4B Hazard Monitoring Equipment Selection, Installation and Maintenance

www.go4b.com

Page 1 of 34

Hazard Monitoring Equipment

Selection, Installation and Maintenance

Johnny Wheat

4B Components Ltd,

East Peoria, IL, USA

Introduction

When selecting hazard monitoring equipment for bucket elevators and belt conveyors,
there is a myriad of choices available. However, before you decide on the specific type of
sensor and control you must first decide on what parameters you are going to monitor. On
certain equipment there are laws and regulations that must be followed. OSHA Standard
– 29 CFR /

Grain Handling Facilities – 1910.272

and NFPA 61:

Standard for the

Prevention of Fires and Dust Explosions in Agricultural and Food Processing Facilities

are two important US documents. However these standards should be used as a starting
point for the minimum requirements. The plant manager or other responsible person
should look at each piece of machinery individually and decide which additional areas
need to be monitored and then look at other machines on the plant which may fall outside
the scope of these documents but which should also be considered. Obviously budget
constraints come into play and acceptable risk for the company and shareholders must be
considered, however as with all insurance policies, you should not skimp and should
purchase as much as you can afford. Typically the cost for hazard monitors is actually
quite reasonable and is good insurance and a sound investment. Many companies will
actually install sensors and controls on every piece of equipment in the plant.
Unfortunately for many plants this is only done after the “horse has bolted”, i.e. after a
major disaster has occurred. Nowadays grain dust explosion reporting is a little more
forthcoming and plant managers/grain companies are becoming more aware of the
frequency of these events and the potential for disaster. Kansas State University

maintains records of explosions reported within the industry (see table 1). Unfortunately
these are only what are reported; many more explosions go unreported and still more
events occur daily which fortunately do not result in explosions but do cause serious
machine down time and lost productivity. Major education at expos and technical
conferences does help to inform users of the hazards and make them more aware of the
need to monitor the equipment and to be more comfortable with making not just the
required investment but also going the extra mile. Some customers install monitoring
equipment in stages and start by meeting the minimum requirements with carefully
chosen equipment and systems that can be easily expanded at a later date to encompass
other machines in the plant as further funds and resources become available. Once a
decision has been made on which machines are to be monitored and which areas on those
machines could cause danger, then the type of sensor and control system can be looked
at. This paper will provide a guide to selecting equipment, taking into consideration these
regulatory and practical factors.

www.go4b.com

Page 2 of 34

Table 1.

Agricultural Dust Explosions in the US

(Source: KSU March 20, 2006 Report)

Year

1996 1997 1998 1999

2000

2001

2002

2003

2004 2005 Total

Number

18 7 8 9 8 8 6 13

106

As with any type of electrical device used in an industrial environment, its usefulness,
service life, reliability and also maintenance cost is very much dependent upon the
quality of the initial installation. This is even more evident when we consider the unique
challenges within the feed and grain industry. There are many outdoors applications
where the equipment and installation must withstand harsh environmental conditions
including wide ambient temperature fluctuation (–40º F to + 120º F), wind, rain, sleet, ice
and snow. Sometimes seasonal workers and inexperienced employees have to work with
and around the equipment. There can also be rodents around the plant, which could gnaw
on cables causing maintenance problems with equipment. We will explore the practical
implementation of installing hazard monitors in the harsh environment of the feed and
grain industry so that they can be reliable and trouble free.

Once installed correctly the hazard monitoring system must be checked and maintained
on a regular basis. The frequency of this testing is determined by the user taking into
consideration the recommendations of the equipment manufacturer. Manufacturers of
specific equipment usually have recommendations on how to maintain and how often to
test equipment, however it is the user of the equipment who must ultimately decide on
how frequent the tests and maintenance is to be carried out. High quality systems, which
have been professionally installed, should not need a great deal of maintenance or
periodic testing. However, external influences on the system can compromise even the
most failsafe designs. As such, periodic system testing is extremely important and should
be made a priority in any plant’s preventative maintenance program. We will look at
some important considerations with regard to maintaining hazard-monitoring equipment.

Equipment Selection

The tables and diagrams below are a guide to Occupational Safety & Health
Administration (OSHA) and National Fire Protection Association (NFPA) regulations
with regard to the monitoring of certain bucket elevators and belt conveyors. As
discussed these should be the starting point and the minimum requirements for each
machine.

www.go4b.com

Page 3 of 34

Belt speed and belt
misalignment sensor
Pos. A for plastic buckets

Plugswitch

chute
blocked

detector

Belt misalignment
sensor and head
pulley alignment
sensor

Bearing

temperature

sensor

Underspeed

/ slip

sensor

Plugswitch /chute
blocked detector

BUCKET ELEVATOR

Belt speed and belt
misalignment sensor
Pos. A for plastic buckets

Plugswitch

chute
blocked

detector

Belt misalignment
sensor and head
pulley alignment
sensor

Bearing

temperature

sensor

Underspeed

/ slip

sensor

Plugswitch /chute
blocked detector

BUCKET ELEVATOR

Belt speed and belt
misalignment sensor
Pos. A for plastic buckets

Bearing temperature
sensor

Table 2.

Guide to OSHA and NFPA requirements for monitoring devices on bucket elevators

Hazard OSHA

Requirement

(See 1910.272 for full
interpretation)

NFPA Requirement
(See NFPA 61 for full
interpretation)

Belt slip

Motion detection device to
provide a shutdown at 20%
reduction in normal belt
speed.

Same as OSHA, but also
activate an alarm.

Belt misalignment

Belt alignment monitoring
device, which initiates an
alarm.

Belt alignment monitors at
head and tail pulleys.

Bearing failure

Bearing temperature or
bearing vibration monitors.

Same as OSHA.

Pulley misalignment

No requirement to date.

Head pulley alignment
monitors.

Plugged spout

No requirement to date.

High-level indicators for
vessels, which the elevator
discharges to.

Figure 1.

