By Greg Gummerson, Electronic Engineering Specialist and Senior Technical Advisor, Relectronic-Remech Inc.
We have investigated thousands of lightning strike claims since we began servicing the insurance industry in Canada in 1992. Throughout this period we have witnessed damages to a myriad of equipment and control systems, including commercial and residential losses.
By and large, commercial claims involving control systems are the most expensive to settle. There are many reasons for this—but before diving into the whys, hows and wherefores of these losses, a brief explanation of lightning should be addressed.
Lightning is an atmospheric discharge of electricity, which typically occurs during thunderstorms, and sometimes during volcanic eruptions or dust storms. In the atmospheric electrical discharge, a leader from a bolt of lightning can travel at speeds of 60 km/second, and can reach temperatures approaching 30,000°C (54,000°F), hot enough to fuse soil or sand into glass channels. There are over 16 million lightning storms every year worldwide. Of these strikes, the resulting damages are generally of two classifications: from direct lightning strikes and from indirect lightning strikes.
A direct lightning strike is normally the type that people most often try to protect themselves against. A direct lightning strike can generate surges up to and beyond 6 kV. A direct lightning strike places an electrical surge on the mains of a power utility and causes damage to equipment that is connected to the grid. Battery backups and line conditioners are used to protect equipment from damages due to a direct strike. Surge protectors that plug into a receptacle are also devices used to protect against these types of strikes and the resulting damages.
In our experience, indirect strikes are more frequently the cause of a loss. Indirect strikes can put surges on utility lines, but more often put surges on data communication lines as well. When lightning strikes, a substantial transfer of stored electrical energy is discharged. When this flow of electrical current occurs, an associated magnetic field is created. When the discharge of electrical energy is complete, the resulting magnetic field collapses. The laws of electromagnetic principles show that when a magnetic field is created and collapsed, a conductor located within the magnetic field has an associated induced current on it. (This is the same principle that an automobile’s charging system uses to keep the vehicle’s battery fully charged.) Following a lightning strike, the induced current searches for a path to ground, usually damaging electronic components along the way.
There are three different recognized means by which these surges are induced: resistive coupling, inductive coupling and capacitive coupling.
Resistive coupling occurs when lightning strikes within close proximity to a facility. This results in a massive rise in ground voltage. The rise in ground voltage affects electrical grounds (buried rods or buried pipe work) and can be conducted back to a building and to its electrical systems. Additionally, any communications cabling connecting the affected building to a second building provides a path for surges, allowing them to damage equipment in the second building as well.
Inductive coupling occurs when a lightning strike hits a conductor forming part of the structural protective system of a building or a structure in close proximity to the building. This generates a large electromagnetic pulse of energy that can be picked up by nearby cables, and cause damages as well.
Capacitive coupling occurs when lightning directly hits the electrical utility grid. High-voltage protection devices that are in place on the power grid dissipate much of the energy caused by the strike, but a large portion still travels along the lines. The high-frequency nature of this surge can couple the low-voltage and high-voltage windings in local transformers at facilities and damage equipment that the transformers normally feed.
So why do adjusters, in many cases after a lightning strike, feel as though they have bought the farm? There are many reasons for this, but in most cases it comes down to the manufacturer’s support for the damaged device, the type of interconnection between devices in a system, and the relatively low level of protection that communication devices inherently have.
For illustration purposes, let’s assume that a claim has occurred at a farm. The insured indicates that the feeding system and phones have been damaged. The feeding system provides feed to livestock, and it does this through a computer that monitors many factors, including the weight and calculated age of the livestock. The computer has to be able to mix food recipes, weigh the livestock and distribute the food to the proper eating areas. The manufacturer of this system has spent a great deal of resources to develop the software to perform these tasks in an accurate manner. All the electronics involved in the system have to be able to communicate with the computer so that the feeding operation can be adjusted appropriately. This means that load cells located inside scales must weigh feeds, water and the livestock, as well as the feeding bins. Load cells are devices that change in electrical resistance based on the force applied to them. They typically operate in the millivolt range (less than one volt) and have their own power distribution, shared from the communications lines back to the computer. However, there is no easy means to connect a load cell directly to a computer, so an interface circuit board is put in line to convert the normal load cell operating voltage to a number that can be communicated to the computer in a meaningful state. Then the computer can evaluate all of the weights required and start producing a feeding recipe to be distributed.
