At last week’s Emerson Exchange, John Rezabek of ISP presented Wireless or Fieldbus? You might recall John from his columns in Control magazine and on the ControlGlobal.com site. He presented a cost comparison for a plant expansion considering Foundation fieldbus instruments versus WirelessHART-based ones.
John values the digital integration of field devices, which both technologies provide. I asked if he minded if I blogged his presentation and he was OK with it, as long as I wasn’t disruptive (only kidding on the disruptive part!)
John did an analysis on a project to purify a co-product using two distillation columns and adding intermediate and project storage as needed. The process required the measurement and control of flows, pressures, levels, and temperatures around the column. The project included 16 control valves, 14 control loops, and 16 indicate-only measurements.
Conduits, wires, and stainless steel junction boxes are expensive and complex. Wireless has the great promise to save design and installation costs associated with wired systems. Wireless does not replace all wires because of the power consumption requirements on some of the devices like control valves’ digital valve controllers. Also, some of the indicate-only devices, such as RTD temperature sensors, required wires and associated cable raceways. This lowers the incremental savings since the tray and junction boxes would still need to be present. John’s analysis also considered conduit and basket tray for the wires in both the wireless and Foundation fieldbus cases.
The project would already run a network of RGS conduit and twisted pair fieldbus cable for the control valves. The analysis compared the incremental costs for the wireless instrumentation-related services against Foundation fieldbus device-related services. Also, a home run wiring infrastructure for both cases was considered pre-existing. Other assumptions used for the analysis were list prices for the instruments and local, union, man-hour labor rates.
John calculated the Foundation fieldbus wired case to be just over $25,000 (USD) and the wireless case to be just over $27,000. The main difference is the difference in list prices of the devices and differences in the conduit and terminations required. In his summary, John noted that the actual price of the devices based on the buying agreement might have an impact on the analyses others might conduct.
From an engineering standpoint, John attributes more spec sheets, more devices, device cost, fewer drawing, lower change impact, less power conditioners, and similar configuration to the wireless case versus the Foundation fieldbus case.
From a design effort standpoint, no junction box drawings, fewer terminations, no termination hardware, no field power because of the batteries, less conduit, and less total copper used for the wireless case.
Points John made favoring a wired approach included: uncertainty about column temperature control, uncertainty about update frequency adequacy, spares diversity, signal integrity, and wireless protocol diversity with ISA100. His points favoring the wireless approach included easier infrastructure for future/unforeseen measurements and with things being more-or-less equal–creating a foothold for future opportunities.
He closed noting that 2-wire multi-drop bussed I/O squeezed the wireless installation advantage and the use of basket tray versus conduit tightens this advantage even more. As long as a wired backbone is needed for control services, wired strategies for indicate-only devices are cost comparable. The case for wireless was not a slam-dunk for this project, but (not unlike fieldbus) we all have to do the math for our individual circumstances.
As I noted in an earlier post with wireless devices used in control applications, technology advancements in control, communications, and power over time might alter the necessity of the wired backbone.