760 36,8640 — 1,536 3,0720
88,2 11,2896 16,9344 22,5792 33,8688 — — — 2,8224 5,6448
96,0 12,2880 18,4320 24,5760 36,8640 — — — 3,072 6,1440
176,4 22,5792 33,8688 — — — — — 5,6448 11,2896
192,0 24,5760 36,8640 — — — — — 6,144 12,2880
Table 1. Typical master clocks and bit clocks required by the codec/DAC/ADC for diff erent audio sample rates.
When designing audio docking stations and accessories for portable digital audio devices, and other digital audio
sources, designers are constrained by cost, while trying to deliver the highest-quality audio playback.
Figure 1. Typical audio application.
Figure 2. I2S format.
complex phase-locked loops (PLLs) to ensure
fl exible sample rates for stable communication
of the sampled audio data.
The USB interface is a convenient interface
for the transfer of audio data. However, to
meet the requirements of professional audio,
the subtle loss of quality due to USB clock and
codec clock mismatch must be addressed.
In this article, the available audio-specifi c
features of the PIC32 MCU are explored to
address these needs. The SPI module supports
diff erent standard audio communication modes
and off ers high bit resolution for high-quality
audio applications.
The fl exible reference clock output feature
of the PIC32 can be used to provide the master
clock to the analog front-end to generate the
diff erent sample rates. The reference clock
output also eliminates the need for an external
crystal/oscillator by a codec.
It also eliminates the need for a PLL on
the codec. The reference clock output can be
tuned to prevent buff er
under-run and over-run
that arise because of
clock mismatches. The
MCU also off ers the USB
Host and Device module
with fl exible PLL clocking
schemes at low power.
Digital audio data
basics
When analog audio is
converted into a discrete
digital format, the analog
signal is sampled at a
frequency of at least twice
the highest frequency
component in the analog
signal, or the Nyquist rate.
Therefore, an audio signal
that spans 0 to 20 kHz
can be sampled at a data
rate of 44,1 kHz, which
in this case is the suit-
able Nyquist rate, so that
the signal can be recon-
structed without aliasing
when converted back to
the analog domain.
In addition to the
![Page 21: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/21.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
feature 19featureAudio & video electronics
Sample Bit clock (MHz) SPIxBRG MCLK rate 32 fs 64 fs 32 fs 64 fs (MHz) (kHz) (16-bit) (24-bit) (16-bit) (24-bit)
11,2896 44,1 1,4112 2,8224 4 2
12,2880 48 1,536 3,072 4 2
32 1,024 2,048 6 3
8,1920 32 1,024 2,048 4 2
Table 2. Baud rate values when the PIC32 SPI
is confi gured as a master.
Table 3. Comparison of master clock
generated using the codec PLL and reference
clock output on a PIC32 device.
8,192 8,192 8,192
11,2896 11,2889025 11,28893
12,288 12,288 12,288
Expectedmaster
clock (MHz)
Average masterclock using
external crystal
(MHz)
Average masterclock using
REFCLKO (MHz)
Table 4. Divisor and trimming values for diff erent master clocks.
8,1920 5,859375 5 440 8,1920 0
11,2896 4,251701 4 128 11,28893 0,005937
12,2880 3,90625 3 464 12,2880 0
16,3840 2,929688 2 476 16,3840 0
16,9344 2,834467 2 427 16,93728 0,01703
18,4320 2,604167 2 309 18,43661 0,02501
22,5792 2,12585 2 64 22,58824 0,04002
24,5760 1,953125 1 488 24,5760 0
33,8688 1,417234 1 213 33,85124 0,051848
36,8640 1,302083 1 154 36,84558 0,049975
MCLK (MHz)
Divisor with
USB-PLL of
96 MHz as clock
source
RODIV
(N)
ROTRIM
(M/512)
Observed MCLK
(MHz) % Deviation
Figure 3. Left-justifi ed format.
Figure 4. Right-justifi ed format.
Figure 5. DSP/PCM format.
sampling rate, data bit resolution can be 16-
bit or 24-bit stereo audio data. For compact
disc (CD) quality audio, the standard is 16-bit
resolution with a 44,1 kHz sample rate. How-
ever, there are higher-performance CD music
options. One such standard encodes the data
with a 24-bit resolution and increases the
sampling rate to 96 kHz.
For professional audio, the audio fi les
are encoded with a resolution of 24 bits per
sample, which provides headroom when the
audio is mixed and manipulated. Also, the
resolution choice allows for the trade-off
of sound quality versus fi le size, even with
compression.
The USB interface can readily handle
the streaming of high-quality audio over
isochronous transfers. Its ability to deliver
high-quality audio is quite evident, as it is
popular among many audio
users. With its universal
ease-of-use, USB audio can
transfer high-resolution and
high-sample-rate audio with
negligible jitter, when pack-
aged with a fl exible audio
interface.
Isochronous data transfer,
amongst its various other
uses, is utilised to stream
audio data to and from a
source at a constant rate in
real-time. Stereo audio data
packets, with size governed
by the sample rate of the
audio stream, are transferred
as part of USB frames every
1 ms on the USB full-speed
link. USB audio also provides
controls for common features
such as volume, tone, gain
control and equalisers,
among many control and
processing units.
The diff erences in bit rates
and sample rates require the
hardware in the playback
system, or dock, to be able
to handle the diff ering rate
data streams. To do that,
the system must either use a
more complex DAC that is expensive and can
phase-lock to each sample rate and adjust
itself to each playback option, or use an
external sample-rate converter IC with the
low-cost DAC, or convert all the streams into
a standard sample rate and bit rate using an
algorithm running on a microcontroller that
a simple low-cost DAC can handle.
The PIC32 MCU off ers a fl exible reference
clock output and audio mode to address
these requirements to achieve high-quality
audio while maintain a low design cost. The
serial interface with the Audio mode and
fl exible reference clock output module are
explored in the following sections.
PIC32 serial peripheral interface
module with Audio mode
Most codecs off er serial communication over
a 4-line serial interface. The transmission and
reception of the stereo audio samples between
the MCU and codec occur over this serial inter-
face. Typical serial interfaces have the following
signals:
• Serial Data Output (SDO) to transmit stereo
audio data to the codec.
• Serial Data Input (SDI) to receive stereo audio
data from the codec.
• Serial Bit Clock (SCK/BCLK) is the required bit
clock provided by the Master.
• Left/Right Clock (LRCK) is the phase clock
provided by the Master for stereo data.
PIC32 devices have a serial peripheral inter-
face (SPI) module with Audio mode.
Audio mode off ers various
interface formats, bit resolu-
tions and Master/Slave con-
fi gurations. The communica-
tion modes supported include
the following:
I2S format
I2S mode is where the most
signifi cant byte (MSB) is avail-
able on the second rising edge
of the BCLK following an LRCK
transition (Figure 2).
Left-justifi ed format
Left-justifi ed mode is where
the MSB is available on the fi rst
rising edge of BCLK following
an LRCK transition (Figure 3).
Right-justifi ed format
Right-justifi ed mode is where
the least signifi cant byte (LSB)
is available on the rising edge
of the BCLK preceding an LRCK
transition and the MSB is still
transmitted fi rst (Figure 4).
DSP/PCM format
In DSP/PCM mode, the left-
channel MSB is available on
![Page 22: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/22.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature20 feature Audio & video electronics
Table 5. Observed THD and SNR using codec PLL and PIC32 reference clock output – sample rate of 32 kHz, expected master clock of 8,192 MHz.
fs = 32 kHz Using external crystal as time base Using reference clock output as time base
Average Average
observed observed
Expected frequency frequency
frequency (Hz) (Hz) THD (%) SINAD (dB) (Hz) THD (%) SINAD (dB)
500 499,9649 0,002 88,62 499,9668 0,00224 88,476
1000 999,9318 0,00224 88,752 999,9342 0,002 88,752
1500 1499,895 0,00208 88,308 1499,901 0,00192 88,308
10000 9999,318 0,002 82,212 9999,335 0,002 82,212
the fi rst rising edge of the BCLK following a
rising edge of the LRC (Figure 5). Right-channel
data immediately follows left-channel data.
Depending on word length, the BCLK frequency
and sample rate, there may be unused BCLK
cycles between the LSB of the right channel
data and the next sample.
PIC32 Audio mode supports 16-, 24- and
32-bit stereo audio data. It also provides
advanced error handling with receive over-
fl ow and transmit under-run fl ags
and control bits to disable them as
required. Also, the clocks gener-
ated by the BCLK and the LRC when
the PIC32 MCU is a serial Master are
free-running, which is essential for
uninterrupted audio data transfer in
streaming applications.
