| by Kenneth Chase | 10 comments

NTSB Board Meeting for Investigation of Silver Spring Maryland Apartment Building Explosion


Morning. Welcome to the boardroom of the National Transportation Safety Board. I’m Robert Sumwalt and I’m honored to serve as the Chairman of the NTSB. Joining us today are my colleagues, Vice Chairman, Bruce Landsberg, Member Earl Weener, and Member Jennifer Homendy. Today we meet in open session as required by the government in the Sunshine Act to consider a gas pipeline related building explosion and fire in Silver Springs, Maryland on the night of August the 10th, 2016. The explosion caused a 14 unit apartment building to partially collapse. A second building sustained substantial damage. Residents of both buildings were evacuated and many residents of nearby buildings voluntarily self-evacuated. Tragically, the explosion took the lives of seven people. Of the survivors, 65 residents and three firefighters were transported to the hospital. On behalf of all of us here at the NTSB, we’d like to offer our sincere condolences for all of you who have been affected by this tragedy. We know that this accident investigation has taken more than two years to complete and that longer investigations can add to frustration on the part of survivors and families during a time already marked by pain and loss but please understand that the goal of the NTSB safety investigation is to improve safety to prevent future accidents so that others will be spared of future pain and loss. Meeting this goal with confidence, it takes time and given the challenges of this investigation, investigators had to work even harder to consider and eliminate all the factors that may have caused the explosion. We hope that our work in this investigation honors those lost and entered in the accident throughout the recommendations which if acted upon would prevent future tragedies that echo the Silver Spring explosion and fire. In a moment investigators will provide details of how the accident unfolded. They will also explain how a communication gap might have resulted in missed opportunities to detect and repair the failure. Six times in the weeks and months preceding the accident, residents reported that a natural gas smell, they reported that to Kay Management. In each case, maintenance staff reported that they did not detect gas or attributed the smell to the painting of apartments. Two weeks before the accident, a resident called 911 to report a gas odor. The fire department unable to access the building’s basement meter room, they did not detect the gas. On August the 10th, on the night of the accident, that same resident again smelled gas but before he could act, the explosion occurred. In the lead-up to the accident and indeed for five years prior there was no evidence that residents, Kay Management, or any emergency personnel notified Washington Gas of a gas odor and generally, Washington Gas technicians have the expertise and the equipment to identify methane at lower concentrations. We will also consider the installation of fixed methane detectors that go beyond the literal smell test. The odorant in natural gas provides some warning but the methane detector alarms can provide a reliable, additional safeguard. And finally we will discuss the placement of mercury service regulators, features intended to safely vent excess gas outside of the building rather than let it build up in interior spaces. Our staff have pursued all avenues in proposing a finding, a probable cause, and recommendations to this board. The order of the meeting today will be that staff will present various presentations, technical presentations, to the board. We on the board will then question the staff and then we will propose any amendments, consider any amendments that may be offered, and ultimately, we will vote on the report. We will vote on findings, the probable cause, the recommendations, and then the report overall. We hope that our report, if adopted, will provide the best opportunity to enhance safety. Our public docket contains thousands of pages of information, information used to support this investigation. That is available at NTSB’s website at www.ntsb.gov. It will have photographs additional information and post-accident interviews. Once finalized in a few weeks after we adopt the report, it takes a few weeks for the report to be formatted and uploaded on our web page and in a few weeks the final report will be uploaded on our web page. Now, Managing Director Sharon Bryson, good morning. If you’ll kindly introduce the staff. Thank you Mr. Chairman. A few announcements before we begin. I’d like to ask the folks here in the boardroom if you’ve not already done so, if you could please silence your mobile phones and other electronic devices. There are two exits in the front of the auditorium, one to my left, one to my right which is across the stage here. If there is, in the event of an emergency, please go down the stairs, out the doors, and follow the illuminated exit signs to depart the facility. You may also exit to the rear of the auditorium and proceeded out the glass doors you entered, go up the stairs and exit straight ahead through the large glass doors to the outside. Once you’ve exited, turn left, follow the sidewalk to the end of the street. In the event of an emergency, please walk quickly to the nearest exit and make your way to the outside following the instructions of NTSB staff personnel. Do not delay. If you have any questions or concerns, please don’t hesitate to contact any NTSB personnel. There are a number of them here with you in the room today. Seated at the panel this morning, unless otherwise noted, are staff members of the Office of Rail, Pipeline, and Hazardous Material. On the first row, to my right, is Mr. Robert Hall, director of the Office of Rail, Pipeline, and Hazardous Materials. Sitting to Rob’s right, is Rachael Gunaratnam, who is the investigator-in-charge on this accident. To her right is Frank Zakar. He is in the Office of Research and Engineering, responsible for metallurgy. To Frank’s right is Joe Pantiogou, from the Office of Research and Engineering, a fire specialist. And to Joe’s right is Nancy Mcatee, also in the Office of Research and Engineering and a fire specialist. Sitting directly behind me is Dolline Hatchet, the director of our Office of Safety Recommendations and Communications. To Dolline’s right is Kathleen, Kathy Silbaugh, our general counsel. To her right is Jim Ritter, the director of the Office of Research and Engineering and to Jim’s right is Bob Beaton, the acting director of our Pipeline Division. In the last row is Alex Coletti, also from Rail, Pipeline and Hazardous and she’s handling visuals and timer for us today. To her right is Gina Evans the editor and to her right is Jeff Marcus and he’s the acting chief in our Office of Safety Recommendations and Communications. With that, the presentations will begin with an overview of the accident by the investigator-in-charge Rachael Gunaratnam. Rachael? Good morning, Chairman Sumwalt, Vice Chairman Landsberg, members of the board. On August 10th, 2016 around 11:51 p.m., a natural gas explosion and fire occurred at a 14-unit apartment building located at the Flower Branch apartment complex at 8701 Arliss Street in Silver Spring, Maryland. Montgomery County Fire and Rescue units responded at 11:56 p.m. and notified Washington Gas Company. Within the next hour, the gas crew arrived and shut off the mainline that supply gas to five apartment buildings along Arliss Street. As a result of the explosion, seven residents died and 65 residents were transported to the hospital. Three firefighters were also treated and released. The explosion and fire also resulted in the partial collapse of the first, second, and third floor including the roof of building 8701. The adjacent apartment building, 8703 Arliss Street, shared a common wall that also collapsed. The next slide will display two silent video clips, one from a witness cell phone and another which is an aerial view from Anne Arundel, Prince George’s County Police County helicopter. This video footage may be upsetting to some members of the audience present here today or viewing this proceeding through the webcast. I will now pause for a moment to allow those who prefer not to view the video to leave the room or temporarily suspend your video feed before we continue. A resident from building 8701 smelled gas between 11 to 11:30 p.m. on August 10th after returning from work. As he entered the building, he smelled gas in the stairwell. He left the building after 11:30 to take the trash out and reported that the smell was stronger as he approached the lower floors. He went to investigate the odor in the basement and heard a hissing noise but could not access the meter room because it was locked. The building exploded before he could call 911. NTSB investigators arrived the next day. Washington Gas conducted pressure tests on the gas main and the two service lines to building 8701 and 8703. They also conducted soil bar hole tests on the main and service lines in the front and back of buildings 8701 and 8703. No leaks were detected. Post-accident odorant testing was also found compliant with regulatory requirements for odorizing natural gas. The Bureau of Alcohol, Tobacco, Firearms and Explosives determined from the blast damage and pattern of debris that the explosion originated from the meter room in the basement of building 8701. They concluded natural gas was the fuel source of the explosion and ruled out other sources. Investigators narrowed down two potential areas where natural gas leaks could have occurred in the basement of building 8701. One area identified was the vent line union connections from the gas meter assembly. Mr. Frank Zakar will discuss this in further detail in his presentation. Investigators also considered a commercial water heater as a potential source of fuel but there was no evidence that the water heater and its associated piping failed prior to the explosion. Here is the floor plan for the basement of building 8701 which included one apartment, a leasing office, a maintenance shop, and the meter room where the gas meter assembly and water furnace was located. Right above the basement level were twelve other apartments located on floors one, two, and three. The arrow indicates the shared wall between buildings 8701 and 8703 where the gas meter assembly was located. Here is the ATF’s reconstruction of the meter room in building 8701 after the debris was removed. The gas meter assembly, the electric meter rack, and water heater were placed in its likely pre-explosion location. The gas meter assembly consisted of 15 meters that serviced 13 apartments, the maintenance shop, and leasing office. Mr. Zakar will also discuss this further in his presentation. The safety issues for this investigation include the location of service regulators within a structure, the inspection of the gas meter assembly, notification of natural gas odor to the gas company, and natural gas detection methods. The parties to the investigation are the Pipeline Hazardous Materials Safety Administration, the Maryland Public Service Commission, the Montgomery County Fire and Rescue Service, the Montgomery County Police Department, and the Washington Gas Light Company. I will turn it over now to Mr. Frank Zakar to discuss the laboratory examination of the equipment. Good morning, Mr. Chairman Sumwalt, Vice Chairman Landsberg, Member Weener, and Member Homendy. In this presentation I will discuss the examination of the mercury service regulators and their respective union assemblies. This photograph shows the reconstructed configuration of the two service regulators, meters, water heater, at 8701 Arliss Street prior to the accident. The two service regulators were referred to as the upper and lower service regulators. This is a close-up photograph of the reconstructed assembly with the two service regulators circled in red. Gas is delivered to the regulators at a pressure of about 20 pounds per square inch, following the path indicated by the white arrows. The purpose of a service regulator is to reduce the gas pressure to a level that is suitable for operating household appliances which typically operate at a pressure of about 0.25 pounds per square inch. From the regulators gas is delivered to the gas meters in the direction indicated by the yellow arrows and thereafter gas is distributed to the apartments. When excess pressure builds up inside a service regulator, gas departs the regulator and flows out of the building by means of a vent pipe, in the direction indicated by the black arrows. Excess pressure buildup is relieved through a mercury cup located at the bottom of the regulator in the area highlighted by a red circle. The excess gas will forcefully flow through a column of liquid mercury in the direction indicated by green arrows and then flow through a vent pipe that leads out of the building. When excess gas flows out of the vent pipe, it is referred to as venting or an overpressure event. Venting can occur when parts within a regulator degrade. For example, insufficient mercury level or a perforated flexible diaphragm can lead to venting. Material build-up between the valve seat and orifice