AutomatedBuildings.com Article - The Construction Process

All right. We finally have a controls contractor under contract. Now all we have to do is build the building. By now, we are either on the path to a well-built building or already on the path to yet another automation disaster.

Jim Henry
National Marketing Manager
Electromation, Australia

Contributing Editor

This article will consider the construction process up to the point of commissioning. You will notice that aspects are written tongue in cheek…

Firstly will be an overview of the construction process. Secondly will be a light-hearted look at a typical catastrophic project. Finally, we will consider proactive actions towards a quality project.

Basically, construction entails roughing in conduit and cable, pulling in wire, mounting devices and controllers, terminating, configuring controllers and programming, and commissioning.

Beyond the direct work, what procedures are associated with these construction steps?

An automation contractor cannot do too much work up front. Planning the project is critical. First, you must have one person who takes sole ownership of the entire project. This individual either needs to be intimately involved in the project, or requires the necessary skills and experience to organize and supervise a construction team to execute it.

Part of the contract negotiation process and tender review should include questioning the work method (approach, philosophy) that will be taken towards executing the project, looking toward best practice.

There are two approaches towards construction generally taken by automation contractors.

The first approach is to load one individual with the complete project. This individual is expected to handle every task for the job. The advantage to this method is that one person knows all aspects of the project. This approach has two glaring weaknesses. Firstly, when you get to the end of the project, there is not enough manpower to handle the variations in manning requirements (commissioning in particular). Secondly, there is no back up if this staff member becomes ill, goes on vacation, or has demands from previous, future, or concurrent projects.

The other approach is to have a team of employees who each specialise in different tasks for the project. One person engineers the project, another does the CAD drawings, another does the programming, another does the commissioning, and yet another does the graphics. Although you get very capable people handling each section of the project, problems will arise when specialists in one area do not understand all aspects of either the project, or the business of automation for that matter.

Automation contractors need to have the expertise and skills in all the following areas:

In most articles in this series I pontificate about the best ways to do things. However, in this instance I will waver and equivocate. Different philosophies are entirely valid. You have to take into account current workload, the skill levels of available staff, and the size of the automation operation. There will be times of heavy workload when the only consideration will be maximum execution.

Good automation contractors will always be thinking in terms of training and the future as they execute projects. This usually leads to having staff learn multiple skills, rather than just doing programming or graphics. This lends flexibility to handling varying workloads at different times

Now, let's take a look at normal construction execution (-execution of the contractor being the desired action the consultant and owner contemplate at completion).

The salesman has been chasing an order for three months. The mechanical contractor keeps on putting him off, while looking for some cheaper company to prostitute the work. Finally, the mechanical contractor gives up when he realises that he needs valves on site next week. So, the salesman is given a Purchase Order and told that he is holding up the project.

The salesman hands over a half-baked estimate full of dreams, concoctions, shortcuts, and misunderstandings along with his original quotation full of unrealistic promises and unachievable commitments. This file has no mention of all the verbal promises the salesman has made to get the project across the line. This sets the stage for a great job.

The project is put into the hands of an inexperienced engineer who takes a flying stab at designing a system, The senior engineer or engineering manager only tells the junior that he has to do the project within the costs allowed, or better yet, he is expected to find ways to shave money off the equipment currently suggested.

Next the junior sends his design to the mechanical contractor who has no idea what all these diagrams mean, who forwards it onto the engineer, who also has no idea, but is fully conscious of the fact that engineers are expected to be omniscient. So the engineer finds one or two things to complain about and puts red ink all over the submission. This means the drawings are suitably decorated, and can be sent back for revision. There is, of course, no actual comprehension of the real needs of the automation system by the salesman, the application engineer, the mechanical contractor, or the mechanical engineer.

Nothing happens until the mechanical contractor notices that the project is mostly finished and no one has roughed in any wiring for the controls. On exceptional projects, the mechanical foreman is oblivious and nothing happens until the builder notices that the drywall is up and there are no thermostats roughed in or any other sign of a control system.

Now the real fun begins. The Projects Manager for the automation company gets an irate phone call and has the sudden realisation that they have missed the start on yet another project. He reads through the design, shakes his head, and rewires the architecture on a scrap of paper and tells the lead hand to get down and get the wiring roughed-in some how.

Okay, by now the BMS panels and associated terminal devices have arrived and are sitting in the job box. Or at least some of them are. In reality the poor bastard who did the last project (with exactly the same problem that this sorry bastard is facing), has already pilfered the job box desperately looking for usable devices.

The project manager takes a look at what he has got, sadly shakes his head, goes to the storeroom and looks to see what is in stock that might actually work.

On site the lead hand is yelling at everybody and everything. The whole site is pretty well built out and the network is going to have to be fished through finished ceilings. Access panels are going to have to be installed and compromises made, such as surface running conduit in stairwells and the parking garage. Some drywall will have to be removed and reinstalled, and the general foreman is swearing about controls contractors in general and at the lead hand in particular. He points out that the issue of controls was raised at the last ten site meetings, but the controls contractor never showed up for one of them. The fact that the automation contract was issued for the first seven of those site meetings is not considered pertinent.

