Control Sheet No. 21

Dragons Unite: Introducing KeZhi (Cosylab China)

By : Klemen Žagar (Cosylab), Dominik Peruško (Cosylab), Yang Wenwei (Indy) (KeZhi Controls)

About the Title

Dragons are a strong element in both Chinese and Slovene, specifically Ljubljana, culture. While in the West dragons are generally negative, in China the dragon is a symbol of power, strength, and good luck for those that deserve it. This positive symbolism is also found in the Ljubljana Dragon, who protects the city and is featured on the coat of arms.

CS21-1CS21-2

There is no denying that China is developing at a rapid pace. Interestingly, in the last three years, China used more concrete than the United States did in the last 100 [1]! The development of Chinese scientific infrastructure is growing just as fast, with projects such as the Chinese Spallation Neutron Source (CSNS, [2]), Chinese Accelerator Driven System (CADS, [3]), Thorium Molten Salt Reactor [4] and Chinese participation in the ITER project [5] being just the beginning of an ambitious long-term agenda that also includes the 52-kilometer electron-positron collider [6] and the Chinese DEMO fusion reactor.

With such an abundance of opportunity, Cosylab’s decision to establish a branch office in China probably does not come as a surprise. The decision was made in 2013, and the first challenge was to pick the city, the office and the name. At least the color-scheme was compatible!

The city we chose was Hangzhou. With a population slightly more than 6 million, it is a modern Chinese city with good airplane and fast-train connections. Geographically located at the center of the east coast, this means convenient and cost-efficient travel to all customer sites. Zhejiang University, one of the top five Chinese universities, is also located in Hangzhou - one of the important sources of future Cosylab engineers.

CS21-3

Cosylab China / KeZhi Controls office building

When non-Chinese companies establish subsidiaries in China, they usually pick a Chinese name, and Cosylab is no exception.

Our name in Chinese is 科挚控制. Roughly translated back to English it would read KeZhi Controls. KeZhi was chosen because it phonetically most closely matches the pronunciation of cosy, while the character 科 means science, and the character 挚 means friendly/sincere, resonating strongly with Cosylab’s mission and culture.

In April this year, we started staffing our office. The employment of administrative assistant Zhong Xiaofang was closely followed by Yang Wenwei (Indy), an experienced software developer and R&D manager. There are many job-seekers in China, (see the picture from a typical job fair), which makes finding those with the right combination of skills even more challenging.

CS21-4

A job fair at a Chinese university. Staffing the office was like looking for needles in a haystack...

For 2015, our goal is to have a team of five talented and well-trained engineers at KeZhi. This team will be able to deliver services in EPICS, C/C++ and Java, as well as in FPGA and electronics development. The local team will be complemented by several Cosylab experts that will relocate to China, as well as by consultants from Cosylab’s ranks on special topics such as kernel driver development and timing system adaptation.

CS21-5

The first KeZhi team building. (The first 8 needles. :))

REFERENCES

  1. http://www.gatesnotes.com/About-Bill-Gates/Concrete-in-China
  2. http://csns.ihep.ac.cn/english/
  3. http://english.ihep.cas.cn/rs/fs/ADS/
  4. http://www.theguardian.com/world/2014/mar/19/china-uranium-nuclear-plants-smog-thorium
  5. http://www.iter.org/
  6. http://www.nature.com/news/china-plans-super-collider-1.15603

ABOUT THE AUTHORS

Klemen Žagar joined Cosylab in 1999 and started as a software developer. From there he continued as a software and systems architect and is now Chief Technology Officer. His professional interests are distributed control systems, real-time and networking. In his spare time, he enjoys hiking, cycling and running.

Dominik Peruško joined Cosylab in 2012 as a hardware developer. Currently, he is a hardware architect and project manager, responsible for the development of safety systems in a medical accelerator. He enjoys brewing homemade beer and traveling.

Yang Wenwei joined the Cosylab China branch office (KeZhi Controls) in 2014 as a project manager and senior software developer. He has more than 10 years experience in embedded systems development, and as a project manager and team manager. In his free time, Wenwei enjoys swimming, playing badminton and reading.

Chinese

English

Name

杭州科挚控制技术有限公司Hangzhou KeZhi Controls Technologies Co., Ltd.