can prevent the valve seat from fully sealing against the orifice also leading to venting. Investigators examined the regulators from the accident. However, they sustained severe damage from the fire, as shown in the left photograph. For example, the aluminum cover and internal components of the service regulators were severely damaged or consumed by the fire. Damage to the regulators prevented investigators from finding any evidence of a specific malfunction. The service regulators are connected to the vent pipe through what is called a union assembly. The union for the upper regulator is shown connected at the upper left side of the photograph and this is how it was recovered after the accident. The union assembly was supposed to be connected to the service regulator through an external threaded pipe extension, referred to as a nipple. However, the nipple was separated from the outlet portion of the gas regulator as shown by the dotted black lines. Detailed examination of the separated nipple revealed the first four external threads were severely deformed. This damage coincided with the first four fractured or stripped threads observed on the mating outlet port of the regulator as shown in the upper right photograph. Therefore, NTSB staff believe that the upper service regulator was most likely connected to the vent line through the threaded nipple and union assembly prior to the explosion in building 8701 and the regulator likely became separated from the vent line due to forces of the explosion or from the building collapse. The photograph on the left shows the union assembly for the lower regulator assembly. In this case, the union assembly was oriented horizontally. The photograph on the right is a close-up photograph of the assembly. The nut was found disconnected from the socket after the accident. When mating threads are pulled apart by force, fracture or deformation by shear of the threads can be expected. However, detailed examination of the internal thread for the nut and external thread for the socket revealed the mating threads showed no evidence of major deformation damaged or stripped threads. This point is further illustrated by an X-ray CT scan of the nut and socket from the lower union assembly. The view on the left side shows the internal threads of the not when unrolled and laid out on a flat plane. The view on the right side shows the external threads of the socket when unrolled and laid out on a flat plane. Note that the internal threads of the nut and the external threads of the socket show no evidence of severely deformed or fractured threads. The threads showed no evidence of scars from thread pullout. Had the force of the explosion, falling debris, or a building collapse caused the disconnect of the lower union assembly, there would be evidence of severe mechanical damage through the engaged threads. Based on the examination, NTSB staff believed that the socket of the lower mercury service regulator union most likely was not threaded to the nut and therefore the lower mercury service regulator was not connected to the vent line prior to the explosion. Malfunctions can and do occur in service regulators and lead to overpressure events. During an overpressure event gas will normally flow through the vent line and terminate outside of the building where it dissipates. However, investigators found that a union assembly for the lower regulator was not connected prior to the accident. NTSB staff believe that the mercury service regulator, past history, the pre-existing unconnected vent line union, the lack of evidence of any pre-existing anomalies in the gas piping or gas appliances indicate that a failed mercury service regulator was the most likely source of the natural gas release in the meter room at building 8701. The gas accumulated in the meter room to an explosive level and found an ignition source. This concludes my presentation. I will turn over the microphone to Miss Gunaratnam. Thank you. In the following presentation I will discuss the operations of mercury service regulators. mercury regulators in service today were mostly manufactured in the 1940s and ’50s. Many of these devices are still installed in residences that were built prior to 1968. In 2011, the U.S. Environmental Protection Agency issued a recommended practice to remove mercury containing devices such as mercury service regulators because of the toxic hazard that mercury poses to human health. mercury regulators are now replaced with a non-mercury service regular that uses a spring mechanism as its safety relief valve. Washington Gas has been replacing their mercury regulators over the years. As of December 2018, the company has approximately 125,000 mercury regulators still in service. While reviewing the Washington Gas records, staff identified errors in the company’s field entry forms which did not correctly identify the type of regulator in service. The company relied on unvalidated information to determine the location and condition of its mercury service regulators. Staff proposes safety recommendations to address this issue. In the Washington Gas operations and maintenance manual, the company’s procedures list the following conditions to replace a mercury regulator: When a mercury regulator has blown or weeps gas through its vent, when a regulator has been tested and malfunctions, when a gas system is upgraded to a higher maximum allowable operating pressure of 25 PSIG, when gas service is being replaced such as a meter change, when connected piping is altered or replaced in such a way that it makes it practical to change the regulator, or when the regulator is scheduled for removal. In the five years prior, Washington Gas replaced 992 regulators due to performance failures. The primary failures of mercury service regulators include gas venting or leaking from a body connection or from debris or blockage to the valve seat or mercury leaking out of the reservoir. Another failure mode Washington Gas has identified can be a torn diaphragm but that type of failure was not recorded for any regulators found in their system. Service records for the five years prior to the accident show that Washington Gas technicians have visited the meter room at building 8701 on six different occasions. Two of which were to turn on gas service. Washington Gas procedures require mercury service regulators to be tested each time a gas meter is set, reset or changed. Part of this procedure calls for the technician to a vent line test which involves disconnecting the vent line to ensure the line is clear of debris and will appropriately direct the gas to the atmosphere. Washington Gas technicians told investigators that this vent line inspection was generally performed on single-family homes and not on multi-family dwelling such as Flower Branch apartment complex. They only perform visual tests for debris and the outdoor venting of the lines. They did not check whether the lines were connected properly. Due to the potential of natural gas leaking from a disconnected vent line combined with a failed service regulator, staff proposes recommendations to address this safety issue. Because both mercury service regulators from building 8701 were significantly damaged from the explosion and fire, they could not be tested to identify what type of failure occurred. Staff participated in field testing at a Washington Gas facility simulating certain failure scenarios using similar models of mercury service regulators. Those failures included a diaphragm tear, obstruction of a valve seat that would cause venting and leaking, and low mercury levels in the reservoir cup. Staff determined gas flow rates for these failed regulator scenarios. The last person to leave the meter room was at 8:42 p.m. No reports of gas odors by residents or the maintenance crew occurred before this time. The explosion happened about 11:51 p.m., leaving a three hour time frame for gas to accumulate. Using the flow rate data from the regulator field testing, investigators estimated the amount of time needed for gas venting from a malfunctioning service regulator to accumulate to explosive levels inside the meter room of building 8701. Using a computational fluid dynamic model, the lowest flow rate identified could release a sufficient amount of gas in the meter room to create an explosive atmosphere within the established three hour time frame. PHMSA regulations allow service regulators to be installed inside a building as long as they are in a ventilated area and at least three feet away from an ignition source. Each service regulator must be installed in a readily accessible location and near the service line entrance. And service regulator vents and relief vents must terminate outdoors. The service regulators installed at building 8701 were connected to a vent pipe that vented outside of the building. However, staff found evidence that a vent line union was likely not connected prior to the explosion and with a failed indoor service regulator, this allowed explosive concentrations of natural gas to accumulate inside the building. Had the service regulators been located outside the building, gas from a failed service regular would have vented safely to the atmosphere. Staff proposes recommendations to address this safety issue. My next presentation will be on natural gas odor detection and notification. About two weeks before the accident, an odor call was reported to the Montgomery County Fire and Rescue. On July 25th, the same resident who smelled gas on August 10th called 911 to report a gas odor at building 8701. The fire department arrived on scene and approached the building using gas monitors. They continued to check inside the building monitoring for gas at each floor, within occupied apartments, and underneath the doors of unoccupied apartments and detected no gas. Firefighters attempted to enter the meter room with a key they would normally have access to however they were unable to enter the room because the locks on the door was recently changed without the replacement of a new key. Firefighters did wedge open the top of the door to the meter room but did not detect gas. Firefighters left the building and did not notify Washington Gas of the odor call. Staff found that Montgomery County Fire and Rescue procedures for responding to a gas incident only requires that the gas company be called when a leak is confirmed. The 911 operator who received the call told the resident on the phone to evacuate himself and others from the building and not to turn on anything that would create an ignition source. Neither the resident nor the 911 operator notified the gas company. 911 operator protocols for gas incidents do not require the gas company to be notified whenever a member of the public calls in a gas odor complaint. Staff believes that whenever a gas odor complaint is called in to 911 the gas company should be notified because they are trained and equipped to detect natural gas at lower concentrations and remedy any leaks. Had they been notified of the July 25th odor complaint, Washington Gas would have had the opportunity to investigate further and potentially prevent the accident. Staff propose a recommendation to address this safety issue. A year before the August 10th accident, the property management company for the apartment complex received six gas odor calls from residents of building 8701. Maintenance staff used their gas detection equipment to respond to these calls and did not detect any gas in or around the apartments that reported the odors. Washington Gas has no record of these or any odor calls from building 8701 in the five years prior to the accident. PHMSA regulations require gas companies to have a public awareness plan to keep the public and local responders aware of pipeline operations and safety issues. This plan includes procedures for reporting gas releases. Washington Gas did have a public awareness plan in place at the time of the accident which included mailing information to its customers in their monthly billing statements on how to report a gas odor. They also have guidance on their website directing individuals to first call 911 and then Washington Gas. The company maintains a gas operations dispatch center that sends the closest service technician when leaks are reported. PHMSA regulations require that natural gas and distribution lines must contain sufficient odorant such that gas can be detected at 20% of its lower explosive limit. The state of Maryland also adopted a more stringent odorant requirement to detect natural gas at a lower threshold of 10% of its lower explosive limit. The purpose of odorant is to act as an early warning of a natural gas release so that steps can be taken to stop the release and prevent an explosion and fire. However, the NTSB has investigated previous pipeline accidents where gas odorant does not always provide sufficient warning to people about hazardous conditions before a fire explosion. This is due to a number of factors such as odorant fade which is a condition where the odorant gets stripped from gas, preventing people from smelling it or individuals were not within close proximity of the odor having been in a different part of the building or they were not aware the odor was gas or they didn’t understand they were