The audio codecs and DACs allow
either or both of these confi gura-
tions: Master or Slave. In the PIC32
SPI Audio mode Master confi guration,
the MCU provides the BCLK and the
LRCK to the codec. This confi guration
is shown in Figure 6. In the PIC32 SPI
Audio mode Slave confi guration, the
codec provides the bit clock (BCLK)
and stereo phase clock (LRCK) to the
MCU. This confi guration is shown in
Figure 7.
PIC32 devices also provide a
control interface channel over the
Inter-Integrated Circuit (I2C) peripheral
module. The control interface is used
to confi gure the codec for a specifi c
mode of operation by confi guring the
control registers on the codecs.
Flexible reference clock output
The PIC32 devices have a fl exible
reference clock output. The reference
clock output module (REFCLKO) can
be used to generate the fractional
clock that can be used by audio
codec/DACs to accommodate vari-
ous sample rates. Typical examples of these
sample rates and associated master clocks are
shown in Table 1.
The REFCLKO module can be used to gener-
ate these audio-specifi c Master clocks and is
not limited to just these. The reference clock
Figure 6. PIC32 SPI-codec confi guration with PIC32 SPI as master
and the codec as slave.
Figure 7. PIC32 SPI-codec confi guration with PIC32 SPI as slave
and the codec as master.
Figure 8. SPI and codec confi guration showing the reference
clock output signal.
can be mapped to any port pin on the device
if the Peripheral Pin Select (PPS) functional-
ity permits on the specifi c device. REFCLKO
can be confi gured to the source clock for the
SPI module on the PIC32 MCU instead of the
peripheral bus clock.
This ensures synchronisation of the USB
clock with the SPI channel clock to minimise
eff ects of clock jitter when the USB PLL clock is
selected as the source clock for the REFCLKO.
The stereo word select signal and the bit clock
will be synchronised with the reference clock
output since all of them are sourced off the
reference clock output.
The clock source for the reference clock
output module has various options. It can be
selected from USB-PLL, Primary PLL, POSC,
FRC, SOSC, LPRC, PBCLK, SYSCLK clocks.
Based on the selected clock source, a frac-
tional divisor needs to be confi gured for the
REFCLKO to generate the needed MCLK based
on the oversampling rate as noted in Table 2.
The fractional divisor needed is confi g-
ured as an integer part and fractional part
determined by RODIV and REFOTRIM bits in
the REFOCON register, respectively. Figure 9
shows the PIC32 oscillator module
with the reference clock output indi-
cating the diff erent clock sources for
the reference clock output and the
divider with the trimming scheme.
When the PIC32 SPI is the Master,
the BCLK and the LRCK are gener-
ated by the device. With REFCLKO
as the clock source for the SPI
module, the BCLK can be confi gured
by loading the appropriate values in
the SPI baud rate generator register,
SPIxBRG. If needed, the REFCLKO
can provide the master clock input
to the codec.
As shown in Table 2, the SPIxBRG
register values and the required
REFCLKO frequencies for the audio
sample rates are 32/44,1/48 kHz for
stereo audio data with both 16-bit
and 24-bit resolution. The bit clock,
BCLK, can be 32 fs or 64 fs. When
the PIC32 SPI is the Slave, the BCLK
and the LRCK are generated by the
codec. The reference clock output
can provide the master clock for the
codec/DAC.
Example implementation
Let us explore an example of
generating a REFCLKO signal. If
the required MCLK is 12,288 MHz
and if the USB-PLL clock of 96 MHz
is used as source for the reference
clock output, the required divisor is
7,8125. The divisor needs to be programmed
for a half period of 3,90625, which is
(7,8125/2). The RODIV bits will be confi gured
with a value of 3 and ROTRIM confi gured for
464/512 = 0,90625.
Table 4 provides the required RODIV and
![Page 23: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/23.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
feature 21featureAudio & video electronics
Table 6. Observed THD and SNR using codec PLL and PIC32 reference clock output – sample rate of 44,1 kHz, expected master clock
of 11,2896 MHz.
fs = 44,1 kHz Using external crystal as time base Using reference clock output as time base
Average Average
observed observed
Expected frequency frequency
frequency (Hz) (Hz) THD (%) SINAD (dB) (Hz) THD (%) SINAD (dB)
500 499,9649 0,00208 88,632 499,9668 0,00168 88,62
1000 999,9318 0,00184 88,332 999,9342 0,002 88,548
1500 1499,895 0,00168 88,38 1499,901 0,00176 88,212
10000 9999,318 0,00192 83,148 9999,335 0,00208 82,968
Example 1. RODIV and ROTRIM values for REFCLKO of
12,288 MHz.
Figure 9. Block diagram of the PIC32 oscillator showing the reference
clock output.
Figure 10. Clock jitter DAC underrun or overrun manifesting as subtle
audible clicks.
ROTRIM values needed for the
standard audio MCLKs previously
listed in Table 1. The clock source
selected here is the USB-PLL
clock. However, any other clock
source can be selected and the
RODIV and ROTRIM values can be
calculated in a similar manner.
The reference clock output can
be changed with a fi ne precision.
The unit resolution is between
approximately 0,02% and 0,05%
based on the generated clock
frequency. This ensures that the
sample rate deviation is well
below the limits required by the
codec when the reference clock
output is only being used as the
SPI clock source and not as the
master clock to the codec. If the
reference clock output is used as
the master clock for the codec, the
deviation is only limited by the
specifi cation of the codec.
Table 3 shows an example of
an audio master clock generated
by an expensive codec with fl ex-
ible internal PLL using a standard
external crystal of 12 MHz and the
master clock using reference clock
output. The data indicate that the
master clock generated by the
PIC32 reference clock output is
very close to the required master
clock.
Also, the reference clock out-
put performs as well as or better
than the master clock generated
using the internal PLL on an ex-
pensive codec. This would elimi-
nate the need for a PLL on the
codec or DAC used as an analog
front end and eliminate the need
for an external crystal used as time
base for the codec. This will result
in cost savings in the audio design
while providing the same or better
quality than an expensive codec.
Table 5, Table 6 and Table 7 show examples
of the signal quality metrics, SNR and THD, for
a set of (pure) audio tones generated using
external crystal with internal PLL on the codec
and using the reference clock output as master
clock on the same codec.
The SNR metric or signal-to-noise ratio
measures the power of the noise
induced compared to the signal.
The higher the SNR metric the
better is the quality of the signal
produced.
The THD metric or the total
harmonic distortion measures
the strength of harmonic distor-
tion caused by the harmonics
of the fundamental frequency.
The lower the THD metric the
better is the quality of the signal
produced.
The SNR and THD data indi-
cates that the quality of audio
signal generated by an AFE with
reference clock output as the
time base performs exception-
ally well compared to that of
a very expensive codec with
internal PLL.
Tuning reference clock
output
The reference clock output has
the ability to be tuned on-the-
fl y, and can be tuned in steps
between a specifi ed range. The
range should be such that it
ensures a swing of the sample
rate, typically about ±0,2%,
which is well below a range
that might introduce audible
artefacts. For example, a swing
of ±0,2% of the data stream
with a sample rate of 48 kHz
requires a tuning of the refer-
ence clock output between the
ranges of 12 263 424 Hz and
12 312 576 Hz.
Table 8 lists the required
RODIV and ROTRIM values for
tuning the sample rate within
±200 Hz of the sample rate. The
required master clock column
indicates that the sample rate
needs to be tuned between the
range of 47,9 kHz and 48,1 kHz, ROTRIM needs
to be tuned between 459 and 468 and the
required RODIV is 3. Similar tuning ranges for
RODIV and ROTRIM to tune the reference clock
![Page 24: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/24.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature22
Table 8. REFCLKO tuning example with RODIV and ROTRIM values for 48 kHz audio stream.
