B513 Planning Meeting, 14th December

Present:           Tony Cass Marc Collignon, Anne Funken, Dick Minchin, Jukka Lindroos, Dave Underhill

1.             Previous Minutes

  • Given the budget situation, it is extremely unlikely that any part of the cost for replacing the lighting in the vault will be covered by the general maintenance budget. Nevertheless, the lighting should be replaced—discussions during the week highlighted the need to improve the provision for emergency lighting. Overall, the lighting level should not be less than in the machine room today. The placement of the emergency lights depends on the placement of the equipment racks—which in turn awaits the provision of a plan at a 1:50 scale.
  • Tony has still to arrange discussion of the pre-heating coils with system managers. Jukka commented that this element has not been included in the tender.

·         The HVAC tender is in the final stages before issuing. Jukka expects it to be sent out today or early next week. This represents a delay of 1 week relative to Anne’s schedule.

  • The UTP boxes have been ordered.
  • Nigel has not yet had chance to study placing robots in vault; this is expected next week.
  • Anne has still to find the date of the next site committee. [Confirmed later as either 13th or 20th February.]
  • The DAI for the removal of two humidifiers in the plant room has been created (and signed by Tony).
  • Further work is needed before a tender for a new chiller can be issued. Tony emphasised the need to have the chiller installed and paid for during 2002.

2.             Dismantling Status and Plans

  • All the UTP cabling has gone. The old coax cable is being removed with the electrical cabling. Dick will move the tape boxes against the North wall to allow the removal of the cable ducts there.
  • Anne asked if we really want to remove the old canalis “normabarres” on the ceiling—in particular as some sockets in the MG room area are powered from one of these bars. This question was considered later by Anne, Tony and David during a visit to the vault. The MG room sockets can easily be removed, but other sockets elsewhere, including some in the machine room, appear to be powered from these bars. Overall, it was felt that, despite the small advantage of leaving the bars as a potential backup supply, it is best to remove these bars now. In this way we can find all the undocumented connections and start with a clean situation. Any normabarres for which there is a clear need (or for which the removal costs would be unreasonable) could be left.
  • Although the old false floor tiles are of no use to IT, they can be reused elsewhere at CERN—and are worth some 100KFF. Anne will check for likely customers. The cloisons from the vault are also available for reuse, most probably in the MG room.
  • The dismantling of the air-conditioning units is scheduled to start on January 7th. The remaining piping in the MG room is associated with these units and will be removed at the same time.
  • It is possible to reroute the UPS cables that run through vault. There are two independent sets of cables which can be taken through the air-conditioning area on an existing cable tray. Moving the cables will not lead to lack of UPS cover, just the removal of the “spare” module needed for redundancy. Anne has asked for an estimate and the work could be done in January.

3.             Installation Plans

  • Following discussions within ST, it has been agreed that the false floor and civil engineering work can be carried out, starting in January, under an existing ST/CE contract. Anne commented that this is a very positive development—the company concerned is well qualified for this work. Jose Miranda from ST/TFM —who was being shown round the vault and introduced to the project by Nigel—will be following this work. The price estimate that is being requested will include costs for a false ceiling in basement corridor.
  • Although TIS have objected to the proposed 10% slope of the ramp from the lift up to the false floor, this is only a problem for wheelchairs and manual transpalettes. A 10% slope is not a problem for powered transpalettes. As we have a powered transpalette and there is no requirement for disabled access, the 10% slope will be maintained.
  • Jukka asked if there would be a workshop area for the air conditioning company to carry out welding and similar operations. It is unlikely that such an area can be provided, but this work can be done outside.
  • A number of cables that pass through walls that will be demolished have been identified. These may have to be moved before demolition works; this may affect operations in the machine room. Anne is investigating the exact use of these cables.
  • Anne is preparing a note on the new electrical distribution arrangements. Anne is still concerned that the addition of supervision and control will reduce the overall reliability. In addition to the possibility of human errors, a local failure of the 48V supply could lead to a loss of power for all equipment supplied by one PDU. The utility of a remote control system needs to be balanced against the likelihood of failure and also the cost.
  • In addition to the direct costs, a remote control solution indirectly increases the costs of the “critical equipment” area as the power for the network equipment should be controllable independently. If there is no remote control then one set of normabarres can serve both network and other equipment, leading to savings at the PDU level.
  • A decision on whether or nor remote control is needed must be made by the end of January.

4.             Asbestos Risk

  • Following a review of the vault and installations, together with Claude Jacot from ST, it is clear that the risk of the works disturbing any asbestos is extremely low. All of the walls to be removed are concrete and equipment that could contain asbestos, such as joints in the air-conditioning ducts, was installed well after the use of asbestos was discontinued.
  • Although the risk is extremely low, ST will nevertheless check against previous studies of the building for asbestos content and will come back if there is any chance that special precautions are needed.

5.             Requirement for differential protection on PC racks

  • Anne and Tony had had a meeting with Guy Salomon of TIS on Monday to discuss the need for differential circuit breakers to be installed on each rack of PCs.
  • Although differential protection is provided at the PDU level, the efficiency of differential circuit breakers depends on the length of the cable to the load. Additional differential protection is therefore provided at the level of the sockets in case long extension cables are used to connect equipment.
  • However, the relevant electrical codes do not require the use of differential circuit breakers at the socket level where it is clear that the sockets are provided only for supplying equipment that is nearby. Although special sockets are preferred in such cases (to prevent the use of normal plugs), if there are clear operational procedures and warnings then we do not need differential circuit breakers on the equipment racks.
  • Guy Salomon will contact firms to see if special sockets (and the associated plugs/cables!) can be provided to prevent connection of non-computing related equipment. In any case, “general purpose” sockets will be installed around the walls of the vault and some barrettes with differential protection could be installed on the side of equipment racks to supply mobile consoles.

6.             AOB

  • The report into the state of lightning protection for B513 has been received. At first sight, the report indicates that the building is well protected, although some minor repair work is needed. The report includes some comments to the effect that the lifetime of the existing UPS could be extended significantly by replacing some capacitors. Anne will investigate. If this is possible then it might be interesting to use this UPS to serve the critical equipment.
  • Next meeting: 2pm, Friday 18th January.