Stabilisation status and integrated simulations

Stabilisation status and integrated simulations

A.Jeremie

FF stabilisation

• Criteria to give a direction of study:– 0.15nm above a few Hz (4Hz to compare to

previous studies) for the mechanical stabilisation

– 0.1nm at 0.1Hz for the integrated study (including mechanical stabilisation, pre-isolator, beam feedback, Adaptive feedback)

2

LAPP and SYMME contribution to a “stable” CLIC accelerator

CostingBeam PhysicsDaniel Schulte

Technical designHermann

Schmickler

Accelerating structures CTF3

Activity 1:Stabilisation

technology =>Sensors, actuators, stabilisation system

(incl.feedback, electr.)A.Jeremie and

K.Artoos

Activity 2:Apply to MB quad

=> MB quad simulations and measurements

K.Artoos

Activity 3:Apply to FF quad =>

QD0 proto measurement,

integration (hardware + beam

physics)L.Gatignon

Activity 4:Integrated feedback

=> Choice of controller, overall performanceB.Caron and A.Latina

Integration, simulations, physics

A.Latina

Stabilisation WGK.Artoos

MDI WGL.Gatignon

CLIC Technical CteeCLIC Beams Ctee

A.Jeremie 10/01/2011

Activity 1Stabilisation technology =>

Sensors, actuators, stabilisation system (incl.feedback, electr.)

4

Stabilisation technology

What we need• Need to measure nm in

accelerator environment without disturbing the structure

• Need to act on the structure to be stabilised (several kg)

• Feedback for vibration control

Issues• Commercial sensors studied,

some identified, but still need to work with accelerator

• Commercial actuators working in the nm resolution are often limited in force and fragile

• Find real time systems ( + electronics) that have a sufficient range and resolution

5

Specific to MDI

What we need• Very compact system to fit

inside the detector

• Very tight vibration criteria

Issues• “absolute” sensors large• compatibility with the other

components in crowded space

• Compromise between total range and resolution

• Non-vibrating QD0 => measurements on future prototype

6

7

Need to be light, compact, low frequency, nm, rad hard, resistant to magnetic field

Instrumentation (measure, act, control)

Passive and active isolation

Real time material

SensorsWe had identified 3 types of sensors for “absolute” measurement:• Endevco (piezo accelerometers): manufacturing discontinued,

replaced by Wilkoxon; seem a little less performing but OK; need to test => but low signal to noise ratio at low frequency

• Güralp (geophones): our sensors had been sent for calibration and repair (some re-centering) but some confusion has delayed the return of the sensors. Should be back today. => but big and heavy

• SP500 (chemical sensors): decided with Kurt that LAPP will re-examine them (for comparison, we prefer to have the Güralp before starting). => seem to degrade with time

• Will we need to develop a dedicated sensor? If we get the French ANR financing, a new partner will look into this issue.

8

Sensors

• Inside the LAPP stabilisation system, for compactness reasons, “relative” sensors have been chosen, the “absolute” measurement being done at a more “practical” location.

9

PI D-015

240mm

Actuators

In the past, because of “historical” contacts, we decided with Kurt that CERN concentrate on PI actuators and LAPP on Cedrat actuators for performance evaluation.

• For the stabilisation “feasibility” demonstration we used APA60S

• For the stabilisation system, we use PPA10M.

CERN is using PI actuators, but in addition have made an extensive mechanical, flexural joint and rigidity study to avoid any shear

Our stabilisation system uses elastomer for guidance.

10

I/O Acquisition system

• We have started the stabilisation system instrumentation tests with dSPACE 16 bit system.

• For the initial tests, sufficient for the measurements, but will need to study an upgrade.

• An electronics engineer will help us (very small part time…!) during the tests.

11

IWLC2010 Geneva 2010 A.Jeremie 12

Mid-lower magnet

1355mm

Elastomeric strips for guidance

Piezoelectric actuator below its micrometric screw

Lower electrode of the capacitive

sensor

Fine adjustments for capacitive sensor (tilt and

distance)

V-support for the magnet

First tests in Annecy

2mV=0.1nm

Next step: add feedback

240mm

Status

• Stabilisation system has moved to SYMME for feedback implementation tests.

• Room has been better adapted to the measurements

• Real Time dSPACE system is in place • Additional sensors are being installed (there

was only one for initial evaluation tests, now the 4 planned sensors have arrived)

• Tests start now

13

Activity 2:Apply to MB quad => MB quad simulations and measurements

Not MDI… so will skip this activityAlthough experience gained through this

activity will benefit MDISome planned measurements have to be

adapted to recent changes in pole procurement: imminent discussion with Kurt 14

Activity 3:Apply to FF quad =>

QD0 proto measurement, integration (hardware + beam

physics)

15

Hardware integrationWhat it looks like in the CDR

16H.Gerwig + N.Siegrist

Beam trajectory control( also activity 4)

17

• Preliminary results, not checked nor digested…• Just to show what we are working on: a more complete

presentation will come in due time• Comparison FB, AFB and noise• Pre-isolator used (very similar to initial filter used)• Independent optimisations: each FB has it’s own

optimal parameters• Compare “CMS” GM and “B” GM

Prelim

inary!!!!!

Comparisons done in simulations

18

Pre-isolator is used in all simulationsThe following GM and noise curves are used:

13 pmnoise

BCMS

Summary of integrated rms at 0.1Hz

19

Prelim

inary!!!!!

The red color indicates that the simulation has been done using ground motion from CMS. The green one corresponds to the model B generated by PLACET.

FB: “classical” PLACET feedbackAFB: FB + Annecy Adaptive feedbackPID: Parameter optimisation for comparisonAPID: PID + adaptive PID

Some sample figures

20

PSD and RMS with PLACET optimization, 13pm white noise, pre-isolator and AFB.

PSD and RMS with PLACET optimization, pre-isolator and AFB.

Prelim

inary!!!!!

Planned activities

• Tests on stabilisation system with 4 sensors are starting now (validate sensors, actuators, feedback and electronics) until at least June 2011

• Work on chemical sensors (calibration, noise measurements, contact manufacturer and users, long term stability measurements) June 2011 and more?

• Start thinking of set-up to test sensors in accelerator environment

• Work on the integrated simulation: when results digested, decide with Daniel and Andrea L. what to include in CDR (before the 2011 CDR correction deadline!)

21