47800 12236800 3,922594 3 472 48010 12290560 3,905436 3 463
47810 12239360 3,921774 3 471 48020 12293120 3,904623 3 463
47820 12241920 3,920954 3 471 48030 12295680 3,90381 3 462
47830 12244480 3,920134 3 471 48040 12298240 3,902998 3 462
47840 12247040 3,919314 3 470 48050 12300800 3,902185 3 461
47850 12249600 3,918495 3 470 48060 12303360 3,901373 3 461
47860 12252160 3,917677 3 469 48070 12305920 3,900562 3 461
47870 12254720 3,916858 3 469 48080 12308480 3,89975 3 460
47880 12257280 3,91604 3 469 48090 12311040 3,898939 3 460
47890 12259840 3,915222 3 468 48100 12313600 3,898129 3 459
47900 12262400 3,914405 3 468 48110 12316160 3,897319 3 459
47910 12264960 3,913588 3 467 48120 12318720 3,896509 3 459
47920 12267520 3.912771 3 467 48130 12321280 3,895699 3 458
47930 12270080 3.911955 3 466 48140 12323840 3,89489 3 458
47940 12272640 3,911139 3 466 48150 12326400 3,894081 3 457
47950 12275200 3,910323 3 466 48160 12328960 3,893272 3 457
47960 12277760 3,909508 3 465 48170 12331520 3,892464 3 456
47970 12280320 3,908693 3 465 48180 12334080 3,891656 3 456
47980 12282880 3,907878 3 464 48190 12336640 3,890849 3 456
47990 12285440 3.907064 3 464 48200 12339200 3,890041 3 455
48000 12288000 3,90625 3 464
Legend: Shaded cells indicate Sample Rate Tuning Range values
LRCK
(Hz)
LRCK
(Hz)
MCLK
(Hz)
MCLK
(Hz) Divisor DivisorRODIV RODIVROTRIM ROTRIM
feature Audio & video electronics
fs = 48 kHz Using external crystal as time base Using reference clock output as time base
Average Average
observed observed
Expected frequency frequency
Frequency (Hz) (Hz) THD (%) SINAD (dB) (Hz) THD (%) SINAD (dB)
500 499,9959 0,002 88,62 499,9965 0,00224 88,476
1000 999,9938 0,00224 88,368 999,9932 0,002 88,752
1500 1499,989 0,00208 88,332 1499,997 0,00192 88,308
10000 9999,915 0,002 82,164 9999,934 0,002 82,212
Table 7. Observed THD and SNR using codec PLL and PIC32 reference clock output – sample rate of 48 kHz, expected master clock of 12,288 MHz.
output can be determined for other standard
audio sample rates.
The tuning capability of REFCLKO prevents
buff er under-run and over-run and alleviates the
audible clicks, as discussed in the next section.
USB clock mismatch
USB specifi cations require a tolerance budget
and a limit on the USB clock frequency as a way
to achieve immunity to radio interference. The
USB clock with the allowed tolerance budget
results in reduced audio quality if there is a USB
clock mismatch.
The real-time streaming audio samples must
arrive at precise, regular time intervals so that
the DAC can convert the digital samples to an
analog signal with the expected constant rate at
which it is confi gured.
The DAC clock that expects and receives
the audio samples at a particular sample rate
cannot miss even a single sample. A missing
sample manifests as a subtle click for the listen-
er, since the DAC fails to generate an accurate
representation of the streamed audio signal.
On a microcontroller or microprocessor with
an embedded USB module, the USB clock is
sourced from an independent clock such as an
on-chip PLL with an external crystal oscillator
of specifi c value. Since the clock is not sourced
from the USB interface, the mismatch in clocks
introduces buff er over-run or under-run, caus-
ing audible clicks as shown in Figure 10.
An easy solution for the audio data under-
run or over-run issue as related to audio-quali-
ty degradation, is to use a good asynchronous
sample rate converter (ASRC), where the input
sample rate is estimated with jitter attenuation,
and the internal fi lters are dynamically tuned
for a new sample rate. However, a good ASRC
is very expensive and the system still requires a
DAC for analog conversion.
As an eff ective low-cost solution, the USB
audio packets are buff ered and the clocks of
the codec or DAC can be tuned to prevent
under-run or over-run using a feedback mech-
anism. The feedback mechanism monitors
the buff er level and ensures it stays within an
acceptable range, while achieving at least the
same quality achieved by an expensive ASRC.
The reference clock output with USB-PLL
clock as its source can be used to generate and
tune the required master clock, as discussed
in the ‘Tuning reference clock output’ section
above. This capability prevents buff er under-run
and over-run while maintaining an acceptable
DAC sample rate with a swing range of 0,2%.
This is the lowest-cost solution compared to
the other two, while still achieving high-quality
audio.
Audio accessory board and application
demonstration
As an example, all of the features discussed in
this article are implemented in the PIC32 USB
Headset application project available for the
PIC32 USB Digital Audio Accessory Board.
For more information contact Tempe
Technologies, +27 (0)11 452 0530.
![Page 25: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/25.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
feature 23
The TPA3116D2, TPA3118D2
and TPA3130D2 feature
a very high program-
mable switching rate and
wide power supply range,
enabling audio output that
is both high-quality and
highly effi cient.
The devices are pin-to-
pin and software compat-
ible and can support a
single-layer printed circuit
board (PCB), enabling cost
savings by allowing designers to reuse a single
layout for multiple designs.
A supply range of 4,5 V to 26 V enables si-
multaneous operation of line power and battery
power on the same board for longer battery life
in soundbars and portable audio docks. Ther-
mally effi cient FETs allow 30 W x 2 operation
without the need for a heatsink, or 50 W x 2 with
a heatsink (TPA3116D2 only).
Programmable switching rates of 400 kHz to
1,2 MHz allow inductor fi lter-free operation and
interference avoidance for clear audio with no
distortions. The ICs’ closed-loop design provides
excellent power supply rejection ratio (PSRR)
Class-D audio amplifi ers
CML Microcircuits recently introduced the
DuraTALK family of digital voice products,
consisting of a wide range of fl exible
technologies supporting voice-data
generation, coding, transcoding, decod-
ing, storage and scrambling functions for
digital communication systems.
The CMX608, CMX618 and CMX638 ICs
are designed for RALCWI, a CML-owned,
low bit rate (2400 bps) vocoder technol-
ogy that facilitates the transmission and
reception of highly compressed voice
over inherently noisy narrowband radio
channels. RALCWI off ers fl exible, highly
integrated, high-performance vocoders,
with near toll-quality voice at very low bit
rates.
The CMX8341 digital PMR baseband
processor system-in-a-chip is aimed at
low-cost, licence-free digital PMR (dPMR)
radios conforming to the ETSI standard, TS
102 490, and analog PMR (legacy mode).
Seen as the digital successor to
PMR446, digital PMR is now a proven
technology targeting highly functional
solutions by using low-cost and low-com-
plexity technology. dPMR is a narrowband
(6,25 kHz) FDMA technology that is 100%
digital and off ers many forms of voice
and/or data applications. RALCWI in digi-
tal PMR will provide savings in cost, size
and power consumption.
For CVSD/PCM/ADM, fl exible coders
and decoders are available for analog-to-
digital and digital-to-analog voiceband
conversion. The CMX639 and CMX649
address this market segment, providing
robust and fl exible coding algorithms, low
cost, low power and small size, ideal for
use in a wide variety of consumer, military
and business digital voice applications.
The CMX7261 is a single-chip G.711/
G.729A/CVSD full duplex multi-transcoder
device, capable of encoding analog voice
into linear, μ-law or A-law PCM, CVSD
or G.729A data formats. It is capable of
decoding PCM, CVSD and G.729A back to
analog voice. It can also transcode data
between PCM, CVSD and G.729A.
The fl exible IC creates toll quality
speech for applications such as software-
defi ned radio (SDR), voice-over IP (VoIP),
wireless private branch exchange (PBX)
and privacy type digital voice communica-
tions.
For more information contact Avnet Kopp,
+27 (0)11 809 6100.
ICs for digital voice communication
A fast laser diode
from OSRAM Opto
Semiconductors has
been used to provide
light for a new 3D camera
from Swedish manu-
facturer Fotonic. The
component was devel-
oped especially for 3D
cameras and facilitates
range fi nding for moving
objects with good depth
resolution at video rate.
The infrared laser di-
ode provides high optical
power of about 0,5 W, even when in continu-
ous wave operation. With a wavelength of
about 845 nm the light is hardly visible to
the human eye, and yet it is well within the
sensibility range of the camera chip.
Three-dimensional cameras superimpose
a two-dimensional picture with information
on the range and surface profi le of an object.
Industrial applications for such systems are,
for example, the gauging of fl uid levels or the
grading of objects according to size or shape.
Three-dimensional sensors also provide
reliable stereoscopic information, e.g. for the
control of robot arms.
IR laser diode used in 3D camera
featureAudio & video electronics
Texas Instruments has introduced three new analog-input, class-D stereo amplifi ers
for use in soundbars, after-market automotive solutions, portable audio docks and
LCD televisions.
and slower edge rates for waveform switching
to improve electromagnetic interference (EMI)
performance. RDS(on) of 120 mΩ at 30 W x 2
improves effi ciency by decreasing resistance
during operation.
An evaluation module and SPICE model are
available for the TPA3116D2, TPA3118D2 and
TPA3130D2, allowing designers to evaluate
system-level performance and quickly integrate
the amplifi ers into product designs.
For more information contact Dirk Venter, Arrow
Altech Distribution, +27 (0)11 923 9600.