Bucket elevator sensor locations

Note: If the elevator has a bend (knee) pulley, those bearings should
also be monitored. Installation of additional belt alignment sensors
should also be considered at elevator idler pulleys.

www.go4b.com

Page 4 of 34

Tail pulley
misalignment
sensor

ENCLOSED BELT CONVEYORS

Level detector /

plugswitch

/ chute

blocked indicator

Bearing

temperature

sensor

Belt alignment
sensors

Rotation

sensor

Bearing
temperature
sensor

ENCLOSED BELT CONVEYORS

Level detector /

plugswitch

/ chute

blocked indicator

Bearing

temperature

sensor

Belt alignment
sensors

Rotation

sensor

Bearing
temperature
sensor

Figure 2.

Enclosed belt conveyor sensor locations

Table 3.

Guide to NFPA requirements for monitoring devices on belt conveyors

Hazard NFPA

Requirement (See NFPA 61 and

NFPA 654 for full interpretation)

Belt slip

Motion detection device to provide a shutdown
at 20% reduction in normal belt speed and
actuate an alarm.

Belt misalignment

No requirement to date however true alignment
must be maintained to minimize friction.

Bearing failure

No requirement to date.

Pulley misalignment

No requirement to date however true alignment
must be maintained to minimize friction.

Plugged spout

High-level indicators for vessels, which the
conveyor discharges to.

Classes, Divisions, and Zones:

Once a decision has been made on which parameters are to be monitored then the
monitoring equipment specification can be considered. The first consideration for this
must be the suitability of the equipment for safe operation in and around the facility.
Areas within and around a grain handling facility can be classified according to the US
National Electrical Code (NEC). The areas concerning combustible dust, for example
grain dust, are all Class II and can be found in NEC 500.5(C)(1) and (C)(2). The areas

www.go4b.com

Page 5 of 34

concerning an explosive gas, such as hexane gas, are Class I. We are dealing primarily
with the dust hazard in this paper, however serious consideration must be made for other
hazards such as hexane gas, which is used quite extensively in certain grain handling
processes. This gas is extremely explosive and because its vapor density is heavier than
air it can accumulate in pockets and low areas such as reclaim conveyors and bucket
elevator boot pits. Quite devastating explosions have occurred in plants where hexane has
leaked from one plant to another and been ignited by a source in the receiving plant.

The dust hazard, Class II, areas are defined by divisions according to their potential
danger with Division 1 being the highest risk, followed by Division 2. NFPA 654:

Standard for the Prevention of Fire and Dust Explosions from the Manufacturing,
Processing, and Handling of Combustible Particulate Solids (2006)

provides guidance to

determining area classifications. Table 4 summarizes this.

Table 4.

Guide to NEC Class II Divisions

Division Definition

Typical

Area/Location

Hazard is present under
normal operation

Inside the head or casing of a
bucket elevator

Hazard is present only under
abnormal conditions

Control room inside the plant

A further sub category defines the type of dust and divides it into three (3) groups

See

table 5.

Table 5.

Guide to NEC Class II Groups

Group

Combustible Dust

Includes these materials

E Metal

Aluminum,

magnesium

Coal

Coal, carbon black, charcoal

Grain

Corn, flour, wood, plastics

All electrical equipment installed in and around a grain handling facility must be listed or
approved for use in the specific classified area including the specific division and group.

Note:
1. Equipment must be approved for the specific Class of hazard and as such Class I (gas
hazard) approved products are not approved for Class II (dust hazard).
2. Equipment approved for use in Division 1 areas can be used in Division 2 areas.

The products must be listed or approved by a third party testing laboratory such as
Underwriters Laboratories (UL), Canadian Standards Authority (CSA), Factory Mutual
(FM), Edison Testing Laboratory (ETL). Tests carried out by the laboratory are specific
to the standard applied and one must be careful to use equipment approved to the
applicable standards. A CSA approved sensor can be used in the US only if the sensor
has passed the required US tests. This is usually denoted by the small US or NRTL

www.go4b.com

Page 6 of 34

(Nationally Recognized Testing Laboratory) mark applied below the logo. The products
are labeled with the appropriate listing mark and Company’s name:

Example of an approved electrical equipment label for use in the USA and Canada:

LR# XYZ Manufacturer

CLASS II DIVISION 1 GROUP G.

European Approvals

In Europe all equipment must carry the CE mark before it can be sold or used. This mark
is applied by the manufacture of the equipment and designates that all relevant European
health and safety standards have been met. Before the CE mark can be applied to
products intended for use in the dust hazard areas within the feed and grain industry the
product must first be certified to the Atex Product Directive (94/9/EC). Atex is a
contraction of “Atmosphere Explosible”, the French term for “Potentially Explosive
Atmosphere”. Atex defines hazardous areas depending on the levels of risk similar to the
NEC hazardous (classified) locations but there are three levels of hazard defined in Atex
whereas NEC only defines two divisions. Table 6 provides a comparison guide.

Table 6.

A comparison guide to Atex Zones and NEC Divisions

Atex Zone

Atex Definition NEC

Division

Dust Hazard continually present

Dust Hazard likely to be present

Dust Hazard not likely to be present

Note: The 2005 edition of the NEC was updated to include Article 506:

Zone 20, 21, and

22 Locations for Combustible Dusts, Fibers, and Flyings.

www.go4b.com

Page 7 of 34

Figure 4.

Example of approved product label for use in Europe

ATEX Coding

Ex II 1 D

Equipment Group: I for mining

II for non-mining

Equipment Category:
1 – very high protection – Zone 20
2 – high protection – Zone 21
3 – normal protection – Zone 22
D – Dust Hazard

Guide to Ingress Protection (IP)

DUST
IP 5x – Dust protected
IP 6x – Dust tight
WATER
Protected against:
IP x4 – splashing water
IP x5- water jets
IP x6 – powered water jets
IP x7 – temporary immersion
IP x8 – continuous immersion

Worldwide Approvals

Due to the many different standards and approvals for different countries the costs of
obtaining and maintaining these approvals can be significant for the equipment
manufacturer and ultimately for the equipment user. As such a global standard that is
applicable across the world would prove a better solution. One such standard that is
starting to gain acceptance in many countries is IECEx.

II 1D T120DegC IP66

Manufacturer Name and Address
Product Identification

Product Serial Number

24V DC 50mA

2007

CE 1180

Baseefa OSATEX0001X

ATEX
Certification No.

