Emergency Stop – What’s so confusing about that?

This entry is part 1 of 11 in the series Emergency Stop

I get a lot of calls and emails asking about emergency stops. This is one of those deceptively simple concepts that has managed to get very complicated over time. Not every machine needs or can benefit from an emergency stop. In some cases, it may lead to an unreasonable expectation of safety from the user, which can lead to injury if they don’t understand the hazards involved. Some product-specific standards

This entry is part 1 of 11 in the series Emergency Stop

I get a lot of calls and emails asking about emergency stops. This is one of those deceptively simple concepts that has managed to get very complicated over time. Not every machine needs or can benefit from an emergency stop. In some cases, it may lead to an unreasonable expectation of safety from the user, which can lead to injury if they don’t understand the hazards involved. Some product-specific standards mandate the requirement for emergency stop, such as CSA Z434-03, where robot controllers are required to provide emergency stop functionality and work cells integrating robots are also required to have emergency stop capability.

Defining Emergency Stop

Old, non-compliant, E-Stop Button
This OLD button is definitely non-compliant.

So what is an Emergency Stop, or e-stop, and when do you need to have one? Let’s look at a few definitions taken from CSA Z432-04:

Emergency situation — an immediately hazardous situation that needs to be ended or averted quickly in order to prevent injury or damage.

Emergency stop — a function that is intended to avert harm or to reduce existing hazards to persons, machinery, or work in progress.

Emergency stop button — a red mushroom-headed button that, when activated, will immediately start the emergency stop sequence.

and one more:

6.2.3.5.3 Complementary protective measures
Following the risk assessment, the measures in this clause either shall be applied to the machine or shall be dealt with in the information for use.

Protective measures that are neither inherently safe design measures, nor safeguarding (implementation of guards and/or protective devices), nor information for use may have to be implemented as required by the intended use and the reasonably foreseeable misuse of the machine. Such measures shall include, but not be limited to,

a) emergency stop;

b) means of rescue of trapped persons; and

c) means of energy isolation and dissipation.

Modern, non-compliant e-stop button.
This more modern button is non-compliant due to the RED background and spring-return button.

So, an e-stop is a system that is intended for use in Emergency conditions to try to limit or avert harm to someone or something. It isn’t a safeguard, but is considered to be a Complementary Protective Measure. In terms of the Hierarchy of Controls, emergency stop systems fall into the same level as Personal Protective Equipment like safety glasses, safety boots and hearing protection. So far so good.

Is an Emergency Stop Required?

Depending on the regulations and the standards you choose to read, machinery is may not be required to have an Emergency Stop. Quoting from CSA Z432-04:

6.2.5.2.1 Components and elements to achieve the emergency stop function
If, following a risk assessment, it is determined that in order to achieve adequate risk reduction under emergency circumstances a machine must be fitted with components and elements necessary to achieve an emergency stop function so that actual or impending emergency situations can be controlled, the following requirements shall apply:

a) The actuators shall be clearly identifiable, clearly visible, and readily accessible.

b) The hazardous process shall be stopped as quickly as possible without creating additional hazards.
If this is not possible or the risk cannot be adequately reduced, this may indicate that an emergency stop function may not be the best solution (i.e., other solutions should be sought). (Bolding added for emphasis – DN)

c) The emergency stop control shall trigger or permit the triggering of certain safeguard movements where necessary.

Later in CSA Z432-04 we find clause 7.17.1.2:

Each operator control station, including pendants, capable of initiating machine motion shall have a manually initiated emergency stop device.

To my knowledge, this is the only general level machinery standard that makes this requirement. Product family standards often make specific requirements, based on the opinion of the Technical Committee responsible for the standard and their knowledge of the specific type of machinery covered by their document.

Note: For more detailed provisions on the electrical design requirements, see NFPA 79 or IEC 60204-1.

Download NFPA standards through ANSI

This more modern button is still wrong due to the RED background.
This more modern button is non-compliant due to the RED background.

If you read Ontario’s Industrial Establishments regulation (Regulation 851), you will find that the only requirement for an emergency stop is that it is properly identified and located “within easy reach” of the operator. What does “properly identified” mean? In Canada, the USA and Internationally, a RED operator device on a YELLOW background, with or without any text behind it, is recognized as EMERGENCY STOP or EMERGENCY OFF, in the case of disconnecting switches or control switches. I’ve scattered some examples of different compliant and non-compliant e-stop devices through this article.

The EU Machinery Directive, 2006/42/EC, and Emergency Stop

Interestingly, the European Union has taken what looks like an opposing view of the need for emergency stop systems. Quoting from Annex I of the Machinery Directive:

1.2.4.3. Emergency stop
Machinery must be fitted with one or more emergency stop devices to enable actual or impending danger to be averted.

