Forensics, Engineering Assessments and Design

Light, Medium and Heavy industrial including chemical processes, fuels, petroleum products, etc.

Commercial

Hospitality

Residential

Healthcare

Power

Controls

Contracting

September 27, 1998

REPORT TO:     Mr. Jones

                      ABC Corporation

                      111 Corporate Road

                       Anywhere, USA, zip code

FROM:              David Burger, P.E.

REFERENCE:    Electrical Fire

                      XYZ, Incorporated

                      Dallas, TX

                      Client Reference:      Policy 8500988-00

                      Forensics File Number: HST00001

            As per your request on September 15^th, 1998, investigation to the referenced incident was initiated. The purpose of the investigation was to evaluate the cause of the September 12^th, 1998 incident that involved the main electrical switchgear in the pole production facility, Building No. XX.

            A visit was made to the site in Dallas, TX on September 23rd, 1998. At that time, I met with Mr. Jones, and other individuals, involved with the owner’s investigation and recovery from the incident, including John Smith. The production facility was toured and available technical information was provided by the owner. Copies of witness statements were made available to read at the site but were not allowed to be copied.

This report summarizes background information and the current status of our investigation of the referenced incident.

A.   BACKGROUND

1. The facility that experienced the incident is XYZ’s pole production facility, Building No. XX. This is a heavy industrial facility with a number of buildings including the building of incident.

2. Building No. XX is used to finish raw stock into a finished pole using various machines, processes and tools.

B.   FACILITY AND AFFECTED COMPONENTS

1. Building No. XX is served from a nearby, on-site, utility owned transformer located approximately 200 feet west of the building. The transformer is served from a nearby, utility owned, power pole with an underground high voltage line. The transformer reduces the high voltage to 460 volts 3-phase which is then used to operate the facility.

2. The power conductors from the transformer to Building no. XX is a set of copper cables run underground to just inside the main electric room. The main electric room is located next to the employee break room. Once inside the main electric room the conductors transition from below ground and connect to a copper buss-bar system.  The copper buss-bar rises up the wall to an overhead aluminum buss-bar system. Both buss-bar systems are enclosed in a steel shroud.

3.  The aluminum buss-bar system runs across the ceiling space to the main switchgear. The switchgear has one main breaker rated 2500 amps. This main gear serves all of the electrical loads in Building No. XX.

4.  The main gear was originally designed by Rautaruukki Oy, a company in Finland, and has no Underwriters Laboratory or equivalent listings. It was provided with the production line equipment when the plant was originally designed. The gear sits on top of a false wood floor that allows conduits and cables to the plant to enter and leave.

5. Within a cabinet immediately adjacent to the 2500 amp main breaker, is a 1000 amp breaker that serves a piece of equipment called the Thermatool Welding Machine. Subsequent cabinets contain other smaller breakers and switches.

6. The main electric room is a lean-to style addition to Building XX. Building XX is a metal building with metal panel walls. Some areas of the building are partially insulated and air conditioned, including the electric room, allowing them to remain cool in the summer. It was noted by the maintenance staff that the electric room in the past has had numerous roof leaks over the years and possible roof leaks at the time of the incident.

C.  INCIDENT

1. On Saturday, September 12^th, 1998, at the beginning of the third shift, an electrical fire occurred in the main 2500 amp electrical switchgear. The fire was located just above the 2500 amp main switch.

2. The time of the incident was about 10:58 PM. Third shift was scheduled to begin at 11:00 PM. Immediately prior to the incident, the facility was not in operation. It had been shutdown at the end of the first shift, roughly 3:00PM. There was no second shift that day and all of the equipment had been shut off.

3.  At the time of the incident, it was reported to be raining. The prior few days of the week and the weekend had been very rainy due to a tropical storm that was passing over the area. Rain began Thursday night, September 10^th, and continued through Saturday night. Rainfall amounts ranged from five to ten inches throughout the area.

