Fujitsu Aizu Wakamatsu Plant
Fujitsu and ON have signed a foundry agreement under which Fujitsu will manufacture wafers for ON Semiconductor at its 8-inch fab in Aizu-Wakamatsu, Japan.
Initial production of wafers is expected to begin within a year from today, and ON will have the opportunity to access additional capacity in the Aizu-Wakamatsu fab in the future.
The two companies have also entered into a definitive agreement under which ON will obtain a 10% ownership interest in a newly formed subsidiary of Fujitsu that will include Fujitsu’s 8-inch Aizu-Wakamatsu fab.
The consideration to be paid by ON for this minority interest will be ¥700 million (approximately $7 million). The transaction is expected to close during the fourth quarter of 2014 or in early 2015, subject to certain regulatory approvals and other closing conditions.
“We believe that growth of the new company will contribute to the development of the region, as well as to maintaining employment,” says Fujitsu CEO Haruki Okada, “we expect that the agreements for the foundry services and the acquisition of minority stake of the 8-inch fab by ON Semiconductor will greatly boost the businesses of both companies.”
“This is a strategic investment for ON Semiconductor to secure additional manufacturing capacity, to support our production needs and revenue growth in coming years,” says Alan’s CEO Keith Jackson, “we believe these agreements with Fujitsu Semiconductor will enable us to maintain our industry-leading manufacturing cost structure and also help us optimize our capital spending in coming years.”
Fujitsu is trying to sell its fab, in Mie, to UMC. Earlier this year it put its semiconductor interests into a jv with Panasonic. Last year it sold its analogue and microcontroller business to Spansion
Inside are 256 current integrators, 256 sample-and-holds, and ADCs with configurable sampling rate and up to 24bit resolutions. 11.656ksample/s (86μs) is available at 20bit resolution.
Integral linearity is +/-0.015% of reading +/-0.4ppm of full scale with one channel active, or +/-0.050% of reading +/-1.0ppm of full-scale with all channels active.
“The signal chain and sampling architecture is designed to guarantee that all channels are simultaneously sampled, and that no charge is lost throughout the sampling process,” said ADI.
Dissipation is 3mW per channel at any throughput.
All converted channel results are output on a single, low voltage differential signaling (LVDS), self-clocked serial interface.
Configuration is over an SPI-compatible serial interface, and integration includes: a temperature sensor, reference buffer, and decoupling capacitors.
Aimed at CT scan inputs and photodiode arrays, devices cost $304 each when buying 1,000.
Support tools include an evaluation board, reference design including layout, and FPGA Verilog code.
The UK’s Centre for Advanced Laser Technology and Applications (CALTA) has supplied a £2.2m gigawatt laser amplifier to the Czech Republic’s Institute of Physics (IoP) for the European Extreme Light Infrastructure (ELI) ‘Beamlines’ facility.
Oxfordshire-based CALTA is part of the Science and Technology Facilities Council (STFC), and exploits technology developed in STFC’s Central Laser Facility (CLF) for industry and big laser facilities.
“The laser platform will benefit new build laser infrastructure projects, including ELI, and will drive new laser-based applications in industry, including advanced material treatments and energy production,” said CLF director Professor John Collier.
The ELI laser amplifier head is based ‘DiPOLE’ – high-repetition-rate super-power pulse laser technology from CLF – and can supply 10J nanosecond pulses at 10Hz. Average power is 100W. Pulses from 2 to 10ns are available, peaking at 1GW at 10ns.
“Each laser burst has power equivalent to that of a full-sized power station,” said STFC.
“DiPOLE technology not only enables high power laser operation at high repetition rates of many pulses per second” said CLF’s Collier, “but also demonstrates high efficiency operation, with at least 10% efficiency in producing laser power from electricity, compared to conventional systems that are typically much less than 1%.”
The amplifier (see photo) will sit part way along an infra-red laser chain. Inside, a doped YAG ceramic block absorbs 10J of pump laser energy over 1ms, CALTA manager Justin Greenhalgh told Electronics Weekly, then releases it at the command of a different frequency 10ns 100mJ trigger pulse from a Czech-made diode laser. Mirrors on the ceramic block zig-zag the trigger pulse six times to allow it to grow into the output pulse.