The Fotonic C70
camera is based on a
CMOS sensor and makes
use of the principle of
time-of-fl ight measure-
ment. A laser diode,
which is integrated into
the camera, transmits
a light pulse which the
target refl ects back. Each
of the 120x160 pixels of
the camera chip records
the distance to the object
by measuring a respective
phase shift compared to
the emitted light pulse.
The precision of this distance measurement
and the range of the sensor increases with
the modulating frequency of the light source.
Due to their fast switching times of only a few
nanoseconds, laser diodes are particularly well
suited for this application.
Within the range of 0,1 to 7 metres, the
camera measures distances with an accuracy
of 3 up 30 millimetres.
For more information contact Ryan Hunt,
OSRAM Opto Semiconductors,
+27 (0)79 525 1779.
![Page 26: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/26.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature24
Historically, testing of acoustic response in
audio playback devices has required criti-
cal performance analysis conducted by
specialised audio engineers. In addition
to design verifi cation, acoustic response
testing is also important in production
line testing of multimedia devices.
Currently, many multimedia devices
such as MP3 players, mobile phones and
car stereos utilise speakers. Accurate
frequency response measurement and
detailed harmonic distortion analysis
during speaker production testing require
high resolution dynamic signal acquisition
modules with fl exible analog input/output
function and high dynamic range.
AO output, for example, generating a
multi-tone or stepped sweep sine signal,
may be employed as a stimulus for a
speaker. To convert speaker output to a
voltage signal, a calibrated microphone
may be employed as a sensor. AI input can
then acquire the signal output from the
microphone and perform a fast Fourier
transform (FFT) to analyse the measured
audio signals, providing frequency and
impedance response and harmonic distor-
tion analysis.
The ADLINK PXI-9527 features a 2-
channel, 24-bit simultaneous sampling
analog input. The Sigma Delta ADC pro-
vides up to 432 KSps sampling rate and
high resolution, ideally suiting it for higher-
bandwidth dynamic signal measurement.
The sampling rate can be adjusted by
setting the onboard DDS clock source to
an appropriate frequency. All channels are
sampled simultaneously and accept inputs
in the range of ±40 V to ±0,316 V. The
analog input supports software-selectable
AC or DC connection, and 4 mA bias cur-
rent for integrated electronic piezoelectric
(IEPE) sensors.
For more information visit
www.adlinktech.com
Signal acquisition card for speaker testing
Housed in 3 x 3 mm packages, the devices
reduce board space compared to fully discrete
solutions and are ideal for space-constrained
applications, such as rear-view cameras.
The ADA4830-1 (single-channel) and
ADA4830-2 (dual-
channel) video dif-
ference amplifi ers,
and ADA4432-1
(single-ended
output) and
ADA4433-1 (dif-
ferential output)
analog video fi lter
amplifi ers, provide
designers with a
complete chipset
for transporting
analog video sig-
nals over shielded
and unshielded
twisted pair
cables.
The ICs can
also be used in-
dependently to provide receiver (ADA4830-1/2)
and transmitter (ADA4432-1/33-1) functions
as part of a video signal chain. They feature
integrated short-to-ground protection, fault
detection, wide common-mode voltage ranges
and robust ESD performance.
The ADA4830-1 and ADA4830-2 video
diff erence amplifi ers are designed for use as
receivers of CVBS (Composite Video Blanking
and Synchronous) signals in standard-defi nition
PAL/NTSC video applications. In addition to
Video amplifi ers
Analog Devices has introduced a 16-chan-
nel audio D/A converter (DAC) capable of
improving audio system performance and
reducing power consumption for profes-
sional, ‘prosumer’ and automotive audio
equipment applications.
The 24-bit ADAU1966 provides 118 dB
SNR (signal-to-noise ratio) performance at
32 kHz to 192 kHz sampling rates. This is
achieved with power consumption of less
than 300 mW for 16-channel operation by
running the digital portion of the chip at a
lower voltage.
The digital supply can be generated on-
chip via a linear regulator, allowing the IC
to operate off a single analog supply. Alter-
natively, the regulator can be bypassed and
a separate 2,5 V to 3,3 V supply provided to
the chip.
feature Audio & video electronics
Hi t i ll t ti f ti i
H
r
s
a
A
c
f
a
(
o
A
f
a
Operating the ADAU1966 with full-
scale diff erential output voltage of 3 Vrms
reduces the need for downstream gain and
preserves a high SNR against electrically
coupled system noise. By using the device’s
onboard PLL (phase-locked loop) to derive
the internal master clock from an external
left/right clock, the IC eliminates the need
for a separate high-frequency master clock
and can be used with or without a bit clock.
The ADAU1966 is qualifi ed for AEC-100
automotive applications and operates over
a -40°C to +105°C temperature range. It also
features an SPI/I2C port which allows an ex-
ternal microcontroller to adjust volume and
read the chip temperature to within ±3°C.
For more information contact Arrow Altech
Distribution, +27 (0)11 923 9600
16-channel audio DAC
Analog Devices has introduced a series of high-speed video amplifi ers that
integrate short-to-battery protection and fault detection for analog video
systems in automotive safety and infotainment applications.
integrated short-to-battery (input) protection up
to 18 V on a single 3 V to 5 V supply, the devices
include a short-to-battery fl ag that indicates the
presence of a fault condition at the input.
The ADA4830-1 and ADA4830-2 also have
ESD tolerance up
to ±8 kV Human
Body Model
(HBM), a wide
input common-
mode voltage
range that ex-
tends to -8,5 V be-
low ground and
55 dB common
mode rejection at
7 MHz.
The ADA4432-
1 (single-ended
output) and the
ADA4433-1 (dif-
ferential output)
analog video
fi lter amplifi ers
are designed
as transmitters for automotive analog video
fi ltering applications. They feature the same 18 V
short-to-battery protection (output), fault fl ags
and ±8 kV HBM tolerance as the ADA4830.
Additionally, they feature continuous output
short-to-ground protection and include a high-
order video fi lter with a -3 dB cutoff frequency at
10 MHz and 45 dB of rejection at 27 MHz.
For more information contact Avnet Kopp,
+27 (0)11 809 6100.
![Page 27: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/27.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 25featurefeatureAudio & video electronics
Silicon TV tuner ICs are rapidly displacing
legacy mixer oscillator phase-locked loop
(MOPLL) CAN tuner technologies to reduce
cost and size and improve performance.
Silicon tuner IC adoption began before 2007
and gained real traction during 2010 when
fl at-panel TVs and set-top boxes (STBs) saw a
signifi cant increase in sales.
Designing silicon TV tuners that matched
MOPLL performance levels was the main
hurdle for adoption, but once semiconductor
suppliers met this performance standard the
road was cleared for silicon tuner IC ship-
ments to accelerate.
Several suppliers now off er a wide variety
of silicon TV tuner products. Navigating the
competing options can be a complex process
given the wide range of issues surrounding
TV tuner IC selection. Putting these technical
issues into perspective will help simplify the
TV tuner selection process.
State of the market
TV market demand has grown to an amazing
230 million units (Mu) per year across a wide
range of regions. The broader TV market
breaks down into two segments relevant to
TV tuner ICs: integrated digital television
(iDTV) platforms (approximately 160 Mu)
and analog-only TVs (approximately 70 Mu),
which consist of both fl at-panel TVs and
analog-processed CRT TVs.
Analog-processed CRT TVs employ an
older image processing technology. In
contrast, fl at-panel TVs incorporate advanced
DSP-based image processor SoCs supplied
by companies such as Broadcom, MediaTek,
MStar, Novatech and Sunplus Technology.
Some major TV brands also have their own
internal semiconductor teams that build
captive fl at-panel SoCs.
Analog-processed CRT TVs use less
capable microcontrollers and cannot easily
accommodate silicon tuner IC programming
and confi guration. These analog-processed
TVs are likely to retain the older MOPLL
CAN tuner implementations, which often
require more than 150 separate components
(see Figure 1).
Flat-panel platforms employ modern
high-performance processors that run a
substantial amount of fi rmware so they are
well positioned to confi gure modern digital
devices such as silicon TV tuner ICs.
Many TVs must support both analog
and digital transmission standards to meet
consumer demands. Examples of analog
transmission standards include NTSC and
PAL/SECAM. In this case, the word ‘analog’
describes the transmission modulation
format and not the specifi c processing
technology used to decode it.
Many countries are in various stages of
executing their analog switch-over plans
where the terrestrial analog broadcasts
will be terminated and replaced with
higher-quality digital broadcasts. Yet even
in countries that have theoretically made
the hard cut over to digital, the analog
transmissions remain active, and consum-
ers continue to demand support for these
transmissions.