to report the gas odor. Staff believes the use of gas odorants alone does not sufficiently mitigate the risk of death and injuries caused by gas system leaks in some cases such as the one that occurred at the Flower Branch apartment complex. The seven residents who were killed in the explosion and fire on August 10th were located in apartments closest to the meter room. Had there been an alarm to warn of a natural gas release, residents could have been notified to evacuate to a safe place away from the building without relying on someone to smell gas odors. Currently there are no federal or state laws in effect to require alarms to detect natural gas detectors that alarm in residential buildings. Natural gas consists primarily of methane gas. Currently the natural gas industry has been researching available technology that would trigger an alarm for flammable gas such as methane when detected inside a building. The Operations Technology Development, which is a not-for-profit corporation that involves a partnership of natural gas distribution companies, initiated the residential methane detectors program. The program researched commercially available natural gas detectors and is currently in the testing phase of this technology. The OTD is advocating for a lower detection limit of 10% of the lower explosive limit to provide earlier warning of a gas leak. Currently there is one product standard, UL 1484 residential gas detectors, that covers electrically operated devices that detect both propane and methane in residential occupancies. This standard is published by UL and provides guidance on the design and manufacturing of the alarm detector and permits alarms to trigger at 25% of the lower explosive limit. The OTD has proposed lowering the detection limit to UL which they have not accepted. Currently the National Fire Protection Association and International Code Council issue national gas standards that have been adopted in all 50 states. These standards provide minimum safety requirements for the design and installation of fuel gas piping systems in homes and other residential buildings but the codes do not include any provisions for methane gas detectors. Some states and city cities are considering requiring methane detection alarms such as the state of New York and New York City who have had their own natural gas incidents. However, they are waiting for adequate industry guidance for this technology before implementing a requirement. Staff believes the development of a national standard covering the performance, installation, maintenance, and testing of methane detectors is needed. Then jurisdictions can require this technology to be installed in residential and commercial buildings. This technology would be a redundant means of leak detection to supplement the use of gas odorants to help identify gas leaks and save lives. This concludes my presentation and staff is prepared to answer questions at this time. We really want to thank you for those presentations as well as a very thorough and comprehensive investigation. We’ll begin now with questions from the board Vice Chairman Landsberg. Thank you, Mr. Chairman. In one of the party submissions there was a second potential gas source cited. I believe it was joint at the hot water heater. How did you come to discount that? I take that. We examined the steel pipe that was connected to the water heater and that pipe had a full fracture and I had read Washington Gas’s submission and one of the things that were noted that there were air-conditioning units placed next to the water heater and there was some suggestion that the air-conditioning units had leaned over and had fallen on the gas pipe and they believed that caused a full fracture separation of the pipe. My thought on that is that the air conditioning leaning over would have bent a pipe not caused any, not caused a full separation of the pipe. I think what what makes more sense is that if you imagine the fire in the building when it eventually starts consuming the floor structure in those first and second floor a lot of the contents from the apartment would be falling through the floor structure. These are heavy items. They could be refrigerators or a couch and falling from the first or second floor they would have a great deal of energy. A lot more energy that I think when they impact the basement they would caused a full separation of that pipe. So, my thought is that the fracture of that pipe was, occurred during the fire and not prior to the fire. I think we also were discussing the fact that nobody had complained that the water heater had been out. Nobody complained about that there was lack of hot water in the hours prior to the accident. So that suggests that the water heater was working fine. Thank you. Is it possible, we’ve had several reports of gas around this meter room on multiple occasions. What makes it possible for, if there’s a leak and subsequent odor to fluctuate? Why would some people notice it, some people not, the fact that it might be more present at some times and other times not as much? Odorant concentration and an individual’s ability to smell it, is just that, individual. Some people cannot smell odorant. It could be that winds were diluting it so it was below the detectable limit for a person’s nose. So there’s a lot of reasons why that odor could have come and gone. Is it also possible that based on the demand, so if you have a small leak, based on the demand from the water heater that that would also cause a fluctuation in the amount of gas that was released? Absolutely Okay, thank you. I think I’ll defer to the next round. Thank you, sir. Thank you, Vice Chairman. Member Weener. Thank you, Mr. Chairman. Determination of the explosion was the meter room. How was that determined that that was where the explosion occurred? The ATF who came at the request of the Montgomery County, they determined based off of the blast debris and the way that the explosion impacted the building, they determined that it occurred in the basement of building 8701 and that the source was natural gas. Okay and that was because of the blast pattern? Yes, that was the blast pattern, the way the doors were pushed out in the building and how far they they went across into the parking lot. So that would have been different had it been occurring, had it occurred on the first floor? Yes. So they also evaluated the basement floor which is also the first floor, ground floor and found that it had flipped over and not fallen in. So the fact that it flipped over showed that the blast had originated in the basement. Thank you. It was mentioned a couple of times that the same resident smelled gas about two weeks before the explosion. What happened in the meantime? In other words presumably there was a meter or a regulator failure two weeks before and then the failure went away for two weeks and then reemerged? Do you wanna answer that? If you look at the postulated failures of the regulator, if we had a piece of debris that was holding the orifice open, the demand on the system when you have high demand, system pressure would remain in a normal range. When you had low demand your system pressure will increase and the relief device within the regulator would then allow venting of the excess pressure gas when you have that condition where the orifice is stuck open. Was there a pattern to the reports of smell of gas? In other words was there preponderance of reports in the evening for example as opposed to the morning? On August 10th or on July 25th? Well I think just in a kind of general case because as Mr. Hall was saying there was some probably relief of pressure because of use of the gas so there wouldn’t be a buildup but that would be dependent, that would show cyclical patterns as the day progressed– Yes.
in 24-hour period. Yes, it would. It would– We only had two.
In other words, were there more reports of gas in the morning, evening, afternoon? Yeah, we only had the one on July 25th and there were no other odor complaints that were found in between that two week period. Okay what are the failure modes of the regulator? Especially what failure modes would cause hissing? Well, there are basically three failure modes that we have in the regulator. One is insufficient mercury in the mercury cup and also I you have a tear– Which would cause venting at a lower pressure, is that correct?
Yes. The amount of mercury you have in the cup determines the setpoint for the regulator and that would determine when you would have venting and if you don’t have sufficient mercury in the cup that would allow the unit to vent. Then the other second possibility is if you have a tear in the flexible diaphragm inside the regulator that could cause severe venting. That’s one of the highest venting rates that would occur that would be audible and then of course if you have material stuck between the orifice and the seat or if you have corrosion or degradation of the orifice you could have venting of the regulator. So, those are several of the malfunctions that can occur within or degradation occur within a regulator that lead to venting. And the hissing sound would come from the vent orifice? Well the gas flowing through the regulator and eventually through the pipe has an effect on the sound and the rate of gas flowing out. So, if you have a small leak, a little bit of corrosion or something stuck between the orifice and valve seat that would cause a low amount of hissing which and on the extreme side if you have a fracture or cut in the diaphragm that would be the most severe type of venting which would clearly be heard. Thank you. Thank you, Member Weener. Member Homendy. Thank you. And normally I would jump right into questions but I actually have a few comments first. And I’ll save my questions for the other rounds. First I want to express my appreciation to the director of our Office of Railroads, Pipelines and Hazardous Materials, Rob Hall, and to the investigators and the materials specialists, fire and explosion experts, and others who work so diligently on this investigation. Secondly, I wanna thank the parties to the investigation for their work including the Montgomery County Fire and Rescue Service and the Montgomery Police Department, County Police Department. This was a serious tragedy. Seven residents were killed. 65 others were transported to the hospital for treatment and three firefighters were injured. In the days and months that followed, Washington Gas should have been laser-focused on the safety deficiencies that led to this incident and what could have been done to prevent it from occurring. On June 17th, 2017 and again in various meetings with board members a few weeks ago, Washington Gas submitted comments to the board which stated that because the ATF concluded that the cause and origin of this accident was undetermined that the NTSB should also conclude that the cause and origin was undetermined. To provide some context, ATF arrived on the scene on August 11, 2016 the day after the explosion to conduct an origin and cause investigation. Their investigation focused on fire dynamics, fire modeling, and fire causation and concluded one week later on August 18th. The report was issued one month later on September 23rd, 2016. ATF did a great job. But the NTSB’s role and our investigative process is far different than ATF’s. It’s more in-depth. It’s more analytical and it’s focused on how a tragedy occurred and what safety measures should be taken to prevent it from happening again. Washington Gas stated in their comments and I will quote, “After gathering available information “and evidence including witness interviews, “making certain assumptions and vetting reasonable theories “as to cause and origin against the available evidence “and assumptions, the ATF concluded that the cause “and origin of this investigation was undetermined.” Washington Gas went on to suggest that NTSB should conclude the same. In fact that’s not what the ATF said. The origin and cause report states, “The three-part union connecting the regulators “to the vent pipe was disconnected prior to the explosion.” Prior to the explosion. “Thus by bypassing a safety feature of the system. “With the unknown circumstances involving the disconnection “of the vent pipe as well as the ongoing NTSB investigation “which will,” and this is a quote, “possibly provide more data useful “to this report, this explosion is to be classified “at this time as undetermined.” Washington Gas continued in their comments to remind the NTSB and again I’m quoting, “Mere suspicion, inference, “and conjecture must not suffice.” Using testimony from another horrific tragedy to seemingly warn NTSB against giving incorrect information or the wrong answer. According to Washington Gas, “The concern of a wrong answer “in any investigation of this nature is “that it may lead to wrong recommendations. “Safety recommendations based on wrong conclusions “not only fail to meet a central purpose “of the NTSB investigation but may actually serve “to alter or divert needed attention on safety improvements “and initiatives that are demonstrated “to improve safety and reduce accidents.” I am astounded. I may be new to the board but I have worked with board members and staffs for many, many years. Never once