4. At roughly 10:30 PM, prior to the incident, the lead-man for the third shift had arrived and began the required normal start-up procedures. This process involved starting each piece of equipment to insure it was operating and ready for third shift personnel to use. His normal procedure started him in the coil processing area. From there, he progressed throughout the facility starting all the equipment. This included hydraulic pumps, air compressors, burnisher, etc.

5. Having finished the start-up procedure, the lead-man went to the burnisher to process a pole in order to determine its quality. At roughly 10:58 PM, while loading a pole into the burnisher, all power to Building No. XX was lost.

6. The lead-man immediately telephoned the on-site maintenance electrician located in the Large-Pole facility and informed him of the outage. Shortly after that, an employee reported smelling smoke in the employee lunch-room. The lunch room is located next to the electric room where the incident occurred.

7. At an undetermined time, shortly after the initial calls, the electrician arrived, unlocked the main electric room and found it filled with smoke. A call was placed to 911. The fire department responded and extinguished the fire. While the fire department was in route to the site, Building No. XX was evacuated for the safety of the personnel.

8. Following the incident, a report was received that the lights in the facility had flickered just prior to all power being lost.

9. As a result of the incident, damage was caused to the 2500 amp main breaker, the adjacent 1000 amp branch breaker serving the Thermatool welding machine, the buss-bar system and a few adjoining motor starters.  Significant physical damage was found on the connecting buss-bar systems.  

D.  APPLICABLE CODES AND STANDARDS(At time of plant construction)

            1. 1981 National Electrical Code

            2. Standard Building Code – Southern Building Codes Congress International

E.  CAUSATION

1. There were five primary possibilities considered as to why the system failed; loss of power to one phase, loose connections on the buss-bars, short-circuit caused by the entry of a rodent, snake, or pest, short-circuit due to entry of moisture into the main switchgear or an unexplained failure of the main switchgear. Each will be examined in the following paragraphs.

2. Loss of electrical power to one phase was a consideration. Loss of power in one phase can happen do to a surge or a lightning strike. It can also happen as a result of a short-circuit, as will be discussed in later paragraphs. While investigating the incident, it was found that three utility company fuses had blown. All were located on the high voltage side of the utility system. Two were at the transformer and one at the utility pole. This suggests that the fuse at the utility pole possibly failed from a surge or lightning which led to a single phase condition and the subsequent short-circuit. The short circuit would have then tripped the remaining two fuses on the transformer. If this had happened, the motors that were running at the time could have continued to run during the single phase condition and would have drawn unusually high currents. In this condition, the over-current protection devices on various motors should have tripped or we should have seen some damaged motors. Neither of these conditions was found during our investigation. Lightning was considered a reason for a possible single phase condition. Global Atmospherics, Inc was contacted and asked to review their records on the night of the incident. Global Atmospherics, Inc provided a Lightning Verification Report for the time and location of the incident and found no such strike in the area. This report has been included as an attachment. In addition, the high voltage fusing was the only over-current protection ahead of the switchgear so the incident could have easily caused the fuses to open, not the other way around. So there is no direct cause and effect that can be drawn from which fuses tripped and their location in the electrical system. Having found no motor damage or lightning present strongly suggests that loss of electrical power to one phase was not likely the cause of the incident.

3. Possible loose connections on the aluminum buss-bar inside the main switchgear was suggested by the owners recovery team. While disassembling the main switchgear and buss-bar, the electricians noted unusually loose connecting bolts. Loose connections in an electrical system can cause joints to arc and overheat resulting in eventual failure. However, upon visual examination of the connection joints, no telltale signs such as discoloration, pitting or burning were seen at these joints. In addition, a severe short circuit, of the level our team feels caused the damage, puts extreme mechanical stress on all parts of the affected gear and buss-bars. This stress could easily have caused the deformation found in the conical washers and the resultant loose bolts. Having found no signs of over-heating buss-bars it is unlikely that this was the cause of the incident.