Japan is perhaps an exception, having
achieved a true conversion to all digital, but
TVs sold in virtually every other region are
forced to retain analog reception capability
to deal with low-power analog broadcasts,
slow-to-upgrade cable networks and legacy
consumer electronics devices.
Tuner support for analog transmissions
will be required by all major TV brands
for TVs outside of Japan for at least fi ve
more years and likely for ten more years.
Numerous TV platforms accept content
from both terrestrial and cable sources.
As a practical matter, these platforms must
support analog transmissions as long as
the multi-subscriber operators (MSOs)
continue to transmit analog content on their
cable networks.
Cable and terrestrial STB segments add
an additional 175 million systems of demand
per year. A recent forecast of silicon tuner
ICs from IMS Research (Figure 2) projects the
market growth for the overall number of TV
tuners for demand TV, STB, cable modem,
satellite receiver and other segments. Sev-
eral multipliers are applied to account for
multi-tuner IC applications, generating the
roughly 700 million units of overall tuner IC
market opportunity shown.
The rapid adoption of silicon tuner ICs is
evident from this data and is even stronger
when focusing on just the TV and cable STB
segments where silicon tuner ICs have made
deeper inroads.
Opportunities going forward
The high-growth markets of mainland China
and Taiwan are taking the lead in new silicon
TV tuner IC opportunities. TV makers in
Figure 1. Example of MOPLL CAN tuner, which requires 150+ components.
Figure 2. IMS Research forecast shows growing market demand for silicon tuners.
Silicon tuner ICs in mainstream television designs Information from Silicon Laboratories.
![Page 28: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/28.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
News26 feature Audio & video electronics
China have rapidly increased TV production
over the last three years, and cost-optimised
silicon TV tuner ICs help them deliver high-
quality TV models at competitive price points.
According to the analyst fi rm Display-
Search, China became the world’s largest
market for LCD TVs in 2011. DisplaySearch
also predicts that fl at-panel TV shipments
(both LCD and plasma models) in China
will nearly double from the 31 million units
reported in 2009 to 59 million units in 2014,
reaching a compound annual growth rate
of 14%.
Top-tier TV brands are accelerating their
adoption of silicon tuners to save
cost, improve reliability, reduce
component inventories and support
thinner TV form factors. These same
advantages are driving signifi cant
opportunities in the second- and
third-tier TV brands where the
adoption of silicon TV tuners is just
starting.
Because today’s silicon tuner ICs
must support all worldwide terres-
trial and cable broadcast standards,
these smaller TV brands benefi t
from the intensive work already
accomplished by the top brands.
Connected TVs expand the range
of program sources beyond just the
antenna and cable, and are gaining
signifi cant traction as ‘apps’ make
their way to large-screen TVs.
New communication interfaces
for TVs such as Wi-Fi, Ethernet and
LTE (cellular) are delivering rich,
diverse content beyond the tradi-
tional broadcast sources. These
interfaces present special technical
challenges for over-the-air (OTA)
broadcast tuners. For example, a special Wi-Fi
immunity circuit is needed to protect against
this interference.
Silicon TV tuner selection challenges
Tuners perform a complex task in that they
must cleanly receive one low-power trans-
mission in a crowded spectrum full of other
content. This puts a great deal of pressure
on the RF front-end of the silicon TV tuner IC.
The complexities present at the geographic
boundaries of two diff ering transmission
standards add additional challenges.
Add to this the architectural complexi-
ties of popular DTV and STB SoCs, and it
becomes clear that a single silicon TV tuner IC
cannot meet the needs of the entire market.
Fortunately numerous tuner IC architec-
tures are available today that address these
varying platform needs. The challenge is to
confi dently select the right device for the
application.
Channel reception: The keys to achieving
clear TV broadcast reception are sensitivity
and selectivity. These characteristics in turn
are based on low noise fi gure (NF), high RF
front-end linearity and high-quality RF front-
end fi ltering.
Low noise fi gure means that the RF
front-end adds very little unwanted noise to
the incoming signal, while high sensitivity
means that there is ample video signal-to-
noise ratio (VSNR) even with weak input
signals to correctly receive the broadcast.
But TV tuners almost always need to
receive a single broadcast in the presence
of many other signals. This is where the
silicon tuner IC’s selectivity – the ability to
block or exclude content on nearby ‘blocker’
channels – is essential for clean reception.
To achieve high selectivity in a cost-eff ec-
tive solution, a silicon tuner IC needs both
high linearity and high-quality fi ltering in the
RF front-end. As a result, silicon TV tuner ICs
with high sensitivity and excellent selectivity
performance, as shown in Figure 3, add up
to more channels received under real-world
conditions.
Field testing: Major TV brands deliver
products adhering to the offi cial broadcast
standards, but they must also address the
many real-world exceptions to these broad-
cast standards or face consumer complaints
and costly product returns.
Unusual broadcast conditions and excep-
tions occur more frequently than one might
realise. This is especially true at the geo-
graphic boundaries between one broadcast-
ing standard and another, where issues can
get very complex. For example, in Europe
numerous analog and digital standards can
be received across international borders by
a single TV.
To address these non-standard reception
conditions, the TV tuner and demodulator
must fi rst detect and then compensate for
the non-standard transmissions. These condi-
tions are corrected on a case-by-case basis
and require extensive fi eld testing to perfect.
Without the extensive control and confi gura-
tion capabilities of the silicon TV tuner ICs,
these increasingly challenging requirements
could not be addressed.
TV manufacturers put TV tuners through
rigorous fi eld testing, highlighting both
standard performance as well as performance
to anomalous broadcasts. TVs destined for
cost-conscious regional markets may choose
a lower-cost TV tuner, but TVs destined for
broader export markets will certainly require
robust worldwide performance.
Flexible architectures reduce cost: Greater
choice in TV tuner architectures has been a
positive development for iDTV and STB sup-
pliers. Several silicon tuner IC variations are
now available to meet the specifi c architec-
ture needs of iDTV and STB platforms.
Modern TV and STB platforms have three
main circuit functions: the silicon TV tuner
(see Figure 4 for example), the demodulators
and the audio/video and processing graphics
SoC. The output of the tuner drives the input
of the demodulator, and the output of the
demodulator drives the input of the audio/
video processing and graphics SoC.
TV and STB platforms have varying needs
for analog and digital tuners. For example, a
simple narrowband cable access box (cable
STB) may need a single digital-only tuner. In
contrast, a high-volume mainstream iDTV will
likely support analog and digital terrestrial
broadcast as well as cable input sources.
Figure 3. The Si21x6 TV tuner demonstrates exceptional VSNR performance in the presence of a blocker
(red trace) compared to alternate tuners (blue trace).
![Page 29: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/29.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
News 27featurefeatureAudio & video electronics
For the larger TV brands, these tuners must
support all analog and digital transmission
standards worldwide.
TV tuner ICs are now available sup-
porting numerous architectural confi gura-
tions such as digital only, multi-channel
digital only, analog and digital tuners,
analog and digital tuners with analog TV
demod ulator, as well as various receiver
confi gurations with a tight linking of tuner
and demodulator.
Most modern TVs support at least one
tuner that can tune both analog and digital
channels. Separate demodulators for the
analog and digital channels are needed. To
support varying digital demodulator stan-
dards, there may even be multiple demodu-
lators such as DVB-T and DVB-T2 within the
same platform.
Demodulators vary in complexity with
ATSC (North America) and DVB-T (digital ter-
restrial broadcast for Europe and many other
regions of the world) being relatively low-
complexity standards compared to DTMB
(China’s new digital broadcast standard)
and DVB-T2 (a new higher-density digital
standard in Europe).
In the case of digital demodulators such
as DVB-S2, DVB-C2 and DVB-T2, the demodu-
lator is typically contained in a standalone IC.
Less complex demodulators can be found
in the audio/video processing and graphics
SoC, or they may be combined with other,
more complex digital demodulators.
STBs support a cable input and in some
cases a broadcast antenna input. Unlike TVs,
the SoCs used by many STB platforms usually
lack an analog TV (ATV) demodulator and are
not well suited to analog broadcast reception.
In these cases, the ATV terrestrial broadcast
demodulator will be external to the SoC, if
such a demodulator is required for the STB.
Demodulators for analog TV (ATV
demodulators for NTSC, PAL and SECAM)
Figure 4. Example of state-of-the-art silicon TV tuner architecture.
are especially prone to poor behaviour
when presented with the many real-world
anomalous transmissions. In practice,
actual demodulator performance is highly
dependent on the experience gained from
prior generations of eff ort. This experience is
concentrated in specialised control fi rmware
running on modern silicon TV tuner ICs and
demodulators.
