deletions | additions       diff --git a/sectionDISCUSSION_We.tex b/sectionDISCUSSION_We.tex  index 0d80b0f..6186fb7 100644  --- a/sectionDISCUSSION_We.tex  +++ b/sectionDISCUSSION_We.tex 
...
We propose a simple, well-defined, formal operation metrics for storage ring light sources, to make the reliability of these facilities comparable.   The metrics itself clarifies for each facility the beam parameter promised to the users, and the statistical data does reveal, how well these promises are kept.  The direct comparison of the failure modes of the different facilities revealed several differences of the current situation with respect to the proposed metrics.  We will highlight these differences now and propose measures to resolve them.  \subsection{Discussion of the Proposed Failure Modes}  {\em 'No-beam':} events are defined to start when the beam current drops below a given limit.  For several facilities this is currently not the case.   Table~\ref{table:pf_limits} shows, that ALBA, BESSY2 and LNLS-UVX do consider a closure of the photon shutters by an interlock equivalent to a loss of the electron beam.   Other facilities, like the SLS do not have these kind of interlocks. 
...
A 'downtime' often only ends after the photon shutter or insertion device control is given back to the users, which is not compatible with the 'no-beam' rule.   Again this can be solved by having additional event types for 'photon-shutter-closed' or 'insertion-devices-blocked-open' while there is not a 'no-beam' event.  This solution requires then to record in addition if the event was preceded by another failure: if the photon-shutters get an interlock, the beam is then dumped and afterwards it takes fife minutes until the insertion-devices are unblocked, then it should be visible from the failure data that this were not three but only one individual interruption in user operation.  We've decided to allow arbitrary limits for 'no-beam' events, but in reality this does not make a large difference: situations are rare where the current drops from 100\,mA 100$\,$mA  below 50\,mA 50$\,$mA  but not to zero. {\em 'Low-beam-current':} events can vary significantly with the definition of the I(tol) limit.  At SPring-8 already a beam decay of about 0.1\% causes a 'low-beam-current' event, while at BESSY2 up to 9\% additional beam decay are still tolerated as insignificant for the users.  The comparison of the number and duration of 'low-beam-current' events for BESSY2 and SPring-8 won't be useful therefore.  But at At  least the comparison of the tolerated beam decay is still useful; and for those facilities where the limits are similar, the failure rates will allow a meaningful comparison of the reliability of the injector system.  But the {\em 'Low-lifetime':} events are apparently rare events at most facilities.   The  example of ALBA shows, that it still  makes sense to record and evaluate these events, too. The normal beam lifetime varies a lot between facilities and operation modes.   what is the nominal beam lifetime at the SLS would be considered a very low lifetime at ALBA or SPring-8.  But a significant decrease in the beam lifetime can cause problems at most facilities and should therefore be recorded to judge the reliability of the facility.  But this {\em 'Beam-blow-up':} events are again infrequent at most facilities.  This  again depends of the betatron coupling the machines operate with: at very low coupling even small effects can lead to large relative changes of the vertical beam size. For some facilities the vertical beam size is difficult to measure, if the emittance is very low the beam size can be very small where traditional beam size monitors become diffraction limited.  Still without a defined limit for the tolerated variation of the beam size a published emittance and coupling becomes meaningless; and the numbers of failures to stay within these limits is an essential measure for the reliability of a synchrotron light source.  Again {\em 'Distorted-orbit':} events are recorded at most of our facilities, but are rarely taken into account in the yearly evaluation of  the failure statistics.  The  limits when an orbit is considered to be distorted are varying by orders of magnitudes between different facilities. Already the publication of these limits would be very useful to compare facilities;   the failure data for this mode would be an essential marker for the reliability of a synchrotron light source.  {\em 'Distorted-filling and Bunch-purity':} events are not relevant at some facilities.  Mostly time resolved measurements are depending on a clean filling or even a high bunch purity.  Very few facilities have the means to measure the bunch purity on-line: a ratio of $10^{-8}$ of the bunch charge of adjacent bunches can only be measured in a lengthy procedure.  Still many facilities do have the means to detect deviations from their nominal bunch charge distribution.  Where these means exist we encourage people to publish their failure limits and the failure data.  This will enable a comparison of the facility capabilities for reliable time resolved measurements.  {\em 'Beam-unrelated':} events should be recorded, whenever they have an impact on a significant number of beamlines.  These events are rather rare and often facility specific.  {\em 'Short-user-time':} are for most facilities just subtracted in the beam availability calculations; they are not recorded as events.  An independent recording of these events would enable to calculate beam availabilities with and without accounting for the 'Short-user-time';   that would improve the comparison of facilities that handle 'Short-user-time' differently. 
...
We want to distinguish between shifts given to the users at short notice and extra shifts announced to the users at least a month before.   While it is a common practice to compensate users for daylong day long  outages by extending the user run, these extra shifts can often not be used by the users: other liabilities often require the users to leave and the extra time is sometimes only used by half of the beamlines. If the additional time is announced month in advance then this allows for some flexibility for the beamline responsible to reschedule their users; leading to a much better utilisation of the extra time.