have I questioned the board’s focus on much needed safety improvements and I’m not going to now. It’s worth mentioning that the fire and explosive investigator and bomb technician who was selected as a party member to this investigation also submitted comments to our docket system which state, “At times during the investigation, “Washington Gas treated meetings and interviews “more like a courtroom cross-examination “than a joint investigation geared “at finding the cause of a tragic accident. “I feel that Washington Gas overlooked testimony provided “to the NTSB during field interviews “and questioned the validity of the facts. “In numerous emails, over several months, “Washington Gas questioned the factual content “of the ATF report and the ongoing investigative efforts. “It was disheartening to think “that party members would take pieces of information “from NTSB meetings then turn around and attempt to cast out “or insinuate criminal nexus to diminish culpability.” He goes on to thank the NTSB investigators for their professionalism and methodical and scientific approach to the accident which I couldn’t agree more with. I strongly suggest that Washington Gas refocus its attention on actions they could have taken to prevent this accident from occurring rather than spend time telling us how to conduct our safety investigation. Don’t misunderstand me I strongly support the party process and allowing party members to submit comments to the docket. But this was, in my view, an extremely disappointing use of the party system. Thank you, I’m sorry, I’m a little bit over my time but I will save my questions for the next round. Member Homendy, thank you very much. Mr. Hall, can you explain why the ATF was involved in this investigation? How that worked? What was their role? The ATF was brought in at the request of the Montgomery County Fire and that’s a fairly common thing to have ATF assist fire departments in determining cause and origin. But part of their primary focus is to determine was there any criminal act that resulted in this explosion. And they took the investigation to the point where they determined that there wasn’t a criminal act. That the explosion occurred in the meter room in the basement and that natural gas was the fuel source. And at that point they really left it to us to further the investigation to determine the probable cause. Thank you. Now, the NTSB, we arrived on scene the morning after on the morning of August the 11th and I believe we could not immediately determine whether this was in our jurisdiction or someone else’s jurisdiction and so can you discuss that? At what point we did actually determine it was within our jurisdiction? Certainly, and this is an issue that we face in all our investigations of distribution systems. By definition transportation of gas and we investigate transportation accidents, the transportation ends at the meter and beyond the meter it’s the consumer’s consumption of gas. And so when we investigate an accident particularly these natural gas distribution accidents, part of our first steps are to determine where the leak occurred. And so, we arrived with an unknown situation and had to determine where the leak occurred before we took it as an official NTSB investigation. Initially we didn’t have that information. The building had collapsed into the meter room. The testing outside of the building did not find a leak in the service pipe outside of the building. So we didn’t have sufficient information. At the point in which ATF identified the disconnected union, we realized there might be a possibility that it occurred before the meter so at that point we began a full investigation. Great, thank you very much and finally I do want to ask you about the length of the investigation and I’ve always said, the investigation has to take as long as it takes. And can you talk about that, the complications that our investigators may have faced which would have protracted the the length of time? Yes, sir. This, in my history, my 40 year career, has been one of the most challenging investigations that I’ve ever faced and my team has faced. In this event we had the explosion and the collapse of the building into the meter room which really compromised evidence that would have led to a quick resolution. Instead, what we had to do is take the available evidence and then start postulating the various types of failures and testing each one of those types of failures against the evidence. So, it became a process of elimination to find what we were left with as the most likely cause of the gas explosion and unfortunately that took a great deal of time, a great deal of work by our research and engineering people in their investigation. Additionally, we were challenged by several retirements through the course of the investigation in which case we had to replace staff members working on this. Thank you very much, including the investigator-in-charge who retired about a year ago. So, Miss Gunaratnam, I want to thank you for picking up the mantle and continuing the investigation. And also I want to echo Member Homendy’s points about thanking the first responders. I know we have the Montgomery County Fire and Rescue Service who are here today and we want to thank you and as well as the Montgomery County Police Department for their efforts. Next round of questions, Vice Chairman. Thank you, Mr. Chairman. On July 25th, the fire department responded to the, and could not gain access to the meter room. So there was a report by the same resident that gas had been smelled and and they were unable to detect the gas I guess for the reasons that we’ve previously discussed. And the reason that they could not gain access, as I understand it, was that the locks were changed on the room and the fire department did not notify Kay Management that they had responded and secondly that they needed access to that room through whatever means. Also as we’ve also pointed out, Washington Gas was never notified of the leaks. I’m curious if the fire department, if we’re aware, has any protocol for this kind of breakdown in communications, kind of a two-step situation? One of if you’re required to have access to the gas service area and for whatever reason you can’t, how does that get communicated and remedied because that might have addressed the issue before it ever happened? And then I think we’ve already talked about the importance of notifying the gas company even if you just suspect that something is not right. Yes. The fire departments are supposed to have access to an apartment building, to their rooms that are locked through the availability of a key in a Knox Box. And so they usually have access to that and that’s required in the Montgomery County Fire Safety Code and so I’m not aware of any new protocols of communication between the fire department and the apartment of Flower Branch complex but it is required by law and so the fire department can enforce and cite the apartment building for not updating their keys because they are supposed, the Flower Branch apartment complex is supposed to be updating the keys whenever they change the locks and do so promptly for the very reasons why on July 25th. But that didn’t happen. No, hat didn’t happen, no. This may seem rhetorical but I’m not sure do we have any idea when the last maintenance was performed in the vicinity of the regulators? Any postulation as to when the union might have been either A, disconnected or B, in the inspection process been able to say, gee, we need to tighten this up? We examined the union assembly for any evidence that might suggest a time– Mr. Zakar, if you would kindly pull that microphone very close please, thank you. Yeah we examined the union assembly to determine if we could specify a time frame when it separated and we cannot come up with a time frame so it could have been disconnected at any time. We don’t have any evidence of a separation time. Okay, thank you. No further questions, Mr. Chairman. Thank you, Vice Chairman. Member Weener. Just to follow on the Vice Chairman’s line of questioning, so the fire department arrived, found the door locked, no key available, they pried a corner of the door open and how reliable would have been a gas measurement at that point? It depends on the concentration of the gas and where it was. I mean, they couldn’t get very accurate, reliable detection there because they’re not close enough and it depends on the instrument they’re working with, the gas detector. So there’s a number of factors here that make a difference. So having access to that room would have been important so that they would have been able to detect any leaks. They would need to get closer. Is there some policy of protocol with regard to breaking a door down to gain access? I’m not aware of the protocol but I know that fire departments have broken down doors. If they had probably detected a fire, I’m sure they would have but there was no fire ongoing on July 25th. It was just an odor complaint. And since their detectors didn’t detect anything they had no reason to block the door down. You mentioned detectors. What is the status of the technology of methane detection? Right now, the OTD that I had mentioned before they have testing commercially available methane detectors that are available and it has grown over the years and now it’s in its more reliable phase so that the detectors won’t get interfered by other household chemicals that could be in people’s homes. And so right now they’re in the middle of testing it and they’ve narrowed it down to certain devices that are much more reliable now. How close are we to having a methane detector like a radon detector or carbon monoxide detector? They are available. But the testing is still ongoing and I expect that it’s probably going to be done this year sometime. So lemme get back to the utility room, the meter room. It had an intake screen or an intake hole of 66 square inches, right? We don’t know exactly how large the window was in the meter room. It was intended to provide combustion air for the hot water heater. 66 square inches would be the minimum it would have to be for that size water heater. But what it actually was in reality, we don’t know. So there was inflow into the meter room but not necessarily outflow? The outflow would have been part of the combustion process? Depending on the function of the hot water heater, that vent could be providing air into the meter room or providing for an escape path for air or gas to exit the meter room. It really depends on what’s going on at the time. If the hot water heater is working at the time, then air would be coming into the meter room through that vent and burning in the hot water heater and exiting through its exhaust system. If the hot water heater is off at the time and depending on other environmental factors, the vent could be allowing gas to exit the meter room. So the methane is lighter than air, so it would have accumulated at the ceiling? Yes, initially upon release it would start to accumulate at the ceiling and then form a gradient from the ceiling down towards the floor. Eventually it would mix but initially there would be some separation. Do we know where the air intake exhaust is located relative to the ceiling? In other words, would it have flowed out of the meter room? Yes, the window providing ventilation into the meter room is up high near the ceiling in the meter room therefore when methane gas would accumulate at the ceiling you would expect it to exit through that window to some degree. Thank you, I will yield. Thank you Member Weener. Member Homendy. Thank you. The draft report states that as of December 2018 Washington Gas has replaced about 175,000 mercury service regulators throughout its system and has about 125,000 regulators still in service. Some of these were replaced due to venting or a leaking body connection. What concerns me is whether the 125,000 figure is accurate. During questioning Washington Gas stated, “We are unable to provide exact numbers “of remaining mercury regulators “due to multiple changes in the means “for which the mercury regulator installation, maintenance, “and replacement data was captured over the years.” Can you talk about this a bit? How does Washington Gas know that there are only 125,000 left to be replaced? So when we. (microphone handling noise) What would a board meeting be without audio problems? (laughs) Thank you. We did inquire with Washington Gas about how they came up with 125,000 and it was an estimate based off of what they installed in the ’40s and ’50s and subtracted that from the ones they’ve replaced. And so, that’s how they came up with the number. They do have the ability to survey as technicians go out and identify what the regulator is and then but it is just, the 125,000 is an approximate number, it’s just an estimate, their best estimate based off of installation back in the ’40s and ’50s. And this regulator though, and I don’t know if it’s both regulators or just the the lower one, was designated by Washington Gas in their records as a non-mercury regulator, correct? That’s correct, yeah. So, they have a field data form that when a technician goes out, they have to fill it out to provide an update, sort of on what they’re looking at or what is there?
