Request for Quotation 34-ID CXD Diffractometer
March 8, 1999
1. Overview
The front experimental hutch (34ID-C) of the APS 34-ID beamline will perform experiments using Coherent X-ray Diffraction (CXD) as its sole function. The unique demand of CXD is that a very small beam (typically 5 microns) be defined a short distance (10 to 20mm) in front of the sample. Diffraction from the sample is angularly resolved by a detector that is typically quite far away, at the end of a detector arm that is up to 3m long. The three roles: i) support of the beam-defining apertures, ii) orientation of the sample in the beam, and iii) angular positioning of the detector arm, are the primary functions of the CXD diffractometer described in this RFQ.
The 34-ID beamline is also special in that it is planned to operate two experiments simultaneously. Its undulator beam is divided by a mirror at 30m into two beams that supply independently shuttered hutches. Since the CXD takes place in the first hutch, centered at 51m from the source, it is inconvenienced by the permanent presence of a shielded pipe carrying the second beam. The diffractometer must, therefore, operate in a space that does not encroach that of the pipe, whose outer wall lies just 180mm from the diffractometer's center. For this reason we will employ the style of the kappa-axis diffractometer for both sample and detector motions.
Because of the space restrictions, the diffractometer must be able to provide both vertical and horizontal scattering-plane geometries. In particular, horizontal and vertical detector motions must be available on an equal footing. In the double-kappa configuration, scattering planes in between horizontal and vertical are also provided. Finally the detector arm should be rotatable about its own axis: this has several advantages such as facilitating polarization analysis or alignment of exit slits or linear detectors along selected reciprocal-space directions. Since it is impossible to make a fully rotatable arm that is also completely rigid, relaxed design tolerances have been specified.
It is expected that the diffractometer specified in this document can be fabricated from a standard off-the-shelf kappa-goniometer attached to the side or base of an L-shaped support. The most suitable position for the detector arms is on the back of the support. Although these arms are in a custom arrangement, the specification can be met using off-the-shelf rotary tables. Because their functions are quite separate, it is conceivable that the sample goniometer and the detector-motion axes might be provided by different vendors.
Figure 1. Layout of hutch 34ID-C for coherent diffraction experiments at APS. The overall outside dimension is 7.5m long by 3.4m wide by 4.25m high.
2. Floor Space Requirements
Figure 1 is a sketch of the CXD hutch layout where the diffractometer will be stationed. The close collision possibility with the microfocus diffraction (MFD) white beam pipe is clearly apparent. When the detector arm is extended to its maximum length of 3m, its angular range is constrained by collisions with the far wall and with the roof. When the arm is shorter, there will still be a collision in the horizontal plane that is internal to the kappa motion, but no restrictions on the vertical angular range.
Figure 1 indicates that the diffractometer is to be movable back and forth along the direction of the beam, with a total range of 1.0m travel. The transport system that allows this and also the installation and removal of the table is not shown. The diffractometer will couple to a UHV sample environment in the front position and operate with samples mounted in air or local enviroments in the back position. Finally, there is need for an optical bench opposite the diffractometer, rigidly connected to the support, and a rigid stand for holding input beam components. The input beam stand must disassemble sufficiently to slide under the UHV sample preparation chamber in the front position. Further details are given in section 6 below.
Figure 1. View 1. View 2. Sketches from two different viewpoints of the `double kappa' diffractometer specified in this request for quotation. The setting shown has omega_D at 80° and kappa_D at -100° . Two beam pipes and a typical experimental vacuum system are indicated schematically for reference.
3. Axis Specification
Figure 2 is a sketch of how we envision the diffractometer will be achieved using a `double kappa' configuration. A `double kappa' instrument has a total of six principal axes, three for the sample and three for the detector. The three sample axes follow a standard kappa sequence: omega, kappa, phi. The three detector axes follow the same sequence: omega_D, kappa_D, phi_D. The angle of inclination of the kappa axes with respect to both omega and phi, here called `alpha', must be greater than 45° and is usually chosen to be 50° or 60° , but could adopt a different value if necessary. The angle of inclination of the kappa_D axes with respect to both omega_D and phi_D, here called `alpha_D', must be precisely 45° because this serves to protect the adjacent MFD white beam pipe from infringement. This means that the detector arm cannot be scanned beyond the straight position in the horizontal direction, except by adjustment of the support table.
The design specifications of the six principal axes are listed in Table I. Standard equations are used to convert between the {omega, kappa, phi} setting and the more commonly used {theta, chi, phi} notation. These conversions will be made in user software, so it is not necessary for this to be a function of the diffractometer control hardware. Because scans of the values of the converted angles will be made, it is necessary that simultaneous motions be possible on multiple axes, as detailed below.
