LIN: The Laser In-Coupling

Overview

The Laser in-coupling (LIN) serves the following functions:

  • Bring the optical PP laser beam into the vacuum environment of the X-ray beam and onto the sample;
  • Maintain the vacuum gradient between the KBS and the FFT chambers, i.e. act as a differential pumping stage. The pressure difference can be up to 4 orders of magnitude (10-9 mbar at the KBS and 10-5 mbar at the FFT experiment chamber);
  • Image the X-ray beam at the exit of the KBS with the LIN-DPS imager. Image the X-ray and optical laser beams in vacuum and check their relative positions with the LIN-LIC imager.

The LIN is situated between the KB mirrors (KBS) and the FFT chamber. The complete assembly comprises two chambers, which are pumped by turbomolecular pumps. They are separated by a low-conductance element (a narrow tube), thus forming a two-stage differential pumping unit. A gate valve is fitted with a zero length aperture, which also serves as a low-conductance element to maintain differential pressure between the KBS and FFT. In the upstream chamber one imager for X-rays will be installed to help align the KBS system.

A set of mirrors is installed on a manipulator in the downstream chamber to reflect the optical beam from PP laser (optical laser) onto the sample. The manipulator moves the mirrors in X, Y, Z. In-vacuum piezo stepping actuators tilt the mirrors about a horizontal axis (lying in the XZ plane), and about the vertical axis (Y-axis). These X, Y and Z motions are remotely controlled with encoders. The piezo motors are also remote controlled but without encoders. An imager actuated by a stepper motor is installed after the optical mirrors in order to visualize the X-ray and optical beams, and to align the low-conductance element placed between the two chambers of the LIN.

Selecting, Positioning and Aligning a Mirror

To move to a different mirror on the LIN mirror stack, you will need to move it along the y axis, and possibly in x and z. Here is how:

IMPORTANT: There are NO LIMIT SWITCHES set for the x or z axes. The y axis has only the lower limit set. Please only move LIN motors from inside the hutch while watching what is going on. Collision and, or, damage to bellows are all possible and will lead to severe delays in operation!

  • The maximum length of the bellows, including its flanges, is 566 mm. This is protected by a limit switch for normal operation but not for maintenance (see below).
  • The minimum length of the bellows, including its flanges, is ?? mm.
  1. Move the mirror stack x and z motors to zero to centre the stack in the chamber. SCS_LIN_LIC/MOTOR/X and SCS_LIN_LIC/MOTOR/Z. There are mm scales marked on the stages. The stages are attached to the Y platform of the LIN elevator. They are from the company Huber.
  2. Move SCS_LIN_LIC/MOTOR/Y to move the mirror stack until the mirror you want is at about the right height.

3. We normally position the top of the mirror about 5mm below the FEL beam. You can check this by clipping the ALAS beam with the top of the mirror and then moving down by 5mm. Switch off the motor when done. Y ‘actual position’ values for the four mirrors positioned about 5mm under the FEL beam are: * 387 mm (lowest mirror stack) 400nm dielectric * 317 mm, 266nm dielectric * 245 mm, Thorlabs UV enhanced aluminium * ca. 175 mm (highest mirror on stack), not occupied as of 16.10.2023 Note: the encoder scaling may be off? Something to investigate 08.01.2021.) The mirror spacing seems to be about 70mm.

