About the Author
Tom Spisak is senior controls engineer at Automated Cells & Equipment, Painted Post, N.Y., which is a Fanuc Robotics-certified servicing integrator that builds systems for machine tending, dispensing, press loading and unloading, palletizing, case packing, robotic spray coating, material removal and material handling.
An issue that isn't being discussed in our robotics cell industry is that safety standards and codes, such as NFPA 79 and 70E, have grown unnecessarily large and complicated to the point that many small manufacturers don't follow them. Most are just unaware that the standards exist or apply to them.
Likewise, for industrial electrical controls, everyone has to follow electrical standards intended for commercial facility wiring or follow standards written to protect workers based on older components. For example, older control panels with open NEMA starters should require full arc-flash protection (hard hat, face shield, ear protection, FP clothing), whereas a modern panel with finger-safe IEC components should warrant less-protective gear.
SEE ALSO: Toward Sane Safety Standards
Another example of more-complex, but less-user-friendly, safety standards is in robotic cell design. We'll design a cell that takes the person out of a hazardous area (loading a forge, lifting a heavy object), but the standards still require all kinds of effort and expense to keep adding layers safeguards to that area even though no one is in it, which means the actual safety return on investment is very low.
The standard committees write more and more standards, but never get together to make safety standards less complex and make safety simpler and cost effective.
No one wants someone to get hurt at work. The industrial controls industry doesn't have its own electrical safety. If you look at OSHA's statistics, electrocutions are just a part of a small slice on a pie chart. Most deaths from electrocution happen in the construction industry, not in industrial control panels. On the controls side, we're not building the old Frankenstein lab panels from 30-40 years ago that weren't safe to walk past, but we still have to follow the same old standards as if we were. You're far more likely to be hurt driving a car or walking from the parking lot to the office. Today our panels are safe enough to walk past, and most industrial controls are available in finger-safe designs.
We need to make safety simpler. If you have a robot with an eight-foot fence around it and a safety-rated, dual-channel door switch wired into a robot with an approved fence circuit, then that's enough. The standard should tell you that you're good to go. Small machine shops that want to buy a robot should only have one or two pages of instructions on creating a safety installation. Simplicity would encourage a lot more compliance and be easier to enforce.
Similarly, on the engineering side, the new Arc Flash 70E standard requires a high degree of skill to do its calculations, and it requires a lot of work and expense. The goal should be to make a safer control system in the most cost-effective manner. If this could be achieved for 90% of the control panels just by using Class J and CC fuses and using high-quality electrical components (contactors, power distribution terminal blocks), then that's what the standard should be.
Unfortunately, there's no advocate on the controls side for making standards simpler and less costly. No one wants to be seen as making safety standards less strict because the perception is that we'll be less safe. No one wants to be the critic and say that many safety standards are out of hand. As a result, we just get bigger and bigger standards, add second fences to cells and maybe a moat, too. But we never ask, “What's our safety objective? What are we trying to do?”
I don't say the existing safety standards should be thrown out, but we should step back, check for overkill and make a simpler formula for what's really needed to ensure safety.
If a panel is designed to be finger-safe, then a lesser degree of arc-flash protection should be permitted if you're only troubleshooting dc circuits. There's no reason to spend 20 minutes putting on safety equipment to troubleshoot the low-voltage, dc-powered portion of a panel, and our standards should reflect that reality. We just need to be more intelligent about what we're doing, use the analytical skills that were used to draft these standards in the first place and achieve practical safety with standards that are simpler, tighter and coherent.
