Static electricity is often perceived as an invisible risk. This case study explains why static electricity provides an ignition source for serious fires and explosions that could occur during everyday operations involving the handling and processing of flammable products.
A company supplying aluminium powder had an order cancelled when the bulk truck transporter carrying the powder was en route to a railcar hopper loading station. The truck driver was instructed to return the aluminium to the plant from where it was manufactured. As this scenario had never occurred before there was no standard operating procedure in place to offload the aluminium from the truck back into the production facility. Shortly after the operators had worked out how to overcome some practical challenges for moving the powder back into the plant, an explosion occurred which propagated throughout the plant.
On return to the plant it was noted that there was no direct loading point for the finished powder to be injected back into the production stream directly from the truck. A decision was taken to convey the powder into the pneumatic transport system entry point of the plant using the 3” hoses on the truck. Unfortunately the hoses could not reach the entry to the plant’s pneumatic system so an additional length of hose from the plant was added to the line of 3” hoses running from the truck. Both hose types were constructed of rubber tubing that contained internal metal helixes that ensured the hose flanges were electrically bonded together.
The truck was grounded so the 3” hoses (assuming they were in good condition) were also grounded, hence the risk of an accumulation of electrostatic charge on the truck and hoses was minimal. One issue that was encountered, however, was that the plant hose that was used to complete the distance between the truck’s hoses and the entry to the plant’s pneumatic system was wider in diameter to the truck’s hoses. This meant that a sealed connection between the hoses could not be made. The operators overcame this issue by stuffing rags into the gap between the hose flanges. This had the effect of electrically isolating the plant hose from the truck’s hose, potentially impeding the transfer of electrostatic charges from the plant hose to ground via the grounded truck. The other end of the hose was assumed to have been resting on the concrete floor inside the plant. The other issue was that the density of the air-powder phase coming from the truck was above the minimum explosive concentration of the aluminium powder.
Although static electricity can be regarded as a difficult subject to grasp, we hope that our series of case studies give you an insight into the reasons why static electricity provides an ignition source for serious fires and explosions that occur during everyday operations involving the handling and processing of flammable products.
This case study investigates the factors resulting in an electrostatic ignition incident involving toluene, a prolific charge generator filling a metal bucket via gravity fed 0.75” metal pipping.
In this scenario, an operator opened a valve to draw toluene into a metal bucket with toluene from an overhead tank by gravity flow at approximately 5 gallons per minute. The operator hung a metal bucket with a wire bail and plastic handle over a globe valve. The plastic handle on the bail isolated the metal bucket from ground.
On opening the valve, the operator backed away from the bucket allowing the toluene to flow as he had previously done several times. Within a few moments the toluene had ignited causing the operator to immediately leave the scene returning with a small fire extinguisher, which proved inadequate to put the fire out. The operator then left the scene returning with a larger fire extinguisher, however by the time he had returned the fire was out of control and he was unable to close the valve to prevent the flow of toluene to the bucket which was already over flowing.
The investigation into the incident outlined that the operator had opened the valve and backed away from the metal bucket. The operator stated “I was just standing there looking at it when it caught fire”. As a result, discharge from the operator could be ruled out as a cause of the incident and the scenario of a streaming current was considered.
(I) Is = 2.5×10-5 ∙ v2 ∙ d2
(II) Is = 2.5×10-5 ∙ 1.10692 ∙ 0.019052
(III) Is = 0.01μA
Is = Streaming Current (A)
v = Velocity (m/s)
d = Pipe diameter (m)
The streaming current was found to be in the order of 0.01 µA were it not for the presence of the in-line filter. The residence time of the toluene between the in-line filter and the exit of the pipe was less than a second, much shorter than the recommended 30 seconds; therefore, a reasonable estimate of the streaming current at the exit of the pipe can be calculated around 0.1 µA. In any case, an estimate for the streaming current can be assumed to be between 0.1 µA and 0.01 µA.
Assuming that the toluene flow had continued for 30 seconds, there would have been a charge of 3 µC on the bucket provided that the bucket was completely isolated from earth.
The potential energy on the bucket can be found using the equation:
(I) Q = Charge on the bucket
(II) C = Capacitance of the bucket
Therefore the potential energy on the bucket:
And the voltage on the bucket can be found using the Equation:
With the breakdown strength of air at 3 x 106 V/m a spark from the bucket could easily jump across a gap of 50 mm (1.96″) meaning it was probable discharge from the wire of the bail to the body of the globe valve could occur.