Maximum
Surface Temperature

Ingress Protection
(see below)

ATEX Notification
Body ID Number

Electrical
Parameters

ATEX Coding
(see below)

Year of
Manufacture

www.go4b.com

Page 8 of 34

Figure 5.

Typical configurations for hazard monitoring systems

System Design

A hazard monitoring system consists of three basic elements: sensors, control device and
alarm/shutdown signal. The sensors are mounted on the machine to detect the potential
hazard and translate it into an electrical signal that is then transmitted to a control device.
The control device then provides a signal to warn personnel and automatically shutdown
equipment. A stand-alone control device dedicated to the hazard monitoring function
designed specifically for use within the industry and for use with the specialized sensors
is preferred. If required, this stand-alone system can be connected to the plant PLC
system to provide visual/graphical indication to the plant operators at a central location
and can also perform hazard logging and trending functions for plant maintenance
personnel. With this type of system design the hazard monitoring function is not
dependent upon the PLC and the system will continue to function quite safely and totally
irrespective of what the PLC is doing. Occasionally the sensors are connected directly to
the plant PLC without using a stand-alone control. This is not preferred as the PLC is not
dedicated to the hazard monitoring function and actually performs many processing
functions around the plant. However, the latest safety classified PLC’s can be used to
give added insurance but the programmer/software engineer must take responsibility for
the program/code written to provide the safety function. (Three configuration diagram
examples are shown in figure 5)

www.go4b.com

Page 9 of 34

Key features to look for in a Hazard monitoring control unit:

Hazardous area approvals (Class II Division 1 or 2, Group G)
Unique serial number identification
Well-documented instruction manual
Relay contacts normally energized for failsafe shutdown
Simple menu driven parameter set-up and adjustment
Visual display
RS485 serial communication output for PLC connection
In-built test and diagnostic functions

System Sensors

The sensors are typically specialized for the feed and grain industry and standard
industrial sensors are not normally used. The sensors need to be able to stand up to the
extreme conditions found in the feed and grain industry and must also be safe to use
within the potentially explosive dust atmosphere.

1. Belt Slip Sensors

Bucket elevators and belt conveyors consisting of two pulleys and a belt are capable of
generating dangerous amounts of heat in the event of belt slip. Belt slip on the drive
pulley can occur when the belt is loose or is overloaded. If a belt could be infinitely tight
and capable of handling an infinite load then it would never slip, and the drive motor
running at constant speed would eventually stall when the load surpassed its full load
rating (as the load increases, 3-phase induction motors slow down only very marginally,
they run at almost constant speed and when overloaded they stall). The motor load
current during belt slip is typically less than the normal running load current and
therefore contrary to common belief; current detectors or amp meters are not a good
indication of belt slip. Slip must be detected by monitoring the speed of the belt directly
from the belt (or bucket) or indirectly from the tail pulley rpm.

Figure 6.

Typical Multi-Hazard
Monitoring System for belt
speed, belt misalignment,
bearing temperature, plug
condition, and pulley
alignment for belt conveyors
and bucket elevators.
(Photograph courtesy of 4B
Components Ltd.)

www.go4b.com

Page 10 of 34

Because a 3-phase induction motor runs at constant speed independent of its load, a
single sensor calibrated to the normal running speed can be used to detect belt slip. This
sensor can be mounted in one of two locations on the bucket elevator. Either by the boot
pulley to detect the rpm of a target attached to the shaft or on the casing to detect the
speed of the passing buckets or bolts. Either location is suitable, although it is sometimes
preferable to install sensors higher up on the casing to prevent possible damage from
water and moisture in the boot area. When installing higher up on the casing, care must
be taken not to be too far from the pulley. The further away from the pulley, the more the
belt flaps and rolls and the more difficult it is for the sensor to detect the bucket bolts
consistently. The types of sensors for both these locations are very different from each
other and are not interchangeable. Because the target for the boot-mounted sensor is
always uniform in its shape and its distance from the sensor, only a simple sensor with a
short fixed range is required (figure 7). The sensor mounted on the casing however
requires a much greater sensing range along with physical and electronic range
adjustment so that it can be set-up correctly to cope with the normal side-to-side and front
to back movement of the bucket bolts (Figure 9). When ferrous buckets are used instead
of plastic buckets, the sensor mounted on the casing must be moved to detect the buckets
instead of the bolts, in order to prevent false readings due to the sensor detecting the steel
buckets through the belt (figure 8). It also important to understand that the upside, or tight
belt side of the elevator has less belt movement and it is easier to set-up the sensor and
provide a more constant speed signal on this side. On belt conveyors the sensor is
normally mounted on the tail pulley, as there are no bucket bolts or steel buckets to
detect.

Pulse Train

Simple Sensor

Shaft

Sensing Circuit

Sensing Field

Target

Pulse train to
external monitor

Figure 7.

Simple shaft speed sensor

www.go4b.com

Page 11 of 34

For monitoring the pulley shaft speed, the traditional system of standard inductive speed
sensor with a fabricated mounting bracket, shaft mounted target and separate guard has
caused some problems over the years with regard to speed sensing reliability. Also, since
installations are rarely identical, there is usually a significant amount of site design and
adjustment required to make the complete system function correctly and for this reason it
was common to leave the details of the target, bracket and guard for the field personnel or
the millwrights to figure out. Recent advances in this area have led to more reliable and
standardized speed sensing installation by using a shaft-mounted sensor system. The
sensor, the bracket and the guard are now available as one complete unit that is attached
directly to the shaft and hangs on the shaft with no additional brackets required to hold
the sensor. There is no on site fabrication or adjustment required and reliability is
improved, as the sensor and target are integrally mounted. A typical shaft mounted sensor
is shown in figure 10.

Figure 10.

Typical shaft mounted speed

sensor (photograph courtesy of Rolfes Co)

Figure 8.

Belt speed sensor
for steel buckets

Figure 9.

