Physical diagnostics - Part II

Columnist John Rezabek says if you’re planning a new project, you’ll have some leverage to bundle the advanced diagnostics capabilities with your power conditioner, field termination and/or DCS.

By Industrial Networking, ControlDesign.com

By John Rezabek, Contributing Editor

John RezabekIn the spring issue, this column [“Physical-Layer Diagnostics”] discussed the need for physical-layer diagnostics, specifically some form of 24/7 vigilance for the cable, power conditioners, terminations and other physical aspects of your fieldbus networks. You apply these to selected segments, depending on the perceived risks and consequences in specific applications. Diagnostics for a refinery or a drilling rig in Siberia might be more extensive than those in a compact, indoor pharmaceutical plant. A few products already are available, and more will be released soon.

The first to market in North America was Pepperl+Fuchs’ ADM (advanced diagnostic module) in both a mobile and fixed version that mounts to P+F power hubs. The fixed version monitors up to four segments continuously, while the mobile handles one segment at a time.

The most detailed information about the segment(s) is revealed using field device tool/device type manager (FDT/DTM) software. Oscilloscope traces can be captured and examined in detail using the DTM. Fieldbus nerds with faster-spinning beanie-copters than mine even can filter transactions by type. This feature can be very useful if your device and DCS suppliers are debating who has an implementation problem.

The software automatically detects excessive noise, missing terminators, jitter (variation/drift in the zero-crossing timing of the waveform) and current/voltage problems. It records a baseline and automatically sets alarm levels for all the variables it monitors, including the voltage and signal strength at each device.

P+F also can help you get funding facts for the ADM. Download its white paper, "Advanced Online Physical Layer Diagnostics," which has very plausible economics for the time savings you’ll experience during commissioning.

MTL-Relcom is close to releasing a diagnostic module for its redundant eight-segment F890/892 power conditioners. The module appears as an H1 device on the first of the up to eight segments powered from the 890/892 power backplane. Peak and average noise levels in up to three frequency bands and signal levels at up to 32 fieldbus devices are updated in the device’s transducer block, along with the DC voltage at each segment.
 
Like P+F, the MTL-Relcom device also has a simple contact closure that can be wired as a “common trouble” discrete input to the host. While the MTL-Relcom solution consumes a “device” slot on one segment out of eight and adds a degree of asynchronous communications load, I don’t see this as a huge issue for most of us. Some argue that sending diagnostics on the same network that’s having problems is a mistake. I’d counter that, if the network is so bad that not even the diagnostics are getting through, you’ll know it and have bigger problems to solve.

MTL also plans to offer a one-per-segment solution that can be attached to any network, regardless of whose power conditioner is used.

At last fall’s ISA, Turck demonstrated a diagnostic module for up to 16 segments, powered from up to four of its four-segment, redundant power conditioner backplanes. It gathers and sends diagnostic information independent of the H1 segment or the DCS H1 interface, and uses high-speed Ethernet (HSE) to interconnect the modules and power backplanes. A Windows PC running DTM software is used for viewing the individual segment data, which includes device signal levels and jitter. Turck sets default alarm levels and also provides a contact closure for sending a “common trouble” alarm to the DCS. The new redundant power conditioner backplanes and HSE diagnostic modules are expected to ship in mid-2007.

MooreHawke released a diagnostic module that monitors power supply voltages from its conditioner backplane and includes fixed alarms for noise on each segment, alarming when noise exceeds 75 mV.

MooreHawke’s plan is for diagnostics to interface with the host over Modbus (serial or TCP/IP), rather than consume H1 resources. I think it’s intended for a sampling of segments selected by the end user as suspect or problems.

Chances are, if you’re planning a new project, you’ll have some leverage to bundle the advanced diagnostics capabilities with your power conditioner, field termination and/or DCS hardware purchase. If you do, I think it will pay off, not only during installation and commissioning, but also a few years down the road, when your local climate starts encroaching on your once-pristine installation.


  About the Author
John RezabekJohn Rezabek is a process control specialist for ISP Corp., Lima, Ohio. Email him at jrezabek@ispcorp.com.
 
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