Yeah, yes. The form does mark on whether they have tested it and whether it’s a mercury regulator and they’re supposed to check that off. And do we know of the 175,000 that were replaced, the numbers of those that were inside structures and relocated outside, do we know? That I don’t know but I do know that they have started relocating regulators outdoors like for the Flower Branch complex they have relocated some already outdoors and they plan to do the whole complex in phases. Okay, so that was my next question. So the whole complex has not been redone yet? No, not the whole complex but it is planned. Okay. Do we have any concerns about their replacement at all? Because I know in their submission they stated they will install, whenever possible, the gas regulators exterior to the building on new multifamily construction projects and then for legacy multimeter sets Washington Gas said that when undertaking the replacement of a service line due to age and conditions, where possible, they will move the regulators that are installed within a building structure to a location outside the building. So my concern is, I mean, sometimes it’s obviously not practicable because of some sort of construction but sometimes they decide it’s cost. Right. And we’re hoping that, staff did propose recommendations to PHMSA to relocate those regulators outdoors so we hope that can be in place so that that can expedite that process and operators all around the country will relocate their indoor regulators outdoors. And we hope that the states will follow up on that, the state utility commissions. Well one question that I have on that, this my last question for this round. I mean, PHMSA, I love PHMSA but they are in a rock and a hard place right now for getting rule making done. They are trying to get other rule makings done which were mandated in 2011 and it’s now nine years later. And so my concern is they don’t move forward on that and meanwhile we don’t have a recommendation that specifically says that Washington Gas needs to move these on the outside, located on the outside of the structures. So I’m just wondering, I know we have a recommendation on establishing a time frame for replacement and prioritizing multifamily residences and prioritizing those that are located on the inside to be relocated on the outside. I’m just wondering if we should have the same mandate that same recommendation that we make to PHMSA to Washington Gas which is when you do a replacement move it on the outside if it’s inside? Yeah, I believe staff would support that recommendation that as they are replacing their mercury regulators to move all their indoor regulators outdoors. And have the Maryland Public Service Commission follow up on that. Thank you. And thank you very much, Member Homendy. So, Miss Covetti, if you could pull up slide 21 for us please. And help me to understand what it is that we’re looking at here, Mr. Zakar. The threads here. This is the upper service regulator and the threads are damaged. And how do you explain that those threads were damaged? Well, testing okay. What we’re looking at is the upper service regulator and what we see here is a thread pull out. If you notice the union is connected but we had separation between the external threaded nipple that extends from the union to the outlet port of the lower regulator. This is an excellent example of thread pull out. We’re fortunate to have this. It shows exactly what happens when you have thread pull out. What we see on the upper photograph, on the upper right, is the external threaded nipple that had pulled out of the outlet port of the lower regulator. We see that the first four threads are fully deformed. They show severe deformation. And the corresponding threads on the bottom show thread pull out or thread stripping. Good. So the bottom one is we can tell that this was, as you said, thread pull out–
Yes. Which came under force which was more than likely a result of the explosion. That’s correct. Okay, let’s go to slide 23. This is the lower regulator union assembly going to the vent line. We don’t see that evidence in this case is that correct? That’s correct. And why not? You saw it on one, why wouldn’t you see it on the other? What we try to prove is whether or not there was thread pull out. Had there been thread pull out we would have seen severely deformed threads or scar from thread pull out or thread stripping. That indicates, the fact that we don’t have any of this on the threads indicates that these parts were not connected at the time of the accident. Okay, thank you. Now, yesterday when I went to the lab, we talked about some of these union assemblies and there was one that you could actually insert for a couple of millimeters before it would mate and then you could thread it. Can you talk about that? Yes, after we completed our physical examination, we decide to do one more experiment or exercise. We decide to take the lower regulator, the union for the lower regulator, that’s the one we had found unconnected, and we decide to put it back together. We attached the nut to the socket and we found that we were able to engage the threads of the union. We were actually able to put it together and once we completed that task, we tried to pull it apart by force, by hand, and we could not pull it apart. That indicates that the union assembly had the capacity to hold together. Now had that union assembly been forced apart, we would have seen a scar from thread pull out or stripping. We did not see that. That tells us again that we had, the union was not connected at a time of the accident. So the bottom line is we could tell that the upper of vent line was connected and the lower vent line, based on this physical evidence that we’re looking at here on slide 23, that it was not connected. That’s correct.
Okay, thanks. We’re through with those slides. Now Washington Gas said in their submission, on page 40, they talked about maybe this being some sort of a simultaneous failure where the meter would have had a failure at the same time that this vent line might have disconnected but does staff believe that it was simultaneous or does staff believe that this vent line may have been disconnected and remained disconnected for a long time? As we previously said there was no physical evidence to tell us when that union became disconnected. It could have been disconnected from the time it was installed. We don’t know. So, there is no way to say that both of these things happen simultaneously. Bottom line is it probably was not a simultaneous failure. If it was simultaneous failure, we would have had thread strip out on the lower, on the lower fitting. So it was not a simultaneous failure. This was a latent condition that was in place for some period of time at least on the vent line and then the active failure was some failure apparently in the meter that led to the venting–
The regulator. I’m sorry, of the regulator itself is that correct? Yes, that’s correct. And one last question, who has the responsibility to maintain the vent line? Washington Gas. Thank you. We’ll take a break. We’ll come back at 11:05. We’re in recess. Now that works. And now it works so. Well. We’ve lost the signal. (laughs) Now it’s. (microphone handling noise) (people chattering) (people chattering) (people chattering) (people chattering) We’re going to start in two minutes. Okay, we’re back in session. We’ll begin with a third round of questions and Vice Chairman Landsberg. Thank you, Mr. Chairman. Roughly how many regulators were in the meter service room? In building 8701?
Yes, correct. Two regulators and 15 meters. Okay, and how many of them, so there were only two regulators and they cover for the entire building?
That building. Okay. And I think it’s been pretty, the Chairman very nicely summed it up, and Mr. Zakar pointed out that the failure signature was pretty conclusive regarding where and how the separation occurred. Am I understanding that correctly? Yes. Thank you, no further questions. Thank you, Chairman. Thank you very much. Member Weener. I’d like to talk about or discuss the notification process of a gas smell. Presumably the only party that can actually do something about a gas smell is Washington Gas in this case. Is that correct? With detection of odor, of the gas, I mean? With rectification of a gas leak.
Yeah, yes. So, that’s Washington National Gas just– Yes, they are the Washington Gas, the gas company is usually the subject matter experts on this, yes. Now if we look at the process from somebody smelling gas to some fix being made, there are lots of people involved in that. There’s 911, there’s the fire department, there’s the management. What is that process? For reporting an odor? I mean, it varies depending on who they reach out to first. Washington Gas procedures require that you should call 911, evacuate, and then call them. You want to get to a safe place first. You don’t want to stay inside the building when you’re smelling the odor. So for the consumer the process is to call both 911, and call Washington National Gas, Washington Gas. Is that correct?
Yes. It seems like there’s a lot of decision makers in that process because the gas was smelled several times but Washington Gas was never informed of it. Is that correct? Yes and normally the fire department would call them if they had confirmed a leak but in this case they didn’t confirm a leak so they didn’t call them. So there are a bunch of decision makers in this process that can stop it.
Yes. Yes, there’s a number people, yes. So what would be the ideal process? The ideal process is that the fire department and the gas company coordinate together and respond efficiently and effectively to remedy any leaks together so that all occupants in the building are safe and they’re gonna need to work together to do that but they need to do that together. So they need to be notified. Washington Gas needs to be notified and also the fire department. The order of that, we leave it up to the jurisdiction so that they can work together on that aspect but they do need to be notified. And what’s the role of property management in this case? Well Kay Management is not legally responsible to notify, I mean if they smell gas, yes and any consumer building owner or so forth, should report a gas odor to 911, and the Washington Gas, or to the gas company so they are also another player in this but they’re more on the reporting aspect. Okay so it’s a very important point that the best protocol is to call 911 followed by calling the gas company. Yes, evacuating as well, not staying inside the building, yes, yeah. Very good, thank you. No further questions. Thank you, Member Weener. Member Homendy. The addition of an odorant to natural gas distribution has been in regulation since 1970 and the odor is often used as a line of first defense when it comes to discovering a leak in a distribution system. Our report states since the regulations were issued, there has been concerned with odor fade. Can you talk about this concern a little bit more? Yeah, so under certain environmental conditions, odorant can get stripped and it can get stripped in the pipeline or it can get stripped in the soil if it’s migrated out. So it depends on the region of the country, depends on the type of pipe, the material. There’s a lot of different factors but it can happen so but it doesn’t always happen, it depends on the conditions. Have we seen this in other accidents? Yes.
Yes. We have seen it in, which is in the report, in appendix where we identified it in a number of cases before and an ongoing investigation that we’re doing right now in Dallas, Texas. Great. And we’ve made recommendations before on methane leak detection? Yes, we made one back in the ’70s out of a building explosion in New York City however the technology was not around. And then we made it again in ’92 in Allentown, Pennsylvania that was another residential occupancy that exploded and I think the the technology, it was starting but it wasn’t yet finely tuned yet. But it is getting there or is there
Yes. There are some that are available, commercially available now? Yes.
I know we discussed it earlier but– There’s a variety that are available right now. The question is are they reliable? And so they, the OTD is determining and they’ve been testing it so that it can report an odor leak ’cause you don’t want false alarms. False alarms, people will ignore them or they won’t take them seriously. We want them to be act on them. You don’t want false alarms but at the same time you want to make sure, some people can’t smell the mercaptan or the natural gas and other people, and if there’s odor fade, then you really need a methane
Yes. detector to kind of fill that gap. As an additional safeguard, yes. Great. Okay, thank you very much. Thank you, Member Homendy. Mr. Zakar, as I understand it, there’s three basic failure modes that can cause a venting situation in a mercury service regulator. One would be the malfunctioning valve seat orifice seal, one would be the perforated diaphragm, and one would be a insufficient mercury. So do we have any evidence? What do we believe would be the failure mode of this particular mercury service regulator? We had fire damage that had destroyed a lot of the evidence for the lower regulator and upper regulator. So we basically have to go through the logic tree diagram to rule out all the possibilities or bring into evidence what we have. We have visual evidence that the lower union, the union for the lower regulator was separated. That’s the first finding. That’s something to go on and then the second thing we had to determine was whether or not we had forceful pull out of the threads and after examining– Yeah, now we’ve already talked about that.