Given the many nuanced interactions
between tuners, demodulators and image
processing SoCs, TV manufacturers must
perform extensive fi eld testing to validate
specifi c combinations of tuners, demodula-
tors and SoCs. This means that in some
cases, a tuner with a higher-performance
internal ATV demodulator will be chosen
even if the SoC integrates its own ATV
demodulator.
The key question for the designer is to
determine which demodulator has been
adequately fi eld-tested relative to the TV’s
or STB’s intended market and use. If fi eld
testing for a specifi c tuner/demodulator
combination has not been completed, it may
be necessary to adopt an integrated and
more thoroughly tested tuner/demodula-
tor combination to meet the development
timeline and cost targets.
Cable STB applications can present a
special challenge for tuner selection. In most
cases, the demodulator will be supported
by the STB SoC. Generally, the transmission
environment for cable is considered less
variable than that of terrestrial broad-
cast. Signal strength and other factors are
reasonably well controlled within a cable
network.
Tuner performance metrics are gener-
ally relaxed for cable STBs applications
using this easoning. However, the ability
to control the cable plant depends on the
specifi c cable provider or multi-subscriber
operator (MSO).
In some cases a given MSO will have
issues with ‘channel tilt’ (diff ering signal
strength across the channel spectrum),
external signal ingress (most likely from local
terrestrial broadcast stations that interfere
with the cable transmissions) or a mixed
transmission mode network that re-transmits
an array of channels from diff erent sources
with diff erent transmission characteristics.
A lower-performance silicon TV tuner IC
may not be adequate in these situations, and
a higher-perform ance TV tuner designed for
terrestrial TV broadcasts may be needed to
receive the full range of channels.
Silicon Labs’ product
off ering
As a practical matter, what
types of TV tuner options
are available for use? A
review of Silicon Labs’ latest
generation of silicon TV
tuner ICs can help answer
this question.
This new TV tuner
family off ers a range of
architectures that combine
easily with popular TV and
STB system-on-chip (SoC)
devices. The Si21x6 TV
tuner family includes the
high-performance Si2176
silicon TV tuner IC with ana-
log demodulator, as well as
the Si2136 complete analog
receiver.
The Si2146 digital-only TV tuner IC is
ideal for cable set-top boxes and multi-tuner
HDTVs. The high-performance Si2156 TV
tuner IC provides a popular confi guration
(with just the digital and analog tuners) with
features similar to the cost-eff ective Si2155
tuner. The Si2155/56 devices easily combine
with popular TV and STB SoCs. These devices
apply equally well to iDTV, cable platforms
and STBs.
The entire Si21x6 family, including the
Si2155 TV tuner IC, delivers exceptional RF
performance and design simplicity off ered
by Silicon Labs’ current-generation TV tuner
ICs that are in high-volume production at fi ve
of the world’s top seven TV brands.
The Si2155 TV tuner is designed to
reduce system costs in high-growth regional
markets such as mainland China and Taiwan
where there is substantial variability in the
technical requirements for TV and STB plat-
forms depending on the needs of regional
and export platforms. TV tuner architectural
fl exibility is especially important to reducing
costs and simplifying designs in today’s
rapidly growing TV markets.
For more information contact
Gary de Klerk, NuVision Electronics,
+27 (0)11 608 0144.
![Page 30: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/30.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature Interconnection, switches, relays, cables & keypads28 regular General
Global concerns over climate change require
using our limited energy resources more
effi ciently. For a cleaner and sustainable
future, energy effi ciency plays a key role.
Approximately 20% of the global electrical
energy is consumed for lighting applications.
The transformation to new, intelligent LED light-
ing technologies for residential and commercial
lighting allows signifi cant energy savings.
With an increasing number of countries
banning incandescent lamps in favour of
energy-saving alternatives, 230 V operated LED
lamps are emerging as the favoured replace-
ment. The charm of these so-called retrofi t LED
lights is that they require minimal or even no
change or modifi cation of the original socket
and control gear.
This advantage is generally accompanied
by the customers’ requirement that the overall
performance of the LED retrofi t must be equal
to or even better than the original incandescent
solution. What this means for the LED driving
solution is that:
• The complete circuitry must be designed in a
space-saving way, to fi t into a standard socket;
• High effi ciency is desired to minimise thermal
design eff orts;
• Easy to design, for a fast time to market.
For the consumer who installs the LED
retrofi t at home, two additional requirements
are important:
Especially for 230 V LED bulb retrofi ts, good
LED lighting solution for lamp retrofi tsdimming performance is one of the most (if not
the most) important features;
The LED solution must not fl icker when there
is a variation in voltage at the supply rail.
Infi neon is off ering innovative solutions
to deal with these challenges. The newly
introduced off -line LED driver ICL8002G
sets high standards with respect to inte-
gration, performance, features and total
system cost. It employs quasi-resonant
operation modes with integrated
power factor control (PFC) for off -
line (230 V) residential LED lighting
applications in particular for 40 W
/ 60 W / 100 W incandescent bulb
replacements.
Applied primary control
techniques result in system ef-
fi ciencies of 90% while external
component count is kept low.
The resulting compact driver design
can easily be embedded within the
standard screw-in form factor. The ICL8002G
outperforms current regulatory requirements
and is prepared for the future. Together with an
external CoolMOS switch, it is easily scalable and
presents a solution to implement multiple LED
bulb designs with one controller.
When it comes to ease of designing, Infi neon
off ers designers an online design tool in the form
of Lightdesk (accessed via www.infi neon.com/
lighting). It is available for the whole ICL(S)8xxx
family and helps customers to design the
transformer, which is the most critical external
component in such a system. Furthermore, the
tool gives guidance on the other external
components and lets users share their
results and designs with others.
As good dimming performance
is a key requirement, Infi neon has
put a strong focus on this issue
during product development. The
ICL8002G has broad compatibility
with leading- and trailing-edge
phase cut dimmers found in
most homes.
To provide smooth dim-
ming, without distortion in the
light output, the IC has a con-
tinuous dimming curve, similar
to dimming an incandescent
lamp. It also incorporates
integrated circuitry to ensure a
stabilised light output when trailing-
or leading-edge dimming is active.
The stabilisation of the light output also
accounts for very good line regulation. With
integrated digital fold-back current correction,
variations in the supply voltage will not have a
visible eff ect on the light output.
For more information contact EBV Electrolink,
+27 (0)21 402 1940, Electrocomp,
+27 (0)11 458 9000.
NXP Semiconductors announced the
SSL2115x series of high-effi -
ciency GreenChip driver
ICs for low-cost LED
retrofi t lamps. The new
SSL21151 and SSL21153 are
designed for non-dimmable
5 W and 10 W LED lamps
respectively, providing good
current control in fl yback and
buck-boost confi gurations.
The SSL2115x enables
small form-factor LED drivers,
using primary-side sensing to
regulate current, and off ering
integrated LED open/short pro-
tections, an internal 700 V MOSFET,
and a circuit enabling startup di-
rectly from the rectifi ed mains voltage.
With the use of an additional valley fi ll
circuit, the series can off er a power factor
of approximately 0,9.
For more information contact Avnet Kopp,
+27 (0)11 809 6100.
LED drivers for retrofi t lamps
T,
tage.
ey fi ll
er factor
![Page 31: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/31.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
featurePower & energy 29regularGeneral
Instrument Plastics’ range of light control fi lm is
designed for a variety of purposes and is avail-
able with a range of options.
For privacy viewing, the fi lm’s microlouvres
allow for controlled viewing so an unauthorised
observer’s line of sight is blocked. It is ideal for
confi dential viewing on CRTs, automatic teller
machines (ATM) and anywhere privacy viewing
is desired.
In terms of sunlight readability, the microlou-
vres help block out annoying off -axis light while
maximising the transmission from the display
to the viewer. The result is improved display
contrast with minimal loss of brightness. Light
is directed to where it is needed or away from
where it is not. This eliminates night-time win-
dow refl ections in automotive and aeronautical
applications. It can also be used to hide the light
source in incandescent lighting applications.
Contrast is improved and glare reduced on
electronic displays by blocking out annoying
off -axis light, making the display more comfort-
able to read. The microlouvres maximise the
transmission from the display to improve the
image with little loss of brightness.
Light control fi lm
TDK has expanded its portfolio of SMD power
inductors with new CLF7045-D types that can
withstand temperatures of up to 150°C.
The new types, which have dimensions of
6,9 x 7,2 x 4,5 mm, off er rated inductance values
ranging from 1 μH to 470 μH. Types are available
for rated currents of 0,43 A to 8,9 A, based on
a 10% drop in initial inductance, and their DC
resistance ranges from 9,6 mΩ to 1,42 Ω.