In the photo, the circular beam splitters reflect at the pump wavelength, allowing energy from pump lasers A and B to directed into the hidden YAG ceramic prism through windows. Transparent at the trigger/output frequency, the splitters allow the trigger pulse to enter and the output pulse to leave through the same windows in a straight line. Cooling helium enters through the top port (capped in the photo).
The resulting 1GW pulse can be used to pump another amplifier which, when triggered by a picosecond pulse, will deliver a petawatt picosecond pulse.
The laser head is the first major contract of its kind between CALTA and IoP.
ELI Beamlines will have four giant lasers delivering petawatt pulses at 10Hz – something that CLF’s venerable Vulcan laser can only do every 20 minutes.
Focused intensities of around 1023W/cm2 will be available.
“Being able to operate high power lasers at high repetition rate is key to developing applications such as medical imaging and radiotherapy for industrial use,” said STFC.
ELI is split between four sites – with Beamlines in the Czech Republic, and others in Hungary, Romania, and one to be decided.
Investment exceeds E850m, mostly from European regional development funds.
Combined heat and power units have been used by large businesses for many years, using internal combustion engines to provide electricity via a generator, and heat from the engine cooling system and exhaust.
On small premises, turning on a pump or kettle can increase the electrical load several fold in a second. Attempts to cope with the now unmatched electricity and heat results in the system running inefficiently.
According to the consortium, the solution is a generator that runs constantly at high efficiency, coupled to an electrical energy store.
“The challenge, was to design a system that could simultaneously satisfy the more predictable needs for heating and hot water, as well as the wildly varying demand for electricity in a small dwelling,” said Professor Tony Roskilly of Newcastle. “Our solution was to incorporate electrical storage into the system, both batteries and supercapacitors, combined with system control.”
Excess heat is stored in hot water tanks for heating and hot water.
Refrigeration and air conditioning operates via an absorption chiller running off heat.
Leeds University, part of the consortium alongside Newcastle, the University of Ulster, and three Chinese universities, has also patented a cryogenic energy store.
This combination has been dubbed ‘trigeneration’.
“We wanted to avoid running the trigeneration system using biodiesel or other highly-processed fuels from raw materials,” says Professor Roskilly. “So instead, we developed a system for using the oils obtained from pressing crop seeds, like those from jatropha and croton. These can grow in harsh environments and on poor-quality land and so could be well-suited to providing fuel in developing countries, as cultivating them would not adversely affect food production.”
£1.13m of funding came from the Engineering and Physical Sciences Research Council (EPSRC) through the RCUK Energy Programme.
Before designing the trigenerator, the team logged the minute-by-minute energy use in households. “Previous studies have lost important detail by averaging demand over much longer timescales,” said the EPSRC. “In a typical UK house, heating demand is largely stable when hot water and space heating is required. Electricity consumption can hover around a hundred watts most of the day, but reach peaks of 7kW in a matter of seconds, and for just a minute or two.”
In a follow-up study funded by the EPSRC, the Department for International Development and the Department of Energy & Climate Change, Roskilly is exploring how the trigeneration can be used on small farms in the developing world to refrigerate and process food crops, to reduce post-harvest losses.
The Newcastle team is examining long-term performance running on raw plant oils, and is discussing commercialising the design with potential manufacturers.
A domestic-scale tri-generation system would be rated between 6kW and 9kW of electricity, equivalent to the combined consumption of: lights, TV, fridge freezer, kettle, microwave, vacuum cleaner, washing machine, and dishwasher.
Did you know you can use your frequency counter to detect gravity? You’ve likely done it before and you didn’t even know it!
Dave demonstrates the phenomenon of 2g-tipover on quartz crystal oscillators in an Agilent 53131A frequency counter.
Related videos:
How a rubidium frequency standard works
FE-5680A Rubidium Standard Teardown
Forum HERE
Doherty amplifiers have two RF amplifiers in parallel (see diagram), one Class-AB for low-level signals and one Class-C for peaks. They save power over single amplifiers that can handle high peak-to-average signals, like those of LTE (8dB) and LTE-Advanced, with sufficient linearity.
Signals going into the two amplifiers have to be adjusted carefully to ensure their outputs combine to create the correct antenna feed.
The firm is aiming its chip at macro and small cell base stations.
“The current solution employs discrete components to tune phase and amplitude for carrier and peaking paths,” said Peregrine, which concedes Freescale is also aiming to replace discretes, this time with a system-in-package containing three GaAs and one CMOS die.