Yes. So, let’s talk about the regulator itself. Yes. All right, so what failure mode do we think occurred in this regulator? We don’t know. It’s because we had a fire damaged and we can’t tell. And what makes us then think that this was in fact a failure of the regulator? The process of elimination. We were talking about the water heater before, why we don’t think it’s the water heater and it’s process of elimination. Our investigators did not find any evidence of any other disconnected or leaking pipes and so it’s process of elimination and the fact that regulators do fail and once we have a open union that possibility of gas leak that goes out of the union will eventually wind up in the basement. So it’s process of elimination. Do we have any evidence? We’ve got a recommendation, proposed recommendation that for Washington Gas to come up with a replacement schedule and recognize the need to expedite the program to prioritize replacement of mercury service regulators. That’s pretty loosely summarized there but do we have any evidence that mercury service regulators have a higher failure rate then other regulators. No we don’t have any failure data rate. We do know that they’ve had failures. Washington Gas did provide us data that of their own mercury regulators, they’ve had almost a thousand that have failed due to those three, one of those three failures that we saw up there. So they do fail and they can cause a lot of damage. But if we replace them with another type of regulator is there also a failure problem, potential failure rates with with those as well? Well that would be newer, a newer device and so mercury regulators are outdated. I mean, they were last installed in the ’50s. I mean, they’re going on 60, 70 years now and so the internal components will wear down. They will break down. The service regulator, the mercury service regulators have a long history so eventually there’s a possibility that they will fail. Most of the new regulator is going to service are fairly new compared to the mercury regulators so I don’t think we have the full numbers on how many are failing but certainly they’re not as old as the mercury regulators. So just because something is old doesn’t necessarily mean that it’s going to fail but when we start talking 60 years old, I start worrying about that one But I can see that you would eliminate one of the failure modes for sure if you did not have a mercury service regulator. You would eliminate the possibility of lack of mercury. You might also eliminate the diaphragm problem because you have a newer diaphragm but Washington Gas did not find any evidence of the diaphragms wearing out in leaking anyway. So I’m just trying to determine if we’re placing, but the idea here is that if we replace them with newer ones at least we’re replacing them with something that is, that has less age and less wear and tear on it. Is that correct? Okay, good thank you. Is that right? Yes, that’s correct. Thank you, thank you. Let’s go to another round. One more question, I promise this is the last, I think. Take as many questions as you need. In all accidents it’s seldom just one thing. There always seems to be a chain of events that occurs and I want to go back to the fact that on July 25th, had the fire department had access to the meter room that was locked because the locks were changed and yeah, they did make an attempt to wedge the door open to get their gas sensor in but they weren’t able to get a good reading. This gets into the whole idea of what authorization that they have for code enforcement in a situation that could potentially be catastrophic and Mr. Hall, this might be a question for you as to how to ensure that if there’s a code violation that has the kind of potential that there is immediate access gained somehow by the authorities to be able to thoroughly check the room. Certainly in this case we had the issue of the keys not being updated. One of the things that the team looked at was and our recommendations that there be joint notification to both the fire department and Washington Gas. It’s the team’s position that had Washington Gas shown up on the scene with a locked door, they would have made, since they routinely have to go in there to access the meters, they would have made the appropriate contacts to get the door open with Kay Management. Well, that doesn’t answer the question though in terms of we seem to have a breakdown between the first responders, if it’s the fire department or if it’s a gas company. If it’s a gas company, that’s good. If it’s a fire department then are they empowered or how does this work in order to, bottom line is being able to get access to the meter room, period, and being able to make a thorough assessment and I’m sensing some variability here. Yeah, certainly Kay Management is required by law to have the keys updated and that is something that the fire marshal in doing fire safety inspections would check to make sure that the fire department could access all those areas that they could access. I mean, that’s fairly standard within a jurisdiction. Don’t know specifically if the fire department made any referrals in this case to the fire marshal about the non-compliance but certainly that’s what the expectation that I think we would have for fire departments to make sure that the fire marshal followed up on. I’m sensing a lack or urgency somewhere. If everything works as it’s supposed to, we don’t have the problem because then the fire department or the gas company has immediate access to the meter room but in the event that happened here, that wasn’t the case and in my observation we missed a very good opportunity to possibly have stopped this whole thing before it got to that point. Because of the lock change and the inability to access the room and do it at that time as opposed to saying well okay, we don’t smell any gas but somebody did and we have a code violation, we really do need to get into the room and we need to get into it now. And I’m trying to get to some point to say there are going to be breakdowns in the system. I don’t suspect this is the first time that this has happened and I’m wondering if there’s something that could be done that’s a little more proactive. Well, one thing in this case is when the fire department was there they didn’t detect any gas. And if you would take it to the step of okay, break down the door in the event there is gas, breaking down the door could be the ignition source. It’s not really the thing that you would want to do in entering a natural gas, potential natural gas release, to go in and bust down a door because of the potential to cause an ignition source. I’m not even suggesting that you have to go that far. I’m just saying somehow gain access to the room. And it seems like, in a perfect world, it would work the way it’s planned but the reality is that it doesn’t and I would kind of hope that we get to someplace where we could figure out to empower the fire department and the management company to work together on this. We do know one additional thing that Kay Management did have a 24 hour number for their maintenance and through the interviews we found out the fire department did try to call Kay Management and couldn’t get through. So much for 24 hour numbers, huh? Yes. Thank you, Member Weener. No further questions.
No further. Member Homendy. I just want to circle back on the regulators. They were replaced in the 1950s, I mean not replaced, installed in the 1950s correct? Yes, they were installed then. And Washington Gas is already in the process of replacing them because we do talk about 175,000 have already been replaced. They have 125,000 more to go. Yes. They started before this accident because of the environmental concerns. Okay and some of those are the environmental concerns, some were from lockout tests and some were from, Failures.
Failures and others were part of replacement of the line but there is already an ongoing process that they’re doing now. Yes, it’s not a formal process though so that’s why we proposed a recommendation to make it more formal. Okay, great, thank you. No further questions. Thank you. We have proposed recommendations to require moving existing or regulators that are inside to moving them outside and it sounds like a good idea because we want to get that possible venting, to get it outside instead of relying on a vent pipe that could potentially be not connected properly. But is there a potential unintended consequence of that? And Mr. Hall in the Minnehaha accident that we’re dealing with was that a situation that they were relocating the meters from inside to outside? Yes, it was. And was that all they were doing or were they doing that for other reasons as well? Why were they, I know we’re not here to talk about that accident but I’m just wondering about the potential unintended consequences of a recommendation like that. I didn’t come prepared to speak about Minnehaha. Well about the unintended consequences of having a recommendation where we are going to move them from inside to outside. Well with any any type of action there can always be unintended consequences. In this particular case, it’s not just the venting to make sure that any venting occurs outside. By moving the regulators outside of the building, you move the piping for all the higher pressure gas outside the building as well. You also put the regulators in a location where Washington Gas can access them, the fire department can access that area without the need of a key to get into the room. So, there are a lot of positive consequences to moving them outside and frankly their utilities prefer to have them installed outside primarily because of the access to their equipment. Thank you very much. And maybe Mr. Hall or maybe anyone else on the panel, I can’t quite figure this out. So I alluded in an earlier round of questioning that this disconnected vent pipe was a latent condition. It had probably been, it had been disconnected for a while, we don’t know how long. And we had previous reports of gas smell, gas odor. What was different about this day that led to an explosion? We have an opportunity where the gas regulator malfunctioned and the opportunity was there. We don’t know the exact malfunction or why the regulator malfunction based on the physical evidence but the opportunity was there. Venting does occur and we feel that it leaked out of the union assembly and filled the basement with gas but again we don’t have a exact reason why there’s a failure but failures do occur. There’s enough of them. So we feel that such circumstances could lead to gas leaking into the basement. Mr. Hall would you like to take a stab at that because again. we had the gas apparently in earlier situations. The gas was there which means that, I’m not worried about the regulator malfunction, I’m not worried about the vent pipe ’cause we know that supposedly that was already disconnected– Just one more, yeah. When you have a lack of mercury in the mercury cup, the situation can also worsen. So as time goes by the more venting events you have, eventually the venting process could be violent enough that it’ll push the mercury out of the cup that it’ll move into the inside of the cup and the venting process will just get worse and worse and the volume of gas also leaving will get, you’ll get much more gas flowing out. So it gets to a point where it’s a more stronger outflow. Okay, thank you. And thank you that does make sense. Mr. Hall. You have to have three conditions to end up with an event like this. You have to have some sort of hole in the container so the regulator failure becomes the hole in the container, you have to build up an explosive atmosphere in the room so the leaking through the union, to the point where it builds up an explosive atmosphere and then you have to have an ignition source. And you have to have all three of those on any given day at the same time to actually result in the explosion. Prior events you may not have made an explosive mixture or perhaps you had an explosive mixture for a period of time but you didn’t have an ignition source. So we really need all the conditions to exist simultaneously for the explosion and fire to occur. Wonderful and so it’s got to be between the lower LEL, the lower explosive level, and the upper explosive level which is just because you have gas and ignition doesn’t necessarily mean that you will have an explosion, if the formula is not exactly correct. [Rob] That’s correct. Thank you. Now I have a few more questions but I want to turn to my colleagues and make sure that I don’t. Thank you. Mr. Hall, do you know, can you tell me the phone number of the gas company that you would call if you were to find gas smell at your house. No sir because I don’t have gas service. Okay. (people laugh) Does anyone, yeah go ahead. If I was in a building and I smelled gas, I would call 911 regardless of whether I knew the gas company’s number or not. And that is exactly my point. Most of us do not know the number of the gas company but we all know 911 and we have seen accidents, in fact the one that I recall vividly because I was there was in East Harlem, New York. Calls were made to somebody but the gas company didn’t get properly notified. So it would seem to me that we want one-stop shopping. Call 911 and let 911 then be the one to deploy the proper assets. And I think we are calling for a recommendation related to that in calling for the IAED to revise their their protocols. Is that correct? Yes, sir. Okay well, with that I think I will cease questions. Anybody else have any further questions? Okay let’s move to the findings