Belt speed sensor
for plastic buckets

www.go4b.com

Page 12 of 34

One of the other considerations when installing a speed sensor is the quantity of targets
required on the shaft. The quantity of targets required depends on the shaft speed and the
reaction time required. For example, a shaft running at 33 revolutions per minute (33
rpm) with 2 targets would produce 66 pulses per minute (66 ppm). If the underspeed
alarm is set at 10% of normal running speed, then the alarm trip speed is 59.4 ppm. The
minimum reaction time for a certain trip point is the time for the sensor to see the next
pulse when running at the trip speed. Therefore at the 59.4 ppm alarm point the reaction
time in seconds is 60/59.4, which is about 1-second. A 1-second reaction time is usually
more than adequate for detecting underspeed on a belt conveyor or bucket elevator. If
you used an encoder with 500 pulses per revolution, on the same shaft, the number of
pulses per minute would be 16,500. This would give us a 10% underspeed alarm point of
14,850 ppm and a reaction time of 60/14,850 seconds (4 mS). Even if the control
circuitry could count this fast, this reaction time is far quicker than what is required for a
belt conveyor or bucket elevator and provides no additional benefit to its protection and
just adds complication and cost. As we can see from table 7, if we require a reasonably
quick alarm reaction time of around 1 second, only 4 targets are required for shafts
running normally at 15 rpm or higher.

Table 7.

Reaction times for different shaft speeds and the

number of targets required to achieve the reaction time

Shaft Speed

(rpm)

Targets

Reaction Time

for 10%

Underspeed

30 to 60

1.1 to 0.56

seconds

15 to 30

1.1 to 0.56

seconds

6 to 15

1.1 to 0.44

seconds

3 to 6

1.1 to 0.56

seconds

1 to 3

1.1 to 0.37

seconds

Speed sensors for monitoring shaft speed are available in three standard formats: a simple
sensor; an intelligent sensor; and a combination of both (see figure 11). The combined
sensor provides both a signal for direct connection to an alarm/shutdown and a pulsed
output for remote display of the speed on a control panel, PLC or tacho display. This
option is usually preferred as it uses its own non-volatile software with output hardwired
to the motor starter. The PLC or control has no effect over the shutdown function and
only acts as a remote monitor or data logger. This arrangement usually provides a safer
option and a very high level of confidence that the system will shutdown during a belt
slip condition.

www.go4b.com

Page 13 of 34

Speed Sensor Options

Shutdown

To Motor

Alarm

Run

Intelligent

Boot of Elevator

OR Tail of Conveyor

Control Room

Alarm

Shutdown

To Motor

Run

Intelligent Sensor

with Pulse

Run

Shutdown

To Motor

Alarm

Simple

Figure 11.

Three standard sensor formats for shaft speed monitoring

(Simple, intelligent, and combined sensors)

Key features to look for in a shaft mounted speed sensor:

Hazardous area approval (Class II, Division 1, Group G)
Unique serial number identification
Relay contacts normally energized for failsafe shutdown
Conduit entry for connection of flexible sealtite
Status LED indication
Optional pulsed output for PLC input connection
Waterproof construction (IP 66 or better)
Sealed construction

Key features to look for in an elevator case mounted speed sensor:

Hazardous area approval (Class II, Division 1, Group G)
Unique serial number identification
Conduit entry for connection of flexible sealtite
Status LED indication
Sealed construction
Test function
Sensitivity adjustment

www.go4b.com

Page 14 of 34

Adjustable mount

2. Belt Misalignment Sensors

Bucket elevators and enclosed belt conveyors have sidewalls, which a misaligned belt
can rub against. Heat generated by this rubbing action can quickly reach a dangerous
level, especially near pulleys where the belt side forces are usually the greatest. Figures
12 and 13 show severe belt misalignment on a bucket elevator. There are a number of
different sensing technologies, available to detect a misaligned belt.

Limit Switches.

Mounted on the side of the elevator casing these devices are activated

when a belt moves over. Wear on the switch, due to belt friction, is kept to a minimum by
using steel or ceramic rollers to activate the limit switch. However these types of switch
are outdated and can be dangerous. With the belt running against the small roller, a
typical roller speed of well over 1400 rpm can be generated. Serious problems can arise
due to the bearings in the roller failing, resulting in dangerous heat generation. The
mechanics of the switch can also wear out or become contaminated with material causing
the switch to stick, since the switch mechanism must move a considerable amount to
activate the contact. This type of system is not failsafe in any way. If a switch becomes
loose and moves away from its mount, there is no way of realizing that the system is no
longer monitoring.

Rub-Blocks.

Placed on the side of the casing these devices incorporate a temperature

sensor, which is similar to the sensors used in bearing temperature monitoring but usually
with a lower trip point. Typical rub-block sensors have a trip point around 120 º F. They
are designed to detect the heat generated when the belt rubs against the brass or

Figure 12.

Inside elevator casing

showing slit in steel casing due to
severe belt misalignment

Figure 13.

Outside elevator

casing showing slit in steel
casing and heat build-up due to
severe belt misalignment

www.go4b.com

Page 15 of 34

aluminum block. However these systems are out dated and can be dangerous. They
require heat from friction of the rubbing belt to detect a belt misalignment, and their soft
brass or aluminum face can wear very quickly. Sometimes a belt misaligns and rubs
against the soft brass face for a short period of time and not long enough for the sensor to
detect any significant heat build-up. Over time, these sporadic misalignments wear
through the soft brass block and render it ineffective. Plant personnel are only aware that
the sensor has been inactive after it has been removed for visual inspection during
planned maintenance. Rub-blocks are also not failsafe. As with limit switch misalignment
sensors, if they become dislodged from their mounting they will not indicate that there is
any type of problem.

Optical Sensors.

Using an Infra Red transmitter and a receiver to detect the belt

misalignment. Sometimes a number of sensor / receiver pairs are used to provide a
warning and then a shutdown as a belt misaligns. However, the set up on these types of
systems can be tedious, as they tend to create false alarms due to sensor alignment
problems and material/dust covering the sensor’s lens. Sometimes air purging can help
keep the lens clean but reliable operation is not assured in the dusty and rugged
conditions found in the feed and grain industry.

Non-Contact Magnetic Sensor.

