A heat exchanger is fouling, and economic conditions are such that even a brief shutdown to clean it is especially onerous. Portable measurements are reliable enough to indicate the problem isn't getting better by itself, but the boss would rather not rely on them. Running with crude measurements isn't going to provide the necessary insight into equipment viability, which could be degrading more quickly-or not. So what's the quickest way to get a handful of new measurements brought into the system?
In the conventional world of 20 years ago, we'd hope we had some spares in a nearby junction box, in sufficient numbers and of the correct signal level and type. Lacking that, we'd be hustling to route a "multipair" cable perhaps hundreds of feet to the vicinity of the process we had to monitor. If there wasn't space in an existing tray or some largely empty conduit (some of which might be underground), we'd erect raceways as well. This scenario assumes the host system in the control house has adequate spare I/O of the types required-not a given. Even with the sensors and transmitters in hand, the time and effort for installing and interconnecting copper could delay the integration of useful measurements for weeks.
In the last 20 years, hazardous-area-capable multiplexers have become commonplace. If your plant has them deployed, it's not inconceivable that a nearby "mux" could have some spares capacity. If not, it's equally viable that I/O bus extenders and modules of the desired type could be purchased and installed for the new points. Since you're in a hurry, maybe you'd have what you need in the storehouse, or you could rob some cards from a lower priority application. Modern offerings like Emerson's Charms and Honeywell's Universal I/O module might further simplify the effort, assuming you had the compatible Honeywell or Emerson host back in the rack room. Even with wireless, speed of deployment also hinges on measurement device availability.
If you have a more generic mux network deployed, there's a reasonable chance you're using an open standard such as Profibus DP, Modbus or Ethernet. Open standards increase your options, and afford you the opportunity to integrate more diverse solutions. If your solution demands sub-second, closed-loop control, you could deploy Foundation fieldbus using Softing's FG-110 FF, and implement control in the field. If your host is modern enough to offer a fieldbus H1 interface card, you can use a single, spare, twisted pair of conventional cable-provided it passes some minimal qualifications and isn't too long-to install a fieldbus segment. Such a fieldbus segment could support as many as 16 devices of all flavors on a single pair. Profibus DP users might have some "purple hose" in the vicinity to which they might add a PA coupler.
"Hey, you're already dealing with Modbus, right? Could fieldbus be that much harder?"
If you're unfamiliar with fieldbus, the pressure to get your solution installed and running fast might drive you to a solution you find less intimidating. But, hey, you're already dealing with Modbus, right? Could fieldbus be that much harder?
If you're comfortable enough with Modbus, you may have a quick path to integrate a wireless solution such as WirelessHART or ISA 100.11a. Both protocols offer Modbus-capable gateways, with the majority available in the "WiHart" flavor.
It's common for such gateways to include a web server, providing a fairly straightforward programming interface for setting Modbus communication parameters, and many of the variables necessary for individual devices. Once the gateway is configured, mounted, powered up and connected to the host system, you can start getting measurements within minutes of connecting field devices to the process.
So does wireless win the race? Depending on your existing spare capacity, storehouse stock and supplier availability, wireless may be the most expedient path to add dozens of new monitor-only indications, with the speed and savings increasing in proportion to the number of measurements you add.