This subproject is one of many research subprojects utilizing theresources provided by a Center grant funded by NIH/NCRR. The subproject andinvestigator (PI) may have received primary funding from another NIH source,and thus could be represented in other CRISP entries. The institution listed isfor the Center, which is not necessarily the institution for the investigator. Objectives In this past year the two automounters on beamlines X12-B and X29 were reliable and attractive to users. Therefore our principal objective has become to develop the automounter-user community further.
We aim to support investigators interested high efficiency data collection by loaning equipment and providing training. We encourage collaborators in our Mail-in program to adopt robot-compatible specimen-mounting techniques. A second objective is to support the broader efforts to offer remote-participation and automatic data-collection capabilities. We work continually to make numerous small improvements. A specific technical objective is to speed up the specimen mount / dismount cycles to 5s. In addition we will deploy a third automounter on beamline X12-C, and more generally will work towards equipping all PXRR facilities with automounters. Results We find the cryogenic specimen-mounting system developed at the Lawrence Berkeley National Laboratory to be safe and reliable. We are part of a national consortium who are adapting the system and sharing improvements. Control of the entire experiment automounters, diffractometer, and beamline is accomplished through our CBASS data-collection suite described above. The modularity and robustness of this architecture allowed us to implement methods that allow users to switch easily between robot-assisted and manual experiment control, as well as to launch the automounter quickly. Finally, the design is reliable, matching that of the ALS program with fewer than 0.1% lost crystals, compared to an observed failure rate of ~2% in manual specimen mounting. Even novice users are accepting the hockey-puck-like ALS cassettes with their simple set of tools. In support of our outreach program we have purchased five loaner kits of apparatus, comprising a shipping dewar, specimen holders (pucks), and tool sets. Demand is high, and two more of these are being assembled now. The kits are on the road nearly all of the time, and 19 different groups have borrowed them in the last year, returning them, filled with crystals to be used in experiments a total of 26 times. Ten of the visits were made by seven mail-in collaborators; the rest visited to do the work themselves. In addition, six groups have visited with their own apparatus on eight occasions. It requires an experienced local staff to train the visitors in use of the apparatus; one well-trained technician and two scientists perform this service. In all, during the last year, there have been 81 robot runs, 1200 crystals mounted on average twice each, about 24,000 diffraction images were measured in an automated way. It should be clear from the discussion in the software and Mail-in sections above, that the automounter program is essential to progress in the sort of remote operations some users would exploit, and the fully automated capability that the dna software will allow. Plans The third PXRR automounter will be installed at beam line X12-C soon. Because the lift table here cannot carry the diffractometer, we are developing it as a quasi-autonomous and self-aligning unit. An equivalent system will be developed eventually for the new X25 microdiffractometer. We believe that the combined mechanism, position sensing, and control software will be useful now, and also for future instrument developments. To support completely unattended operation we are developing an interactive status handler for our automounters. When fully developed, not only will it keep experimenters informed about the current status of the robot and its payload, but also it will provide detailed escape and recovery pathways when unexpected events occur such as program glitches, remote re-connections, or power or mechanical failures. While the first goal of the envisioned state-diagram analysis tool is to preserve specimens, it also will assist facility operators in decision making under normal and recovery conditions. Once we have gained experience in use of such methods with the automounter, we imagine extending them to include additional facility sub-systems. The incorporation of 3D machine vision is a long-term goal. A broad array of technical improvements will sustain and improve robot-assisted data collection. We will strengthen reliability further by building critical spares, refining operational protocols, and translating operational experience into hardware and software upgrades. We are working now to improve the speed and accuracy of automatic crystal centering with both software and optical improvements. A significant effort is work to accelerate the crystal mounting / dismounting cycle time. We are evaluating the performance of data-code readers. Significance PXRR user community demands access to the intense undulator-derived beams at X29 and X25. To increase their access, we work to couple the availability of our dipole beamlines with the power of the undulators. Automounter-aided work at the X12-B dipole allows easy, sometimes unattended screening of multiple specimens for quality. The best of these can be taken to the undulators. In addition, the automounter at X29 increases the throughput there for crystallographic projects for which the dipoles are too weak. In summary, the availability of these automounters support the optimal use of all of the PXRR facilities by increasing the pace and quality experiments performed there.
Showing the most recent 10 out of 167 publications