One of the most common belt misalignment sensing

devices for bucket elevators, these extended range proximity sensors are mounted on the
elevator casing and detect the passing buckets or bucket bolts continuously. When the
belt is tracking normally, each sensor produces a signal as the bucket or bolt passes
through the sensing range. When the belt misaligns, one of the sensors begins to miss
pulses and the control unit determines this as a belt misalignment. Since all belts misalign
to some extent without contacting the side of the casing, the better systems use two
sensors so that this normal belt wander does not cause false alarms. The sensors also have
a range adjustment so that they can be “dialed in” to the normal running / wandering of
the belt, and the control unit to which the sensors are connected, usually has parameter
adjustments for accurate set-up. The active continual sensing of these devices provides
the only real failsafe solution for belt misalignment detection on bucket elevators at this
time.

Solid State Force Activated Switch.

These devices can be used for belt misalignment

detection on bucket elevators or belt conveyors. They measure the force applied to them
by the belt as it touches their hardened stainless steel face. Even the smallest deflection
can be detected immediately so that a control unit can be signaled and the belt can be
stopped without delay. Since they need no heat build-up from belt friction to activate,
they detect immediately and they can be useful in logging and trending belt
misalignments for predictive maintenance purposes. These sensors can also be used to
detect the edge of the pulley, if it misaligns. Unaffected by material or dust build-up, no
site adjustment is ever required. Units are available with a built in test feature to ensure
operation of the sensor and the control circuits, and an integral status lamp shows when
the sensor is operated.

www.go4b.com

Page 16 of 34

Figure 15.

Force activated belt

misalignment switch.
(Photograph courtesy of 4B
Components Ltd)

Figure 16.

Force activated belt misalignment

switches installed on enclosed belt conveyor

Figure 14.

Non-contacting magnetic misalignment sensors

detecting the steel bolts on plastic buckets in a bucket elevator

www.go4b.com

Page 17 of 34

Key features to look for in a contact style (force activated) belt misalignment switch:

Key features to look for in a non-contact magnetic style belt misalignment sensor:

3. Bearing Temperature Sensors

All bearings create heat due to friction when running. When well maintained and
lubricated this heat is minimal and well below the lower ignition temperature for the
grain dust. However, if the bearing or lubricant fails in any way, rapid heat build-up can

Figure 17.

Belt misalignment comparison guide

www.go4b.com

Page 18 of 34

occur and the bearing housing can reach a temperature high enough to ignite any dust
accumulated on or around the bearing. Table 8 shows the ignition temperatures for
various dusts.

Table 8.

Guide to

Ignition Temperatures for various dusts (source NEC)

Dust

Layer or Cloud Ignition Temperature

Wheat

220º C (428º F)

Rice

220º C (428º F)

Corn

250º C (482º F)

Wheat Flour

360º C (651º F)

All bearings on the machine should be considered a potential ignition source. Even
outside bearings at the head of outside bucket elevators have caused serious explosions.
The shaft can conduct heat into the head or the smoldering/burning grain dust on the
bearing housing can be drawn into the head. All bearings will eventually fail and even the
very best preventative maintenance programs won’t catch all bearings before they begin
to fail. The only safe approach is to install automatic bearing temperature monitoring
systems, which monitor the bearing temperature continuously. These systems incorporate
a bearing sensor mounted to the bearing housing and wired to an alarm control panel.
Control relays within the panel provide warning and shutdown contacts when the bearing
exceeds a user defined trip point.

Figure 18.

Continuous bearing temperature-monitoring

system (Photograph courtesy of 4B Components Ltd)

CONVEYOR 27 BEARING
TEMPERATURE = 084º F
Amb 069º F : Relay 01

Rel 30: Abs 176: NTC

(Explanation)

This hazard monitor is measuring
a bearing temperature of 84º F
using an NTC sensor and the
ambient temperature close to the
location of the bearing is 69º F. If
the bearing temperature was to
reach 30º above ambient (i.e. 99º
F) then the system will alarm and
relay 1 (R1) would activate to
shutdown the machine or warn
the operators.

www.go4b.com

Page 19 of 34

Typical trip points for bearing temperature sensors are 176º F or lower.
However, more common nowadays are continuous bearing temperature sensors that can
provide the operator with a visual readout of actual bearing temperature in real time on
the control unit or monitor (See Figure 18)

These sensors and controls are able to

provide relatively failsafe monitoring of bearings compared to the trip point sensors and
are also able to provide bearing temperature logging and trending for preventative and
predictive maintenance. Some systems can provide trip points relative to local ambient
temperature and are therefore able to detect a bearing problem much earlier than a sensor
with just an absolute trip temperature setting. Some systems also provide rate of rise of
temperature monitoring and comparison between different bearing temperatures, all
helping to provide an early indication of a problem.

Many systems are now utilizing some form of serial communication system allowing
much reduced installation and maintenance costs. A single cable is used around the plant
and connects to nodes with unique digital addresses. The nodes collect and transmit the
temperature information along with local ambient temperature and sometimes belt
misalignment information too. The network, or “Bus System” can then be connected to
control devices for automatic alarm/shutdown or to the plant PLC.

Sensor manufacturers use a number of different technologies to convert the bearing
temperature to a voltage or current signal, which can be read by a control unit.
These technologies include:

Thermistor sensors

: these are thermally sensitive resistors and are available as Positive

and Negative Temperature Coefficient versions (PTC and NTC respectively). The PTC
sensor’s resistance increases exponentially as the temperature rises above a fixed trip
point, and the NTC sensor’s resistance changes inversely to a temperature rise. PTC
sensors are usually used to indicate when a certain temperature has been reached and can
be called trip point sensors. They are a solid-state version of the old thermostat sensors.
NTC thermistor sensors provide a proportional signal, which is used for continuous
temperature sensing. Their reaction time is quick, they are very stable and they are ideally
suited for the temperature ranges involved for bearing monitoring in the feed and grain
industry. Also, large changes of resistance allow the monitoring system to easily
differentiate open/short circuit conditions (see figure 20).

Thermocouple sensors

use a junction of dissimilar metals, which produces a small

voltage in proportion to the temperature sensed. The technology has been used for many
years and is quite reliable and cheap, however special thermocouple wire and wiring
techniques must be used since every connection becomes a potential thermal junction.

www.go4b.com

Page 20 of 34

Thermocouples can be susceptible to electrical interference due to the very low signal
levels that they operate at and the wires are a very small diameter (see figure 19) and
must be adequately protected. It is also not possible to monitor for short circuit conditions
(see figure 21) and any short in the thermocouple cables becomes a new thermal junction.
There are however many plants in the feed and grain industry still using thermocouples
with old multiplexing technologies. But as maintenance costs rise, plants are switching
these old systems out to more modern and more versatile serial communication bearing
temperature systems.