Address

杭州市滨江区六和路 368 号

海创基地一幢(北)三楼 B3033

310053 杭州,浙江

Liuhe Road 368

Third floor (North), room B3033

Binjiang District, 310053 Hangzhou, Zhejiang province, China

Office phone

+86 571 8510-2120

Office fax

+86 571 8538-9097

Contact e-mail

info@kezhi-controls.com

ITER: Power for the Future

By: The Cosylab ITER Team

CS21-6aEnergy consumption is growing around the world [1]. At the same time, there is a body of evidence supporting the hypothesis that human beings are adversely affecting the environment and climate. The search for sustainable sources of energy that can meet growing demands while still being conscious of the environment is therefore of extreme importance. Wind-, hydro-, and solar-power have potential, but each comes with their own set of weaknesses, and it is still unclear whether they can meet the necessary energy demands. While nuclear fission-based energy sources can meet the demands, the problem of guaranteed safe, long-term fuel-disposal remains unsolved. Enter the other nuclear process: fusion.

Introduction

Standard nuclear power stations are based on the process of splitting of heavy, radioactive elements releasing energy in the process. At the opposite end of the periodic table, energy is released when lighter elements fuse together. The inputs of a fusion reaction are usually isotopes of hydrogen (deuterium and tritium) and the outputs are helium or other light elements, thereby making fusion a viable source of clean energy. An example of a benevolent fusion reaction is that which drives the Sun, where hydrogen atoms fuse into helium, releasing light and heat in the process.

ITER is the Way

Research of fusion has been going on for many years [2], and the discussions of what is required to turn fusion into a commercial energy-production method started more than 30 years ago. The result of these discussions is the ITER project [3], which is managed and globally funded by six countries (India, Japan, People’s Republic of China, Russia, South Korea and the United States) and the European Union, each represented by their respective Domestic Agency. The ITER tokamak will magnetically confine a deuterium-tritium plasma with the aim of achieving a 10-fold gain in energy - from an input power of 50 MW, the ITER tokamak aims to produce 500 MW of fusion power.

CS21-6b

An artist’s impression of what ITER will look like. The large orange building in the centre is the Tokamak hall. © ITER Organization

Cosylab Involvement

Cosylab has been involved in the ITER project since 2008, working closely with the ITER Organization’s CODAC (COntrol, Data Access and Communication) Section. Cosylab has contributed to or is currently working on three main areas:

  • core controls tools and support
  • diagnostic controls
  • central controls

Core Controls Tools & Support

The central control system responsible for operating the ITER machine is called CODAC. Each of the plant systems (e.g. vacuum, heating, magnets) that make-up the ITER machine will be driven by a local instrumentation and control system called a plant system I&C. Ultimately, ITER will consist of more than 700 plant systems that are being developed by various partners and contractors. These plant systems will all have to be integrated into the ITER machine. ITER has developed a set of standards for the instrumentation & control. Furthermore, to ensure a smoother integration process for the plant systems, the CODAC Section has developed a control system framework called the CODAC Core System. The CODAC Core System is a software package that provides the required environment for developers to develop and test the software in a way that complies with the ITER standards [4, 5].

CODAC Core System

The CODAC Core System was first released in 2010, with stable versions released since 2012. There are 2 minor releases each year that include new components, improvements and bug fixes. If necessary, there are maintenance releases to address any critical bugs that may appear in between the regular release cycle.

The CODAC Core System comes packaged with the Red Hat Enterprise Linux (RHEL) operating system, EPICS [6] and various tools for developing and testing the software for the controllers, central servers and operator terminals [5].

Many of the packaged components have been adopted from the EPICS community and adapted for ITER, in collaboration with other EPICS users. This applies to the CODAC services for operation (e.g., operator HMI, alarms, archives), while other components have been custom-developed for the ITER project (e.g., Self-Description Data configuration toolkit) [5].

The CODAC Core System has thus far been used for the production of the first I&C applications in Cadarache for the monitoring of power stations.

The CODAC Core System has a user-base of around 60 institutes and companies [5].

The current version of the CODAC Core System is 4.3 and version 5 will be released early in 2015. Version 5 will include RHEL 6.5 and EPICS 3.15.

Cosylab has been involved in various aspects of the development of the CODAC Core System.

Training & Support

CODAC Training happens regularly at the ITER Organization in France, as well as at the various Domestic Agencies. There are also various training videos available for registered CODAC users.

Cosylab has developed the bulk of the training materials and is also handles all support queries.

Using the CODAC Core System

If you are interested in using the CODAC Core System, visit the ITER CODAC Core System page for more details [7].

