2001 ATF Newsletters

Jan | Feb | March | April | May | June - July | Aug | Sept | Oct | Nov - Dec

 

Contents

1. Introduction

5. Computer Control System
6. Diagnostics Test
3. VISA Experiment  
4. Fast Optical Experiment  

Greetings everybody,

The big news of this ATF Newsletter is saturation and measurement of harmonics in the VISA experiment, AE24. The announcement came in an email sent by representatives of the four laboratories that collaborate in this experiment, headed by its spokesperson, Prof. Claudio Pellegrini. The text of the announcement  follows below.

The ATF Computer Control system is making excellent progress, as reported below. One feature of the ATF's computer control system, innovative use of Mathcad as on-line software tool,  generated a cover story for the April issue of Scientific Computing and Instrumentation magazine. See Bob's report below.

Ilan Ben-Zvi.

Back to Top

VISA Announcement

On behalf of the VISA collaboration, we are very pleased to announce that VISA has recently demonstrated a gain of 2x10^6, and an exponential growth rate (intensity gain length) of 18.5 cm. There is also evidence of  saturation of the FEL intensity at about 3.6 m from the undulator entrance.

The experiment was done using a 71 MeV high brightness electron beam, with peak current of 80 to 90 A, normalized projected emittance of about 1.5 mm mrad, at a radiation wavelength of 830 nm. The results indicate that the gain depends on the slice emittance, and that this is substantially smaller than the projected emittance.

Spectral measurements of the radiation at saturation show large second and third harmonics, in addition to the fundamental. The third harmonic is many orders of magnitude larger than the corresponding spontaneous intensity.  Additional measurements will be done during the next few months to obtain more data and more information on the spectral and angular characteristics of the FEL radiation.

The VISA (Visible to IR SASE Amplifier) experiment was designed, built, installed and carried out by a BNL-LLNL-SLAC-UCLA collaboration. The experiment is being done using the Accelerator Test Facility, National Synchrotron Light Source, Brookhaven National Laboratory.

Claudio Pellegrini, UCLA, Spokesperson for the VISA collaboration

Max Cornacchia, SLAC

Karl  Van Bibber, LLNL

X.-J. Wang, BNL


Back to Top

VISA Experiment, AE24 (Reported by Aaron Tremaine)

 
Some good results over the last few weeks.

During the last two run weeks energy vs. distance measurements were done after the system was peaked up. A power gain length of 18.5cm was measured and a gain of 2*10^6 attained. Also, through Section 4 the energy was observed to deviate from exponential, indicating the onset of saturation. The system was run ~70A and emittances 1-1.8mm-mrad were measured. Beam measurements are done after the system is peaked and will be continued in the future.

In addition to the energy results above, harmonics were measured. We have captured in a single shot spectrum not only the fundamental (840nm)  but the 2nd (420nm) and 3rd (280nm) harmonics. Initial energy measurements show the 3rd harmonic to have ~1% of the fundamental.

The system is very repeatable. After ending a night with strong SASE lasing, we can  return to good lasing in a few hours once the beam is propagated to the undulator. On the harmonic measurement run it took less than 10 minutes to attain the gain described above. A little luck was probably at hand here, but I think this is evidence to the good performance of the ATF photoinjector.

Back to Top

Fast Optical Detector Experiment, AE23 (Reported by Don Lazarus)

Fast Detector Group Progress in February-March 2001 Efforts in this period were focused on streak camera measurements of the rise time of electro-optical signals induced in LiNO3 crystals by the ATF beam. The goal was to obtain measurements with close to 2ps resolution. Although calculations indicated that our sensitivity would be inadequate for single shot signal detection with the streak camera, we had detection capability with signal averaging. This would entail a 20 ps jitter induced by triggering the streak camera with a digital pulser which was in turn triggered by the stripline signal. It was decided to mix slightly out of time light from the ATF source laser with our our light signal in order to provide a time reference marker which would be used to align streak camera sweeps off-line to overcome the trigger jitter problem. We failed to observe a signal in March running. Cumulative light losses at the many optical surfaces in our setup reduced our sensitivity to a marginal level. In addition, laser noise was significantly worse than previously experienced. We will have most optical surfaces coated prior to the next run. More than a factor of two improvement in transmission is expected and transmission measurements will begin as soon as the coating process is completed. More laser power and hence a greater signal to noise ratio will eventually be achieved with a pulsed laser. In addition to the signal to noise problems, the source laser reference marker suffered dispersion on the order of nanosecond in the 65 meters of multimode fiber from the laser room to our apparatus. Signifcant time was spent attempting to understand this, including a check on the quality of the coupling into our fiber by M. Babzien. Subsequent discussions with the fiber manufacturer indicate that this degree of dispersion is consistent with specifications. We have ordered fiber with radial gradient in index of refraction to minimize dispersion. We have also discovered an on site streak camera with more relaxed triggering requirements that may be available to us.

