HIE-Project-note acceptance test actuator · the HIE-DB and its instruments can be found in Ref. [1]. Figure 1 — Layout of the prototype HIE DB. A first prototype HIE-DB was designed

EUROPEAN ORGANIZATION FOR NUCLEAR RESEARCH

CERN-ACC-NOTE-2014-0099 HIE-ISOLDE-PROJECT-Note-0036

Geneva, Switzerland October 2014

This is an internal CERN publication and does not necessarily reflect the views of the CERN management.

Acceptance test for the linear motion actuator for the scanning slit of the HIE-ISOLDE short diagnostic boxes

E.D. Cantero, W. Andreazza, E. Bravin, A. Sosa

Abstract

We performed experimental tests to characterize the mechanical accuracy of a linear actuator designed by the company AVS for the movement of the scanning slit of the HIE-ISOLDE short diagnostic boxes. The mechanism consists of a linear actuator composed of two guiding rods and a lead screw, with a full stroke of 135 mm. A specially designed blade was mounted on the actuator and the transverse positioning of the blade was monitored with a camera-based optical system while moving the actuator at speeds of up to 10 mm/s. The repeatability of the positioning of the blade after several cycles around predefined positions was also measured. The results of the measurements and a general inspection of the device show that the proposed solution fulfils the specifications. A full prototype of short diagnostic box for the HIE-ISOLDE project can now be built for testing.

2

Contents 1.   Introduction ................................................................................................................................................ 3  2.   Experimental setup ..................................................................................................................................... 4  3.   Procedure .................................................................................................................................................... 6  4.   Results and discussion ................................................................................................................................ 8  

4.1.   Visual inspection ................................................................................................................................ 8  4.2.   Amplitude of oscillations for full stroke movements ......................................................................... 8  4.3.   Movement back and forth around each hole (6 mm, 15 cycles) ....................................................... 11  4.4.   Limit switches precision ................................................................................................................... 19  4.5.   Evaluation of counts losses ............................................................................................................... 20  4.6.   Movement back and forth from hole 1 to hole 6 (100 mm steps, 15 cycles) ................................... 20  4.7.   Movement back and forth around hole 3 (3 mm steps, 50 cycles) ................................................... 20  4.8.   Preliminary stress test of the mechanism (full stroke, 100 cycles) ................................................... 20  4.9.   Overall results overview ................................................................................................................... 20  

5.   Conclusions .............................................................................................................................................. 21  6.   Acknowledgements .................................................................................................................................. 21  7.   References ................................................................................................................................................ 21  

3

1. Introduction The beam instrumentation devices of the REX accelerator for the HIE-ISOLDE project will be installed in several diagnostic boxes (HIE-DB) located both in the LINAC and in the HEBT lines. The main structure of the HIE-DB is an octagonal-shaped tank with 6 ports available for the insertion of instruments or collimators. In particular one port is used to insert a Faraday cup (FC), the opposite port is used to insert a blade with a V shaped slit and a third port is used to connect a vacuum pump, see Fig. 1 for the layout of the first prototype HIE-DB. A stepper motor allows moving the FC in the radial direction with a stroke of ~50 mm. The scanning slit is also controlled by a stepper motor, its stroke being ~150 mm. The scanning slit is used in combination with the FC to measure the horizontal and vertical beam profiles. More details about the HIE-DB and its instruments can be found in Ref. [1].

Figure 1 — Layout of the prototype HIE DB.

A first prototype HIE-DB was designed and constructed as a result of collaboration between CERN and the Spanish private company Added Value Solutions (AVS) [2]. The movement for the scanning slit of that prototype used an in-vacuum guiding system, based on metallic slides with DICRONITE coating, and a commercial linear motion actuator. That prototype did not survive a stress test, performed at CERN in April 2013, consisting of 10000 IN-OUT cycles of the scanning slit movement [3]. AVS then proposed a new solution based on a custom linear actuator with an out of vacuum guiding system. Given the relatively large lever arm between the guiding slides and the slit position it was necessary to verify that the actuator provides the required positioning accuracy.

4

Vibrations of the slit in the transverse direction during a scan can in fact reduce the precision of the beam profile measurement. This document describes the tests performed between 3 and 6 February 2014 at the headquarters of Added Value Solutions in Elgoibar, Spain. It follows the main guidelines of the test procedure described in Ref. [4].

2. Experimental setup The prototype HIE-DB was installed in an experimental hall (grey room) of AVS. The body of prototype HIE diagnostic box was used as vacuum tank for the experiment. The new linear actuator under test was installed on the scanning slit port of the tank by means of an ad hoc adapting piece. A dedicated metallic blade was mounted at the end of the rod of the actuator (vacuum side) and was scanned across (what would be) the beam axis during the tests (there was clearly no particle beam involved in this test). Two optical viewports were mounted on the beam pipe flanges. A light source (Edmund MI-150 high intensity illuminator) and a CCD camera (IDS UI-2210SE) were also installed facing the mentioned viewports on either side of the tank, their supports being independent and mechanically detached from the DB support. A thin (2 mm thick) teflon (PTFE) plate was placed in front of the lamp to diffuse the light in order to have a homogeneous illumination of the viewport. The actuator for the faraday cup was also installed in the DB, as well as a high-vacuum pumping group with a turbo molecular pump. A picture of the experimental setup is presented in Fig. 2.

Figure 2 — Experimental setup.

5

In Fig. 3 detailed pictures of the linear actuator and its limit switches are shown. The mechanism has two guiding rods and the linear motion is driven by a ball-bearing screw connected to a stepper motor. The screw pitch is 1 mm and the full stroke of the mechanism is about 135 mm. The forward limit switch (bellow fully compressed) is used only as a safety interlock to prevent damage to the mechanical parts. The rear limit switch (bellow extended) is used as safety interlock as well as reference for the homing position. The control unit for the stepper motor, including the limit switches, was provided by AVS together with a custom-designed software utility. The external temperature of the stepper motor was monitored by a thermocouple connected to a digital multimeter, during the tests the temperature of the motor never exceeded 40 ºC.

Figure 3 — Linear translation mechanism of the scanning slit.

In Fig. 4 a picture of the scanning blade used in these tests is depicted, together with a drawing showing the main geometrical parameters. The blade has two slits (0.2 mm

6

width) at 45 degrees from the axis of movement (these slits are similar to the ones that will be used for the beam profiles measurements). Six holes (0.1 mm diameter) were also drilled in the blade axis and were used for monitoring the blade transverse position during the measurements. For more details about the scanning blade geometry, please refer to [4].

Figure 4 — Blade with two slits and six drilled holes used for determining the accuracy on the positioning of the linear actuator. Holes are labelled as hole 1 to hole 6 from the innermost to the outermost position.

3. Procedure The experimental procedure consisted of tracking the positions of the drilled holes for different blade transverse positions, either with the blade at a fixed location or while it was moved at speeds of up to 10 mm/s. When the scanning blade crossed the beam aperture the light passing through the drilled holes (or the slits) was detected by the camera. By analysing the size and position of the light spots frame by frame, the displacements of the blade due to mechanical vibrations were determined. The exposure time of the camera was set to texposure = 100 µs with a frame rate of 37.6 FPS. The size of the frames was 640 x 480 pixels with 8 bits grey levels. The experience was carried out with high vacuum conditions (P ~ 2. 10-6 mbar). In order

7

to simplify the data analysis the camera was installed at 45 degrees, i.e. with the horizontal axis aligned with the scanning slit direction.

An example frame is shown in Fig. 5-a, the shapes of a slit and of a hole are visible. A small area around the hole is isolated and magnified, see Fig. 5-b. The profiles of the light intensity in the horizontal and vertical directions are computed, from these the position and size of the spot are determined.

a) Image containing a slit and a hole. b) Image and profiles of the spot relative to a hole. A Gaussian fit is used to determine the size and position.