4. A short-circuit caused by a rodent, snake or other pest entering the main switchgear was consider a possibility but was dismissed. The gear sits on top of a false floor. It has significant openings in the bottom for cables to enter and leave. These same openings could have allowed the entry of pests. For a short-circuit of the severity here to have occurred, a very large rodent or snake would have had to enter the main gear and short it out. No evidence of remains was found. The severity of the fault could have vaporized the remains so that lack of remains is not conclusive. However, the short circuit took place above the main breaker, high above the floor. This would have required an unwanted pest to make its way through all of the buss-bars at the bottom of the switchgear without shorting them out nearer to the floor. It was also observed that the two buss-bars believed to have initiated the short-circuit were likely in a phase to phase short. So in addition to making its way through exposed buss-bars, the pest would have then had to climb onto a single buss-bar without causing it to short-circuit to ground. It appears very unlikely that a rodent could have achieved such a feat, therefore, it seems very unlikely that a pest caused the short-circuit to occur.

5. Moisture entry into the switchgear is believed to be the reason for the incident by causing a short-circuit. Two possible sources for moisture entry into the gear were found; the buss-bar shroud and conduit system, and leaks from the roof structure. It was found during examination of the damaged buss-bar that there was corrosion on the aluminum lugs and the bolted connections above the main 2500 amp switch. It had been raining for some time before the incident causing high humidity for numerous days. The conduits entering the buss-bar shroud were not sealed. This could have allowed the entry of hot moist air into the switchgear. With the gear located in an air conditioned room, the moisture could have condensed and penetrated the insulating materials or bridged the mounting insulators, eventually resulting in a short-circuit between buss-bars. Another likely method of moisture intrusion was through roof leaks. It was noted in an attached engineering letter that leaks had re-developed recently during the tropical storm and had been a reoccurring problem for years. If a leak went unnoticed above the buss-bar shroud, water could have followed the electrical pathway into the gear and eventually shorted the buss-bars. Additionally, photographs have been included showing leaks near the new replacement switchgear. This is the most likely cause of the incident.

6. An unexplained failure of the switchgear was possible. However, no mechanical damage or failure above or below the short-circuit location was found. The buss-bar failure location and the loss of metal during the short-circuit strongly supports failure of the insulation material caused by moisture.

F.  CONCLUSIONS

1. It is the opinion my opinion, that the cause of the incident was failure of the insulating materials between the buss-bars in the switchgear. This caused the subsequent severe short-circuit. The insulation failure was caused by entry of moisture into the gear from either roof leaks, condensation or both.

2. My recommendation is that the replacement of the switchgear should allow only bottom entry of electrical raceways. This will prevent openings into the top of the gear that can allow pathways for water to enter. All conduits entering the gear should be sealed with non-hardening electrical putty having a chemical compatibility with the insulation on the electrical conductors. This will limit moist air from again entering the new equipment.

3. Pest control measures should be undertaken, establishing a program to limit the possibility of pest intrusion.

4. The bottom of the new switchgear and the entire electrical room perimeter should be inspected and sealed to further limit pest intrusion into the area.

5. The maintenance department should talk to the local utility to determine the maximum power draw and the possibility of reducing the utility transformer fuses to the lowest practical value. This will provide faster short-circuit interruption and limit damage in the event of a future incident.

6. The maintenance department should also initiate an infrared and ultrasonic scanning program to monitor for electrical system hot-spots that may lead to failures. This will provide a record of system performance that can be used to rule out or support theories in the event of a future incident.

March 10, 2005

To: Client

XYZ Street

Orlando, FL 328XX

Re:      

Resort at Captiva

Forensics Project # 05093

Attn:    John Doe, President

Dear Mr. Doe:

As per our telephone conversation of February 28th, I was retained by your company to provide a review of the engineering reports and damage to the above site to render my opinion of the extent of the reconditioning required on buildings 1700 and 1800.

I have had the opportunity to read the report prepared by XXX Consulting, Inc. on behalf of the insurance company and the electrical reports prepared by XXX Professional Engineering on behalf of the ownership group.  I also performed an on-site survey on Tuesday March 1, 2005.