Diagnostics Controls

Diagnostic systems are a critical part of any big science machine, and no more so than at ITER. Extensive diagnostics are needed to meet the requirements for machine operation, protection, plasma control and the physics experiments that will be carried out [8].

The 45 ITER diagnostic systems will need to satisfy strict requirements with respect to nuclear safety, operating in very harsh environments and instrumentaiton & control. Most of the diagnostics systems will be provided by the ITER Domestic Agencies. From an I&C point of view, a large number of high performance fast controllers must be delivered and the choice of these is based only on guidelines published by the ITER Organization [8].

Therefore, at the request of the Domestic Agencies, ITER Organization has created several diagnostics use case examples to assist with the understanding of diagnostics Plant System I&C and the associated deliverables. The use cases are fully implemented and documented to further help the Domestic Agencies and their partners to satisfy the ITER diagnostics requirements [8].

This will ultimately simplify acceptance testing and commissioning for both Domestic Agencies and the ITER Organization [8].

Cosylab is working on use cases for multi-channel high performance data and image acquisition, data processing as well as real-time and data archiving aspects [8].

Central Controls

The general scenario during the operation of the ITER machine is that the physicists design the experiments (“pulses”) in consultation with the plant system specialists. The experimental parameters are then “translated” into machine parameters, i.e. specific plant system configurations, which are applied during the various stages of the experiment.

Cosylab is involved with the development of CODAC Operation Applications. These applications include Supervisor Applications to configure and coordinate plant systems during machine operation, Scheduler Applications to compose pulse schedules and schedule-specific plant system configuration, Remote-Participation Applications to supports off-site collaboration, and Real-time Software Applications to operate the real-time pulse schedule and perform other real-time plant system tasks.

Conclusions

Cosylab is a long-time partner of the ITER project and we have helped lay the controls groundwork; specifically, we have helped ensure that the infrastructure for the control system is completely pluggable. We believe our contribution will bear fruit as ITER gets closer to being fully realized.

However, CODAC is not just useful for ITER, and other large experimental physics facilities, such as the European Spallation Source (ESS), are also considering using it.

CS21-7

Tall cranes going up: In one corner of the Seismic Pit, a tower crane is going up to prepare for the next phase of work: the pouring of the B2 slab that will act as a floor to the three buildings of the Tokamak Complex. © ITER Organization

CS21-8

A detailed cutaway of the ITER Tokamak, with the hot plasma, in pink, in the centre. © ITER Organization

REFERENCES

  1. World energy consumption. http://en.wikipedia.org/wiki/World_energy_consumption
  2. http://en.wikipedia.org/wiki/Timeline_of_nuclear_fusion
  3. http://www.iter.org/
  4. Approaching the Final Design of ITER Control System, Anders Wallander, Lana Abadie, Franck Di Maio, Bruno Evrard, Carlos Fernandez Robles, Juan Luis Fernandez-Hernando, Jean-Marc Fourneron, Jean-Yves Journeaux, Changseung Kim, Kirti Mahajan, Petri Makijarvi, Sopan Pande, Mikyung Park, Vishnukumar Patel, Pierre Petitbas, Nicholas Pons, Antoni Simelio, Stefan Simrock, Denis Stepanov, Nadine Utzel, Antonio Vergara-Fernandez, Axel Winter, Izuru Yonekawa. ICALEPCS ‘13, October 8 2013, San Francisco. http://accelconf.web.cern.ch/AccelConf/ICALEPCS2013/talks/tucoaab03_talk.pdf
  5. The CODAC Software Distribution for the ITER Plnat Systems. Franck Di Maio, Lana Abadie, Changseung Kim, Kirti Mahajan, Denis Stepanov, Nadine Utzel. 14th International Conference on Accelerator and Large Experimental Physics Control Systems (ICALEPCS 2013) 6-11th October 2013, San Francisco, U.S. http://accelconf.web.cern.ch/AccelConf/ICALEPCS2013/posters/moppc079_poster.pdf
  6. EPICS - Experimental Physics and Industrial Control System. www.aps.anl.gov/epics/
  7. CODAC Core System https://www.iter.org/org/team/chd/cid/codac/coresystem
  8. Diagnostic Use Case Examples for ITER Plant Instrumentation and Control. S. Simrock, P. Patil, V. Martin, L. Abadie, R. Barnsley, L. Bertalot, P. Makijarvi, D. Makowski, J.-Y. Journeaux, R. Reichle, D. Stepanov, G. Vayakis, I. Yonekawa, A. Wallander, M. Walsh. 14th International Conference on Accelerator & Large Experimental Physics Control Systems, 7-11 October 2013, San Francisco, USA. http://accelconf.web.cern.ch/AccelConf/ICALEPCS2013/posters/moppc014_poster.pdf