Resistance Temperature Detector (RTD)

: The most commonly used RTD has a

Platinum element with a resistance of 100 ohms at 0ºC. The resistance increases linearly
with temperature rise. 3 or 4-wire RTD’s are very accurate as they compensate for the
resistance of the cable connecting the sensor to the control unit. These sensors are
relatively expensive and although they are sometimes used in bearing temperature
systems they are more geared towards more sophisticated process temperatures. Also,
large changes of resistance allow the monitoring system to differentiate open/short circuit
conditions.

Inches

Figure 19.

Photo of thermistor on left and thermocouple on right

showing the bead size and the difference in cable diameters.

www.go4b.com

Page 21 of 34

Figure 20.

NTC Thermistor

0.5

1.5

2.5

3.5

4.5

-50

100

150

200

250

300

Temperature [degF]

age [Vdc]

NTC Voltage

NTC SHORT CIRCUIT

NTC OPEN CIRCUIT

NTC

NORMAL OPERATING

RANGE

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

5.00

-50

100

150

200

250

300

Temperature [degF]

e [

]

T Thermocouple

THERMOCOUPLE OPEN CIRCUIT

THERMOCOUPLE

NORMAL OR SHORT CIRCUIT

RANGE

Figure 21.

Thermocouple Sensor

www.go4b.com

Page 22 of 34

Figure 22

. Continuous bearing temperature technology

comparison guide

Semiconductor Sensors

: The latest development in temperature sensing uses silicon

semiconductor technology. Usually manufactured as an integrated circuit (IC), these
sensors are small, accurate, linear and low cost. As more and more facilities being built
are using serial communication links instead of hard wire the digital sensor with
individual address capability is becoming more popular. Instead of thousands of cables
running through a plant, one main 4-wire communication cable is installed, with branches
and nodes to individual machines. These sensors have a unique identification number,
which can be addressed by a central computer or plant PLC. When addressed,
temperature data is sent from the sensor along with the sensor ID number. The ID number
and temperature is then displayed and used to alarm and shutdown the machine at a user
defined trip point.

Some of the serial communication systems being designed today will accept the older
RTD, Thermocouple or Thermistor technology by converting the signal to a digitally
addressable data format. Thus enabling users to couple existing sensors to a data network,
and also mix the old sensors with the new.

www.go4b.com

Page 23 of 34

Key features to look for in a bearing temperature sensor:

Hazardous area approval (Class II, Division 1, Group G)
Unique serial number identification
Conduit Entry for connection of flexible sealtite
Robust construction
Positive mounting to the bearing (with grease through ability)
Open circuit detection (detects a wire break/disconnect)
Closed circuit detection (detects a wire short)
Visual temperature indication

Additional Sensors to consider for bucket elevators and belt conveyors

1. Head Pulley Misalignment Sensors:

The same sensors used for belt misalignment detection can usually be used for head
pulley misalignment detection also. If the sensors are installed on the top half of the
pulley, they will detect both belt and/or pulley misalignment. Sometimes non-contacting
inductive style sensors are installed to detect the position of the steel pulley edge.

2. Plugged Spout Sensors:

It is recommended that sensors are installed in the discharge chute and are connected to
shutdown the elevator or belt conveyor immediately upon a plug condition. When
excessive delays are incorporated the boot of the elevator or drive end of the conveyor
can quickly fill up with material, which can lead to belt slip, belt misalignment and other
serious problems.

Mount the plug sensors as close to the discharge as possible but outside the normal
material flow. Be careful not to impede material flow with the sensor. Consider mounting
additional plug sensor on the downside of the leg near the boot. If using rotary style
indicators, use only units that are failsafe in their design.

3. Tail pulley misalignment sensors:

On many enclosed conveyors, the tail pulley incorporates a belt re-loading mechanism
that runs very close to the casing. Any pulley misalignment can cause the pulley to rub
against the casing and create significant amounts of heat. The two common methods of
detecting this involve an inductive sensor mounted on each side of the pulley or lug/plate
style continuous temperature sensors mounted to the outside of the casing.

Equipment Selection Summary:

There are many manufacturers of hazard monitoring equipment. Be careful to choose a
well-respected reputable company that specializes in equipment for the feed and grain
industry. Also check that the manufacturer is able to provide after sales service and
technical support, along with help in product selection and installation advice.
Remember, hazard monitoring equipment should be designed to provide many years of

www.go4b.com

Page 24 of 34

service so it is important to be confident that the original equipment supplier will be
available to help with system support should it ever be needed in the future.

Most approved equipment should have a unique serial number that allows it to be traced
throughout the manufacturing process to the sale of the product and installation. All
equipment should be provided with a detailed installation and operation manual that
should be available for the customer to view prior to purchase.

Whenever possible choose equipment that allows for fail-safe installation and offers a
high degree of confidence. In general, only equipment that is designed to automatically
shutdown the machinery when a hazardous condition is detected or when any sensor fails
or wire breaks can claim to be failsafe. An alarm horn or alarm lamp is not failsafe and
should only be used as an early warning. As such, always install a failsafe shutdown
mechanism in addition to any alarm or warning device.

The installation of the hazard monitor should not affect the way your plant functions with
regard to equipment interlocking. If you already have a system that automatically stops
any equipment feeding the monitored machine (i.e. all upstream equipment) then the
installation of a hazard monitoring system with automatic shutdown should have no
effect on the way this operates.

Installation

General:

Choose a professional electrical installer who is familiar and has experience with
installing hazard-monitoring equipment within the feed and grain industry. Many good
electricians are not experienced with installing these types of sensors and controls. There
are challenges specific to our industry, including potentially explosive atmospheres in
which the installer has to work safely and in which the equipment has to be installed
correctly following all applicable laws and regulations, confined entry hazards and
moving machinery hazards.