Notice the words “…actual or impending danger…” This harmonizes with the definition of Complementary Protective Measures, in that they are intended to allow a user to “avert or limit harm” from a hazard. Clearly, the direction from the European perspective is that ALL machines need to have an emergency stop. Or do they? The same clause goes on to say:

The following exceptions apply:

  • machinery in which an emergency stop device would not lessen the risk, either because it would not reduce the stopping time or because it would not enable the special measures required to deal with the risk to be taken,
  • portable hand-held and/or hand-guided machinery.

From these two bullets it becomes clear that, just as in the Canadian and US regulations, machines only need emergency stops WHEN THEY CAN REDUCE THE RISK. This is hugely important, and often overlooked. If the risks cannot be controlled effectively with an emergency stop, or if the risk would be increased or new risks would be introduced by the action of an e-stop system, then it should not be included in the design.

Carrying on with the same clause:

The device must:

  • have clearly identifiable, clearly visible and quickly accessible control devices,
  • stop the hazardous process as quickly as possible, without creating additional risks,
  • where necessary, trigger or permit the triggering of certain safeguard movements.

Once again, this is consistent with the general requirements found in the Canadian and US regulations. The directive goes on to define the functionality of the system in more detail:

Once active operation of the emergency stop device has ceased following a stop command, that command must be sustained by engagement of the emergency stop device until that engagement is specifically overridden; it must not be possible to engage the device without triggering a stop command; it must be possible to disengage the device only by an appropriate operation, and disengaging the device must not restart the machinery but only permit restarting.

The emergency stop function must be available and operational at all times, regardless of the operating mode.

Emergency stop devices must be a back-up to other safeguarding measures and not a substitute for them.

The first sentence of the first paragraph above is the one that requires e-stop devices to latch in the activated position. The last part of that sentence is even more important: “…disengaging the device must not restart the machinery but only permit restarting.” That phrase requires that every emergency stop system have a second discrete action to reset the emergency stop system. Pulling out the e-stop button and having power come back immediately is not OK. Once that button has been reset, a second action, such as pushing a “POWER ON” or “RESET” button to restore control power is needed. Point of Clarification: I had a question come from a reader asking if combining the e-stop function and the reset function was acceptable. It can be, but only if:

  • The risk assessment for the machinery does not indicate any hazards that might preclude this approach; and
  • The device is designed with the following characteristics:
  • The device must latch in the activated position;
  • The device must have a “neutral” position where the machine’s emergency stop system can be reset, or where the machine can be enabled to run;
  • The reset position must be distinct from the previous two positions, and the device must spring-return to the neutral position.

The second sentence harmonizes with the requirements of the Canadian and US standards.

Finally, the last sentence harmonizes with the idea of “Complementary Protective Measures” as described in CSA Z432.

How Many and Where?

Where? “Within easy reach”. Consider the locations where you EXPECT an operator to be. Besides the main control console, these could include feed hoppers, consumables feeders, finished goods exit points… you get the idea. Anywhere you can reasonably expect an operator to be under normal circumstances is a reasonable place to put an e-stop device. “Easy Reach” I interpret as within the arm-span of an adult (presuming the equipment is not intended for use by children). This translates to 500-600 mm either side of the center line of most work stations.

How do you know if you need an emergency stop? Start with a stop/start analysis. Identify all the normal starting and stopping modes that you anticipate on the equipment. Consider all of the different operating modes that you are providing, such as Automatic, Manual, Teach, Setting, etc. Identify all of the matching stop conditions in the same modes, and ensure that all start functions have a matching stop function.

Do a risk assessment. This is a basic requirement in most jurisdictions today.

As you determine your risk control measures (following the hierarchy of controls), look at what risks you might control with an Emergency Stop. Remember that e-stops fall below safeguards in the hierarchy, so you must use a safeguarding technique if possible, you can’t just default down to an emergency stop. IF the e-stop can provide you with the additional risk reduction then use it, but first reduce the risks in other ways.

The Stop Function and Control Reliability Requirements

Finally, once you determine the need for an emergency stop system, you need to consider the system’s functionality and controls architecture. NFPA 79 is the reference standard for Canada and the USA, and you can find very similar requirements in IEC 60204-1 if you are working in an international market. EN 60204-1 applies in the EU market for industrial machines.

Download NFPA standards through ANSI
Download IEC standards, International Electrotechnical Commission standards.

Functional Stop Categories

NFPA 79 calls out three basic categories of stop. Note that these are NOT reliability categories, but are functional categories. Reliability is not addressed in these sections. Quoting from the standard:

9.2.2 Stop Functions. The three categories of stop functions shall be as follows:

(1) Category 0 is an uncontrolled stop by immediately removing power to the machine actuators.

(2) Category 1 is a controlled stop with power to the machine actuators available to achieve the stop then remove power when the stop is achieved.

(3) Category 2 is a controlled stop with power left available to the machine actuators.

This E-Stop Button is correct.
This E-Stop button is CORRECT. Note the Push-Pull-Twist operator and the YELLOW background.

A bit later, the standards says:

9.2.5.3 Stop.
9.2.5.3.1 Each machine shall be equipped with a Category 0 stop.

9.2.5.3.2 Category 0, Category 1, and/or Category 2 stops shall be provided where indicated by an analysis of the risk assessment and the functional requirements of the machine. Category 0 and Category 1 stops shall be operational regardless of operating modes, and Category 0 shall take priority. Stop function shall operate by de-energizing that relevant circuit and shall override related start functions.

Note that 9.2.5.3.1 does NOT mean that every machine must have an e-stop. It simply says that every machine must have a way to stop the machine that is equivalent to “pulling the plug”. The main disconnect on the control panel can be used for this function if sized and rated appropriately. For cord connected equipment, the plug and socket used to provide power to the equipment can also serve this function. The question of HOW to effect the Category 0 stop depends on WHEN it will be used – i.e. is it being used for a safety related function? What risks must be reduced, or what hazards must be controlled by the stop function?

You’ll also note that that pesky “risk assessment” pops up again in 9.2.5.3.2. You just can’t get away from it…

Control Reliability

Disconnect with E-Stop Colours indicates that this device is intended to be used for EMERGENCY SWITCHING OFF.
Disconnect with E-Stop Colours indicates that this device is intended to be used for EMERGENCY SWITCHING OFF.

Once you know what functional category of stop you need, and what degree of risk reduction you are expecting from the emergency stop system, you can determine the degree of reliability required. In Canada, CSA Z432 gives us these categories: SIMPLE, SINGLE CHANNEL, SINGLE CHANNEL MONITORED and CONTROL RELIABLE. These categories are being replaced slowly by Performance Levels (PL) as defined in ISO 13849-1 2007.

The short answer is that the greater the risk reduction required, the higher the degree of reliability required. In many cases, a SINGLE CHANNEL or SINGLE CHANNEL MONITORED solution may be acceptable, particularly when there are more reliable safeguards in place. On the other hand, you may require CONTROL RELIABLE designs if the e-stop is the primary risk reduction for some risks or specific tasks.

To add to the confusion, ISO 13849-1 appears to exclude complementary protective measures from its scope in Table 8 — Some International Standards applicable to typical machine safety functions and certain of their characteristics. At the very bottom of this table, Complementary Protective Measures are listed, but they appear to be excluded from the standard. I can say that there is nothing wrong with applying the techniques in ISO 13849-1 to the reliability analysis of a complementary protective measure that uses the control system, so do this if it makes sense in your application.

ISO 13849-1:2006 Table 8
ISO 13849-1:2006 Table 8

Extra points go to any reader who noticed that the ‘electrical hazard’ warning label immediately above the disconnect handle in the above photo is a) upside down, and b) using a non-standard lighting flash. Cheap hazard warning labels, like this one, are often as good as none at all. I’ll be writing more on hazard warnings in future posts.

Use of Emergency Stop as part of a Lockout Procedure or HECP.

One last note: Emergency stop systems (with the exception of emergency switching off devices, such as disconnect switches used for e-stop) CANNOT be used for energy isolation in a Hazardous Energy Control Procedure (a.k.a. Lockout). Devices for this purpose must physically separate the energy source from the down-stream components. See CSA Z460 for more on that subject.

Read our Article on Using E-Stops in HECP.

Pneumatic E-Stop Device
Pneumatic E-Stop/Isolation device.

Standards Referenced in this post:

CSA Z432-04, Safeguarding of Machinery

NFPA 79-07, Electrical Standard for Industrial Machinery
Download NFPA standards at ANSI

IEC 60204-1:09,  SAFETY OF MACHINERY – ELECTRICAL EQUIPMENT OF MACHINES – PART 1: GENERAL REQUIREMENTS

Download IEC standards, International Electrotechnical Commission standards.

ISO 13849-1-2006, Safety of machinery — Safety-related parts of control systems — Part 1: General principles for design

See also

ISO 13850:06, SAFETY OF MACHINERY – EMERGENCY STOP – PRINCIPLES FOR DESIGN

Download IEC standards, International Electrotechnical Commission standards.
Download ISO Standards

Why Conventional EMC Testing is Insufficient for Functional Safety

At the recent PSES Symposium, I attended a couple of interesting workshops on EMC and Functional Safety. One was called “Workshop on EMC & Functional Safety” presented by Keith Armstrong, Bill Radasky and Jacques Delaballe. The other was a paper presentation called “Why Conventional EMC Testing is Insufficient for Functional Safety” presented by Keith Armstrong.

