Power & Grounding

From my RSS feed, an Automation World article, Wireless Eliminates Ground Loop Noise, caught my attention. It brought back memories of being a systems engineer on offshore oil and gas production platforms. These platforms are huge steel structures and as such, have a nice solid ground.

These structures also act as antennas for electromagnetic interference (EMI) from the starting and stopping of large motors and generators. I became very familiar with the filtering features on the inputs to the automation systems we had at the time.

Unfortunately, I didn't then know ModelingAndControl.com's Greg McMillan who pointed out some downsides with signal filtering:

...signal filtering can be used to smooth out the noise but this adds a lag that reduces the ability of the flow loop to deal with pressure disturbances and valve issues.

The article identifies what will warm the hearts of Electrical Engineers but may be overlooked by other engineers:

What's more, the tests turned up an added bonus. "Because the wireless transmitters are battery powered, this totally isolates them from spurious ground-loop potentials that get introduced into any reaction vessel from a number of different sources, and which drive pH sensors crazy," Broadley observes. During the test, when a ground-noise-induced spike appeared on the wired pH signal, it was not present in the wireless transmitter output.

I know how much time I spent chasing spurious, phantom issues caused by EMI or ground loops on onshore facilities. I imagine a few of the plant engineers who come upon this post, may have similar stories to tell. For some plants, this may turn out a bigger advantage than the wiring savings or trapped diagnostic information in HART field devices.

MP3 | iTunes

I know from my early days as a systems engineer, that power and grounding can sometimes cause vexing, intermittent issues. It's like chasing ghosts, which is an apropos analogy with Halloween occurring yesterday here in the Western world.

I caught up recently with Dan Jacobsmeyer, a specialist for Emerson's FlexConnect solutions. These solutions connect pre-engineered cables to the existing automation system termination panels and marshall the field signals to DeltaV I/O.

Dan shared a recent story where a process manufacturer was having some grounding issues with 1-5V analog I/O cards. Ground loop currents were causing a voltage offset, increasing the error on what the I/O card was reporting to the control system, versus its actual measured value. A solution discussed was to convert the I/O over to 4-20mA.

Dan pointed out that replacing the 1-5V cards with 4-20mA cards is a way to solve the immediate voltage offset problem. However, not fixing the ground issue may result in future problems that may be very difficult to diagnose.

Dan recommended the following course of action based upon DeltaV site preparation documentation.

A proper earth ground is critical. It safely conducts stray electrical current to earth for personal safety and good electrical noise control. Building steel must be part of a good earth ground system to obtain an equal-voltage-potential ground between the steel and the automation system ground networks.

Automation system faults are often the result of poor or faulty grounds. IEEE Standard 1100-1992, Recommended Practice for Power and Grounding Sensitive Electronic Equipment describes industry-accepted methods.

Isolation transformers should be used for power transfer from plant A/C service to the individual automation systems. A single point ground eliminates ground loops and provides excellent isolation between the automation systems and other plant systems.

Dan sums up his recommendations to use the isolation transformer between the A/C power source and automation system and have each system with a dedicated connection to a true-earth plant ground grid. Make sure no other systems share this connection point.

Paying close attention to these recommendations and not masking over them will save a lot of time chasing ghosts and delivery more reliable and accurate system operation.

I recently saw a post from one of the process automation world's newest bloggers, Dave Harrold, on the topic of power and grounding. The post, A Non-sexy Means to More Reliable Operations spoke to the importance of power and grounding. Dave summed it up well:

I've been around this stuff for a very long time and I'm confident that if you spend eight hours per month for the next six months finding and fixing instrumentation and control system power and grounding problems that your instrumentation, control, and safety systems will perform much more reliably and you know what that means - when the boss is happy, everyone is happy.

I caught up with Dave Brown, a Principal Engineer in Emerson's Process Systems and Solutions technology organization for his thoughts on Dave Harrold's post. He gave me a great analogy. It's like building a house. Your power and grounding is like the foundation. The better it is designed and built, the fewer problems you will have with everything else. Automation systems, safety systems and all your instrumentation and plant equipment will perform better.

So what does all this have to do with too much heat? Dave has also just written a whitepaper entitled Effects of Heat and Airflow inside an Enclosure. It discusses how mean-time-between-failure (MTBF) numbers are based on Mil standard specs of 25 degC. For every 10 degC rise above this amount, the MTBF cuts in half. Although equipment can be rated much higher, like 60 degC, its likelihood of failure rate will increase.

Hmmm... this makes me want to check the temperature in my audio/visual cabinet at home.

Equipment with several years of run-time use actual field data for the MTBF calculations. The numbers usually improve given the conservation nature of the Mil standards.

Dave's whitepaper offers some practical solutions for heat dissipation from the placement of the equipment within the cabinet to the auxiliary designs for cooling and air flow.