LCLS Undulator SupportMover System Testing Status Jeff T

LCLS Undulator Support/Mover System Testing Status Jeff T. Collins 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing History: LCLS • We Assembled a “Breadboard” Undulator Support/Mover System Using Existing Parts at the Early Stages of the Design Process • To Get an Early Start on Motion Testing, we Specified, Procured and Installed a Very Accurate Measurement System for the “Breadboard” Support/Mover System • A Labview-Based Data Acquisition and Control System was Developed In-House for Command Control of all Motors with Feedback and Monitoring from the Measurement System • The “Breadboard” System was Used to Learn How to Control and Characterize the Performance of the Motion System • Results from Numerous Tests were Used as Feedback for Refinement of the Design Process. Problem Areas have been Identified and Solutions have Been Implemented in the Actual Support/Mover System Design • Measurement Results are Very Encouraging with Overall Excellent Repeatability 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Support/Mover System Testing Our Present “Breadboard” Lacks Many Features that the Prototype will Include: • Does not have “Expanded Wedge” on Upstream Double CAM Mover • Aluminum Frame instead of Steel Frame • Bearing Surfaces are not Hardened or Polished • 100: 1 Gearbox instead of 250: 1 Gearbox • Less Precise Rotary Potentiometers • Only Four Linear Potentiometers • Lack a Precision Temperature-Controlled Environment 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Support/Mover System Testing LCLS Salient Support/Mover System Physics Requirements: • Quadrupole Motion Positioning Repeatability • Short-Term (1 h) BPM and Quad Stability • Long-Term (24 h) BPM and Quad Stability ± 7 µm ± 2 µm ± 5 µm • Horiz. Segment Pos. Repeatability in Roll-Away Cycle • Vert. Segment Pos. Repeatability in Roll-Away Cycle ± 10 µm ± 5 µm • Quad Transverse Position Change in Roll-Out Condition • Quad Position Reproducibility after Roll-Away Cycle • BPM Transverse Position Change in Roll-Out Condition • BPM Position Reproducibility after Roll-Away Cycle ± 25 µm ± 2 µm 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS System Measurement Devices: • Five Keyence Laser Displacement Sensors (0. 05 micron repeatability). Two at each end measuring X and Y Displacement at the Quadrupole Position, and one located mid-plane of the Undulator on one side to measure Roll. • Four Linear Potentiometers (1 -2 micron repeatability) measuring X and Y Displacements on one side of the Girder located near the CAM Movers. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Support/Mover System Testing CAM Mover System Degrees of Freedom: • X-Translation (Inboard or Outboard Translation) • Y- Translation (Up or Down Translation) • Pitch Motion (Upstream or Downstream Pitch) • Roll Motion (Inboard or Outboard Roll) • Yaw Motion (From Either End or Both Ends) Notes: • The System can Move with Five Degrees of Freedom using any Combination of these Motions. • The Z-Dimension is Restricted but not Constrained via the Recessed Flat on the Single CAM Mover. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Support/Mover System Testing LCLS CAM Mover Calibration Procedure: 1. 2. 3. 4. 5. Rotate one CAM through a Full Revolution Measure Displacement as a function of Angular Position using either the Keyence Sensors or the Linear Potentiometers Plot the Displacement as a function of Angular Position Curve Fit the Data to Locate the “Zero” Position Repeat this Process for each of the Five CAM Movers Problems with this Approach: 1. 2. 3. 4. 5. 6. Rotating one CAM introduces Pitch, Roll, and Yaw into the System These Motions add Stress into the Girder causing Distortion These Distortions become part of the Calibration The Calibration is not Decoupled from the Girder The existing Rotary Potentiometers have 1° of Backlash adding to Calibration Uncertainty 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Case 1: X-Translation +/- 1. 5 mm Cyclic Motion Repeatability Location Laser Displacement Sensors Linear Potentiometers Upstream X ± 2. 5 µm ± 1. 5 µm Upstream Y ± 1. 0 µm ± 4. 0 µm Downstream X ± 2. 0 µm ± 2. 5 µm Downstream Y ± 0. 5 µm ± 4. 0 µm LCLS Target Repeatability < +/- 7. 0 µm 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS Infrared Image Prior to Motor Operation, Downstream End Double CAM Mover with Pedestal Base 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS Infrared Image Prior to Motor Operation, Downstream End Double CAM Mover with Pedestal Base 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS Infrared Image After Motor Operation, Downstream End Double CAM Mover with Pedestal Base, Motor on for Approximately 2 Hours 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Case 2: Y-Translation +/- 1. 5 mm Cyclic Motion Repeatability Location Laser Displacement Sensors Linear Potentiometers Upstream X ± 1. 0 µm ± 2. 5 µm Upstream Y ± 2. 5 µm ± 1. 0 µm Downstream X ± 0. 5 µm ± 4. 0 µm Downstream Y ± 1. 5 µm ± 1. 0 µm LCLS Target Repeatability < +/- 7. 0 µm 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Case 3: Roll +/- 1. 5 mm Cyclic Motion Repeatability Location Laser Displacement Sensors Linear Potentiometers Upstream X ± 1. 0 µm ± 2. 0 µm Upstream Y ± 0. 5 µm Downstream X ± 0. 5 µm ± 4. 0 µm Downstream Y ± 0. 5 µm ± 1. 5 µm LCLS Target Repeatability < +/- 7. 0 µm 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Case 4: Upstream Pitch +/- 1. 5 mm Cyclic Motion Repeatability Location Laser Displacement Sensors Linear Potentiometers Upstream X ± 1. 0 µm ± 2. 0 µm Upstream Y ± 1. 5 µm ± 1. 0 µm Downstream X ± 0. 5 µm ± 2. 0 µm Downstream Y ± 0. 5 µm LCLS Target Repeatability < +/- 7. 0 µm 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Case 5: Both Ends Yaw 1. 5 mm in Opposite Directions Location Laser Displacement Sensors Linear Potentiometers Upstream X ± 15. 0 µm ± 2. 5 µm Upstream Y ± 5. 0 µm ± 2. 5 µm Downstream X ± 25. 0 µm ± 2. 5 µm Downstream Y ± 2. 5 µm ± 5. 0 µm LCLS Target Repeatability < +/- 7. 0 µm Note: Yaw Motion is Incidental and is Not Required for Actual Operation. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS CAM Mover System Observations: 1. All Motions that allow the CAM Bearings to Freely Rotate (X-Translation, Y-Translation, and Roll) are Repeatable to better than +/- 2. 5 microns at the Quadrupole Position (Target Repeatability < +/- 7. 0 microns). 2. Though Pitch Motion is Overconstrained the Bearing on the Single CAM Mover only needs to “Slip” a few microns Across the Recessed Flat and therefore the Motion still Shows Excellent Repeatability. 3. Yaw Motion is very Overconstrained and therefore the Bearing on the Single CAM Mover Must “Slip” Hundreds of microns Across the Recessed Flat. This Slippage, the Process of Overcoming Static Friction, Distorts the Girder Causing Large Repeatability Errors. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing CAM Mover System Observations: 4. Some Repeatability Error is Added to the Linear Potentiometers from Sliding Across the Contact Surfaces causing Bending of the Sensor Arms. 5. 6. 7. 8. The Ball Bearing Tips of the Linear Potentiometers Wore Flat after Many Cycles from Sliding Across the Contact Surfaces. 6. From a Cold Start, Heat from the Motors can Cause Tens of Microns of Change from the Original Start Position. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov LCLS

