Latest from Building Automation

Control Specifications: Ask for What You Need; Enforce What You Ask For

Jan. 1, 2005
During the late 1980s, I was working for a consulting engineering firm for which detailed control strategies and heavy field involvement were the norm.

During the late 1980s, I was working for a consulting engineering firm for which detailed control strategies and heavy field involvement were the norm. Drawing on some less-than-ideal experiences I had working for a design/build contractor, and with the controls industry moving from pneumatic to direct digital controls, I undertook the task of rewriting the firm's control specifications. The result was three specifications: one for pneumatic controls, one for electric/electronic controls, and one for direct digital controls. All shared several characteristics:

  • A fairly detailed statement of scope of work, including requirements for interfacing with other trades for the installation of sensors, dampers, valves, and interlocks.

  • A very specific (down to make and model) list of acceptable components. This was fairly radical, especially in the pneumatic specification, which included only components that had proven successful. As a result, many of the required components could be supplied from only one or two commercial product lines (all could be supplied from process-control product lines), and no manufacturer could meet the spec by supplying only products from a proprietary product line.

  • Very specific fabrication and installation details. In essence, this portion of the specs was written to prevent problems on past projects from recurring. For instance, having had trouble with NEMA ratings and compliance with NEC Article 725, I included very specific language regarding the fabrication and arrangement of control panels.

  • An augmented library of standard details.

  • Control drawings showing the location of sensors, control panels, and other major equipment.

  • Detailed narrative operating sequences with system diagrams showing how systems were to work.

The direct-digital-control spec also was supplemented by a fairly detailed points list.

Each of the specs was more than 100 pages long. When I showed them to the firm's senior project managers and principals, the reaction was: “We need to talk about this. Technically correct or not, we can't issue specs that are 100 pages long.” After we met, the specs grew to 125 pages, as it turned out I had forgotten some stuff.

Having concluded that perhaps we should issue specs — edited to reflect project requirements — that long, we faced the challenge of obtaining bids. Responses from the bidding community typically ran along these lines:

  • “Nobody does things this way.” (My response: “I know. That's why I wrote it all down for you.”)

  • “It will cost more if I have to use products from another manufacturer.” (My response: “That's OK. I think my project budget supports the spec I wrote.”)

  • “If I bid to this spec, somebody else will low-ball it and get the job.” (My response: “I hope not because I plan to enforce the requirements.”)

Enforcing the specs was not always easy. First, some of those who took exception were quite good at arguing their points. Then, there was the emotional pressure that came with forcing someone to do something I knew, while legal and technically correct, would cause him or her to lose money. (I had a dream in which a controls salesman I knew was selling pencils out of a tin cup on a street corner. He said, “I had a bright future ahead of me until you enforced the spec on that project I low-balled.”) I stuck to my guns, however, and enforced my specs — or at least their intent. As a result:

  • I started to get superior control systems. Commissioning became less a matter of getting things to work and more one of optimizing how they work.

  • The costs of my control systems were higher than those of other control systems; however, my control systems tended to produce better results that persisted longer. I budgeted for the higher costs, my belief being that the best machinery could be reduced to an inefficient mess (and, occasionally, debris) by a bad control system.

  • Some controls salespeople began to actually like the specs, which they believed leveled the playing field and gave them a foundation for doing what they were capable of doing.

Which brings me to my point: The control problems many of us so often complain about are not entirely the fault of controls manufacturers. Some can be traced to the early phases of projects, when budget and technical details are established (or not). Using vague, unenforced (and, perhaps, unenforceable) performance specifications in a competitive-bidding environment is an invitation to disaster.

What is needed is a change in culture:

  • Owners need to provide higher fees and more time. Consultants are underpaid to the point they leave much of the design function to controls vendors; additionally, the design window is so tight, they have little time to think. I have seen mechanical engineers hired mid-construction — the mechanical design issued as an add-alternate-bid package (with an addendum for controls) and bids due the following day.

  • Owners and design teams need to develop construction budgets significant enough for good equipment to be purchased and configured in a manner allowing design intent to be realized. That means sensors with accuracies greater than ±5 degrees, network configurations supporting the data handling necessary for program execution and trending, points for commissioning and operation, quality final control elements, etc.

For these changes to take place, the design community will need support in terms of education, training, and tools. Some of this support can be found in the form of “Large Commercial System Design Guide” (, “Control System Design Guide” (, and CtrlSpecBuilder (

In short, many product-quality and application problems exist because our current design and construction process will support nothing better. Just as we did not arrive at this situation overnight, neither will we emerge from it.

Supplementing the design effort will need to be a rigorous construction-observation, start-up, and commissioning process. Otherwise, our investment in high technology will be wasted.

I know many designers and controls professionals who are as alarmed as I am by the current state of affairs. Their alarm is compounded by their frustration at not being able to apply the things they know and the technology available to them because the realities of contract documents, the competitive-bidding environment, and the need to stay in business force their hand.

The bottom line: We in the design and construction community need to take the time to understand and ask for what we need and then enforce it. If we can do that, everyone will win: Designers will realize the intent of their designs, manufacturers will get to supply the best that technology has to offer, and building owners will realize the benefits of properly applied equipment and systems.

A member of HPAC Engineering's Editorial Advisory Board, David A. Sellers, PE, is a senior engineer specializing in commissioning and energy efficiency. Over the course of his career, he has worked in the design, mechanical- and controls-contracting, and facilities-engineering fields. Contact him at [email protected].