Figure 5 — Example of acquired frame. The horizontal axis of the camera is aligned with blade axis (direction of movement). The movement of the blade towards the interior of the box was observed as a displacement of the holes from right to left.

A calibration of the camera scale was done by measuring the steps needed to centre the different holes in the camera. The obtained value is δscale = 43.42 pixels/mm. All the data presented below are scaled from pixels to distances using the previously mentioned δscale value. The positioning of the detection system was such that the blade axis was aligned within 0.3º from the horizontal axis of the camera (referred as x in the forthcoming analysis; the vertical direction of the camera will be referred as y). A movement of the blade towards the interior of the box was observed as a displacement of the image towards the left direction.

After the calibration of the scale, a series of tests were performed, which included:

1) Continuous full movement of the blade in both directions (IN → OUT and OUT → IN), tracking the position of the 6 holes. The blade speed was set to 10 mm/s and 5 mm/s. This test allowed determining the maximum (peak to peak) amplitude of the movement of the holes in the y direction of the transverse plane during to the movement of the blade.

2) Movement of 6 mm of the blade in both directions started with a hole centred on the camera, with 1 s pause at each end for 15 cycles. This test was done to measure the repeatability of the blade positioning during several scans. The

8

blade speed was set to 10 mm/s. For holes 1 and 6, the test was also performed at 5 mm/s.

3) Movement of the full stroke of the blade touching both limit switches, without updating the home position, for 10 consecutive cycles. This test was done to measure the accuracy in the position of the blade by using the limit switches as reference. Blade speed was 10 mm/s.

4) Starting with the blade fully OUT, move in steps of ~7 mm until the forward limit switch was reached, then back to the position full OUT in one move. This sequence was repeated 10 times in order to verify that the motor was not losing steps along the movement.

5) Movement of the blade in both directions, starting with hole 1 centred on the camera (i.e. in the beam axis) and until hole 6 was centred (total movement length 100 mm), for 15 cycles. This is a repetition of point (2) but with a longer stroke.

6) Starting with hole 3 centred on the camera, move 3 mm back and forth for 50 cycles to test repeatability of the positioning for a large number of cycles.

7) Preliminary stress test of the mechanism: movement of the full stroke of the actuator for 100 cycles, with 1 s pause between movements. The temperature of the stepper motor was monitored during the test to evaluate the correct dimensioning of it. A general inspection of the device was done in order to check for signs of wear or any issues related with the design and implementation of the proposed solution.

The results of the previously enumerated measurements and other examinations of the device are presented in the following section.

4. Results and discussion

4.1. Visual inspection The design of the system with two guiding rods and the moving force applied directly on the centre of the actuator looks reliable. The use of a ball screw to reduce the friction in the conversion from rotational to linear motion is a good solution, increasing the smoothness of the blade movement. The screw pitch of 1 mm is adequate and, together with the stepper motor, provides sufficient resolution in the positioning of the blade. Limit switches are placed correctly and allow a full stroke of ~135 mm which covers the complete scan of both slits (vertical and horizontal profiles). The device is compatible with ultra-high vacuum standards. For the production devices the lubrication of the screw should be done with grease approved by CERN considering that the boxes will be used in a radiation environment.

4.2. Amplitude of oscillations for full stroke movements The blade was moved at 10 mm/s over the full stroke (OUT à IN) and the position of the 6 holes, while passing in the field of view of the camera (~14 mm wide), were

9

tracked. The position of the centre of mass of the hole image in x and y are shown in Fig. 6. A small correction was applied to the camera coordinates (subtraction of a slope) due to a 0.3º misalignment of the camera w.r.t the direction of movement of the blade. The first detected object is hole 1, moving from right to left in the plot, and showing very small oscillations in the y direction (perpendicular to the movement). After that, all the other holes were detected in succession and their position recorded by the camera. The lowest x position that hole 6 could reach was around x = 4 mm, this corresponds to the blade fully in (touching the forward limit switch). The maximum overall excursion in the y direction for each hole is of the order of 20 µm.