In most cases, there is more than one feeding area, so a network of piping is built throughout the building to accommodate the distribution of feed to each area. Normally this is controlled through a series of valves in the piping network that open and close. Each valve has its own unique ID or address so that the computer can access it independently of other valves. Just like the load cells, these valves need to talk to the computer, so again more interface circuitry is required to provide communications with the computer. For example, feeding area A may require 10 kg of feed and feeding area B may require 15 kg. The valves would close so that the feed in the piping would be dispensed to area A; then the load cell at A would report back once the 10 kg has been reached, and the computer would close off the valves to A and then open valves to area B, and so on. Throughout this process, the computer also has to control a drive system that pushes the feed through the piping, meaning that even more circuitry is required to drive the food delivery system.
As you can see, there are many aspects of the system that are depended on to perform their operations correctly. Independently, each section of the system is relatively simple. Therefore, the communication lines required are generally simple as well. The most common form of communication lines installed are serial communications RS232 or RS242. Serial communication allows many devices to be chained together on one line, but each of these devices, be it a valve control or load cell control, has a unique address. From a lightning claim perspective, this complicates things two-fold — first, because the devices on a communication line that are shared with more than one device are at risk of being damaged, and second, because the labour involved in troubleshooting this communications circuit can become expensive, depending on the number of devices there are to test on a particular circuit. Serial communications depend on at least three conductors, one for sending data, one for receiving data and a common or ground cable. Serial communications generally operate in the 0 V to 12 V range; this is a low-voltage cabling that is highly susceptible to induced currents. Furthermore, the electrical grounding of each device is tied along the communications line, which exposes all the devices, even if only one was initially subjected to a surge.
So, what about the computer itself? We have seen many of these damaged. Generally they are little different from standard personal computers, except they may in some cases have an interface card to allow communications. However, the software they run is normally proprietary and available from a single source, and has a relatively expensive replacement cost. To add to this expense, the manufacturer of the system may insist on supplying the computer itself, including the software. This makes supporting the system easier, because all of the manufacturer’s customers will have relatively the same configuration.
Finally, what if the system fails to operate properly? Almost all feeding systems of this nature are tied into a phone system or alarm panel, to allow the owner to be notified of a problem. So an additional set of communication lines has been added to the mix. We have seen cases in which phone lines had induced charges from lightning that travelled through the alarm system dialer to the computer, and then out to the feeding system itself.
The electronics involved are not overly sophisticated; in fact, they use technologies that were introduced into the market over thirty years ago. However, when combined, a sophisticated network of switches and sensors is created, all controlled by a computer running specialized software. The production volume of these systems is generally low, driving up the cost to supply the specific market.
In our experience of claims ranging from farms to factory industrial control systems, the costs can be high, but they can also be controlled if repair and service vendors are available and co-operative. If service vendors understand the process from an insurance point of view, they will try to repair and service the system cost-effectively. Furthermore, a third party’s involvement can aid in lightning repair issues where warranties are still in place on equipment. For example, a repair vendor may be indicating that the entire installation is suspect, due to a lightning strike, and apprehensive about making repairs. Allowing and paying for a recertification process can alleviate the vendors’ and suppliers’ concerns. Allowing a grace period for functional equipment that may have suffered marginal damages, but are not complete failures, may also assist in warranty matters.
When the repairs are made and the invoices are submitted, it’s not uncommon to think that you just have “bought the farm.” This is where a consultant should be able to assist in the settlement process, by reviewing repair invoices to confirm that the invoices submitted are directly related to the loss. If any upgrades have been supplied, these would be noted and valued. Having a consultant involved as early as possible is beneficial, as the steps to a complete repair will be clearer for both the insurer and the repair vendor. A consultant may also be able to provide input on possible alternative systems and repairs for obsolete equipment. The coverage can be better confirmed to the insured, and repair vendors will know that they are to be accountable.