For more information contact Electrocomp,
+27 (0)11 458 9000.
SMD power inductors
Vectron’s new TX–801 is a TCXO off ering
±100 ppb stability in a 5 x 3,2 mm foot-
print. The device operates from a 3,3 V
supply over an output frequency range of
8 to 26 MHz.
IP-based applications have inherited
legacy timing requirements, such as
Stratum 3 levels of accuracy. Services that
are now carried over packet switched
networks need to comply with levels of
performance that used to be delivered
over the circuit switched networks of the
past.
The TX–801 is optimised for support-
ing timing requirements over modern
IP-based networks and meets the reduced
cost and footprint expectations set by
emerging timing applications.
In addition to the TX–801 TCXO,
Vectron also off ers a range of precision
oscillators designed to support timing
requirements over IP-based networks,
along with innovative clock path technol-
ogy including OCXOs, TCXOs, XOs VCXOs,
VCSOs and jitter attenuators that provide
low-jitter timing solutions.
For more information contact Andrew
Hutton, RF Design, +27 (0)21 555 8400.
Stratum 3 TCXO
The product range off ers choices for
customers in terms of optical quality level,
louvre angle, viewing angle, abrasion/solvent
resistance and surface fi nish.
Instrument Plastics, though its agent Actum
Electronics, is able to handle anything from a
one-off prototype to bulk orders of 100 000
or more. Lead times are typically three to four
weeks.
For more information contact Kevin Klaff ,
Actum Electronics,
+27 (0)11 608 3001.
![Page 32: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/32.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature Interconnection, switches, relays, cables & keypads30 regular General
Since the introduction of the Subscriber
Identifi cation Module (SIM) card technology,
the size of smart cards has greatly shrunk
to pack more on-
chip memory for
provider applications;
increased security,
control and data
processing func-
tions; and to fi t newer
but smaller future
devices.
The migration of
developing markets
from 2G and older networks to advanced
SIM-based 3G networks led to the develop-
ment of a new SIM card
format that enabled
smartphones, tablet
PCs or other devices
to transfer digital
photos and huge data
fi les over high-speed
3G networks. This
format is known as
3FF (3rd form factor),
Mini-UICC (Universal
Integrated Circuit Card)
or more popularly as
micro-SIM.
Measuring only 12
by 15 mm, the micro-
SIM card is 52% smaller
than its 15 by 25 mm
predecessor, the Mini-
SIM (or 2FF) card. It re-
tains the same electric
contacts and circuitry
of previous SIM cards,
has added functional-
ity and is backward
compatible with larger
SIM card holders and
readers via additional
surrounding plastic
cut-outs.
Ideal for ultra-slim
smartphones, tablet
Micro-SIM card sockets
The 55116 connector from Bernier is cylin drical,
has 5 or 6 contacts and is connected with a
bayonet system. This connector is designed for
the transmission of low-intensity signals, and is
often used in the fi eld of military and civil radio
transmissions and on data transmission systems.
It is also used for measurement apparatus,
sensors and testers.
The connectors comply with MIL C 55116
standard. The shells are constructed from
stainless steel and aluminium alloy and insu-
lated with Thermoset. With self-cleaning, gold
Bayonet connectors
Accord has developed a RoHS-compliant keypad
with blue backlighting. The AK-207-N-SSL-WP-
MM-BL has a contact rating of 20 mA, 24 V
d.c. and consumes an LED forward current of
240 mA. Contact resistance is 200 Ω maximum
and the keys can survive 2 million cycles each.
The keypad measures 72 mm high by 56 mm
wide and can withstand operating temperatures
in the range of -20°C to +60°C.
For more information contact Sivan Electronic
Supplies, +27 (0)11 887 7879.
Blue-backlit keypad
PCs, GSM/UMTS modems, PC cards, WLAN
(wireless LAN) cards, as well as machine-to-
machine (M2M) systems, Molex’s micro-SIM
card sockets are
developed for
PCB real estate
and vertical space
savings.
A distinguishing
feature of Molex’s
micro-SIM card
sockets is their
built-in anti-short
design. The risk
of electrical shorting is particularly high
when users of larger SIM cards attempt to
force-fi t these into the shape of a micro-SIM
card by trimming around the plastic edges
of the contact pads (the actual working
area of all SIM cards). This may leave the
pared SIM card with an exposed pad which
may potentially cause shorts with all other
contact pads when inserted into the metal
socket.
Shorting is prevented in Molex’s series
78646 socket by a shell design feature that
confi nes any lateral movements of the SIM
card when inserted, even if improperly cut.
The series 78727 socket has raised plastic walls
that act as contact barriers to insulate the pads
of the inserted SIM card from the metal shell
of the socket.
The socket also has a card polarisation fea-
ture to prevent the card from being inserted
in the wrong direction; and a detect switch to
enable card detection when fully inserted. The
sockets’ round contact terminals provide ex-
cellent electrical contact; their reverse terminal
design with gradual lead-in enables smooth
card insertion and withdrawal.
The sockets have LCP housings that can
withstand high temperatures and feature wide
‘fi nger’ areas for push-pull actions on the SIM
card. Available in 6-or 8-circuit options, these
sockets are ELV- and RoHS-certifi ed.
For more information contact Avnet Kopp,
+27 (0)11 809 6100.
plated contacts, the connectors can withstand
temperatures between -40°C and +85°C, can
be submerged under 2 m of water for 48 hours,
and can mechanically endure 3000 operations.
Voltage rating is 60 V maximum; current
rating is 0,5 A; contact resistance is ≤50 mΩ
mated; dielectric strength is 500 V a.c. RMS;
and insulation resistance is 5000 MΩ under
500 V d.c.
For more information contact Kevin Klaff ,
Actum Electronics, +27 (0)11 608 3001.
![Page 33: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/33.jpg)
Dataweek, 25 July 2012www.dataweek.co.za
featurePower & energy 31regularGeneral
Vishay recently released the fi rst
devices in its next-generation D Series
of high-voltage power MOSFETs. The
new 400 V, 500 V and 600 V N-channel
devices combine low specifi c on-resis-
tance with ultra-low gate charge and
currents from 3 A to 36 A in a wide
range of packages.
Based on a new high-voltage
stripe technology, the MOSFETs’ stripe
design − with a smaller die size and
terminations − lowers the total gate
charge by 50% compared to previous-
generation solutions, while increasing
switching speed and reducing on-resistance
and input capacitance.
The 400 V, 500 V and 600 V devices feature
on-resistance down to 0,17 Ω, 0,13 Ω and
0,34 Ω, respectively. This translate into low
conduction and switching losses to save energy
in high-power, high-performance switchmode
applications, including server and telecom
power systems, welding, plasma cutting, battery
chargers, ballast light, high-intensity discharge
N-channel power MOSFETs
(HID) lighting, semiconductor capital equip-
ment and induction heating.
With gate charges down to 9 nC for the
400 V devices, 6 nC for the 500 V devices
and 45 nC for the 600 V devices, the D Series
MOSFETs off er gate charge times on-resistance
down to 7,65 Ω-nC, 15,6 Ω-nC and 12,3 Ω-nC,
respectively.
For more information contact Dirk Venter,
Arrow Altech Distribution, +27 (0)11 923 9600.
Key features of the
new models include
high power densities,
no external tuning
components (plug
and play), low profi les,
high effi ciencies and
land grid array (LGA)
footprint. They are suit-
able for harsh telecom
and broadcast mast-
mounted applications,
as well as industrial,
test and measurement,
routers and medical
equipment.
The DOSA-2 product
line is comprised of four series with output
current/power ratings as follows, iCF (3 A /
16,5 W), iCG (6 A / 33 W), iBF (12 A / 66 W) and
iAF (20 A / 110 W). All models operate from an
input voltage range of 4,5 V to 14 V and fea-
ture adjustable output voltages from 0,7 V to
5,5 V with conversion effi ciencies up to 94,5%.
The devices range in size from 12,2 x 12,2
x 4,45 mm to 33 x 13,46 x 7,75 mm. Optional
EPC (edge plated castellation) layout is also
available. Featuring fully optimised control
loops, all models within the series provide
excellent transient response without external
loop-tuning components.
These PoL converters operate at a constant
SMT DC-DC converters
switching frequency of 600 kHz, which pro-
vides a low noise profi le and facilitates easy
EMI fi ltering. All models are safety approved
to UL/CSA60950-1 and VDE CB scheme per
IEC/EN60950-1.
Remote on/off , remote sense and power-
good signal are standard. In addition, all
models can operate in ambient temperatures
from -40°C up to +115°C and include a 3-year
warranty.