and so Miss Bryson, if you would kindly read us the proposed finding. Yes, sir. As a result of this investigation, staff proposes 17 findings. Number one, the post explosion responses by both Montgomery County Fire and Rescue Service and Washington Gas were prompt and adequate. Based on the pattern, number two, based on the pattern of debris and the location of the victims, the explosion radiated from the meter room where the gas meter assembly was located. Number three, the upper mercury service regulator was most likely connected to the vent line through the threaded nipple and union assembly prior to the explosion in building 8701 and the regulator likely became separated from the vent line due to forces of the explosion or from the building collapse. Number four, the external threaded socket of the lower mercury service regulator union most likely was not threaded to the nut and therefore the lower mercury service regulator was not connected to the vent line prior to the explosion. Number five, without a requirement that technicians verify the connection of vent lines for indoor service regulators such vent lines could inadvertently be left open following service work. Number six, the gas piping to the water heater was most likely separated at the control valve due to overstressed forces resulting from the explosion. Number seven, the water heater did not fail prior to the explosion and was not the source of the natural gas release in the meter room. Number eight, the audible hissing noises heard by a resident, the mercury service regulator failure history and the pre-existing unconnected vent line union and the lack of evidence of any pre-existing anomalies in the gas piping or gas appliances indicate that a failed mercury service regulator was the most likely source of the natural gas release in the meter room of building 8701. Number nine, a low leakage rate of venting natural gas from a mercury service regulator within an obstructed orifice and other regulator failure scenarios with higher leakage rates could have allowed the gas concentration to build up to an explosive atmosphere in the meter room between 8:42 p.m. and the time of the explosion at 11:51 p.m. Washington Gas, Number 10, Washington Gas relied on unvalidated information to determine the location and condition of mercury service regulators. Number 11, the failure of a service regulator combined with an unconnected vent line poses a significant threat to people and property with little warning. Number 12, had service regulators been located outside building 8701 the explosion would not have occurred because gas would have vented to the atmosphere and dissipated. Number 13, had Washington Gas been notified of the gas odor call on July 25th, a service technician may have had the opportunity to enter the meter room of building 8701, identify the unconnected vent line, and remedy the situation potentially preventing the gas release and explosion that occurred on August 10th. Number 14, the use of gas odorants alone does not effectively mitigate the risk of death and injuries caused by gas system leaks such as the undetected leak that occurred at the Flower Branch apartment complex. Number 15, had methane detectors been installed at the Flower Branch apartment complex, an alarm would have alerted residence to a gas release on either July 25, 2016 or August 10, 2016, reducing the potential and consequences of a natural gas explosion. Number 16, the development of a national code that establishes methane detection performance criteria and requires the technology to be installed in residential and commercial buildings with natural gas service would provider redundant means of leak detection to supplement the use of odorants. And number 17, the scope of National Fire Protection Association 54, International Fuel Gas Code, and their widespread adoption by local authorities appear to be the most appropriate standards for requiring methane detector alarms with local jurisdictions enforcing their requirements and making them feasible. Miss Bryson, thank you very much. Do any of my colleagues have any proposed amendments to the findings. I have one that I’d like to offer and my motion is, and this has been handed out and it’s the same as what was emailed out yesterday afternoon. My motion is to amend finding eight to read as follows, the audible hissing noises heard by a resident comma, the evidence of past mercury service regulator failures comma, the pre-existing unconnected vent line union comma, and the lack of evidence of any pre-existing anomalies in the gas piping or gas appliances comma, indicates that a failed mercury service regulator was the most likely, was most likely, was the most likely source of the natural gas release in the meter room of building 8701. That is my motion. Is there a second? Second. It’s been moved and seconded. As far as discussion, the way that the current finding, draft finding is it talks about a mercury service regulator failure history and to me when I read that it implied that there’s this widespread history of mercury service regulator failures. Now, the reality is that of the 12,000 or so that Washington Gas has replaced only about a thousand of them have failed. And so that’s, the actual percentage is about 8.5%. So to me that did not imply a huge history. So I came in the office yesterday and saying what history are we talking about? But then in a meeting that I had yesterday afternoon, Paul Stencil pointed out well, that’s still a thousand regulators that have failed. There’s thousand opportunities for a regulator to fail. Don’t get hung up on the percentage, 8.5%, think more about a thousand failures. So with that, we have proposed to reword it to say the evidence of past mercury service regulator failures instead of using the word history. So that would be my justification. I’d love to hear comments from my colleagues as well as staff. We’ll start with the colleagues. Mr. Hall, what would staff feel– Staff has no objection to the changes. Thank you. Any further discussion? It’s been moved and seconded, to adopt finding number eight as amended. All in favor, please signal with a hand and say aye. Opposed, there are none. We have a unanimous adoption of finding number eight. Now let’s go back and adopt the findings. Overall, do we have a finding, do we have a motion to adopt the findings as we just amend it? So, moved. It’s been moved by the Vice Chairman Landsberg and seconded by Member Weener. Any discussion? Seeing none, all in favor of adopting the findings as amended please signal with a hand and say aye. Opposed, there are none. We have unanimous decision to adopt the findings as amended. Miss Bryson, if you’ll please read the proposed probable cause. Yes, sir. Staff proposes the following probable cause. The National Transportation Safety Board determines that the probable cause of the explosion in building 8701 of the Flower Branch apartment complex was the failure of an indoor mercury service regulator with an unconnected vent line that allowed natural gas into the meter room where it accumulated and ignited from an unknown ignition source. Contributing to the accident was the location of the mercury service regulators where leak detection by odor was not readily available. Thank you. Is there a motion to adopt the probable cause as proposed? I so move. It’s been moved by Member Homendy and seconded by? Second. By the Vice Chairman. Any discussion? When she read the probable cause, she added a word, was the failure of an indoor mercury ’cause that’s not in my text. (woman speaking faintly) In the executive summary I forgot to transfer in a word. Okay, so can you key the mic, Miss Gina and please say that. It was discussed to revise and add an indoor. I made the change in the revision in the executive summary however multitasking I did not transfer it over properly into the probable cause section. Member Weener, thank you for catching that. So let’s do this again. Let’s read it exactly as staff is proposing and then we can see what we’ve got. So kindly read it one more time. Absolutely. Staff proposes the following probable cause, the National Transportation Safety Board determines that the probable cause of the explosion in building 8701 of the Flower Branch apartment complex was the failure of an indoor mercury service regulator with an unconnected vent line that allowed natural gas into the meter room where it accumulated and ignited from an unknown ignition source. Contributing to the accident was the location of the mercury service regulators where leak detection by odor was not readily available. All right, thanks. Now from a process point of view, we have a motion that’s been seconded to adopt what she originally read which is exactly what she read just then. Is that still you motion and that still your second? Thank you, so we’re good on the motion. Is there any discussion about the motion and the second? You’re okay? Yes, I’m fine with this. Okay, thank you. Again, thank you for catching that. Okay, given that we have a motion and a second and there’s no discussion. All in favor of adopting the probable cause as proposed, please signal with a hand and say aye. Opposed, there are none. The ayes have it, the motion passes unanimously. We have a probable cause. Miss Bryson, if you’d please read the proposed recommendations. Yes, sir. As a result of this investigation staff proposes 11 new safety recommendations. Two recommendations to the Pipeline and Hazardous Materials Safety Administration. Number one, require that all new service regulators be installed outside occupied structures. Number 2, require existing interior service regulators be relocated outside occupied structures whenever the gas service line, meter, or regulator is replaced. In addition, multifamily structures should be prioritized over single-family dwellings. Two recommendations to the Public Service Commission of Maryland, the Commonwealth of Virginia State Corporation – Commission Division of Public Utility Regulation and the Public Service Commission of the District of Columbia. Number three, following Washington Gas’ successful completion of the previous recommendation, audit and verify the performance of Washington Gas’ mercury service regulator replacement program including its record keeping. Number four, oversee the replacement process for the mercury service regulators that Washington Gas has in service. One recommendation to the International Academies of Emergency Dispatch. Number five, revised protocol 60, gas leak, gas odor, parentheses natural and liquefied petroleum gases, to direct dispatchers to notify the gas company when any odor call is received. One recommendation to the International Code Council which would be recommendation number six. In accordance with the Gas Technology Institute and National Fire Protection Association, incorporate provisions in the International Fuel Gas Code that requires methane detection systems for all types of rent residential occupancies with gas service. At a minimum the provisions should cover the installation, maintenance, placement of the detectors, and testing requirements. One recommendation to the National Fire Protection Association which is recommendation number seven. In coordination with the Gas Technology Institute and the International Code Council, revise the National Fuel Gas Code, comma, National Fire Protection Association 54 to require methane detection systems for all types of residential occupancies with gas service. At a minimum the provisions should cover the installation, maintenance, placement of the detectors, and testing requirements. One recommendation to the Gas technology Institute, recommendation number eight. In coordination with the National Fire Protection Association and the International Code Council, work to develop standards for methane detection systems for all types of residential occupancies in both the International Fuel Gas Code and the National Fuel Gas Code comma, National Fire Protection Association 54. At a minimum, the provision should cover the installation, maintenance, placement of the detectors, and testing requirements. There are three recommendations to Washington Gas. Number nine, implement an audit program to verify the data on the service forms used to determine the location and condition of mercury service regulators to ensure the accuracy of this safety critical data. Number 10, revise your procedures and field forms to require technicians to verify the integrity of vent lines following the testing of indoor service regulators. Number 11, establish a time frame with specific dates and milestones for the replacement of your mercury service regulators that recognizes the need to expedite this program and that prioritizes multifamily dwellings where mercury service regulators are located inside the buildings. Thank you, Miss Bryson. As far as amendments, I believe we have several and if we were to do them numerically, Vice Chairman, do you have a recommendation, an amendment to the proposed amendment–
I do, Mr. Chairman. I’m a little confused as to who to direct it to, so I’ll defer to the staff. But I’d like to have something that says that would require fire departments that if access to meter service rooms is required by the fire code and that access is not immediately available to notify building management to provide access at that time. Okay, please state your motion. Well, my motion is to, I’m not sure if it’s the International Code Council or the National Fire Protection Association, Mr. Hall would you have some suggestions as to who this might be directed to. I would actually need to research. There’s more than two organizations that deal with fire services and it would, to do it right, we would have to research the available organizations and (coughs) select the one best suited to implement it. Okay. Why don’t you Vice Chairman, my suggestion to you would be serve to make a motion, let’s see what the motion is, and then if the motion passes, staff, it could be up to staff to determine–