In figure 2, the detector motions are shown on the outside of the main vertical stand. This is to facilitate and conserve the mutual alignment of the omega and omega_D axes, since these have to be the most precise of all combinations. The reversed stacking also allows the detector arm to be supported as far away from the instrument as possible, which will minimize its gravitational deflection (see below). This will also reduce the size of counterweights associated with the detector arm.
|
Axis Name |
Accuracy |
Resolution |
Max speed |
Settling time |
Encoder resolution |
|
Omega |
±0.001° |
±0.0002° |
10° /sec |
0.5sec |
0.0002° |
|
Kappa |
±0.005° |
±0.001° |
5° /sec |
0.5sec |
|
|
Phi |
±0.001° |
±0.0002° |
20° /sec |
0.5sec |
|
|
Omega_D |
±0.001° |
±0.0002° |
3° /sec |
1.0sec |
0.0002° |
|
Kappa_D |
±0.005° |
±0.001° |
3° /sec |
1.0sec |
|
|
Phi_D |
±0.005° |
±0.001° |
10° /sec |
0.5sec |
|
Table I. Critical parameters, resolution, speed, etc, for the six principal axes of the double-kappa diffractometer for coherent x-ray diffraction at 34ID.
Notes on the axis specifications in Table I:
i) It is desirable, but not a requirement, that motor drives can be disengaged for manual positioning. If the motors are permanently engaged, it is necessary to provide for some conveniently-located manually-operated actuation method, such as by joystick.
ii) Optional on-axis angle encoders have been included on omega and omega_D axes only. Space is too confined for this to be possible on kappa and phi.
iii) The settling time refers to small moves less than 1° , or to the final deceleration period only following a large move.
All of the sample axes must be suitably counterbalanced so that large static torques are avoided on their drive mechanisms. This is not required for the temporary loading of the sample axes during installation of the vacuum system (see below). The counterweights of the detector arms should be coarsely adjustable by adding and subtracting weights or by moving them. The maximum detector-arm load is an X95 rail used as a flight path (see below) plus a 10kg detector located up to 3m from the center point.
4. Sample Mount Requirements
The sample mount requires a motorized XYZ stage attached to the phi axis, which is removable in order to provide space to couple to the vacuum system. The range of motion should be ±5mm in each direction with 1m m resolution. The positioning stage should be sufficiently compact to allow a Huber 1004 (or equivalent) goniometer head, mounted on a standard ACA/IUC fitting, to operate at its nominal working distance.
For connection to the vacuum system, when the XYZ stage is removed, the phi axis should present a flat mounting surface and bolt circle, preferably on both faces. A clear hole of at least 40mm diameter must pass to the back side. This hole will be used for attachment of a counterweight to offset the vacuum loading.
Clearance around the sample position (with the XYZ stage dismounted) is mainly determined by the vacuum system connections, which are shown schematically in Figure 3. The phi axis must couple (through a vacuum feedthrough) to the vacuum chamber roughly with the dimensions shown. This requires clearance of a cylindrical space of length A=109mm out to a radius of A2=50mm above the face of the phi axis. The location of the lowest vacuum flange (254mm diameter) constrains the amount of space that can be used for construction of the Kappa axis to a clearance of B=75mm out to a radius of B2=127mm. This latter constraint also applies when kappa moves away from the fully extended position to displace the phi axis from the vertical by up to ±20° , meaning that the body of the kappa arm must be suitably rounded. Finally, there must be at least C=200mm clearance out to a radius of C2=75mm between the face of the omega circle and the sample position.
Figure 3. Critical clearances around the sample, arising from the coupling to the vacuum system. The layout of the diffractometer axes is schematic. The minimum distances are cylindrical exclusion volumes defined by: A=109mm to radius A2=50mm, B=75mm to radius B2=127mm, C=200mm to radius C2=75mm.
Loading on the sample axis requires a double specification. Under normal operation, and when the sphere of confusion (SOC) is determined, the maximum sample weight is 1kg (10N) at the center of the axes. However, the construction must be suffiently strong that temporary loads of up to 12kg (120N), either upwards and downwards with respect to gravity, must be tolerated without permanent degredation of the SOC. This is required because vacuum loading of the vacuum system will be counterbalanced with a weight during operation, but it is not possible to apply the weight and vacuum at exactly the same time.