On new plants, make sure that hazard monitor installation is planned well in advance.
Too often it is left to the last minute and there is a lot of pressure on the electricians and a
big rush to get the installation complete. When purchasing the elevator or conveyor
discuss the hazard monitors with the manufacturer so that pre-installation can be
accomplished in the shop. Most equipment manufacturers already offer optional hazard
monitoring equipment and their machines have been designed to accommodate the
mounting of the sensors allowing easy and simple on site installation.

On existing plant, give the elevator or conveyor an overhaul prior to installation of
sensors making sure that belts are not slipping under load, belts are not misaligning and
bearings are running within normal temperatures. Take readings and make note of the

www.go4b.com

Page 25 of 34

normal operating conditions including shaft speeds and bearing temperatures. Paint over
any old rub or burn marks on casings prior to installing belt alignment sensors.

When installing an alarm device, consider the location carefully. Audible alarms must be
loud enough to be heard over the background plant noise and since plant operators may
be wearing hearing protection it is advisable to install a flashing beacon lamp in addition
to the audible alarm. Consider special alarm signaling devices such as an automatic pager
or GSM device.

Mechanical:

Locate monitors in a suitable control room close to operators. Mount the units at eye level
so that operators can readily read the display. Do not locate them inside control panels
where they cannot be seen. Do not locate them outside in direct sunlight as elevated
temperatures can degrade some displays.

Install belt misalignment sensors for enclosed belt conveyors on the topside of the belt
and inline or very close to the pulley. Make sure that when the belt misaligns it will
contact the sensor and not ride over it. When installed on the return (underneath) side,
hardened grain can cause the belt to ride above the sensor.

If installing rub-blocks, it is sometimes a good idea to install on a hinged door since
access will be required frequently to inspect the sensor face.

To help prevent mechanical damage and protect from rodents, install sensor wiring inside
rigid metal conduit and where flexibility is required use short liquid tight flexible conduit
or sealtite with fittings approved for the area.

One of the common problems with conduit systems is the ingress of water. Many
electricians understand that no matter how well a conduit system is installed, at some
stage a cover could be left loose or condensation can accumulate. This moisture can be
channeled to sensors and over time can accumulate and eventually damage the wires or
sensor. As such, low conduit drains, approved for the location (see figure 24) should be
installed and sensor wiring should be “Teed” with an adequate wiring loop so that water
following the wires is not channeled to the sensor. Part of the regular system maintenance
should include the cleaning of any accumulated debris from around the conduit drains
and the inspection of conduit systems for water ingress. Figure 23 shows hazard monitors
installed on the boot of an outside bucket elevator. Belt alignment, shaft speed, and
bearing temperature sensors are installed using flexible sealtite with adequate loops and
the steel conduit is installed with low point drains.

www.go4b.com

Page 27 of 34

Network and computer wiring:

Special attention should be paid to the installation of systems with communication cables
and systems with low voltage wiring. Shielded cables should have only one connection to
ground (usually at the control unit) and an open circuit ohm reading would verify this
when the shield is removed from the ground connection and the meter is placed between
the shield and ground.

Small stranded wires connected to terminals should be installed using a booted ferrule
system (see figure 25) to ensure the mechanical connection of the wire to the terminal
and to contain the strands. When connecting sensor cables using wire “lugs” make sure
that the correct size is used.

Figure 25.

Photograph of network wires connected to terminals with

a booted ferrule wiring system.

Special attention should be made to the shield wire in shielded cables. A braided shield
made up of many fine conductors should be carefully handled so that there is no chance
of them touching the other conductors and causing a short to ground. Insulated the shield
wire using a wire sheath (which can be the outer PVC jacket pulled from standard copper
cable). When the cable includes a drain wire, cut back the shield, insulate the drain wire
and use it as the shield connection. The shield should be one continuous connection
through the system and should be connected to ground at one point only. This one ground
connection should be made at the end of the cable, and is normally done at the control
unit end.

www.go4b.com

Page 28 of 34

Figure 26.

Customers wiring diagram detail

(Partial schematic shown)

Network and computer wiring should be segregated from higher voltage wiring for safety
and functionality.

Some sensors and systems use Intrinsically Safe (IS) wiring which must be installed
carefully and following NEC Article 504. All Intrinsically safe wiring must be physically
and electrically separated from nonintrinsically safe wiring and special grounding
techniques apply.

Equipment instruction manuals normally include installation-wiring schematics. However
these are general and not specific to the facility where the equipment is being installed.
The Electrical installer should produce a professional set of wiring schematics specific to
the installation, which should include wire numbers, equipment labels, and other specific
information (see figure 26).

When planning the wiring runs keep to a standard color system. If your plant does not
already have a standard system, then decide on one. Try to maintain the same wiring
color through connections and terminals. Use wire numbers on the wiring schematics and
on the physical wires. The example in figure 26 shows part of a customer’s wiring
schematic. The wires have wire numbers, colors are noted, the location of the sensor is
noted, and the sensor has been given a unique identification number and description. The
type of field cable being used is noted. The manufacturer’s part number is noted.

www.go4b.com

Page 29 of 34

Do not use larger than required signal cables. Larger cables take up more room, are more
expensive, and are more difficult to connect into small terminals.

PLC ladder logic diagrams and Input/Output data sheets should be included where
applicable, along with back-ups on CD’s or hard drives.
Never run the machine without an active hazard monitoring system. A number of spare
sensors and components should be kept at hand so that plant down time is kept to a
minimum.
During installation, the plant manager should periodically inspect the wiring and conduit
installation, making sure that the connections are neat and tidy and a high level of quality
is being maintained (figure 27 shows a poor wiring install, figure 28 a good wiring
install). Ideally the installer should test for continuity as each section is completed.
Whenever possible the installers should remain on the same job from start to finish.
Whenever multiple crews are involved the standard of installation can suffer and the time
taken to complete the job can be longer.

Speed Sensor Installation

When installing speed sensors to monitor shaft speed, use either a shaft-mounted sensor
as described earlier or install an inductive speed sensor using a universal sensor mount
(figure 29).

Figure 28.

Good wiring installation

Figure 27.

Poor wiring installation

www.go4b.com

Page 30 of 34

Cover

Base Plate

Sealed
Bearing

Dual
Target

Threaded
Mounting
Holes

18/30mm Mount

Figure 29.