Back to Top

ATF Computer Control System (Reported by Bob Malone)

As reported in an earlier ATF Newsletter, the new control system computer has arrived.  The new machine is an IBM Netfinity 8500 server which eventually will replace our almost 11-year old VAX 4200.  The following table compares the two machines and illustrates at least an order of magnitude improvement in all key components:

Item Present VAX system  New IBM replacement
     
CPU (1) VAX KA660 CPU (2) Pentium III Xeon CPUs operating in SMP (symmetric multiprocessor mode);Expandable to 8 processors
Clockrate 114 MHz 700 MHz (each CPU)
Cache 256 kB 2 MB
Bus  32-bit Q-bus; 64-bit PCI
Expansion slots  12
Main memory 56 MB (64 MB max.) 1500 MB (16000 MB max.)
Disk space 1.7 GB fixed 18 GB Hardware hot swap RAID; Expandable to 4 channels RAID level 5E supporting up to 60 drives
Network (1) 10 MHz Ethernet               (3) 100 MHz Ethernet NICs
Power supply (1) fixed  (3) hot swap, redundant fail-over
Cooling (2) fans (fixed)  (6) fans; hot swap
Tape storage 7 GB max. / tape 80 GB max. / tape
Data Acquisition CAMAC 5 MHz serial highway CAMAC; 100 MHz Ethernet
Operating System OpenVMS Linux with SMP support
Control Tools Vsystem Vsystem

So far the performance of the new system has been excellent.  For example, the time needed to build the 8,000 channel main accelerator database on the VAX took 14 minutes while on the new IBM machine, only 5 seconds are required.

The very large task of migrating all the software developed over the last 10 years continues.  Some of the new software tools which have been developed to help in transferring the existing control system to the new hardware platform include:

·        New scripting tools which can read ASCII configuration files and automatically generate new database input description files as well as associated control system display windows.  This tool reduces the time needed to produce many of the drawings needed for operator displays.  Rather than manually drawing/editing many windows which have similar structures, the script repeatedly edits template displays, making the necessary substitutions.

 This technique has been demonstrated in migrating the Darlington power supplies, where the script constructed 200 new displays and supporting database descriptions in ~10 minutes.  Only two drawings needed to be created manually as input to the script, the rest being developed automatically from the input descriptions.  This single script can handle 20% of the displays needed for the ATF control system.  The same technique will be applied to other ATF subsystems.

·        A quality assurance script which scans all operator displays in the system looking for non-existent database channels.  The script also detects broken links from one display to another and will flag out any “orphan” displays which have no references from any other display.

 Personnel:

Karl Kushe, longtime ATF collaborator from STI has joined the ATF staff part-time and will devoting 50% of his efforts on control system migration issues.  Karl is developing a spreadsheet-based management system which will become the input for the automatic database/drawing generation script described above. The result will be a new control system which will be easier than ever to update and maintain.  Karl has also handled the ordering and installation of new Ethernet hubs which will drive the CAMAC Ethernet crate controllers which will eventually replace the serial highway.  In addition, he as done great work in establishing a better working area for ATF experimenters using the space in the main control room.

ATF control system in the news:

The cover story for the April issue of Scientific Computing and Instrumentation magazine highlights ATF’s innovative use of Mathcad as on-line software tool for control and analysis.  You can visit SC&I’s website at www.scimag.com.  One of the pictures used in the article is at:

ATF_control 

Back to Top

Diagnostics Test (Reported by Alex Murokh)

 Vitaly and myself did another round of diagnostics/YAG test.  The results are presented in the file below.  Most of the questions that motivated us for the test remain wide open, at least till the thorough data analysis is done.  I hope we’ll find some time to get a closer look at the data soon.

diagnostics_test 

Back to Top

 

Top of Page

Last Modified: December 3, 2007
Please forward all questions about this site to: Vitaly Yakimenko