The phi and kappa motions must provide sufficient torque to meet the speed and settling-time specifications above when the following circumstances exist (simultaneously):
i) moment of inertia of the sample up to 0.1 kg-m2.
ii) frictional torque of 1.3 Nm (1.0 ft-lb) from the rotary feedthrough connecting to the vacuum.
The sphere of confusion (SOC) of the alignment of the sample axes {omega, kappa, phi}, under 1kg load at the sample position, shall be smaller than 40 microns. The SOC is defined as the three-sigma diameter of the distribution of excursions of the center point under all possible combinations of angle settings. A 200-point random sample of the angles within their operating range is sufficient to establish the distribution.
For purposes of defining the SOC, the ranges of operation of the sample axes in degrees are the following:
-90° < omega < 90°
(such that the body of the kappa drive always lies below a horizontal plane passing through the center of the instrument)
0 < kappa < 360°
0 < phi < 360°
Figure 4. Plan view of the detector motions of the double kappa diffractometer. The counterweights are drawn schematically, as are the approximate locations of the kinematic mounts.
5. Detector Arm Specification
The detector motions are illustrated in figure 4. The need for the entire detector arm to rotate about its own axis in all possible orientations with respect to gravity places considerable limitations on its rigidity. It is envisaged that interchangable arms made from X95 profile (Newport/ Klinger/ Microcontrole corp.), up to 3m in length, will be used. This material has a cross-sectional moment of 2.2´ 10-6m4 and Youngs modulus 70GPa, which determines its flexion. The X95 flight path must be supported at two points, approximately 0.9m and 1.8m from the center of the instrument. Each support will consist of an opposing pair of X95 clamping shoes through which the detector arm can slide. One support is the motorized rotary table providing the phi_D motion, while the other is a freely-turning follower that will constrain the flexion of the X95 beam under gravity. The clearance between the centerline of the arm and the outer edge of the white beam pipe is 180mm; the radius of the phi_D rotary table therefore must be less than this. The free end of the longest (3m) arm will be 1.2m from the outer support; this end is calculated to deflect by 0.5mm and tilt by 0.04° under a load of 10kg applied at the end of the beam. Such large deflections will be corrected in software, and are not considered to be part of the alignment specification.
The sphere of confusion (SOC) of the alignment of the detector axes {omega_D, kappa_D, phi_D}, carrying a detector of mass 10kg at a distance of 2m from the center, shall be smaller than 100 microns, defined in the same way as for the sample axes. It is recommended that the SOC be measured with an autocolimator rigidly attached to the phi_D axis focussed on a precision ball fixed at the center of the diffractometer's sample axes.
For purposes of defining the SOC, the ranges of operation of the detector axes are the following:
0 < omega_D < 180°
(such that the body of the kappa_D drive is always above a horizontal plane passing through the center of the instrument)
-115° < kappa_D < 115°
(where kappa_D = 0 is the fully extended position)
0 < phi_D < 360°
Combinations of omega_D and kappa_D which lead to collisions of the detector arm with the diffractometer support are excluded from these ranges (see below).
Note that the `zeros' of omega_D and kappa_D are defined differently from omega and kappa.
Whereas the detector arm can swing to any angle in a vertical plane by motion of omega_D alone, the range of horizontal motion is limited by collisions. It is necessary that the arm can swing out to 45° horizontally before the X95 flight path collides with the diffractometer support frame. This happens when omega_D=65.5° (up) and kappa_D=-114.5° (down) contracting like a scissor from the fully extended starting position. While maintaining this 45° inclination to the vertical plane, the arm must then be able to rotate continuously around by increasing omega_D further. To maintain enough clearance will require rounding and possibly bevelling of the top of the diffractometer support.
Figure 5. Side view of the CXD diffractometer showing the locations of the optical bench, kinematic mount and transport table. Dimensions are schematic. The white beam, wall and floor locations are marked. Dashed lines indicate:
(i) the alternate mounting position of the sample goniometer and
(ii) the positions of the phi axis tilted by 10° and 20° .
6. Incident Beam and Optics Benches
The optical bench facing the sample goniometer, indicated in Figure 5, has several functions. It is an attachment point for an (optional) long-working-distance microscope for centering samples and an autocollimator for checking alignment. It will also be used to attach precision beam-defining slits used for the coherent diffraction experiments, the motions of which may be passed inside the vacuum system. When the vacuum system is not in use, a removable stand for holding input beam components is to be attached to one side of the diffractometer support stand. Both of these components must be sufficiently stable that optics can be attached to them which must deflect less than 0.5 microns with respect to the center of the instrument. This deflection could be due to internal vibrations or to small changes of settings or loading over the course of measurement. When the diffractometer is in the front position, the optics will be positioned inside UHV, but still rigidly clamped and referenced to the optical bench; in the back position, apertures, slits, mirrors and beam monitors will be aligned with respect to the bench and to the incident beam stand.