Made on a 0.35µm silicon-on-sapphire process, Peregrine’s PE46120 monolithic phase and amplitude controller (MPAC) includes: a 90°hybrid splitter, two phase shifters, two amplitude controllers, and an SPI digital interface.
Both paths can be optimised on-the-fly through the SPI interface to compensate for operational or environmental changes. Switching time is 840ns.
The firm has based the design on existing products, using switches and passive components on-die. Phase shifting is through combined low-pass and high-pass filters. “It is more of an RFIC approach, said Mark Moffat, MD Peregrine UK. “For an X band product [which is mooted], we would probably use more of a MMIC approach.”
Insertion loss is 6.5dB including the 3dB splitter, and operation is from 1.8 to 2.2GHz in 5.6° steps across an 84.4° range, as well as 0.5dB steps across 7.5dB range. Linearity is 65dBm IIP3, return loss 20dB, isolation 28dB, and maximum power 29dBm.
The chip will run off 2.7-5.5V (at 300µA), and over -40 to 105°C.
It comes in a 6x6mm 32 lead QFN, and needs no external dc blocking capacitors.
Other applications proposed are: micro and pico cells (10W and above), electronically-steered radar, and synthetic aperture radar.
Three other versions are in the pipeline: 2.4-2.8GHz, 700MHz-1GHz, and one for LTA-A bands.
At the same time as announcing the intention to introduce the Doherty chip, Peregrine also revealed what it claims are the first monolithic 50W power limiters.
Power handling is +47dBm (50W) pulsed, +40dBm (10W) CW.
PE45140 (20MHz-2GHz) and PE45450 (9kHz-6GHz) are intended as an alternative to discrete GaAs PIN-diode limiters in ‘land mobile radio’ (LMR), test-and-measurement, wireless-infrastructure and radar.
Linearity (IIP3) is over 40dB and limiting threshold can be as low as 22dBm and as high as 35dBm.
Response-and-recovery is under 1ns.
There are two modes: power limiting and power reflecting, with threshold selectable through an analogue control pin voltage.
For example, in limiting mode (Vcontrol = -2.5 to -0.5V) PE45140 performs as a linear power limiter with adjustable P1dB / limiting threshold.
Power reflecting mode requires an external power detector to sample the RF input power and a microcontroller to toggle the limiter control voltage between +2.5V and -2.5V. At +2.5V, limiter impedance to ground is less than 1Ω and most incident power will be reflected back to the source. At -2.5V, the device operates as in power limiting mode.
Package is 3x3mm 12 lead QFN
Dave looks inside the most popular microcomputer of the 1970′s, the Radio Shack / Tandy TRS-80 Model I
And also a look at the TRS-80 Model 102.
TMS4116 16Kb DRAM
Level II BASIC Reference Manual
Forum HERE
China could change the way the mobile chip world works by its treatment of Qualcomm.
China has declared Qualcomm a monopoly in China and may be about to declare that Qualcomm has abused that monopoly. If it does, Qualcomm could be fined up to $1.2 billion.
Qualcomm is complaining that it’s Chinese licensees are under-stating sales so as to reduce royalties, while other Chinese companies are delaying signing licences until the abuse of monopoly legal process is concluded.
China has a big interest in promoting the local consolidated mobile phone company Unigroup-RDA-Spreadtrum. High licensing fees would make it uncompetitive.
The danger for Qualcomm is that the China government will not acknowledge the validity of the system under which Qualcomm operates and that Qualcomm will find its IP unprotectable in China.
Qualcomm can’t walk away from China without incurring a big penalty – half of its $25 billion revenues came from China last year,
ST’s Executive VP Francois Guibert
ST has opened a MEMS Microphone Lab (Anechoic Chamber) in Taiwan to test and analyse high-performance audio applications built with ST’s MEMS microphones.
ST’s lab will focus on all-level audio performance testing from components (microphone or acoustic components) to modules and systems, including smart phones, tablets, notebooks, TVs, and remote controls, ensuring superior recording and sound quality, shorter debugging period, and faster time to market for end applications.
“With the increasing demand for advanced microphone applications in the Greater China region, it is imperative for ST to strengthen its local technology and application support,” says ST’s Francois Guibert, “establishing a new advanced testing lab facility in Taiwan will allow us to work even more closely with our key customers and partners in the region to optimise the quality of their products at all levels and develop innovative applications with excellent audio performance.”