Somehow, using about five times the number of expected hours, the project is roughed in. Now, there is a serious look for additional savings to recoup the losses on the install.

Next the equipment arrives on site. The builder (general contractor) threatens liquidated damages if the project isn't ready for turn over to the client the following week. The installation manager pulls every available hand off all the other sites, which are only just approaching this level of desperation.

Among all of this swearing, demanding that controls not delay the project, when the controls contractor is finally ready to start commissioning in just three days (what the programme / schedule show to be a three week process) until it is time to turn the project over, we find that half the switchboard wiring still isn't complete.

Okay, we have successfully reached the commissioning stage. This will be the subject of the next article in this series.

Okay, you get the general idea. Things can go dreadfully wrong during construction and you may never see it. It is easy to point out weaknesses in the process. What is useful is to identify actions that the client and consultant can take to ensure competent construction will occur. We have looked at what makes a really bad project. What are some heads-up actions that make a good project?

First, we must backtrack. We are assuming that we have followed the procedures of the previous three articles and selected BAS technology that is relevant to our project, have written effective contract documents and tendered the project so that we have a reasonable level of confidence that we can expect a quality BMS that meets our needs.

The only way to ensure you are going to get a well-built project is to do one (or both) of two things. First, hire a BMCS contractor who has an exemplary track record. The alternative is to do lots of up front work. This is instead of simply reacting when pushed, when it is too late and painful. Remember; Proactive = Successful.

The controls contract should be let as early in the process as possible. Certainly not after the concrete has been poured for half the building. How the hell are you going to rough in conduits in the slab for the CO monitoring system in the parking garage and the stair pressurisation system in the stairwells otherwise?

Contractors should clearly spell out what they are providing prior to being awarded a contract. They should give a hardware points list. This should be thoroughly reviewed by someone who understands building automation. About ninety percent of the intent of a project can be ascertained from the points list.

Meetings between the construction teams of each trade should establish clear delineation of responsibilities (primarily between the mechanical electrical contractor and the controls contractor).

A realistic programme with milestones and targets should be established. Payment can then be matched to these targets.

A large part of our confidence stems from our specification and tendering procedure. However, the success of the project largely depends on several factors during construction. In reality, we have only specified that we are getting the correct architecture platform. If we have based part of our decision on the references and history of the contracting firm we can be reasonably comfortable that they will perform to the same standard again.

If we do not have this luxury we need to take actions that ensure the project is installed well. Due diligence is what will get us there.

If the contractor has not worked for the client or consultant before, they should be required to mock up sections of the BMCS and demonstrate the working architecture. This is one hell of a lot easier before than trying to rectify or improve the BMCS architecture after it has been installed.

We need to thoroughly review the submittal drawings to ensure we are getting a solid architecture and that the I/O points covers the bases for a complete system. We need to verify that the contractor has understood our design intent and that the terminal devices are of correct quality and size. Valves have been sized correctly; damper actuators have sufficient torque to meet our requirements.

Correct inputs and outputs should be straight forward if the specification included a points list.

The engineer needs to investigate that the correct quality of devices has been proposed. For example, differential pressure sensors for filters instead of air flow switches, modulating 0-10vdc actuators instead of floating actuators, actuator feedback signals where requested, and current sensors instead of switches.

Pretty simple. However, complexities arise, in that the controls contractor must execute the BMCS contract in harmony with the rest of project.

In reality, if the builder (general contractor in North America) and the mechanical contractor use experienced and competent Project Managers, the process will probably go well. If they are idiots, inexperienced, and/or assholes (the proper technical term in construction) then the chances are high that the project is going to be a series of problems.

There is no single factor more important than the builder (general contractor) having a savvy Project Manager on any building site; More important than the architect, the mechanical engineer or any other person. A highly competent Project Manager can take control of the whole project, fix mistakes, achieve realistic schedules, and control subbies.

We need to perform each operation at the correct time to meet the schedule of the construction of the building. Equipment needs to be ordered in time that it can be delivered when needed. Valves, wells, VAV controllers will be needed early. Computers need to be ordered right at the end of the project. Ordering equipment to late delays the project. Ordering equipment too early causes cash flow problems.

We need to coordinate our work with other trades. In contracting it must be realised that there is an entire art of sloughing off one's work on other contractors. Traditionally the controls contractor attempts to see how much of their work they can suck the electrical contractor into taking responsibility for.

We need to ensure other trades understand the work they need to do for the BMCS contractor. The mechanical contractor will need to install valves and pipe wells for us.

Regular attendance at weekly site meetings to ensure coordination of the project is crucial. (Exceptions to this are large projects where there may be as much as a six-month period where there is no action by the controls contractor, after rough in and prior to mounting devices.)

The BMCS contractor must push the electrical and mechanical contractors to ensure that equipment will be in place early enough to connect to and still leave the necessary period to commission the project.

Okay, assuming we have followed the recommendations above and have assembled the correct staff, then we should get the wires in the right places, have the right points allowed for, and mount all the devices in the right places, and be ready to commission the project three weeks prior to completion.

Okay. We have probably all had enough of construction. The next article will move us forward to the commissioning process.

Envenergy	Paxton Control Valves
Our Sponsors	Resources