Cosylab Develops Prototype Low-latency Digital Video for ITER Remote Handling

By : Kevin Meyer (Cosylab), Tom Slejko (Cosylab)

The ITER Remote Handling division is responsible for an extremely critical phase of the ITER assembly and maintenance: the task of remotely installing and maintaining core components of the ITER machine, inside the vacuum vessel. They need to manipulate and position components weighing up to 50 tons with high precision in an environment that does not allow direct operator access.

While the core responsibility of this precise control rests with the design of the robotic manipulators, an important aspect of any remote handling activity relies on being able to see what’s going on - remote vision.

CS21-9

The remote handling system is a critical part of the ITER machine. © ITER Organization [1]

Let Me Be Your Eyes

Earlier this year Cosylab was contracted to develop a prototype remote vision system solution for ITER remote handling. The requirements were simple: Develop a system that would be able to support remote handling operators to select from 100’s of cameras and display their video signals on any of dozens of viewing monitors, while ensuring that the video latency (the delay between what the camera sees and what is shown on a monitor) is < 100 ms over digital networks.

After some initial research where we examined current trends in the digital video industry, we identified the H.264 video standard as a likely candidate, because it has been fine-tuned for low latency.

To address the low-latency requirement, we selected appropriate parameters in the x264 encoding library to enable the so-called “zero-latency” mode. The H.264 video compression format supports many compression modes, ranging from a low-bandwidth mode that uses information between many buffered frames to achieve the lowest bandwidth at the expense of latency to a low-latency mode that encodes frames without buffering, at the expense of bandwidth.

We also realised that to solve potential bandwidth issues, we should use UDP (User Datagram Protocol) multicast. This transport protocol allows multiple listeners to “subscribe” to a multicast address (multicast group), and the network hardware (routers) would manage the sending of data only to subscribers. This would allow multiple receivers (monitors) to subscribe to the same video source (a camera) without increasing the network load from the source.

The resulting compressed video frame is still larger than a single UDP packet, which means that multiple UDP packets are needed to transport a compressed frame. While UDP supports multicast and has lower overheads than connection-based network transport protocols (e.g. TCP/IP sockets), it is vulnerable to packet loss or packets arriving in a different order than they were sent. If either of these events were to happen, it would disrupt the video frame decoding, causing video display corruption and delays while the data streams resynchronised.

To address these issues while still preserving efficient transport, we adopted the standard practice of using RTP (Real-time Transport Protocol) to encapsulate the data stream in multiple UDP packets. RTP ensures that packets are received in the correct order, with lower overhead than connection-based protocols.

Getting It Under Control

Having adopted the industry-standard solutions of RTP encapsulated H.260 compressed video as a viable digital video solution, we next needed to implement a control protocol to support configuration and control of the camera drivers (to control parameters like frame rate and image size) and the monitor applications (to control parameters like which camera to subscribe to).

Since we knew that some final camera driver solution implementations would be expected to be implemented on embedded systems, we wanted a lightweight protocol that could be implemented with minimal overhead, while still supporting atomic configuration (that is, setting multiple parameters simultaneously, such as frame-rate, video width and height). For this reason we adopted the JSON (JavaScript Object Notation) to represent configuration commands and results. JSON is a very lightweight text-based mechanism to exchange attribute-value pairs, and can be implemented with a simple socket-based client server architecture (each configurable application implements a server to which clients connect and send & receive JSON-formatted commands and results).

The resulting full-stack solution therefore consists of client control applications (GUIs) that connect to camera drivers to control camera operations (in principle including pan, tilt & zoom) and to video monitor applications to control which camera it should subscribe to.

Multicast Magic

The multicast solution takes advantage of modern routers that support IGMP (Internet Group Management Protocol) to automatically manage multicast traffic flow. By monitoring multicast subscription requests (sent from monitor applications when they subscribe to a multicast address), routers are able to route traffic only where there are active subscribers. No matter how many multicast group subscribers there are, only one stream is created at each branch in the network route - network traffic is never duplicated on any route segment - and if there are no subscribers to a particular camera address, the multicast traffic does not travel beyond the first router.