Newson Gale’s latest series of articles that contain case studies of fires and explosions caused by static electricity draw attention to the wide range of processes that are susceptible to electrostatic charge generation and accumulation on portable and fixed plant equipment.
This case study investigates the factors behind the ignition of a combustible dust cloud during a manual powder processing operation. In this example a process operator was tasked with manually tipping approximately 18 kg (40 lbs) of powder from a plastic drum, constructed from polyethylene, into a metal process vessel. The plastic drum contained a combustible powder that had a minimum ignition energy of 12 milli-joules. A metal chime was positioned around the circumference of the top of the plastic drum to provide it with impact protection from daily usage in the plant.
The operator tipped the powder into the process vessel, resting the drum on the edge of the vessel. As he removed the drum from the vessel when the powder was fully deposited there was an ignition of the dust cloud that had formed at the top of the vessel.
It was postulated that the accumulation of electrostatic charge on the chime resulted in a static spark discharge from the chime as it came into close proximity with the vessel when the drum was removed. The vessel was grounded through its own fixed connection to the plant.
In order to verify this theory an experiment was conducted to determine how much electrostatic charge could have been generated by the movement of the powder. 18 kg (40 lbs) of the same powder was tipped from a similar drum into a Faraday cage from which electrostatic charge measurements were taken.
A charge of 3.6 micro-coulombs was measured on the Faraday cage which received the powder. In this case the powder was charged due to the friction caused between the powder and the plastic drum as the powder slid down the inside surface of the drum. A field meter reading of 500 KV/m (the maximum voltage the meter was capable of measuring) was recorded on an isolated area of the plastic drum which would have had the effect of charging the metal chime by induction.
Given the high rate of charge generation caused by frictional charging, the amount of electrostatic charge that could have been induced on the chime would have been limited by the surface area of the chime. In this case the surface area of the chime was approximated to 0.0641 m2 (99 in2).
If the total quantity of electrostatic charge (3.6 micro-coulombs) created by the movement of the powder was induced on the chime this would have exceeded the maximum charge density any surface can hold in air. The maximum charge density of a surface in air is equivalent to 27 micro-coulombs per square metre. The total charge density of the chime in this case, theoretically, would have been 56 micro-coulombs per square metre.
(I): Charge density (σ) = Total charge (Q) / surface area (A)
Charge density (σ) = 3.6 x 10-6 / 0.0641
Charge density (σ) = 56 x 10-6 C/m2
It can be assumed that the maximum charge density, i.e. the total possible amount of charge that could be held on the chime, was achieved through the simple and rapid act of tipping the powder from the drum into the vessel. In this study the capacitance of the chime was estimated to be 71 pico-farads. Knowing these values it is possible to estimate what the potential energy of the spark discharge was.
Taking the above formula (i), Q = σA, the maximum charge on the chime can be calculated:
=> 27 x 10-6 x 0.0641 = 1.7 x 10-6 C
Therefore, the total charge on the chime would have been close to 1.7 micro-coulombs. Hence the voltage of the chime would have been in the region of 24,000.
(II): voltage = total quantity of charge / capacitance of charged object
V = 1.7 x 10-6 / 71 x 10-12
V= 24 KV
The average breakdown voltage of air is 3000 volts per milli-metre, therefore the voltage of the chime would have been capable of discharging an electrostatic spark from a distance of at least 8 mm (0.3”) to the grounded process vessel.
The potential energy of the chime can be calculated from:
Potential energy (W), = Q2/2C
• Q = charge on chime
• C = capacitance of chime
Therefore the potential energy of the chime:
= (1.7 x 10-6)2 / (71 x 10-12).(2)
= (2.89 x 10-12) / (142 x 10-12)
= 20 milli-joules.
This exceeds the minimum ignition of the powder which was 12 milli-joules.
Given that the minimum ignition energy of the powder dispersed in air was 12 milli-joules and that the circumstances of the process proved there would have been significant electrostatic charging of the equipment, and other sources of ignition being eliminated, a static spark caused the ignition of the dust cloud that formed around the grounded process vessel.
Our series of case studies are intended to provide an insight into how, and why, static electricity provides the ignition source for serious fires or explosions that occur during everyday operations, involving the handling and processing of combustible products. Although static electricity can be regarded as a difficult subject to grasp, we hope that these case studies give you a better insight into the reasons why static electricity is a credible ignition source and what practical measures, based on internationally recognised codes of practice, can be taken to remove the fire and explosion risk it represents for your operations. The case studies cover a range of operations that involve flammable liquids and gases and combustible dusts when used in EX / HAZLOC areas.
Operation: vacuuming of off-spec toluene from a sump.