Taiwan, home to a number of top-tier global ODMs and some of the world’s largest EMS companies, is the world’s second largest producer of IT products. With a comprehensive semiconductor-industry supply chain, from IC design and manufacturing to packaging and testing, accounting for almost 70% of the world’s contract IC chip output and a quarter of the global IC design market, Taiwan is the ideal location.
Equipped with the APx525 Family Audio Analyzer from Audio Precision, ST’s lab complies with the ISO 3744/3745 industrial acoustic standard and environmental noise regulation, as well as Intel’s Speech and Voice Recognition Standard.
According to IHS’ latest report[1, global MEMS microphone shipments rose by 37% year-on-year from 1.9 billion in 2012 to 2.6 billion units in 2013. By 2017, the shipments are forecast to reach 5.4 billion units.
The S6AP412A series — an expansion of the Spansion MB39C031 family of DC-to-DC buck-converter, programmable PMICs — supplies three channels of power on a single chip to cutting-edge system-on-chip (SoC) products, memory and peripheral products that make up the core of office automation equipment and network devices with advanced image and voice processing functions.
Systems equipped with the latest SoCs require control of a 3-channel power supply sequence when starting up or disconnecting the power supply. The Spansion S6AP412A uses software to enable flexible and precise power supply sequence control. In addition, by enabling 3-channel power supply within a single chip, the BOM for the power supply block can be reduced and the board surface area can be reduced by approximately 30 percent compared to conventional products.
The new product also uses the common Inter-Integrated Circuit (I2C) communication interface to enable switching to an optimum voltage according to the SoC operation mode and process conditions while maintaining the output, and is thus able to support both Dynamic Voltage Scaling (DVS) and Advanced Supply Voltage (ASV) technologies. The S6AP412A series also supports reduced power consumption by systems through precision control of the voltage used by cutting-edge SoCs.
“This new PMIC expands our power management portfolio, providing the freedom to design with software, and delivers high performance and a lower power consumption and BOM than previously required to support high-end graphics and voice functions of large-scale systems in office automation systems,” says Spansion’s Tom Sparkman, “the multi-channel power supply eliminates the need for dedicated power supply lines to each component and aids in the trend of miniaturisation in system design.”
The chips support advanced power supply sequence control, and therefore complex power supply startup and disconnect sequences of systems with various part architectures can also be flexibly supported with software.
Easy preparation of startup and disconnect sequence for the power supply line: The latest systems require optimal control of the 3-channel power supply sequence when starting up and disconnecting, and this product can freely and flexibly control various sequences through software control. In addition, a method to step-up and step-down the voltage is adopted for the channel used for peripheral devices, and a stable output can be achieved even with a single cell lithium-ion battery and a 3.3V input.
By adopting this product, systems that require different sequences can be supported while maintaining the same hardware, the need to prepare new hardware when expanding a derived family can be eliminated, additional investments can be reduced, design resources can be effectively utilized, and added value can be easily provided to the user.
Supports DSV and ASV control using an I2C interface: High-performance processing is realized by dynamically controlling the power-supply voltage of the latest SoC that use advanced processes, thereby enabling a reduction in system power consumption.
Other features and specifications include:
1. 0.7 to 1.32V for SoCs (with an electric current supply of up to 4Amps); 1.2 to 1.95V for memory and 2.8~3.5V for peripheral devices with a single chip.
2. Built-in passive components (output voltage setting resistance, phase assurance circuit) reducing the BOM necessary to configure the power-supply block and reducing the surface-mount area by approximately 30 percent.
The Spansion S6AP413A series also is available with dual-system output support: A 2-system voltage output option is also available to enable support of systems equipped with dual SoC (with a maximum of 2-Amps output for each system with a supply voltage range of 0.7 to 1.32V). This series can also supply power with a single chip to systems equipped with two CPUs. In addition, a lineup of preset values for output voltages tailored to the main SoC voltages is available.
Most of the new sensors are related to exhaust aftertreatment because of new emissions laws with NOx reduction a focus alongside that of carbon dioxide.
As a pollutant, NOx has long been a stronger focus for US legislation, which also dictates that the emission parameters are measured under realistic driving cycle conditions.