For readers who are new to the idea of Functional Safety, this field deals

At the recent PSES Symposium, I attended a couple of interesting workshops on EMC and Functional Safety. One was called “Workshop on EMC & Functional Safety” presented by Keith Armstrong, Bill Radasky and Jacques Delaballe. The other was a paper presentation called “Why Conventional EMC Testing is Insufficient for Functional Safety” presented by Keith Armstrong.

For readers who are new to the idea of Functional Safety, this field deals with the ability of a product or system to function in it’s intended use environment, or in any foreseeable use environments, while reliably providing the protection required by the users. Here’s the formal definition taken from IEC 61508-4:1998:


3.1.9
functional safety
part of the overall safety relating to the EUC and the EUC control system which depends on the correct functioning of the E/E/PE safety-related systems, other technology safety-related systems and external risk reduction facilities

3.2.3
equipment under control (EUC)
equipment, machinery, apparatus or plant used for manufacturing, process, transportation, medical or other activities

NOTE – The EUC control system is separate and distinct from the EUC.

Table 1: (E/E/PE) electrical / electronic / programmable electronic

Reliability requirements are found in two key standards, ISO 13849 and IEC 61508. These two standards overlap to some degree, and do not define reliability categories in the same way, which frequently leads to confusion. In addition there is a Machinery Sector Specific standard based on IEC 61508, called IEC 62061, Safety of machinery – Functional safety of safety-related electrical, electronic and programmable electronic control systems. These three standards make reference to EM effects on systems but do not provide guidance on how to assess these phenomena. This is where IEC TS 61000-1-2 comes into play.

All three experts are members of IEC TC 77 and are directly engaged in writing the second edition of IEC TS 61000-1-2 (more info on this at the bottom of this post). This IEC Technical Specification deals with electromagnetic (EM) effects on equipment that result in functional safety problems, like failures in guarding circuits, or failures in some of the new programmable safety systems. This is becoming an increasingly important issue as programmable controls migrate into the traditionally hardwired safety world. In fact, Keith pointed out that EM effects are present even in many of our “tried and true” circuits, but the failures have been incorrectly attributed to other phenomena because most electrical engineers have not been used to thinking about these phenomena, especially in 24Vdc relay-based control circuits.

In the workshop, the presenters discussed a typical product life cycle, then went on to explore the typical environments that a product may be exposed to, including the EM and physical environments. They went on to discuss the need for an EMC-related Risk Assessment and then finished up by looking at Electromagnetic Safety Planning. The whole workshop took the entire second day of the Symposium.

A key point in the workshop is that conventional EMC testing cannot practically prove that systems are safe. This is due to the structure of the EMC tests that are normally undertaken, including the use of fixed modulation frequencies during immunity testing, failure to assess intermodulation effects and many other issues. In addition, EMC testing does not and cannot test for aging effects on performance, wear & tear and other use-related conditions. The presenters discussed a number of ways that these problems could be addressed and ways that testing could be extended in selective ways to attack predicted vulnerabilities. EMC testing does not consider the reliability requirements of the tested product (i.e. IEC 61508-1 SIL-3 or SIL-4).

On the following morning, Keith Armstrong presented his paper. In this paper, Mr. Armstrong went into considerable detail on the shortcomings of conventional EMC testing when it comes to Functional Safety. He suggested some approaches that could be used by manufacturers to address these issues in safety critical applications.

The workshop presentations and Mr. Armstong’s paper can be purchased through IEEE Xplore for those that did not attend the Symposium.

The IET has published a new book, available for free from their web site, entitled Electromagnetic Compatibility for Functional Safety. This guide will be reviewed in a future post, so keep reading!

Keith Armstrong, Bill Radasky and Jacques Delaballe are members of IEC Technical Committee 77, writing IEC TS 61000-1-2 Ed 2.0, ELECTROMAGNETIC COMPATIBILITY (EMC) – PART 1-2: GENERAL – METHODOLOGY FOR THE ACHIEVEMENT OF THE FUNCTIONAL SAFETY OF ELECTRICAL AND ELECTRONIC EQUIPMENT WITH REGARD TO ELECTROMAGNETIC PHENOMENA. Edition 2 of this standard should be published by Mar-2009 according to the IEC.

Keith Armstrong is Principal Consultant at Cherry Clough Consultants in Brocton, UK.

Bill Radasky works with Metatech Corporation from his office in Goleta, California.

Jacques Delaballe works for Schneider Electric Industries SAS in Grenoble, France.