  1. Swap the Smaract controller to the current mirror. IMPORTANT: Switch off the Smaract controller in the rack room first. It is located near the Kepco. Make sure it’s the one for the LIN, not for the FFT Smaract stages. I believe that disconnecting these actuators with the power on can damage the piezo. We have killed two so far.
  2. At the bottom of the LIN mirror elevator is a CF40 feedthrough for the Smaract motors with a single black cable attached to it. Move the cable appropriate connector. Turn the controller back on.
  3. Now you have to align the PPL beam from the LIN into the FFT. Put the LIN-LIC imager in. On the Karabo scene it says that the IN position is -20. Go instead to -37. See post https://in.xfel.eu/elog/SCS+Instrument/2031. That way you can see the ALAS beam and the PPL beam on the same screen. Turn on ALAS and the PPL (single pulse, 0% transmission). The two spots should be aligned horizontally, i.e. one is directly above the other, separated vertically by a few cm. If not, move the Smaract yaw (AY) to align them. Don’t worry about the height yet.
  4. Move the LIN-LIC screen out. (Close the PPL beam shutter to avoid burning the camera with the PPL when it reflects of the YAG!). If you are lucky the PPL beam will be on the sample pBN. In this case you are done.
  5. If you are less lucky it will be somewhere on the sample card. In this case, use the Smarct pitch and yaw to overlap with ALAS at the sample pBN. If you are in a rush you can stop now and do your experiment. Otherwise, go back to the LIN-LIC imager and check that the beams are still horizontally aligned. If not, move the X and Z motors (SCS_LIN_LIC/MOTOR/X and ../Z) to align the beams and then the Smaract yaw to get back on the sample. You should not have to move more that 2mm with these motors. You are only compensating differences in the thickness of the mirrors. (The beam entering the LIN should be horizontally centred on the window, btw. Vertically it is near the lower edge.)
  6. If you are unlucky, the PPL beam will not be entering the FFT at all. When this happens, close the guard stage around the ALAS and use the camera mounted on the door side of the LIN looking at the upstream side of the guard stage in the reflection of the LIN mirror mount (SCS_LIN_SCR/CAM/DOORSIDE) to find the PPL beam. Correct the horizontal position on the guard stage with the the Smaract yaw and with the Z and X motors at the LIN-LIC imager. Iterate until the beam is on the sample and the PPL is vertically below the ALAS beam on the LIN-LIC imager. Turn off motors X and Z.

In-coupling window

The in-couping window is mounted in a custom modified DN100CF flange. The window is a separate optic with an o-ring seal and is held in place by an aluminium ring screwed to the flange body. This was done so that the window can be easily and cheaply exchanged. It can accomodate optics with a diameter between 75 mm and 80 mm, including therefore 3” (76.2 mm)

Current in-coupling window is shown on LIN mirror stack scene.

Table of available windows:

Date Manufacturerer Material Surface Diameter/mm Thickness/mm Coating Wedge Elog
28.01.2020 Eksma Optics UVFS l/10 76.2 3 none no https://in.xfel.eu/elog/SCS+Instrument/2998
23.05.2020 Unknown Unknown ? 76.2 5 (not 3) none no https://in.xfel.eu/elog/SCS+Instrument/993

Interlocks

The LIN has pretty standard vacuum interlocks. We want to isolate it once something goes wrong. But first let’s protect the cold cathode gauge CC13.

No Property Op Value
c1 SCS_LIN_VAC/GAUGE/PCC14.Value GT 1e-6
No Logic Action
a1 c1 COff

We close GV7 and GV9 if the pressure is bad (GV8 is not actually a valve, it is an aperture). GV10 could be closed instead of GV9, but we consider it more part of the FFT.

No Property Op Value
c1 SCS_LIN_VAC/GAUGE/CC13.Value GT 1e-6
c2 SCS_LIN_VAC/DCTRL/CC13.state.Value EQ FALSE
c3 SCS_LIN_VAC/VALVE/GV1.state.Value EQ FALSE
c4 SCS_LIN_VAC/VALVE/GV2.state.Value EQ FALSE
No Logic Action
a1 c1 or c2 or c3 or c4 CClose

Both GV1 and GV2 are closed if their turbo pumps have problems, or their roughing vacuum goes bad.

No Property Op Value
c1 SCS_LIN_VAC/TPUMP/TP1.state.RPM80 EQ FALSE
c2 SCS_LIN_VAC/VALVE/RV1.state.IsOpen EQ FALSE
c3 SCS_LIN_VAC/GAUGE/P5.Value GT 1e-2
No Property Op Value
c1 SCS_LIN_VAC/TPUMP/TP2.state.RPM80 EQ FALSE
c2 SCS_LIN_VAC/VALVE/RV2.state.IsOpen EQ FALSE
c3 SCS_LIN_VAC/GAUGE/P5.Value GT 1e-2

As an exception: if the pressure is really bad, the turbo pumps are not running and the LIN is isolated, one may pump it down. This complicated conditions are explained in the section about the DPS.