But European legislators have also become tougher on this gas in recent years. IHS forecasts that the market for NOx sensors will grow at a CAGR of 9.3% during the next five years from 2014 to 2019.
The global market for sensors used in internal combustion engines (ICE) is on the road of steady growth for the next few years, propelled by increasing utilisation in engine management and exhaust after-treatment.
Sensor shipments for engines will top 1.34 billion units in 2019, up from about 1.08 billion in 2013, as shown in the attached figure. Overall, the six-year compound annual growth rate (CAGR) from 2013 to 2019 will equate to 3.6 percent.
“Shipments of ICE sensors are growing slightly faster than car shipments,” says IHS’ Richard Dixon, “the main reason is that new concepts in emissions mitigation in the engine and in exhaust after-treatment systems require advanced sensors for their operation. Added to that, emission legislation in some major markets of the world, like China, is beginning to catch up with that of mature markets like the U.S., Europe and Japan, at least in the larger cities.”
IHS examines more than 20 sensors attached to the engine, fuel and exhaust systems of passenger vehicles. The list includes pressure sensors, devices to monitor flow and temperature, ceramic sensors for the gases nitrogen oxide (NOx) and oxygen, in addition to knock sensing, position and speed.
Among the 24 applications identified for sensors used in ICEs, several measurements have been essential to electronic fuel management systems for more than 20 years: the position of the throttle and crankshaft, the absolute air pressure of the intake manifold and the residual oxygen in the exhaust. And multiple sensor insertions are possible—depending on the pipe configuration, a gasoline engine can feature four oxygen sensors, two of which serve an on-board diagnostics function to check for correct operation, although a diesel engine still has only one oxygen sensor, located before the diesel oxygen catalyst.
The biggest category for sensing is temperature measurement, with multiple sensors to be found on exhaust systems. The technology used is typically platinum-based resistance temperature detector (RTD) sensors to withstand temperatures of up to 1,000 C (1,832 F). On average, approximately two temperature sensors are used per vehicle. At lower temperatures, negative-resistance sensors (NTC) or semiconductor integrated circuits are also deployed. Examples include engine-coolant monitoring to protect against overheating, intake air measurement or in exhaust-gas recirculation systems used to lower NOx output from the engine.
At other points in the powertrain, the fuel injection system and cylinder pressure are proving to be interesting new opportunities for pressure sensor suppliers. Although cylinder-pressure sensing has been largely too expensive for car manufacturers to adopt, IHS expects significant growth in this area in coming years, led by Volkswagen and Daimler in diesel engines. Owing to the low penetration today, growth rates in the coming five years to 2019 are high at more than a 40 percent CAGR.
The type of engine has an impact on exhaust systems. Lean-burn engines require more sensors than standard, stoichiometric engines because the exhaust aftertreatment is more complex. The high residual oxygen content in the exhaust of these engines makes it impossible for conventional three-way catalysts to reduce NOx pollutants.
Diesel and some direct injection gasoline engines fall into this category. These engines require feedback on NOx concentration in the exhaust stream as part of the emission control system. Stoichiometric engines—some direct injected engines and all modern port injected engines—do not need a NOx sensor.
In addition, many vehicles today feature some kind of forced induction—turbo- or supercharging—of the fuel/air mix. The increased use of turbochargers is the main factor enabling engines to be made smaller and less polluting.
As these boosters become more sophisticated, such as having a variable geometry, position sensing becomes important. So far, though, turbocharger speed sensing is mostly confined to the large commercial vehicle segment and is not yet widely found in cars.
For the first time, European Euro 6 legislation specifies a particle number per kilometer as a key measure, with phased-in adoption. Direct injection gasoline engines produce more hazardous particles than standard gasoline multi-port injection engines. This means that gasoline particle filters on the exhaust will be adopted in the former, along with the means for self-monitoring, which will drive the use of particle mass and possibly pressure sensors.
Particle filters have been used on diesel engines to reduce NOx since legislation in Euro 4. In particular, diesel particle filters have pressure sensors to monitor the filter when it is clogged, indicating to the engine ECU when to regenerate the filter—a process that also requires temperature sensing.
Sensor adoption also has a strong regional emphasis. As emissions standards are not yet harmonized worldwide, less stringent emission standards than those affecting U.S., Japanese or EU standards are in play, which leads to lower sensor content in emerging markets. A gasoline-engine vehicle in Iran or Malaysia, for instance, will have a much lower sensor requirement than an engine in Japan. The difference can be as many as 10 sensors.
