Contents 

	problems with spectrometer reservations
	disk usage information
	FAQ 2004-03.1: uninterrupted power supply (UPS) - why do our spectrometers not have it?
	FAQ 2004-03.2: in case of a power failure can I get my sample out of the magnet?

The number one problem by far is the fact that there are still users who reserve time but never use the spectrometer at all during the reserved time slots. Some do that several times a day! Needless to say that this behavior is completely unacceptable, neither during regular working hours nor on weekends.

As there are obviously individuals in the Department who do not care enough about this problem despite several warnings, much more drastic measures are being implemented starting immediately: users who reserve but do not remove unneeded time will be warned once in writing, then their name will be removed from the user data bank, making it impossible to reserve time for a month, except by using another individual's username and password in which case this individual would suffer the same consequences.

You are responsible for what happens when your username and password is used for reservations - think twice before you share it with others. For those who think "they will never get me": the on-line reservation and LINUX systems keep extensive records of user activity or inactivity. This has already resulted in suspensions.

Another reoccurring problem is pooling of instrument time by groups. This also has been the subject of discussions several times. It is entirely feasible to make this impossible by disallowing adjacent bookings by members of the same group. However, as it is a computer that applies the rules, this becomes very strict and excludes legitimate instrument use by two members of the same research group, measuring different samples in adjacent time blocks which is perfectly okay. However, the following, for example, is unacceptable: when 3 members of the same group pool their time and run one sample right across the pooled time.

Overall, it is quite sad that in a University environment it is necessary to introduce such drastic measures. But there is little choice in the absence of better usage of the instrument time by individual users.

Disk usage information
When VNMR is started the user is given information about disk usage relative to the set disk quota. So far this information was based on a check of usage and quotas once every 24 hours. The process to do this has now been changed and it is done on an hourly base. The date and time of the last check is provided with the information.

Like in the past, data over 90% of the quota trigger a warning message in yellow to archive data and remove them from the server. Above 95% the same happens but in red and in words that make it clear that very soon there will be no more NMR measurements as the quota is full. Please note that the end user cannot change quota limits.

FAQ 2004-03.1: uninterrupted power supply (UPS): why do our spectrometers not have it?
This summer, the thunderstorms seem to be particularly strong, resulting in power failures such as last Saturday night. Interestingly, only the spectrometers in EB-44 stopped working (ibd5, u500) while all the others in the sub-basement were fine.

Neither the computers nor the NMR consoles are connected to UPS devices. The reason for this is quite simple: NMR consoles consume a large amount of electricity and consequently a UPS needs to be huge to provide power for any meaningful length of time. Not only would the batteries be extremely heavy and bulky but they also need periodic replacement. Overall, the cost and other inconveniences outlast the benefits. The computers, on the other hand, could be connected to much smaller UPS units. However, preventing the computer from going down in a power outage does not prevent the failure of an on-going experiment when the console quits, and hence this is of no benefit either.
The LINUX computers boot up without human intervention when the power comes back but the NMR consoles do not which can led to difficulties to remove a sample.
 

This latter scenario is typical if a power failure stops the console over night while an experiment is running, the power is restored and the user returns in the morning to a non-functioning spectrometer that has compressed air (compressed air pumps re-started).

Please note that this sample eject method is only meant to be used for this situation. If you use it to insert a sample it has a high probably to break because the air is cut off very rapidly and the sample drops "like a stone".