Universal shaft sensor mount

This type of speed monitoring system is much easier to install, much safer, and more
reliable. Since many machines require periodic belt tensioning that involves moving the
monitored shaft, a traditional speed sensing installation requires careful consideration for
the attachment of the bracket, which holds the sensor to the machine. The bracket must
be able to move with the shaft as the belt is tensioned so that the target on the shaft
remains within the sensors sensing range. By design the shaft mounted sensor installation
requires no such consideration as the whole assembly is attached to the machines shaft,
and therefore moves with it. Also, machine vibration can sometimes cause problems with
a traditional shaft speed sensing system for the same reason. The typical sensing range
for the inductive sensor is 7/16”. If there is a clearance of ¼” between the rotating target
and the face of the sensor then the absolute maximum tolerance is 3/16”. Under heavy
loads and machine vibration the clearance between target and sensor could exceed 3/16”,
resulting in periodic false alarms or nuisance shutdowns. When a shaft mounted sensor
installation is used, the sensor and target are mounted on the same plate, so that the
distance from the target to the sensor remains the same no matter how severe the
vibration is.

Since the sensor, the target, the bracket and the guard are all one assembly that hangs
from the machine’s shaft only a single hole drilled in the end of the machine’s shaft is
required for installation. Some systems are also available with a magnetic attachment so
that no drilling or threading of the shaft is required.

www.go4b.com

Page 31 of 34

Bearing

Shaft

Target

Bracket

Sensor

Traditional Installation

(Shown without guard)

Shaft Mounted Sensor
Installation

Sensor

Whirligig

Bearing

Flexible Metal
Conduit

Shaft

¼” Typical

1/16”

Figure 30. Speed sensor mount.

Traditional vs. Shaft Mount

Testing and system handover:

One of the most important parts of the installation will be the final system test. Good
installers should have a full understanding of how the system works and will have tested
the system passively as the installation progresses. When it comes time to run the system
a good electrician will be confident that the system will work as expected, but will
nevertheless perform a full and thorough test of the system in the presence of the plant
manager and other key personnel.

This test must be done in a safe manner, and without the introduction of conveyed or
elevated material.

Ideally, real life conditions should be forced so that the correct operation of the
equipment, wiring, and auxiliary equipment can be confirmed.

Upstream equipment should always be interlocked with downstream equipment so that
automatic shutdown will not result in a plug condition.

After successful testing, the installation is complete and the system should be handed
over. A final walk through by the plant manager and the installers will ensure that the
manager has a full understanding of the sensors and controls which have been installed

www.go4b.com

Page 32 of 34

and the capability of the system. The manager will also ensure that the work areas have
been cleaned and no debris has been left. During this handover there should be allowance
for training of operators and key personnel. Instruction manuals, installation schematics,
spare parts, emergency contact numbers and any outstanding issues should all be
addressed.

Maintenance

Even a solid state electronic monitoring system with failsafe design requires periodic
maintenance to ensure that you will have trouble free monitoring and so that you can be
confident that the system will perform when required. Some of the maintenance checks
you can easily perform include:

Follow manufacturers test procedures and record results. Remove from service any
machine that shows a problem, until the monitoring system is up and running again.

Physical inspection of the sensors and controls can provide invaluable information as to
the status of the system. A sensor could have come loose from its mount and may need
attention. Bearing sensors or belt misalignment sensors do no good when hanging in mid
air! Rub sensors can be rubbed through and only visual inspection can catch this.

Repair any damaged conduits or wiring connections found. Replace missing junction box
lids.

On a regular basis inspect contact style belt misalignment sensors for wear or damage.
Physical checks of contact sensors should be made periodically, the frequency depending
on the application and the amount and duration of detection.

Check bearing temperatures using a hand held IR temperature sensor and compare this
with what the system is indicating.

Check pulley speeds and compare to initial system start-up values. If no values were
recorded then make sure the belt is tight and not slipping, and compare unloaded and
loaded rpm. There should be negligible reduction in speed when fully loaded.

Where practical, compare resistance values of temperature sensor circuits to known start-
up values.

The hand held devices in figures 31, 32, 33, 34 are essential tools for hazard monitoring
maintenance, inspection and testing. However, be sure to use either approved equipment
or only when the dust hazard is not present.

www.go4b.com

Page 33 of 34

Figure 31.

Hand held tachometer

Figure 32.

Hand held temperature

indicator

Figure 34.

Hand held stroboscope

Figure 33.

Digital multi-meter

www.go4b.com

Page 34 of 34

Summary

The safe operation of plant and machinery within the grain and feed industry
requires a conscientious effort from plant designers, machine manufacturers,
installers, plant managers, operators, and maintenance personnel. The threat of a
catastrophic event is always present and a well designed, professionally installed,
and well maintained hazard monitoring system will help to make the plant safer and
more productive.

References:

1. NFPA 49:

Hazardous Chemicals Data (2001)

2. NFPA 61:

Standard for the Prevention of Fires and Dust Explosions in Agricultural

and Food Processing Facilities (2002)

3. NFPA 69:

Standard on Explosion Prevention Systems

4. NFPA 70:

National Electrical Code (2005)

5. NFPA 499:

Recommended Practice for the Classification of Combustible Dusts and

Hazardous (Classified) Locations for Electrical Installations in Chemical Process Areas

6. NFPA 654:

Standard for the Prevention of Fire and Dust Explosions from the

Manufacturing, Processing, and Handling of Combustible Particulate Solids (2006)

7. NMAB 353-4:

Classifications of Dusts Relative to Electrical Equipment in Class II

Hazardous Locations, published in 1982 by the Committee on Evaluation of Industrial
Hazards, National Materials Advisory Board, Commission on Engineering and Technical
Systems, National Research Council, and National Academy of Sciences.

8. OSHA Title 29:

Code of Federal Regulations,

Part 1910.272.

National Fire Codes Vol. II. The Prevention of Dust Explosions (1946) – NFPA

10.

KSU – Dust Explosion Report (March 20, 2006) – Robert Schoeff

11.

Dust Explosions in the Processing Industries (1999)- Rolf K Eckhoff