B513 Planning Meeting, 23rd September 2002

Present:           Nigel Baddams, Tony Cass, Anne Funken, Jose Miranda

1.             Previous Minutes

  1. Approved.

2.             Substation planning

  1. Anne has reviewed the UPS and secured (diesel) supply requirements to cover 200kW for 2 hours, including support for the air conditioning unit, pumps and lighting. Two 300kVA UPS modules would be required and, in the worst case scenario, the diesels would need to support a total load of 1.4MVA. Space for the two 250kVA modules has been allocated in the substation layout and Anne will check whether or not the 1.4MVA load is acceptable (this is above the 1.2MVA that is known to be available).
  2. Anne commented on two aspects of the UPS for the physics load
    1. The MGE specifications require a maximum phase-neutral distortion of 7% downstream. The only measurements we have are at the PC rack level. This is a long way away, but the distortion of 11% means that compensatory modules may be required between the load and the UPS. These will need space, but sufficient should be available.
    2. The MGE and Gutor solutions differ in the way the UPS modules are connected to the 400V supply. Whilst the MGE solution has one connection per UPS module, the Gutor solution has one unit which connects to the switchboard and to which each UPS modules is connected. The Gutor solution thus has a single point of failure, but the MTBF of the connection unit is extremely high.
  3. Anne presented an initial substation layout which required a 21.4m´10.7m bunker. The following points were made in the discussion that followed.
    1. As yet, no attention has been paid to constraints imposed by ducts and possible door placement.
    2. The requirements for false flooring in existing rooms needs to be understood. This has an impact on accessibility—more space might be required to lift equipment.
    3. Nigel commented that there would be no impact on the car park if the bunker width could be kept to 10.5m.
    4. There is no easy way to arrange the rooms in the substation such that they can be reuded directly in a 4MW solution. Some remodelling would be required unless the bunker is built at full size initially. Allowance can be made now for any future remodelling by using beams to support the roof instead of load bearing walls. The interior walls could then be cloisons which would be easy to remove later—although it was noted that these could not be built up from the false floor but would have to extend the full height of the bunkerfor safety reasons.
    5. Safety reasons also require the air conditioning units to be placed outside the substation rooms. The required units can probably fit in the existing technical areas, avoiding an additional dedicated room in the bunker. However, ducts would have to run across the corridor. The corridor will be fairly cluttered at ceiling level given the ducts and the many cables that are needed. An underfloor caniveau would be possible but costly. This should therefore be avoided if at all possible.
  4. With the proposed layout, at least 1.2MVA of UPS capacity can be installed in the areas housing the transformers and low voltage switchboards (S-414/418/419). This allows an installation schedule that would spread the costs of the physics UPS over a number of years:
    1. Construct the bunker, install new low voltage switchgear and relocate the transformers.
    2. Remodel S-414/418/419 to house UPS modules and batteries.
    3. Install 1.2MVA of UPS capacity to replace the current UPS. (This step can be delayed until early 2006).
    4. Remove current UPS and remodel S-412 as required.
    5. Upgrade UPS capacity by 1.2MVA in 2007 and again in 2008.
  5. Tony queried the installation on batteries and UPS modules in the same room. Anne replied that although not preferred, this arrangement is now allowed as battery technology has developed and there is no risk of gas leaks.

3.             Supporting a 4MW load

  1. In addition to the point above (2.3d), Anne pointed out some other obstacles to supporting a 4MW load:
    1. The Isolde 18kV loop does not have sufficient capacity to support the overall load of 10MVA (total capacity is only 15MVA).
    2. It looks rather difficult to accommodate a UPS to support 4MW for 10 minutes.
    3. The siteside diesels cannot support anything more than the predicted 200kW load for critical equipment. There would thus be problems if the power demand here were to increase as well.

4.             AOB

  1. Next meeting: 14:00, Tuesday 8th October.