Figure 6 — Tracking of the holes positions in the x-y (camera) plane. The blade speed was 10 mm/s, and the movement was from the fully OUT to the fully IN positions. Holes are labelled as hole 1 to hole 6 from the innermost to the outermost position.

The measurement was repeated for the reversed direction of movement, i.e. with the blade moving the full stroke (IN à OUT) at 10 mm/s. The results are presented in Fig. 7. The maximum overall excursion in the y direction for each hole is again very small, being of the order or lower than 20 µm for all the holes.

Similar tests were also performed at a lower speed (5 mm/s). The results are presented in Figs. 8 and 9. The general conclusions are similar to the presented in the previous paragraphs. The maximum overall excursion in the y direction for each hole is of the order of 20 µm, and it does not seem to be dependent on the blade speed.

In the analysis of the results presented in Figs. 6 to 9, it is evident the presence of small vertical offsets of the graphs. These offsets are systematic and probably due to the finite accuracy in the drilling of the holes on the blade. The offsets do not influence

10

the results on the overall excursion of the slit in the y direction, and therefore have not been considered as relevant in the present analysis.

Figure 7 — Same as Fig. 6, but with the direction of movement changed (IN à OUT).

Figure 8 — Same as Fig. 6, but at a lower blade speed (5 mm/s).

11

Figure 9 — Same as Fig. 7, but at a lower blade speed (5 mm/s).

4.3. Movement back and forth around each hole (6 mm, 15 cycles) Several cycles were performed moving back and forth with a blade stroke of 6 mm to determine the repeatability of the positioning system in the x direction (direction of the movement). This measurement was done with the blade moving at 10 mm/s (top speed) and a pause of about 1 s at each end. For holes 1 and 6 the measurement was also repeated at 5 mm/s. The results of the x position of the hole versus time and of the x position versus the y position are presented in Figs. 10 to 25. During all the test presented in this section, the accuracy on the positioning of the scanning blade is of the order of 1 µm, which might as well be influenced at this level by the uncertainties introduced by other possible artefacts on the experimental setup (vibrations of the camera support, resolution of the acquired images, accurate level of detection of the optical recognition software).

12

Figure 10 — Tracking of the position of hole 1 while moving the slit 6 mm back and forth at 10 mm/s. Central plot shows the time evolution of the x coordinate (direction of movement) for the detected hole image, while top and bottom plots represent the same data in a magnified scale around the limit values in order to display clearly the repeatability of the positioning of the blade during the different cycles.

Figure 11 — Tracking of the position in the x-y camera plane for hole 1 while moving the slit 6 mm back and forth at 10 mm/s.

13

Figure 12 — Idem as Fig. 10, for hole 2 at 10 mm/s.

Figure 13 — Idem as Fig. 11, for hole 2 at 10 mm/s.

14

Figure 14 — Idem as Fig. 10, for hole 3 at 10 mm/s.

Figure 15 — Idem as Fig. 11, for hole 3 at 10 mm/s.

15

Figure 16 — Idem as Fig. 10, for hole 4 at 10 mm/s.

Figure 17 — Idem as Fig. 11, for hole 4 at 10 mm/s.

16

Figure 18 — Idem as Fig. 10, for hole 5 at 10 mm/s.

Figure 19 — Idem as Fig. 11, for hole 5 at 10 mm/s.

17

Figure 20 — Idem as Fig. 10, for hole 6 at 10 mm/s.

Figure 21 — Idem as Fig. 11, for hole 6 at 10 mm/s.

18

Figure 22 — Idem as Fig. 10, for hole 1 at 5 mm/s.

Figure 23 — Idem as Fig. 11, for hole 1 at 5 mm/s.

19

Figure 24 — Idem as Fig. 10, for hole 6 at 5 mm/s.

Figure 25 — Idem as Fig. 11, for hole 6 at 5 mm/s.