On the Shoulders of Giants

In order to deliver most efficiently, we selected the gstreamer framework to do the bulk of the heavy lifting. The gstreamer framework is a well-supported open source framework for multimedia applications. Its pluggable architecture means that it is very easy to write plugins to provide new functionality and the numerous existing plugins provide access to a wide range of existing hardware, such as Video4Linux, IEEE-1394 FireWire, etc.).

We used existing gstreamer plugins for each stage of the video transport pipeline: For video compression and transport, we used v4l2src to acquire images from USB cameras, x264enc for H.264 encoding, rtph264pay for RTP encapsulation and udpsink for packet transport. Likewise for video decoding, we used udpsrc for transport, rtph264depay to decapsulated the RTP data, ffdec_h264 to decompress the H.264 video and xvimagesink for display.

When combined with the JSON protocol developed for the purpose, the gstreamer framework allowed us to develop a complete low latency digital video transport solution.

Initial Results

For a 25 fps 800x600 pixel test image (colour bars with a section of random static) (Figure 2), the initial testing at ITER demonstrated that the compression-transport-decompression delay is approximately 70 ms, with a compressed bandwidth of about 250 kB/s per camera.

Rough calculations with these figures indicate that ITER remote handling should be able support 100 active cameras (actually, 100 active monitors) on a gigabit network backbone.

CS21-10

The test image.

REFERENCES

  1. http://www.iter.org/mach/remotehandling

ABOUT THE AUTHORS

Kevin Meyer joined Cosylab in 2012 as a senior software developer. He has a PhD in physics and extensive experience in many diverse fields, a real Jack of all trades. Before joining Cosylab, Kevin worked for a number of different companies, either as a software developer or specialist software consultant. Currently, he is project manager for the ITER Operations project, that is responsible for working with plasma fusion experts to design the future ITER plasma pulse schedule and scheduling system. In his free time, Kevin enjoys reading, walking, and generally enjoying the little things in life.

Tom Slejko joined Cosylab in 2009 as a software developer, and then moved through the ranks of control system engineer, control system architect and now project leader and architect for the PSI SwissFEL project. Tom’s professional interests include distributed real-time control systems, Linux kernel and embedded development. In his free time Tom enjoys paragliding, motorbikes, hiking, climbing and anything else that spikes his adrenaline levels :)

Partner News

First Future Circular Collider Study Annual Meeting

CS21-11oPAC (optimization of the performance of any Particle ACcelerator) falls within the FP7 Marie Curie Initial Training Network (ITN) scheme. oPAC is working with CERN on the development of the next generation of accelerators within FP7 ITNs. The Future Circular Collider study is closely linked to this effort.

The first Annual Meeting of the Future Circular Collider Study [1] will take place from 23 to 27 March 2015 in Washington D.C.

This meeting will be an important milestone to conclude the first, exploratory phase, leading to the identification of the baseline for the further study. Organized as an IEEE conference, it will provide the opportunity for re-enforcing the cohesion of the community and to catalyse cross-fertilization within the FCC study.

This event will follow the traditional layout of plenary and parallel sessions with invited contributions. Plenary sessions will give an overview about the ongoing activities across all parts of the study and serve informing study members about the main boundary conditions and working hypothesis. Parallel sessions will focus on specific areas of the study and a limited number of contributed 10’ presentations are foreseen, to enable communication of key findings of ongoing work with significant impact on the subsequent study phases in an efficient way. We encourage submission of proposals which will be reviewed by the organising committee.

Satellite meetings for related projects and governance bodies will be included the program. Participation of industry is highly encouraged and supported via a dedicated industry track and a micro exhibition, focusing on superconducting cable technology. Communication and equal opportunity aspects will be addressed in dedicated working group meetings.

The detailed agenda is under preparation and a preliminary version is available on the FCC Week website [1] where some practical information regarding travel and accommodation can also be found. Registration [2] is mandatory and has already started. Please note that the early registration deadline is 5 January 2015.

REFERENCES

  1. http://cern.ch/fccw2015/
  2. https://indico.cern.ch/event/340703/registration/register#/register

Combining Lasers and Accelerators to Train the next Generation of Researchers

CS21-12LA³NET focusses on the exploitation of lasers for applications at accelerator facilities for ion beam generation, acceleration and diagnostics. LA³NET is part of the FP7 Marie Curie Initial Training Network (ITN) scheme.