This case study investigates the causes behind a fire that occurred during a vacuum truck operation. The vacuum truck was deployed to a below grade sump that contained mostly of “off-spec” toluene. As the vacuum truck operation was nearing completion of the removal of the toluene from the sump an ignition of the vapors occurred resulting in a fire. In the ensuing investigation of the incident it was determined that the vacuum truck had not been grounded by the operator. Although other ignition sources would have been considered, the fact that the truck was not grounded and the material being transferred was toluene, it was highly plausible that a static spark was the cause of the fire.
For a static spark to be discharged from the surface of a metal object there needs to be a voltage on the charged object that exceeds the “breakdown voltage” of the surrounding atmosphere. This voltage results from the presence of too many positive or negative charges on the object and simply means that the voltage of the charged object is strong enough to create a conductive channel through the air, to a secondary object. The conductive channel provides a path for the static charges to flow through. In the split second that the channel is formed, the excess charges rapidly pass through the gap releasing energy in the process. The energy released results in a static spark and if a flammable atmosphere is present in the “spark gap” there is a high probability the energy of the spark will exceed the minimum ignition energy (MIE) of the vapor, gas or dust present in the spark gap.
To create a voltage there needs to be a constant supply of electrical charges to the object being electrified, which in this case is the vacuum truck. Effective grounding of the truck would have provided a means of sending the excess electrical charges to the general mass of the earth (grounding) removing the risk of the truck becoming electrified. In electrical terms this means there was a very high resistance from the chassis/tank of the truck to earth. A constant stream of electrical charge is a current so the more current flowing to the truck, the greater its voltage. But where does this electrical current come from? This is where the vacuuming operation combined with the suction of a material like toluene would have created a “streaming current”. Toluene has a very high resistivity with the effect that when it comes into rapid and repeated contact with other objects, especially conductive objects like metals, it strips electrons from the other material. This means the toluene carries more negative charge than positive charge. When the charged toluene makes contact with the truck it causes the outer surface of the truck to carry the same amount of negative charge.
In the case of this incident a neoprene hose with an embedded metal wire helix was used. As the metal helix would have been connected to the truck via the metal couplings of the hose, the full length of the hose would have been at the same voltage as the truck. To put some “hard numbers” on the case for static being the source of ignition we need to look at some of the physical characteristics of this operation.
A streaming current for a resistive liquid flowing through a pipe (including hoses) can be estimated from the equation:
Streaming current, IS = (2.5×10-5)(v2)(d2)
v = velocity of liquid (metres per second)
d = internal diameter of hose (metres)
It is known that the suction flow rate was 500 gallons per minute which is equivalent to 3.9 metres per second. The internal dimeter of the hose was 4 inches which is equivalent to 0.102 metres. Hence the streaming current would have been somewhere in the region of 3 micro-amps (3×10-6 A). It was estimated that the resistance between the truck and ground was at least 1×1010 ohms. Most of the resistance would have been provided by the truck’s wheels and the asphalt on which the truck was positioned.
The voltage of the truck could be assumed to have reached a level of at least:
V = R x I
Where V = the voltage of the truck-hose system
I = the streaming current provided by the charged toluene
R = the resistance to ground of the truck-hose system
V = (1×1010).(3×10-6)
Minimum truck voltage = 30,000 volts
As the ignition of the vapors occurred in the sump, the discharge must have been from the hose to the metal rim of the sump. However, in order to prove this theory we must investigate the hose itself. As stated earlier the hose was made from neoprene with an embedded metal wire helix. This means that the metal spiral was not in direct physical contact with any external objects. In addition, neoprene is a high resistivity material with a dielectric strength that is several times greater than air with a typical value of 10,000 volts required per mm of spark gap. The neoprene layer was 2 mm (0.08 in.) in thickness. This would mean that for every millimetre of distance between the metal helix and the metal sump there must have been a high enough voltage to discharge a spark with enough energy to exceed the MIE of the toluene. In order to breakdown the neoprene, at least 20,000 volts must have been present to achieve this. Given that the minimum voltage on the truck-hose system is estimated to be at least 30,000 volts it means an extra 10,000 volts would have been present to discharge from the neoprene surface to the sump. So, at some point in the operation, a gap between the surface of the hose and the sump must have been created resulting in a complete spark being discharged across a gap that had toluene vapors present.
The final part of the jigsaw is the potential energy of the spark itself. We can estimate the amount of energy available for a discharge via the spark from the equation:
Potential energy of spark (Joules) = ½ x capacitance of
charged object x square of the object’s voltage.
E = ½ CV2