Machine Safety: Thinking of bypassing an e-stop? Have you talked to an attorney?

Is it ever permissible to install a bypass switch or to disable an e-stop? For some reason, some Internet “safety forums” have had this continuing discussion. See the one requirement consistently missing from these international discussions. And ask these 4 questions.

10/25/2013


Jokab Safety, part of ABB, showed a giant e-stop at a Pack Expo show a few years ago. CFE Media photo

Is it ever permissible to install a bypass switch or to disable an e-stop? For some reason, a few “safety forums” on the Internet have had this discussion going for months. These are international discussions, yet one requirement in the analysis is missing consistently.

On this subject, can you actually imagine several engineers corroborating and contributing argumentation on how to install and why not to install a bypass switch on an e-stop circuit. The discussions veer off into SIL levels, PL (Performance Levels), MTTFd (Mean Time To Fail dangerous), B10 values, and more. Sometimes there’s even a brief discussion about the Machinery Directive or different standards like IEC 61508-1, IEC 62061, ISO 13849-1, and IEC 60204-1, and different ways to derive calculations based on different interpretations.

Just when it looks like the discussion is about to arrive at a conclusion someone asks something like, "Well, just what is it you’re trying to accomplish? And, why is this question on the table?" Then a strategy discussion launches on why and how a safety function can be bypassed.

So, what is the one requirement not even hinted at so far? That’s right, the risk assessment!

Why is it that these discussions can go on for months back and forth and no one interjects the "risk assessment" discussion?

4 questions to ask if you're thinking of bypassing an e-stop

Here are some of the questions I can think of related to this topic:

1. Does the current risk assessment allow for a given hazard to become unmitigated?

2. Does your company safety policy allow for a given hazard to become unmitigated?

3. Do machine safety application standards like IEC 60204-1, ISO 13849-1, or in the U.S. – NFPA 79 allow e-stops to be by-passed or disabled?

4. Is there any established and documented compliant basis whereby it is okay to bypass an e-stop?

Well, if the answers to these questions are generally, "NO," then, why would anyone even consider designing and engineering an e-stop bypass? 

Additionally, I’d want to know if the individual asking the original question is talking about an application based on international standards or domestic U.S. standards. The answer to that question might then promote discussions involving risk assessment, that is: ISO 12100 versus ANSI B10.0. I’m sure you get the point!

J.B. Titus, CFSE

Has this presented you with any new perspectives? Do you have some specific topic or interest that we could cover in future blog posts? Add your comments or thoughts to the discussion by submitting your ideas, experiences, and challenges in the comments section below.

Also see additional posts linked at the bottom of this article.

Related articles:

Calculation of MTTFd for a pushbutton

Machine Safety – consequences of not performing risk assessments!

Machine Safety – does a risk assessment need to be updated for a minor modification to a machine?

Machine Guarding & The Hierarchy of Measures for Hazard Mitigation



BRUCE , MO, United States, 11/02/13 08:14 AM:

Gentlemen: The correct answer is: Expert Judgment. PERIOD.

No standard or group of standards and safety committees can trump expert judgment.

We should look at standards as GUIDELINES. These guidelines could be envisioned as boundary conditions within which good process procedures and operations are expected for normal good operations.

Instead, design and execution of good and safe process control should EXCEED standards - particularly where serious safety factors are concerned.

I am seeing too many cases of XYZ Standard Documents being used in discussion groups to replace expert judgment. We have a plethora of standards piled so high now that we've forgot what trumps all of them. Now you know.

Back to the answer. Instead of worrying about whether a particular e-stop can be bypassed for some special need such as machine setup and whether the risk is acceptable and follows all the XYZ standards -- focus on changing the situation to provide a safe operation. If there is ANY foreseen risk of serious injury when this estop is bypassed (based on Expert Judgment not standards)then work on reducing that serious risk to zero or at the vary minimum to a level where Expert Judgement plus process experts are in conclusive agreement that the risk is now nil and you are safe.

There are the unpredictable risks that must also be dealt with and Experts will spot this. Often the machine and process knowledge and discipline of the operator, setup tech, (or anyone who touches the equipment) is the highest risk after making the control factors safe. Be sure to take care of this safety factor -- no matter what the standards say.

The real solution to this question should not have to consult a single standard document to decide on how to make the required special operation with a bypass of the main E-stop button safe. The solution should please the safety guys as they will find it meets or exceeds all standards.

With greatest respect for all the hard working standards and safety engineers please seek out true hands-on experienced and knowledgeable experts and put Expert Judgement as the first requirement in all safety standard documents and standards as the determining factor - don't rely and piles of standards to guide accepatble safety questions - there are too many variables which no number of documents will cover.
GREG , WA, United States, 11/15/13 03:47 PM:

I can see the need to have a bypass for guards to allow access while running. What would be the need to bypass an E-Stop?