What cost pervasive networking?

Comparing prices of HART, WirelessHART and fieldbus can lead to surprising results

By John Rezabek

The plant had determined that monitoring the bleed line of a double-block-and-bleed isolation assembly would easily reveal whether either of the block valves were leaking. Since this was a key to validating the effectiveness of the isolation system, Sandra, the new instrument specialist, was tasked with bringing these points into the DCS. Problem was, there were 50 new temperatures to monitor.

For this back-of-the-envelope comparison, Sandra made a few assumptions. The cost of sensors was assumed to be the same, whether they were connecting to HART, WirelessHART or fieldbus. Sandra included the labor and material to bring the necessary signals to the host system, but made no attempt to compare incremental host I/O infrastructure—the cost of serial I/O or a Modbus TCP card is not a deal breaker compared to Foundation Fieldbus (FF) H1 or Profibus cards.

To get a generic quote for hardware pricing—no special discounting—Sandra used the “Rosemount Online Store” for the transmitters and Allied Electronics for as much of the bulk hardware as she could find there (cable, conduit, terminals, junction boxes, and so on).

2017 State of Technology Report: maximize I/O flexibility

Errors and shabby workmanship have had an unfortunate impact on some early fieldbus installations, where a loose termination or a compromised transmitter can impact reliability.

For the HART 4-20 mA transmitters, Sandra priced the single-input 248. The Rosemount Online Store prices 50 at about $33,000. The wireless version of the 248 came in at over $83,000 for 50, so instead she opted for quantity 13 of the four-sensor 848TX, saving about $15,000. This choice makes wiring of sensors a little more complex, since the 248’s could have been mounted integrally to each sensor.

But it was comparable to the fieldbus option, which uses the eight-input fieldbus 848T. The FF option reduces transmitter hardware cost to only $24,000 for quantity 7. After adding batteries and a gateway, Sandra wonders whether the wireless option will pay for the $37,000 premium relative to wired HART, and nearly $45,000 compared to wired fieldbus, for her 50-point project.

Copper prices are volatile and sometimes cable prices can fluctuate weekly, but when Sandra searches the web for 12-pair, individually shielded, Div. 2- and tray-rated cable, she finds it for about $2 per foot (ca. $1,000 (list) per 500 ft. on Allied). To span the 700 ft. to each field junction box, she’ll need (5) 12-pair cables for the HART solution, or about $7,000 in multi-pair home run cables – and maybe, say, $10,000 to cover waste and rework. From the junction box to devices, Sandra budgets for 16 AWG twisted, shielded pairs—at $0.33 per foot, she figures another $1,000 for 3,000 ft.

The question becomes, with prevailing labor rates, how much conduit, wire and terminations will the remaining $25,000 buy before the wireless option begins to pay out? Setting aside $15,000 for conduit, supports, tools and equipment, there’s only $10,000 left for labor. In Sandra’s area, that only gets her two electricians for about two weeks. You might see the scale tipping distinctly toward the wireless approach.

But fieldbus came in very competitive. Cable is $0.99 per foot (list), and Sandra only needs 2,000 ft. for the whole job. She finds modular Phoenix junction boxes, power supplies, and device couplers online for less than $1,500 total for seven spurs and a home run. The remaining difference in cost to the 13 wireless transmitters—about $40,000—is more than enough for the relatively light infrastructure for single-pair cable. Labor costs would need to be astronomical and productivity dismal before wireless breaks even.

For this case, fieldbus wins. But wireless often wins not because it’s cheaper, but because it’s easier. Errors and shabby workmanship have had an unfortunate impact on some early fieldbus installations, where a loose termination or a compromised transmitter case can impact the reliability of an entire segment. There’s incremental engineering—some might over-analyze and obsess about segment loading, but the same folks might also invest in overzealous wireless site evaluations.

Sometimes wireless is a slam-dunk. Next month’s column will continue the discussion of paths to pervasive networking, and when the choices—despite costs—are more obvious.

Show Comments
Hide Comments

Join the discussion

We welcome your thoughtful comments.
All comments will display your user name.

Want to participate in the discussion?

Register for free

Log in for complete access.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments