aveform at the digitizing rate of the acquisition hardware.  A significant portion of the spectrum consists of zeros, but there is some risk in setting a threshold value too high and neglecting to store some of the signal in the data file.  If the data burden is not too great, it would probably make sense to store each entire waveform rather than trying to filter it.

  The 10bit 8GS/s digitizer boards from Acqiris are capable of achieving the required specifications, although a single board will need to be dedicated to each time-of-flight channel in order to achieve the 100ps temporal resolution.  The base memory on the DC222 single channel board is capable of acquiring 1024kpnts, enough for 128μs at 8GS/s, more than adequate for this application.

  The digitizer boards will be triggered by a pulse from the event receiver that is delayed an appropriate amount from the primary trigger to correspond to the arrival of the FEL pulse in the experiment’s interaction region.  Jitter on the order of one time channel, or ~100ps, will not have a serious impact on the data. The delay may be varied to zoom the time window into a particular region, so should be an easily changed variable.  The delay value is typically set to something close to the desired value and then set empirically though examination of the data.

  Spectra will be grouped according to specific criteria, summed, displayed and potentially stored and/or analyzed.  The sorting criteria may be changed at later times, so all raw spectra should always be saved except for those rejected by some independent criteria.  Save/Reject criteria will be something like whether or not the LCLS lased on a given shot.  Binning criteria will be things like power density, wavelength, etc.  

  Five electron spectrometers will be used in the high-field physics end-station and each will eventually require an individual 8GS/s digitizer.  The diagnostics chamber will also have a magnetic bottle electron time-of-flight spectrometer that will have the same detector requirements although with different operating parameters (i.e. maximum duration, trigger delay, etc.).

Ion Time-of-Flight Spectrometer (Wiley McLaren type)
 

  The ion time-of-flight spectrometer acquires data using the same principle as the electron time-of-flight spectrometers; a signal is acquired as a function of time following photoionization of the sample by the FEL radiation.  In the case of the ion TOF, the flight time is proportional to the square root of the mass-to-charge ratio, m/q, of the ion.  The flight time for ions is considerably longer than it is for electrons (depending entirely on the length of the flight tube and the accelerating voltages used), and subsequently, less temporal resolution is required from the digitizing electronics.  Flight times of up to 1ms should be accessible with the electronics, although in most situations, much shorter flight times will be necessary.  Temporal resolution of 0.5ns should be sufficient.
 

  Figure 4: Ion time-of-flight spectra for Ar atoms and clusters measured with 2Ũ1013 W/cm2 of 32.5 eV FEL radiation.

  A further complication arises when we want to use the electron and ion spectrometers at the same time.  A high voltage pulse must be applied to the interaction region at some time (on the order of 100ns) after the FEL pulse photoionizes the sample.  Electrons will quickly depart the interaction region while no electric field is present.  Ions, which move much more slowly, will remain in the interaction region in the absence of electrostatic repulsion arising from high charge densities, and can be swept out into the ion spectrometer with a high voltage pulse. 

  A delayed trigger must therefore be implemented for the pulse generator.

  The Ion TOF Spectrometer is used to measure when the ions arrive to within 25 psecs. Peak power is increased as the focus is made tighter. An 8 GHz digitizer is used to sample the signal over approximately 10 usecs of time. Each Ion arrives at some energy at a given point in time. The resultant data is stored as a sparse matrix, with representative time of arrival / current for each data point that surpasses a threshold setting. This is the first instrument to be used. A typical pulse is expected to yield around 100 data points.

  This instrument may also be used for a final focus of the optics. After the diagnostics have been used to focus the beam in the interaction region, a scan may be done using the ion TOF as the read-back for the focus optics.

Ion Imaging Spectrometer
 

  The imaging ion spectrometer measures the momenta of ions by imaging the intersection of the expanding sphere of ions with the flat detector.  The image consists of rings of intensity corresponding to the velocity of the ions (with ions expand