Undulator Roll-Away Cycle Testing LCLS Undulator Segment Roll-Away Cycle : +/- 10 mm Cyclic Translation Upstream Translation Stage Note: Downstream Translation Stage Motions were Driven Purely by Motor Steps with no Feedback from the Linear Encoders. Many Changes will be Made to these Stages for the Actual Prototype. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 1. Improved Translation Stage Design for Roll-Away Cycle: • Stages will be Preloaded to Decrease Play in the Slides • Gearboxes Replaced with GAM Angular Gearboxes with Backlash Less than 15 arcsec. (Compared to 40 arcsec. on Existing Stages) • Table Length Increased from 6” to 8” and Ball Bearing Separation is Increased Accordingly • Size of all Mounting Holes have been Increased for Added Stiffness 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 2. Steel Girder Replaces the Aluminum Girder. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 3. Expanded Wedge on the Upstream Double CAM Mover for 1. Increased Stability. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 4. “Inverted Wedge” for the Single CAM Mover to Constrain Z-Dimension 1. Movement. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 5. Two Side-by-Side Bearings for the Single CAM Mover to Eliminate “Slip” across the Bearing Surface. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 6. Spherical Bearings for Both of the Double CAM Movers to allow 1. Rotational Play on these Bearings. The Single CAM Mover will 2. still use a Roller Bearing. High Precision Bearings will Replace the 3. Existing Standard Bearings. Note: Changes to the Bearings on the Double CAM Movers, the “Inverted Wedge” Bearing Surface, and the Side-by-Side Bearings on the Single CAM Mover will Eliminate Bearing “Slip” and will Allow the Girder to Pivot about the Single CAM Mover Contact Surface. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 7. The Addition of Four more Linear Potentiometers to Provide X and Y Displacement Information at the Four Corners of the Girder near the CAM Movers. Note: With Eight Linear Potentiometers Defining the Girder Position, the Effects of Girder Distortion can be Decoupled from the CAM Mover Calibrations allowing Precise CAM Mover Calibration. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Implemented Changes for the Actual Prototype LCLS 8. Replace the 100: 1 Gearboxes with 250: 1 Gearboxes. This will Eliminate the need for Brakes on the Motors, add Better Resolution to the System, and Simplify Control System. 9. Replace existing Rotary Potentiometers (1° Backlash) with more Precise Rotary Potentiometers with better than 0. 1 ° Backlash. This will Improve CAM Mover Calibration. 10. All Bearing Surfaces shall be Hardened and Highly Polished. This should add to Long-Term System Stability and Repeatability. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov

Support/Mover System Testing LCLS Conclusions: 1. The LCLS Undulator Support/Mover Tests on our “Breadboard” are Very Encouraging with Overall Excellent Repeatability. 2. We have Identified Problem Areas and have Implemented Solutions that will Improve Performance. 3. Testing To-Date has Provided Valuable Information on System Dynamics, Environmental Influences, Calibration Procedures, and Feedback and Control Algorithms. 4. Similar, but more in-depth, Tests shall be Performed on the Actual Prototype once the System has been Assembled. 20 October 2005 Undulator Support/Mover System Testing Jeff T. Collins@aps. anl. gov