For more information contact
Accutronics, +27 (0)11 781 2645,
Vepac Electronics, +27 (0)11 609 7122.
TDK-Lambda has released the iCF, iCG, iBF and iAF series of non-isolated SMT DC-DC
converters that comply with the latest 2nd-generation DOSA standards for point-of-
load (PoL) converters.
![Page 34: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/34.jpg)
Dataweek, 25 July 2012 www.dataweek.co.za
feature Interconnection, switches, relays, cables & keypads32
HotHot Chips
32
ON Semiconductor off ers two new multiple-
channel diff erential switching ICs in single-pole,
double-throw (SPDT) switch confi gurations
for high-frequency signals such as PCI Express
3.0 and DisplayPort 1.2 I/O. The NCN3612B is a
6-channel diff erential SPDT switch that supports
8 Gbps data transfer with an on-state capaci-
tance (CON
) of 2,1 pF and an on-state resistance
(RDSON
) of 8 Ω. This is complemented by the
NCN3411, which is a 4-channel diff erential
switch also capable of supporting 8 Gbps opera-
tion. It has CON
of 2 pF and RDSON
of 7,5 Ω.
The NCN3612B has a power supply voltage
range of 3 to 3,6 V and draws a supply current
of just 250 μA, while the NCN3411 has a power
supply voltage range of 1,5 V to 2,0 V and draws
a supply current of 200 μA.
Avnet Kopp, +27 (0)11 809 6100
The BGU8006 from NXP Semiconductors is
a GPS low-noise amplifi er packaged in a tiny
WLCSP package with a footprint of just 0,65 x
0,44 x 0,2 mm and requiring only two external
components. Featuring a low noise fi gure of
0,60 dB, the device off ers enhanced reception
for weak GPS signals by dynamically suppress-
ing strong cellular and WLAN transmit signals.
It uses adaptive biasing techniques to instantly
detect any output power from jammers, and
compensate by temporarily increasing the
current. With the BGU8006, adaptive biasing
improves linearity with a 10 dB better IP3 under
-40 to -20 dBm jamming conditions and pro-
vides eff ective GPS output with jammer power
up to -15 dBm.
Arrow Altech Distribution, +27 (0)11 923 9600
RFMD’s new
RF5836 provides a
complete inte-
grated solution in
a single front-
end module for
Wi-Fi 802.11a/n
systems. The
small form factor
and integrated
matching minimise layout area and reduce
the number of external components. The
IC integrates a power amplifi er, single-pole
double-throw switch, and a power detector
coupler for improved accuracy. It is provided
in a 3 x 3 x 0,5 mm, 16-pin package. The chip
operates from a single supply voltage of 3,0 V to
4,8 V, and a control voltage above 1,6 V. Power
output is 15,5 dBm (11a, 54 Mbps at 4% EVM) or
14,5 dBm (11n, 65 Mbps at 2,8% EVM).
RF Design, +27 (0)21 555 8400
Multiple-channel switching ICs
Low-noise amplifi er for GPS
Wi-Fi front-end module
Texas Instruments introduced the fi rst in a
family of multi-chemistry, multi-cell battery
management gas gauge circuits with propri-
etary ‘Impedance Track’ technology which
maintains up to a 94% accurate capacity mea-
surement for the entire life of the battery. The
bq34z100 supports a wide range of lithium-ion
and lithium iron phosphate chemistries in 2- to
16-cell battery packs, extending battery run-
time in applications like medical instruments,
power tools, e-bikes and uninterruptible power
supplies. TI also plans to introduce new gauges
this year to support lead acid, nickel cadmium
and nickel metal hydride chemistries.
EBV Electrolink, +27 (0)21 402 1940
Battery gas gauge
AMD intro-
duced the
latest entry in
its embedded
G-Series pro-
cessor family
with the
G-T16R Accelerated Processing Unit
(APU). It is targeted at cost-sensitive
embedded designs that require a
combination of x86 compatibility
and graphics. The chip boasts power
consumption of just 2,3 W on average, or 4,5 W
thermal design power (TDP). It fi ts into small
form factor boards by implementing a two-chip
platform – the APU and its companion control-
ler hub. Legacy I/O card support is based on a
full 32-bit PCI interface and an ISA bus solution
with DMA support. The APU supports a full
range of display technologies, with analog VGA
and LVDS support for legacy applications and
DVI, HDMI and DisplayPort interfaces for the
latest display technology.
www.amd.com
Low-power embedded processor
International Rectifi er has introduced its third
generation of SupIRBuck synchronous buck
voltage regulators. The new single-output
devices feature a newly patented modulator
scheme claimed to generate the industry’s
smallest, jitter-free pulse widths. By reducing
jitter for pulse widths as small as 50 ns, the tech-
nology enables higher closed-loop bandwidth
for better transient response and lower output
capacitance. In addition to switching frequency
up to 1,5 MHz, pre-biased startup, input volt-
age aware enable, tracking and programmable
power-good, the new devices off er reference
voltage accuracy of 0,5%, enhanced line/load
regulation with input voltage feed-forward and
thermal compensation for current limit.
Future Electronics, +27 (0)21 421 8292
Synchronous buck regulators
![Page 35: Dataweek Electronics & Communications Technology | 25 July 2012](https://reader031.fdocuments.net/reader031/viewer/2022020912/620348ca24f6b61e9c6636e2/html5/thumbnails/35.jpg)
Dataweek, 25 July 2012 IBC
* denotes advertisement
QUICKFINDCompany Telephone Pagese-mail Website
More suppliers, more products, more info:
www.ebg.co.za
www.dataweek.co.za
Accutronics +27 (0)11 781 2645 [email protected] www.accutronics.co.za 15,31
Actum Electronics +27 (0)11 608 3001 [email protected] www.actum.co.za 29,30
Arrow Altech Distribution +27 (0)11 923 9600 [email protected] www.arrow.altech.co.za 6,8,23,24,31,32
Avnet Kopp +27 (0)11 809 6100 [email protected] www.avnet.co.za 9,23,24,28,30,32
Centurion Micro Electronics +27 (0)12 666 9066 [email protected] www.cme.co.za 7*
Comtest +27 (0)11 608 8520 [email protected] www.comtest.co.za OFC*,10,11,12,13,14
Conical Technologies +27 (0)12 345 1545 [email protected] www.conical.co.za 15
CST Electronics +27 (0)11 608 0070 [email protected] www.cstelectronics.co.za 4
Current Automation +27 (0)11 462 4253 [email protected] www.rectifi er.co.za Stitched insert*
Denver Technical Products +27 (0)11 626 2023 [email protected] www.denvertech.co.za 13*,16,17
EBV Electrolink +27 (0)21 402 1940 [email protected] www.ebv.com 28,32
Electrocomp +27 (0)11 458 9000 [email protected] www.electrocomp.co.za 28,29
Electrocomp Express 0860 10 20 20 [email protected] www.eexpress.co.za 6
Electronic Touch Systems +27 (0)11 782 3346 [email protected] www.electouch.co.za 9*
Electronica +27 (0)11 486 2775 [email protected] www.electronica.de 5*
Future Electronics +27 (0)21 421 8292 [email protected] www.futureelectronics.com 8,32
Keystone Electronic Solutions +27 (0)12 460 4135 [email protected] www.kses.net 5
Lambda Test Equipment +27 (0)12 349 1341 [email protected] www.lambdatest.co.za 11*
Logica Technologies +27 (0)12 654 4174 [email protected] www.logica.co.za 14*
M/A-COM Technology Solutions +27 (0)11 908 6679 [email protected] www.macomtech.com 7
Mantech Electronics +27 (0)11 493 9307 [email protected] www.mantech.co.za 12*,29*
Mateq Systems 0861 373 777 [email protected] www.mateq.co.za 15*
Newelec +27 (0)12 327 1729 [email protected] www.newelec.co.za 30*
NuVision Electronics +27 (0)11 608 0144 [email protected] www.nuvisionelec.co.za 25,26,27
OSRAM Opto Semiconductors +27 (0)79 525 1779 [email protected] www.osram.co.za 23
RF Design +27 (0)21 555 8400 [email protected] www.rfdesign.co.za 7,8,29,32
Sivan Electronic Supplies (SES) +27 (0)11 887 7879 [email protected] 28*,30,31*
Technews Publishing +27 (0)11 543 5800 [email protected] www.technews.co.za OBC*
Tempe Technologies +27 (0)11 452 0530 [email protected] 18,19,20,21,22
TRX Electronics +27 (0)12 347 4419 [email protected] www.trxe.com IFC*
Vepac Electronics +27 (0)11 609 7122 [email protected] www.vepac.co.za 31
Webb Industries +27 (0)11 719 0000 [email protected] www.webb.co.za 15