The Spanish Pulsed Lasers Centre (CLPU) in Salamanca, Spain recently hosted LA3NET’s Advanced School on Laser Applications at Accelerators attracting over 70 participants from all over the world.

Cosylab’s Frank Amand presented a talk on Scientific Entrepreneurs and sat on a panel for an interactive session with the audience to explore aspects of funding proposals. He also represented Cosylab on the supervisory board looking into training of the fellows and future events.

CLPU Director Prof. Luis Roso opened proceedings along with Dr. Enrique Conejero with introductions to Salamanca, the hosts and the school itself. This was followed by a brief overview of LA³NET before the main lecture programme started. The first day included talks on an introduction to lasers, the history of accelerator development in Europe, accelerator applications, as well as beam generation, acceleration and diagnostics – all given by internationally renowned lecturers.

Day two featured lectures on laser ion sources, photo injectors and Free Electron Lasers, in addition to a two-hour study session giving delegates a “hands-on” look at some of the topics covered. An outreach talk about “attosecond science” by Prof. Luis Plaja in the evening on the main University of Salamanca campus attracted more than 100 students from the university and local high schools in addition to the school participants.

The following days covered advanced topics in ion and electron acceleration, commonly used simulation codes for accelerator design and optimization, as well as industry applications of accelerators and lasers. This was complemented by a Laserlab-sponsored visit to the facilities at CLPU, a second study session and a lively poster display and industry exhibition, sponsored by Danfysik. The School drew to a close with talks on THz applications, compact X-ray sources and the Extreme Light Infrastructure project. The School stimulated many fruitful discussions throughout the week and was an excellent addition to the many scientific events the network has organized to date. All presentations can be found on the School’s Indico site.

The LA³NET project is coordinated by the Cockcroft Institute/University of Liverpool and led by Professor Carsten P. Welsch with assistance in delivery from Dr. Rob Ashworth and the EU TEAM of the Cockcroft Institute. LA3NET is indebted to CLPU staff and in particular Ms. Yaiza Cortés for the local organization and support. The effort of all lecturers for their hard work and excellent contribution to make the School such a success is also appreciated.

The Photo Board

CS21-13

Erik Steinman (Danfysik) borrows a plug adapter from Cosylab at LINAC14 in Switzerland. While the roller banner advertises “Full of power”, clearly this power cannot be accessed without Cosylab’s support :)

CS21-15

George Neil (Jefferson Laboratory) armed with a Cosylab mug and kitted out in a Cosylab T-shirt appears ready for anything that his day of kayaking might throw at him.

CS21-16

Han-Sung Kim (KOMAC) sent in this picture of his wife, Hee-Young Ju, in our T-shirt.

CS21-14

Cosylab signs multi-year contract with PSI for contributions to the SwissFEL control system: The photo shows the leader of the SwissFEL project, Dr. Hans-Heinrich Braun, shaking hands with Cosylab CEO Mark Plesko, after a long and exhaustive negotiation session for the framework agreement for "control system services for SwissFEL". Both were so tired that they had to relax on the CosyCouch, conveniently provided at the Cosylab booth during the FEL conference in Basel. We can only speculate what was negotiated. Maybe the PSI wanted to pay more for Cosylab services, but Cosylab stood firm and didn't want to raise it's price. Or was it the other way around? In any case, both look very happy and friendly, so they must have reached an amiable solution.

Did you receive a Cosylab T-shirt at a conference? We want to see you, your husband or wife, children or pets(!) in the T-shirt. We will send you another T-shirt (there are many designs!), and we will publish your picture in the next issue of Control Sheet.

Send pictures to controlsheet@cosylab.com.

New Year's Message

As 2014 ends, we say to all our colleagues and friends that once started out as our “customers”:

  • Thank You
  • Merci
  • 谢谢
  • ありがとうございます
  • 감사합니다
  • Danke
  • Dziękuję
  • Mulțumesc
  • Tack
  • Dank u wel
  • Grazie
  • Hvala

To our new customers, we look forward to working with you in 2015, to help you bring your machine under control.

May the beam with you!

Best wishes

The Cosylab Team

p.s. We hope you enjoy our New Year Card at http://www.cosylab.com/ny2015/.

back to previous content

Download a printable (pdf) version of Control Sheet no. 21 here.

To subscribe to Control Sheet, send an email to controlsheet@cosylab.com with "Subscribe" as the subject.