Dick Smith Electronics (known colloquially as Tricky Dicks here in Australia), a once (in)famous electronics components retailer, but long since been just a consumer electronics store, have released some excellent videos highlighting three local Sydney makers as part of their Unleash Your Smith advertising campaign shown on Australian TV and media.
One can’t help but wonder why they didn’t call it the Unleash Your Dick campaign?
It’s rather perplexing why Tricky Dicks would be promoting makers when they don’t sell anything the least bit Maker related any more?
But hey, any company spending money to promote Makers is all good!
Anyway, I was asked to help organise this at the last minute, and they have shot three excellent local makers, two of which you may have seen on the blog before.
Malcolm Faed with his home made Segway
Alex McClung with his Alex Bot
Tom Hetherington and crew from Optic Soup with their video projection art.
The videos are well produced, enjoy:
Dave shows some techniques on how to build and mount usable PCB based front panels user interfaces with LCD displays and push buttons and capacitive touch buttons onto small cheap extruded aluminium enclosures.
In particular for the µSupply project. This is Part 14 of that series.
The uCalc project gets another look in, and does the Sharp Memory dot matrix LCD.
Forum HERE
The SLG46116V and SLG46117V are the first GPAK devices to enable 1 A P-Channel mosfet soft-start power switching, controlled with a mixed-signal matrix, in a 1.6×2.5×0.55 mm seven-GPIO STQFN package.
The soft-start power switch has fixed slew rate control, and is available with an integrated discharge path (SLG46117V) or without (SLG46116V).
SLG46116V and SLG46117V projects use GPak development hardware and GPak Designer.
“Silego’s latest devices add intelligent power switching functionality to the analogue and digital functions provided by GPAK,” said Silego’s Nathan John.
Pete Hutton – EVP and President of Product Groups
ARM has licensed processor and security IP to Atmel for imaging, video and display applications.
Atmel plans to integrate the technology into ICs for wearable devices and automated factory tasks reliant on image processing such as the identification of faulty products.
The license includes the ARM Cortex-A7 processor, ARM Mali-V500 video accelerator, Mali-DP500 display processor, and ARM TrustZone technology.
“Atmel and ARM have a successful history of collaboration,” says ARM evp Pete Hutton, “that partnership continues to build with Atmel now expanding its ARM IP portfolio to prepare the way for products in emerging markets such as IoT, wearables and factory automation.”
ARM Mali-V500 and Mali-DP500 enable HD 1080p60 resolution capabilities on a single core.
Both the ARM Mali-V500 and Mali-DP500 incorporate support for ARM TrustZone technology.
“As IoT and wearable devices become smaller, more sophisticated and integrated, the SoCs used in the devices will need to offer more features and functionality in smaller packages,” says Atmel’s Reza Kazerounian, “the small area footprint of the ARM Cortex and Mali multimedia solutions will allow us to offer HD video and display processing in unprecedented sizes.”
With Mali-DP500, Atmel SoCs will have the capability to deliver UI functionality such as multi-layer composition, scaling and post-processing with support from ARM’s Frame Buffer Compression (AFBC) protocol. This technology is unique to ARM and is capable of delivering a 60% reduction in system bandwidth for video playback.
ST CEO, Carlo Bozotti
Thanks to a payment of $100 million from the French government under the Nano2017 R&D support programme, STMicroelectronics recorded its first quarterly profit for 11 quarters in Q2.
Q2 revenue was $1.86 billion and net profit was $38 million.
Free cash flow was minus $99 million compared to minus $51 million in Q1.
ST’s net financial position was $423 million at the end of June ($1.55 billion cash less $ $1.13 billion debt). At the end of March the net financial position was $612 million.
The proceeds from ST’s recent sale of $1 billion worth of bonds will be accounted for in Q3′s balance sheet.
There is another $300 million due to come from Nano2017.
ST recorded a loss of $52 million on exiting the 3Sun solar power jv with Sharp and Enel which made solar modules in ST’s 300mm Catania fab which was never facilitised for chip production. Both Sharp and ST have now transferred their shares in 3Sun to Enel.
“During the second quarter we made positive business and financial progress in key areas,” says ST CEO Carlo Bozotti.
