This website is no longer maintained. Its content may be obsolete. Please visit http://home.cern for current CERN information.
The orbit at TCDQ/TCSG in IR6 during a single fill (injection, ramp, squeeze, collide) should be stable to within about 0.2 mm, compared to the established reference, by means of the global feedback. The same holds true through the injection sequence for the TDI/TCLI in IR2/IR8.
For the reproducibility between fills of the orbit at these devices, it is anticipated that this will be at the 0.1-0.2 mm level. This will require some setting-up of the orbit at each injection to recover these reference conditions. Some slow drift of the orbit away from the reference may be experienced over a timescale of a few days; in this case a 'reset' with some correction of the orbit may be needed when this gets to the tolerance limit.
Ralph (S) will have a look through the simulations he has made to see what the exact numbers are at these devices. There is no real way to prevent a global orbit correction from changing the beam position at the protection devices - therefore, orbit correction must be subject to a strict procedure once these devices have been set in place.
As you know the TCDQ is moved by DC motors. We have had a lot of discussion with Michel and Robertor trying to achieve the same low-level interface to the overall LHC control system as the collimators (even to the extent of seriously exploring whether we can replace the DC motors with stepping motors), but time has now run out (HWC in a few months...) and we have to put something in place. This is our proposal - basically the high level application of Stefano takes care of generating and loading the correct movement and reference functions, as it will be doing for the other collimators. The low level is different, but this should not compromise functionality. We (BT) will use the LVDTs as a separate acqusition channel for position interlock generation.
Control of TCDQ diluter Motor drivers
- Each TCDQ diluter has two DC motors, driven in a servo-loop using position measurements from potentiometers as feedback signals.
- The low-level motor drive control is via Siemens PLCs
- The control methods available will be LOAD POSITION, LOAD FUNCTON, LOAD REFERENCE FUNCTION, DRIVE and READ.
- The DRIVE command will be triggered (assumed to be by the start of ramp timing event). This will require a standard timing card in the TCDQ Motor Control PLC.
- The control methods will be managed via FESA contracts.
Position interlocking
- Each TCDQ diluter has two LVDTs for independent position acquisition.
- A separate acquisition and interlocking module will be implemented in a different PLC to the motor driver.
- The control methods available will be LOAD REFERENCE POSITION, LOAD REFERENCE FUNCTON, DRIVE and READ.
- The references will be managed by MCS.
- The DRIVE command will be triggered (assumed to be by the start of ramp timing event). This will require a standard timing card in the TCDQ Interlock Control PLC.
- The control methods will be managed via FESA contracts.
TCDQ adjustment:
the betas at the TCDQ are pretty favourable wrt spot size (beta_x ~550m, beta_y ~180m) TCDQ will need to be adjusted to 1 sigma i.e. 0.6 mm at top energy.
At injection it will not come out but will go in later from 20 to 5 mm. A good orbit measurement will be required i.e. to 0.5 sigma or 0.25 mm.
The TCDQ will have move during the squeeze.
TCDQ has 2 motors each with DC motors and potentiometers/12 bit ADC (or possibly 16 bit). PID Feedback loop.
Possible use of LVDT - again calibration - possibly remote is an issue.
Follow a function - ramp & squeeze
- errors during function following possibly an issue
- no so critical TCDS behind etc.
- test in LSS6
- interlock is not following
- piecewise function proportional to sqrt(E) in ramp, for example
reference in volts - who's going to generate this
state, status, settings, acquisition, post-mortem, auto-calibration and function
Smallest step size 0.1 mm (better needs examination - oscillaition loop)
Fold in calibration/Non-linearity
Reproducibility 0.1 mm
Accuracy 0.2 mm ish