How can we install and commission the BMS faster and reduce the risk of not achieving practical completion?
The days of practical completion dates being pushed back to allow for Building Management Systems to be properly commissioned are over (read more here..). In this article I will describe an alternative delivery method that should allow installation to start earlier and create some breathing room between commissioning and practical completion.
In my opinion, one of the bottle necks delaying the start of BMS installation is the MSSB (mechanical services switch board), which is installed in a central location in each plant room, requiring all BMS instrumentation (sensors and valves etc), to be wired back to one location in the plant room. So, lets get rid of the MSSB altogether.
Current BMS installation and commissioning method
If we take a plant room with four AHU's for example, the AHU's are usually delivered early, as they are large long lead items, perhaps 3 months before power on is scheduled. The following is generally the sequence of who has installation priority in the plant room.
- Mechanical contractors have first priority to install duct and pipe through the plant room.
- At the same time the main electrical contractor installs the main power feed to the location of where the MSSB will be installed, and what ever other main electrical cables need to run through the plant room to other areas.
- The mechanical electrician then has priority to install the power feeds to the AHU VSD's.
- During the same time frame, the BMS installer is either waiting or is starting minor rough-in on the AHU's, but is mostly unproductive as there is either no route back to the MSSB yet or other trades have priority.
- The MSSB is then delivered and the mechanical electrician has priority to complete cable terminations. The rush is then on to get the MSSB powered up for air and water balancing to start, which should have started a month ago.
- BMS installation time just got compressed from 2 weeks to 2 days.
- Once air and water balancing are complete, mechanical witnessing starts, and more often than not BMS witnessing in the next couple of days, because practical completion is the following week.
- BMS commissioning just got compressed from 2 weeks to 2 days.
Distributed control Panels
Rather than install all four BMS controllers in the panel next to the MSSB and wait to wire across the entire plant room, install a small 500x500 control panel on each of the four AHU's. This will allow BMS installation to be started as soon as the AHU's are delivered. If you coordinate with mechanical, you may be able to get a couple of meters of supply and return duct work, and the heating and cooling valves installed early. You will then likely be able to install the following:
- Outside air and return air damper actuators
- Heating and cooling valves
- Supply and return air temperature and humidity sensors
- Supply air pressure transmitters
- Dirty filter pressure transmitters
If you coordinate with the mechanical electrician to install the supply fan VSD early, then you can wire the speed signal and HLI communications cable.
You will still need a small control panel at the MSSB for start/stop and status signals. I used to avoid network communications between start/stop controllers and temperature/pressure controllers of the same system, however, this reliability risk can be easily mitigated with good network design (read more here..).
Temporary power for commissioning
It is not unusual to temporarily power a control panel from the plant rooms temporary power board, but we can't always get this over the line with the Electrical Trade Union as the 24Vac circuit is energised throughout the entire plant room. However, in this case we are only temporarily powering one controller on one AHU. We should be able to get the following tasks completed:
- Controller addressed, configured and software program downloaded
- Any wired sensors commissioned and calibrated
- Any wired outputs (dampers and valves) stroked (leave the valves 100% open in preparation for water balancing)
- If you had a network controller and mini server on a trolley you could check the AHU's graphic, tests some alarms and trends etc.
We have now precommissioned at least 50% of each AHU months before power on dates.
Distributed controls are easy to implement as the BMS contractor is in control of this component and can make good progress without relying on the collaboration of other trades. A major game changer is distributed power. Rather than work around the bottle neck of the MSSB, just get rid of the MSSB all together.
Lets take a step back. Why do we have MSSB's?
- Before VSD's we needed a panel to house the star/delta contactors and overloads. We mostly now have VSD's, which need to be permanently powered, so there is no longer a need for contactors.
- We needed Run and fault lamps, because before BMS systems, that was the only way to check that the motors were running and that the overload on the contactors hadn't tripped. Facilities managers no longer walk through plant rooms at 6am to make sure everything is running; general status checks are done at the BMS. If you have a major failure of the BMS you can interrogate the VSD keypads for Run and Fault status.
- We needed Hand/Off/Auto switches because there was no other way to override the contactors. Overrides can be made at the BMS or at the VSD key pads.
- Fire and smoke control introduces a challenge as the MSSB houses these interlocks. Rather than running the fire rated communications cable to MSSB and installing the fire interfaces at the MSSB, simply run the communications cable round the plant room past each AHU and install the fire interfaces next to the VSD's. It's a bit more complicated than touched on here, but it can be done. We are going to need to run a BMS communications cable round to each AHU at some point so we can run the fire communications cable with it.
There are a couple of ways to move forward without an MSSB.
Install a small distribution board next to the distributed control panel on each AHU to house the following circuit breakers:
- 3-phase supply to the supply fan VSD
- 3-phase supply to the return fan VSD
- 1-phase supply to the AHU lighting
- 1-phase supply to the BMS control panel
Other than wiring all available instrumentation on the AHU three months early, we can now also pre-wire power to the VSD's with isolators etc. As there is no longer an MSSB, the start stop signal now comes directly from the BMS controller, so we can pre-wire those.
The original intent when I started looking into this in 2009 was to have AHU's fully wired with power and controls at the AHU factories. This didn't work within the tight time constraints of an AHU manufacturing line, but it was worth a try.
Even though we have distributed the power out to the AHU's, we still need a main distribution board in the plant room to supply power to each AHU's mini distribution board, or it can come from the main electrical board. At least now we only have one 3-phase power supply to each AHU rather than three to four separate power supplies.
The purpose of removing the MSSB should not be seen as a potential cost saving; distributing power and controls will likely increase the cost of the installation. The purpose is to de-risk the delivery of the BMS on practical completion and not to save money. Unfortunately, we are still restricted by what can be achieved within the limitations of 'lowest price wins'. I believe that in the future the focus will shift from who can build the cheapest to who can build the fastest.
In 2008-2010 I worked for Lang O'Rourke in London designing BMS systems for offsite manufacture. Here we have a control panel installed on each AHU. This AHU was for an operating theatre, so we had duty/standby supply and return fans. Notice how the AHU is still sitting on wooden blocks. The BMS installation started months before scheduled install start dates.
In the picture below I was testing the first AHU control panel at the panel manufacturer to make sure it was 100% correct before approving mass production (52 AHU's I think). Note the knife edge terminals at the top. We used these because any sensors not capable of being fitted off early would have their 24Vac power and signal isolated. The relays in this panel would have traditionally been in the MSSB but have now be moved into the control panel.
In this photo, the first of 4 offsite manufactured skids making up a complete chilled water plant room (excluding the chillers) is being positioned. Water balancing started 3 days after the skids were delivered.
In this photo the skid is still on the back of the truck. You can see the control panel, power board and 2 smaller distribution boards on the left. The remaining 3 skids housed the primary and secondary pumps, heat exchanges, pressurisation units and VSD's.
Here I am testing the first FCU, chilled beam, PAC unit controller at the panel manufacturer, before approving mass production. These control panels were issued to the FCU supplier for installation at their factories. Note the knife edge terminals, in this case, not to assist in temporary power and early commissioning, but to assist the FCU supplier in easily pre-wiring fan and valve actuators.