The optical bench will also carry an (optional) long-working-distance microscope of sufficient quality to center samples (outside vacuum) with an accuracy of 5 microns. The optical bench must also be able to accomodate an autocollimator for verification of the mutual alignment of the diffractometer axes and their sphere of confusion (see below). To fit in the space below the MFD white beam pipe, shown in Figure 5, the surface of the optical bench should be inclined down at 25° and ground to a planar surface that is 57.5mm below the line of sight to the center of the diffractometer. This surface should be finished with an array of M6 or 1/4-20 tapped mounting holes on a 31.5mm grid spacing over an area at least 200mm (along the beam) ´ 150mm. This mounting surface should be placed as close as possible to the sample goniometer that its phi-axis can just swing past with the kappa arm fully extended.
7. Limit Switches and Collisions between Arms
All motions must have failsafe electrical limit switches, with closed contacts for safe operation and open circuit upon reaching a limit condition. One user-adjustable limit in each direction must be provided for each diffractometer axis rotation. In addition, compound limits associated with collisions between the moving parts of the detector arm must be separately provided. These potential collisions requiring protection are:
i) The positive and negative upper limits of kappa_D will cause the X95 detector arm to be infringed by the kappa_D arm. This is adequately served by the conventional kappa_D limit switch described above.
ii) Omega_D must turn freely in the range 0 < omega_D < 180° . When the omega_D axis runs negatively or beyond 180° , the omega_D arm will eventually infringe the main body of the diffractometer. This is protected by the conventional omega_D limit switch described above.
iii) Even with omega_D and kappa_D within these allowed ranges, the X95 detector arm can still intersect with the main body of the diffractometer. This must be prevented by a relative limit switch. One way this can be achieved is by a strip sensor surrounding the main body of the diffractometer. This is preferable to protecting the X95 detector arm itself because of its rotation by phi_D and because many user-defined configurations are envisaged. This problem can be largely avoided by appropriate conical contouring of the top of the diffractometer support; this collision will then occur at a fixed value of kappa_D, which would be ±115° to meet the specification for the range of travel of the arm out of the vertical scattering plane.
8. Table Specification
An adjustable motorized three-legged support table is required to hold the center of the instrument at the 1.4m beam height, forming a kinematic mount. Two symmetrically-placed legs are required under the detector-drive side of the main support and one under the optical-bench side. The necessary degrees of freedom for the kinematic mount are listed in Table II. It is expected, but not required, that these will be achieved with motorized vertical and horizontal translations under the "cone" and "groove" legs of the kinematic mount and a vertical translation under the "plane" leg.
Under the kinematic mount should be an integrated transport system, specifically designed to allow 1m of travel along the beam direction. This need not be a precision motion provided there is some means of rigidly docking the instrument at its front and back working locations. This might be realized with air-pads or wheels and a guide rail or else with linear ball-slides.
|
Table Motion |
Travel Range |
Reproducibility |
|
Height |
±10mm |
±5µm |
|
Position across beam |
±10mm |
±5µm |
|
Tilt across beam |
±0.5° |
±0.001° |
|
Tilt along beam |
±0.5° |
±0.001° |
|
Rotation about vertical |
±1.0° |
±0.001° |
Table II. Specifications for the diffractometer positioning table motions.
9. Controls and Interfacing
If servo motors are used to achieve the performance specification, suitable controllers should be supplied. Interfacing by RS232, IEEE488 or TTL signals would then be acceptable. If stepper motors are selected or recommended, options for suitable drivers may be provided. Our preference is for stepper motors interfaced at the TTL (pulse, direction, limit) level, but this is not a requirement.
10. Quality Assurance
Written documentation, such as autocollimator trajetories obtained by moving axes (with appropriate callibrations), is required for the sphere of confusion (SOC).
All manuals and technical specifications must be provided for 3rd party components, such as encoders.
11. Contractual Options
Vendors may bid on the complete diffractometer package as specified, but must bid separately on the following subcomponent breakdown:
i. Sample goniometer (omega, kappa, phi axes) attachable to the support in two positions. Motorized, removable XYZ stage. Optional encoder.
ii. Main L-shaped support, detector arms and motions (omega_D, kappa_D, phi_D axes). Optical bench and incident beam stand. Optional encoder.
iii. Motorized positioning and transport table under support.
iv. Motors and drive electronics for each of the above.
v. High magnification long-working-distance microscope for centering samples.