In addition to the review of the above documents, I have downloaded and provided copies of the following NEMA Publications:

Guidelines for Handling Water Damaged Electrical Equipment

“AB4-2003” Guidelines For Preventative Maintenance of Molded Case Circuit Breakers.

As I have stated, I have reviewed the XXX report and will address each building as it relates to this report.

In the discussion, on sheet seven on the XXX report, the definitions for dry location, damp location and wet locations are given as defined by the National Electrical Code.  The intent of this discussion is to proffer the position that the water from wind driven rain falling on the electrical enclosures, devices, equipment, receptacles, outlets, light fixtures, circuit breakers, panels, fire alarm devices, etc. constitute a dry location with temporary dampness.  I have reviewed these definitions and do not believe that the roof, having been severely breached, allowing wind driven rain into the building as well as the storm surge flowing through the ground floor would constitute a temporary damp location in an otherwise dry space.  Nor do I believe that this is considered a damp location as defined in National Electric Code.  These definitions, although helpful, are not what the National Electrical Code envisioned them to be used for.

The National Electrical Code defines damp locations are those that may be subject to inadvertent, light or moderate water spray from rain or hose spray or from condensation on devices due to differences in temperature between the device and the surrounding area.  Wet locations, as defined in the XXX report document, are areas subject to saturation with water or other liquids.  It is my reasonable belief that with parts of the roof being damaged allowing water to run down through the building, wind driven rain coming through sliding glass doors and in some instance flowing storm surge, that this would constitute a wet location for electric devices.  The XXX report, on page seven in the last paragraph, states “therefore a building which is temporarily subject to moderate degrees of moisture does not necessarily constitute a replacement of electrical equipment”.  I do not believe that wind driven rain within a building during a hurricane constitutes a moderate degree of moisture and had there been a moderate degree of moisture I do not believe that all the gypsum walls and ceiling would have had to be removed entirely in building 1700.  It is my professional opinion that there was significant water in many rooms which damaged the electrical distribution system.

In the first paragraph of sheet eight, the XXX report states that the wiring is designed for wet locations.  I agree with this statement but take must take exception to its use in this situation.  The NEMA publication Guidelines For Handling Water Damaged Electrical Equipment indicates that even wet location wiring, subject to water at the ends, where wicking can occur, should at the very least be tested to ensure that it maintains its integrity for reuse.  Turning your attention to the same NEMA guideline, page five,  it further states “Items which may possibly be reconditioned by trained personnel and consultation with the manufacturer:

Any wire or cable product, not containing fillers, that are suitable for wet location, whose ends have been exposed to water, may be considered candidate for purging,  using an inert gas under pressure to remove water contained in the product, under engineering supervision.  This procedure, if employed on a cable, should be tested prior to energizing.  As a minimum, an insulation resistance test with a MEGA-OHM Meter should can be conducted.”

As previously recommended at our meeting on site, I suggest that insulation testing be done on at least 15% of all circuits within each of the two buildings and 100% of the service entrance equipment and conductors.  If any significant problems are found, all conductors should be tested per the NEMA guidelines.

I also reviewed page five of the Guidelines for Handling Water Damaged Electrical Equipment relating to wiring devices and lighting fixtures. Here it clearly indicates that it should not be attempted to reuse end devices that may have been subjected to contaminants or water.  Refer back to page three of the NEMA requirement under “Items Requiring Complete Replacement.”  These items include electrically controlled solid state contactors and starters, components containing semi-conductors and transistors, overload relays, molded case circuit breakers and switches, and fuses.  This would indicate that fire alarm devices, smoke detectors, thermostats, etc., that have been exposed to salt spray, will likely fail prematurely and should be replaced.