4.4. Limit switches precision The blade was moved back and forth the full stroke (138.7 mm) at 10 mm/s for ten cycles. Without resetting the reference position (homing), and using the value read by the control system from the stepper motor counter, the repeatability of the position

20

had a standard deviation of 3 counts for each limit switch. The translation of this value to distances in the linear movement gives the result σRMS

limit switch repeatability ~2 µm.

4.5. Evaluation of counts losses The blade was moved IN in several steps of 7 mm with pauses of 1 s until the forward limit switch was reached, and then moved fully OUT in one large step. This test was used to evaluate the possibility of lost counts/steps during the acceleration and deceleration processes, and was repeated 10 times. The standard deviation on the number of counts during the 10 cycles was of a few counts (out of a total of 277000 for the full stroke), which indicates that the loss of counts was inexistent or completely negligible.

4.6. Movement back and forth from hole 1 to hole 6 (100 mm steps, 15 cycles) Starting with a position where hole 1 was centred in the camera frame, the blade was moved back and forth 100 mm for 15 cycles. The blade positioning in the direction of the movement changed less than 5 µm between the first and last cycle.

4.7. Movement back and forth around hole 3 (3 mm steps, 50 cycles) The test described in section 4.3 was repeated using hole 3, a blade speed of 10 mm/s and a shorter stroke (3 mm), for 50 continuous cycles. The repeatability of the hole position after these cycles was of the order of 2 µm.

4.8. Preliminary stress test of the mechanism (full stroke, 100 cycles) The blade was moved back and forth between both limit switches for 100 cycles, with a pause of 1 s before reversing the direction of motion. The temperature of the stepper motor was monitored and did not rise above 30ºC.

4.9. Overall results overview The proposed mechanical design for the scanning slit actuator for the HIE-ISOLDE short boxes is a solid solution that has been thoroughly tested using a HIE-DB prototype. The repeatability on the positioning of the blade after several scans and changes in the direction of the movements should be better than 100 µm, according to specifications [1], in order not to worsen the resolution for the beam profiles measurement. The repeatability in the positioning of the blade measured indirectly by tracking a series of holes drilled on it is better than 20 µm, a value that fulfils the requirements. The chosen stepper motor can provide the needed torque and move the slit at the reference speed of 10 mm/s without problems. The system is quoted to be UHV compatible, although a proper vacuum test should be done following CERN standard procedures before final acceptance of the prototype. A preliminary stress test was performed (100 continuous cycles, full stroke) and no issues were observed. The procedure and materials used for lubrication should be considered carefully as the device is to be used in a radiation environment.

21

5. Conclusions The tests consisted of performing several scans of the blade movement in both directions (IN → OUT and OUT → IN), at speeds of up to 10 mm/s and determining the transverse position of the blade by using an optical system. All the results show that the accuracy in terms of the blade transverse positioning meet the acceptance criteria defined in the functional specification for the HIE-ISOLDE beam diagnostic systems [1]. The test procedure of Ref. [4] was fully completed and the results were satisfactory, as a consequence the status of this acceptance test is: APPROVED.

6. Acknowledgements The authors would like to thank the technical support provided by José Miguel Carmona, Julio Galipienzo and Juan Reyes during their time in AVS.

7. References [1] HIE-B-FS-0001, M. Fraser et al, "Beam Diagnostic Boxes for HIE-ISOLDE" CERN EDMS 1213401. https://edms.cern.ch/document/1213401/1.0  

[2] Added Value Solutions. Pol. Ind. Sigma Xixilion, Kalea 2, Bajo Pabellón 10. 20870 Elgoibar, Gipuzkoa. Spain. http://www.a-v-s.es/

[3] HIE-BDB-TN-0001, E. Bravin et al, "Failure on the scanning slit movement of the prototype HIE-ISOLDEDiagnostic Box", EDMS 1284254. https://edms.cern.ch/document/1284254/1.0

[4] HIE-BDB-PRD-0001, E. D. Cantero et al, “Mechanical positioning test procedure for the scanning blade of the HIE-ISOLDE diagnostic boxes”, EDMS 1333101. https://edms.cern.ch/document/1333101/1