No Property Op Value Hyst Timeout
c4 SCS_LIN_VAC/GAUGE/PCC14.Value LT 1e-4 1 100000
c5 SCS_LIN_VAC/VALVE/GV7.state.IsClosed EQ FALSE    
c6 SCS_LIN_VAC/VALVE/GV9.state.IsClosed EQ FALSE    
No Logic Action
a1 (c1 or c2 or c3) and (c4 or c5 or c6 or not c1) CClose

Maintenance

Venting

  • Close and disable the gate valves to the KBS (GV7) and end station (GV9). (GV8 can be open or closed because it is not a valve but an aperture.)
  • Switch off the cold cathode gauge in the DPS section (CC13).
  • Connect venting N2 to venting port at low pressure (0.2 bar). Let it purge for a while.
  • Disable all interlocks on gate valves GV1 and GV2 between the chamber and the turbos so that they remain open while the pumps run down and the roughing valves are closed.
  • Close roughing valves RV1 and RV2
  • Turn the turbos off and let them run down to 70% rotation speed.
  • Open the manual venting valve slowly. There should always be a little gas escaping from the over pressure valve, but not so much that the chamber is pressurised.

Pumping down

  • Make sure all flanges and windows are tightened appropriately
  • Close venting valve
  • Open roughing valves RV1 and RV2
  • Start the turbo pumps TP1 and TP2
  • Check that they run up to full speed and that the pressure is okay.
  • Reinstate the interlocks on gate valves GV1 and GV2
  • Reconnect beamline gate valves GV7 and GV9

Accessing the in-vacuum mirror stack

  • Vent the LIN according to the procedure above.
  • Move the mirror stack to x = z = 0 to centre the stack in the vacuum pipe.
  • Move mirror stack up to about y = 398 mm.
  • Attach the 3 x M8 support rods to the bellows. DO NOT move the y motor with these rods in place until the vacuum flange between chamber and bellows is open. Something will break if you do!
  • Undo the CF160 flange between the chamber and the bellows.
  • Turn all screws except 2 or 3 until they are inside the bellows flange. This is to prevent them damaging the knife edge on the chamber flange. The 2 or 3 screws left sticking out act as guides as the flanges come apart (next step).
  • Lower the mirror stack with motor y until the top of the in-vacuum stack is clear of the vacuum flange on the chamber (ca y = 80 mm).

As the flanges move apart, be careful that the bellows does not move sideways and drag screws across the knife edges. * The lower limit switch on the y axis is set for normal operation. It will have to be be physically moved to allow the stage to moved down far enought. Before moving it, mark its position, if there is no existing mark.

Changing mirrors

Mirrors can be exchanged without removing the stack completely. However, you may prefer to remove the stack to get a more convenient working place. (See below for removal instructions.)

  • Undo the 4 x M6 Ti screws holding the stack to its pedestal. Put some isopropanol on the threads first to stop them galling.
  • Rotate the stack so you can access the mirrors.
  • Replace one or more of the M6 Ti screws to secure the stack while you work on it.
  • Remove the 3 x M1.6 screws holding the mirror clamp ring. This is very fiddly in the vertical orientation - you may prefer to remove the stack and lay it down.
  • Exchange the mirror.
  • Put everything back.

Removing the mirror stack

  • Open the CF40 Lemo feedthrough flange at the base of the bellows
  • Disconnect the in-vacuum Lemo connectors. Be sure to note where they are connected! Push the connectors into the chamber.
  • Undo the 4 x M6 Ti screws holding the stack to its pedestal. Put some isopropanol on the threads first to stop them galling.
  • Lift the stack off its pedestal and remove, taking care to bring the cables with you.

Cabling

The following information is also in the eLog: https://in.xfel.eu/elog/SCS+Instrument/3015

The in-vacuum 2” mirror mounts are motorized with piezo motors on the Ry and Rxz tilt axes. The controller for these motors is in the rack room. A cable from the rack room to the Exp. Hutch is labeled LIN.LIC.W19 and terminated with a smallish Lemo connector. An adapter/extension Y-cable is connected to the cable from the rack room. It has a Lemo on one end and 2 x sub-d (grey plastic with black tape in the photo)) on the other. The adapter/extesion cable connectes to a similar Y-cable (metal sub-d to Lemo), which in turn connects to the vacuum feedthrough via the Lemo.

All these cables should stay at the LIN, but you never know.