The XXX Consulting Engineers report, page eight, states “therefore lacking this installation need, it is estimated that the remaining cost to refurbish and/or replace the damaged electrical systems ensuing from storm related activities, will not accumulate to more than 30% of the entire electrical system.”  However, no section of the two NEMA Guidelines or manufacturers recommendations support the XXX reports conclusion.  Nor does the XXX report quote any industry recognized guide to support their position.  It is my determination that, based on industry guidelines, more than 30% of the electrical system needs to be replaced. This would include but not be limited to all of the thermostats, fire alarm devices, outlets, lights, etc., since they were exposed to salt spray.  An individual listing of damage begins on page thirty-three of the XXX report for buildings.  I have reviewed each of the bullet items with the following recommendation:

• Exterior Lighting:  I agree that the exterior perimeter wall mounted light fixtures, oval shaped, were found to be mostly intact and designed for damp locations. I agree with the statement that the damaged fixtures can be replaced and matched to the existing fixtures.  I would also agree that most of the walkway and landscape lighting that were not directly attached to the building were not damaged, however, in order to maintain continuity of the architectural appearance, all of there fixtures need to be replaced.

• Electric Service Equipment:  I have reviewed the parallel over-current protection described in this section of the XXX report as well as the XXX report.  In my review of The National Electrical Code on this condition I have found that since at least 1993 this type of installation has not been allowed.  It is likely that it was never allowed under the conditions currently being used.  As I do not have information regarding how this came about, I cannot speculate why this condition exists.  As a result of the damage done to the building, the code official will require that this be brought up to current code.  This will require the installation of a service disconnect rated at least 600 amps to re-feed the building 1700 main service distribution system.

• Panel-boards, switchboards and load centers:  In my review of building 1700, it appears that water had migrated into most of the top floor panel-boards and possibly the 1^st floor panel-boards.  At least some circuit breakers show evidence of recent water contamination such as water marks and salt residue.  I would direct your attention to the NEMA Publication on water damage, page three, ‘Items Requiring Complete Replacement, molded case circuit breakers and molded case switches”.  Further, I would direct your attention to NEMA publication AB4-2003 for molded case circuit breakers, page nine, paragraph 3.6 , labeled “Water Damaged Molded Case Circuit Breakers,” “Circuit Breakers which have shown to have been subjected to water damage, e.g. by flooding or sprinkler discharge, should be replaced.  For additional information refer to NEMA document Guidelines for Handling Water Damage Electrical Equipment.”

There is a reference in both reports on the sub-fed panels on the ground floor.  They talk about panel taps. The feeder between the first and second floor panels does not comply with the tap rules quoted by XXX because The National Electrical Code requires they terminate in an over-current device rated for the wire size used.  The main breaker in the sub-fed panel is rated much higher than allowed by Code.  This will need to be brought up to current standards before returning the building to service.

• Branch circuit wiring methods:  The XXX report only recommends devices in rooms 31A through 31G of building 1700 be replaced when in fact the devices throughout the entire building likely were made wet with salt water and rain.  It is my opinion that, based on the NEMA Publications and Guidelines and manufacturers recommendation, that all devices in this building must be replaced.

• Interior Lighting Systems:  Once again I believe that most of the light fixtures on the 2^nd floor, where the roof was damaged and the items were exposed to direct water intrusion, have been damaged and that NEMA Guidelines For Handling Water Damaged Electrical Equipment clearly states that these all must be replaced.

• Egress and Life Safety Items:  I am in agreement with the assessment by XXX Consulting Engineers on this item.

• HVAC Equipment:  I reviewed the equipment safety disconnect on the exterior walls.  The final conductor terminations must be replaced when putting the units back in service due to wicking of water into the wire.

• Low voltage systems:  I agree with the XXX Consulting report relative to the existing telephone terminal board and equipment room.  However, the rest of the building was not addressed in the report.  It is clear that most of the end devices were subjected to water intrusion. Therefore the cables, especially on low voltage systems, must be considered for replacement or at least 50% of the cables tested per the Guidelines For Handling Water Damage Electrical Equipment to ensure that these low voltage systems cables were not damaged. Failure to do this could result in costly repairs and on-going failures for years.
• Fire Alarm Systems:  Once again, those devices within the rooms and some of the devices on the exterior appear to have been exposed to significant amount of water intrusion and  need to be replaced. Iit states under the NEMA guidelines, on page six, “Electronic Products, Including Signaling, Protection, Communication Systems and Industrial Controls:

Equipment used in signaling, protection and communication systems generally contain electronic components, and the exposure of such equipment to flooding by water can adversely affect the reliability those systems.  Contamination by pollutants or debris in flood waters may cause corrosion of components of the system, shorting of printed circuits, or alteration of circuit characteristics.  Since some of these types of installations are classified as life safety systems, it is important that the reliability of those systems be maintained.”

It is not my belief that these items were flooded, but instead, many of the items had significant wet conditions, possibly with water running through them during the storm due to the severe roof damage.  Continuing on with the NEMA Guidelines, “where such systems are damaged by water, it is recommended components of these systems be replaced or returned to the manufacturer for appropriate cleaning, recalibration, and testing.  Manufacturers of these systems should be contacted for information on specific equipment.”

Much of the NEMA guidelines recommend that if any of these electrical systems are going to be put back into service, including receptacles, light fixtures, fire alarm devices, smoke detectors, etc., that the manufactures be contacted for approval and that much of the equipment be sent back to the manufacturer for cleaning and recalibration before being put back into service.  This would be very costly to do for light switches, dimmers, receptacles, fire alarm, etc.  Therefore, it is my opinion that these items must be replaced.

Building 1800:

It appears that the building 1800 did not receive as significant damage as building 1700.  It is my opinion that, like building 1700, a selected amount of branch cabling must be tested with the MEGA OHM Meter to verify that it has not wicked moister making persons susceptible to electrocution or shock.  It is my recommendation that the service entrance be completely tested. 

It is only the degree of damage that should be in contention, not the fact that devices need to be replaced and the wiring needs to be tested.

In conclusion, I believe that, although all the wiring within the building does not need to be replaced, certainly a select amount of it needs to be tested, including all of the service entrance wiring for both buildings. Many of the end devices need to be replaced, including the outlets, receptacles, lights, light switches, dimmers, smoke detectors, thermostats, cable television terminations, etc., that have been exposed to water.  It is my opinion that, certainly in building 1700, many of these devices were in wet conditions as recognized by NEMA and the National Electric Code.  This is not a dry or damp situation as described in the National Electrical Code or the XXX Consultants report.  It is my opinion that a substantial amount of the system needs to be tested and reconditioned and that any reconditioning of panel boards, circuit breakers, etc. be done by the manufacturers under their supervision or replaced in their entirety.  I further recommend that a selected number, 15% – 20%, of the circuit breakers be taken out of the building load centers and sent out for destructive testing to determine if the lubricants have been displaced, debris accumulated in them or corrosion has taken place on the internal workings of the circuit breakers.  If this found to be a problem, it is my opinion that many of the circuit breakers must be replaced.

The NEMA publications referenced in my report require qualified testing by manufacturers of their products and qualified installers perform testing and reconditioning of any electrical components that are going to be put back in the service.  The NEMA of publications, although just guidelines, are put together by highly qualified people within the industry, including electricians, professional engineers, manufacturers, etc., who are well aware of the consequences of not following clearly defined guidelines.  Failure to follow the guidelines exposes all parties involved to potential lawsuits and claims.  It is my opinion, therefore, that the NEMA of publication, Guidelines For Handling Water Damaged Electrical Equipment should be followed exclusively.

If you have any questions concerning this matter please do not hesitate to contact our office at your convenience, I will be happy to discuss these items with you personally.

Sincerely,

David A. Burger, P.E.

DAB:ss

Enclosures:      NEMA Standards Publication AB 4-2003

Guidelines for Inspection and Preventive Maintenance of Molded Case Circuit Breakers Used in Commercial and Industrial Applications.

NEMA Standards Publication AB 4-2003

Guidelines for Handling Water Damage Electrical Equipment.