Okay. to whom to issue it. Please go ahead with your motion. Okay, I would like to make a motion to add to the recommendations to the appropriate authorities to require fire departments that if access to meter service rooms is required by the fire code and is not immediately available at the time of a call of notification to notify building management to provide access at that time. Okay, there’s a motion. Is there a second? Okay, I will second for the purpose of discussion but Miss Bryson, you have a look and I think your concern is that it was not read exactly as it was written. Is that correct? Correct, yes sir. What I have in front of me uses the word remind as opposed to required which are two very different words. I understand. My thought was to put a little more force to this because again, I feel a sense of urgency that if the code is required and there is an emergency or potential emergency situation and it is not acted upon at that time then we have the circumstance that we have here. And by empowering the first responders to be able to gain access at that time, and I’m not necessarily talking about knocking down the door, but that they get access to the area that they need to gain access that might have forestalled this and it would give everybody the impetus to say we need to have access to the room and we’re going to wait here until we get it. So that’s the thought process and that’s why I substituted the word require. So for clarification, it is required that you’re suggesting. I think require would be more suitable. Okay, thank you. So we’re talking about building management. And so notify building management. So is it possible that a private residence could have a meter indoors? Absolutely, a private residence can have a meter indoors. So, how would that square with your proposed recommendations Vice Chairman? Maybe we could say to multifamily dwellings and perhaps that would resolve it. The point is that in this circumstance, had the fire department had access it’s entirely possible that they would have been able to forestall this entire situation and to say well it’s in the code, we can’t get access, we’ll notify the fire marshal. This doesn’t serve the public well. Okay. So we have a motion. It’s been seconded. (Jennifer mumbles) Please. I do have one question, and I know your intent, I think it’s a good intent. My question though is what happens if providing access is not safe for the fire department? ‘Cause this says require building management to provide access. So, if for some reason it’s, I know that in this situation that wouldn’t occur but there could be situations where it’s not safe to actually provide access. I think I would leave that to the first responders. They deal with that circumstance all the time as to whether it’s safe to go into a particular area or not. And if, I’m speculating now, but if they had come down into the basement before they got access to the room and they had a strong smell of gas, chances are pretty good they would have exited the area rather than then press on, I think. We can deal with all kinds of hypotheticals here, the point is that if the code requires it and the building owner is not complying with the code then there should be an immediate remedy as opposed to saying well, we can’t get there, too bad, we won’t go in. The question that is on require. I mean, I know you’re trying to require access but you’re requiring then building management to allow access when it may not be safe but it’s a requirement. Well, if you have to change it to allow access and then it’s up to the first responders, again, I think from a practical standpoint, the first responders will know how to manage that. Again, they deal with this sort of thing all the time. They have access to any building. If the building is burning they’ll make the decision. They have access to it, whether they choose to exercise it is up to them. Thank you. Member Weener. I was just gonna point out there were other discrepancies between what I have as written here and what you read. I’m sorry, what Vice Chair read. We will make sure, we’ll make sure when we vote to have the motion clearly stated but you are correct. So, anything else? I hear a lot of well, we could or maybe we could do this. I think when we offer a motion, we’ve gotta, we have to be open to revising it but we want to have a clear idea of what we’re trying to do here and I hesitate to craft amendments on the fly although by nature of the deliberative body we will do that from time to time. We can do that whenever we need to. But we don’t even know at this point who this recommendation would be issued to. I agree with Member Homendy that this is a good intent but I hate crafting things hastily and coming up with something that we don’t want. How would you like to proceed? I’m a little surprised that we can’t figure out who to make a recommendation to when there’s a fire code and it’s these things have been, if you’ll pardon the pun, codified for decades that the fire department shall have access to areas that they need to get into and meter rooms and if they can’t gain access then the response is just to walk away? Well you say you’re surprised that we can’t figure out who to issue the recommendation to but yet you’re the one issuing the recommendation and didn’t know who to issue it to. So how would you like to proceed? I’d like to say to the appropriate local fire code authorities then. Okay, good. If you would kindly restate your motion and we will, if there’s, is there any further discussion? Let’s see what the motion is. Please restate your motion and unless there’s further discussion we’ll take a vote on it. All right, to local fire code authorities to require fire departments that if access to meter service rooms is required by the fire code and is not immediately available to notify building management to provide access at that time. All right, so I hear words, like Member Weener said, I hear words that are being added that aren’t written– Well, this a work in progress, sir. Require fire departments that if likewise– Yeah, I’m sorry, if access to meter service rooms is required by the fire code and is not immediately available to notify building management to provide access at that time and perhaps that addresses the fact that at an individual dwelling or something they won’t necessarily have a meter service room and they won’t have a building management per se. It would be a homeowner. I don’t want to put too fine a point on this. Yeah I want to know what it is that we’re adding. We’re adding here where we say something about access to a meter service room is not required by code, is that what you said? I said and to require fire departments that if access to meter service rooms is required by the fire code and is not immediately available comma, to notify building management to provide access at that time. Thank you. And to Member Homendy’s concern, they provide access that doesn’t mean they have to go in if they determine that it’s unsafe for them to do so. Okay. I want you to read it one more time and then we’re going to vote for it. We’re going to vote for it at this point. Please state your motion which I have seconded for discussion. Thank you, sir. To local fire code authorities to require fire departments that, to require fire departments that if access to meter service rooms is required by the fire code and is not immediately available to notify building management to provide access at that time. Thank you. Does everyone understand what the exact wording is of this motion? Okay, we’ll take a vote. All in favor of this proposed amendment, please signal with the hand and say aye. Opposed? Raise your hand and say nay, please. Nay.
Nay. The motion is two out of two. The motion fails. Okay we’ll go to the next amendment. And Member Homendy, I think you have proposed amendments for nine, 10, and 11 and if you would kindly advise the council. We will take them individually but if you want to explain what you’re doing to each of them in one fell swoop, go ahead. Well, I’ll start with the first three, I have three amendments to the safety recommendations to Washington Gas. Like I said, I will speak about them collectively but then I’ll offer them separately. Okay, so first I offer a motion to amend recommendation number nine so it will now read number nine, throughout the Washington Gas network comma, implement an audit program to verify the data on the service forms used to determine the location and condition of mercury service regulators to ensure the accuracy of this safety critical data, period. Okay, and that’s your motion. So there’s a motion. Is there a second? Seconded. It’s been moved by Member Homendy and seconded by Member Weener. And discussion? Thank you. I’m offering this amendment and the two subsequent amendments to make crystal clear that we are recommending that Washington Gas implement changes across their entire network including Washington D.C. and Virginia, not just Maryland. I feel it’s important that all customers of Washington Gas be provided with the safest system, equipment, and service and while earlier we do issue recommendations to public utility commissions in Maryland but also Virginia and D.C. I thought that it needed some clarification in the recommendations to Washington Gas that we mean the entire network. Thank you, thank you very much. Further discussion, questions? Mr. Hall, what would be staff position on this? Staff concurs with the proposed amendments. Thank you very much. Any further discussion? It’s been moved and seconded. All in favor of Member Homendy’s amendment to recommendation number nine, please signal with a hand and say aye. Opposed, there are none. The motion carries. The amendment has been adopted. And you have recommendation 10 amendment as well I believe. Thank you. I offer a motion to amend recommendation number 10 as follows number 10, and this is to Washington Gas, number 10, revised your procedures and field forms to require technicians to verify the integrity of vent lines following the testing of indoor service regulators throughout the Washington Gas network period. There’s a motion. Is there a second? Second. It’s been seconded by the Vice Chairman Landsberg. Any discussion? Same rationale and– Staff concurs. Thank you. Further discussion? Seeing none, all in favor of amended recommendation 10 as moved and seconded please signal with a hand and say aye. Opposed, there are none. The motion carries. Recommendation 10 has been amended. And now recommendation 11. Thank you. My amendment to recommendation number 11 would read number 11, established a time frame with specific dates and milestones for the replacement of mercury service regulators throughout the Washington Gas network that recognizes the need to expedite this program and that prioritizes multifamily dwellings where mercury service regulators are located inside the property, period. Thank you, thank you. There’s your motion. Is there a second for that? Seconded. Member Weener seconds. Any discussion? Staff concurs. ‘Kay, thank you very much. All in favor of adopting amending recommendation 11 as proposed as moved and seconded. All in favor of that please signal with a hand and say aye. Opposed, there are none. Recommendation 11 has been adopted as amended. And Member Homendy, I believe you have a proposed amendment to recommendation 12. Well, there’s no 12. This would add two additional recommendations to Washington Gas. My amendments, these would be two new recommendations, number 12 and number 13, oh sorry. No, that’s good, that’s perfect. Go ahead and just to do 12– 12 would read, it essentially mirrors the PHMSA recommendation but specifically for Washington Gas. Number 12, install all new service regulators outside occupied structures period. Okay, that’s a motion. And is there a second? Seconded. It’s been moved by Member Homendy, seconded by Vice Chairman Landsberg. Any discussion? Let me see what we’re doing here. Go ahead, any discussion? Yep, just to explain a little further. We make recommendations to PHMSA that all new service regulators should be installed outside occupied structures and those recommendations are for PHMSA to issue a requirement which would require a gas distribution pipeline operators or just pipeline operators to install these new service regulators outside the occupied structures. So we don’t go the next step of actually requiring that of Washington Gas which PHMSA would be doing through regulation anyways and my concern is that if rulemaking is delayed, these safety improvements aren’t made and that Washington Gas is already in the process of installing 125,000 additional mercury service regulators so we would want to see those safety improvements now rather than having them go back and change them in the future. So, bottom line is you’d like to make sure that they’re installed, everything is installed outside of occupied structures. Yes. And as you pointed out that does mirror exactly the recommendation to PHMSA. Member Weener. Mr. Hall. Staff concurs. No further discussion for number 12, adding number 12? None? Okay, it’s been moved and seconded to adopt a new recommendation to Washington Gas. All in favor, please signal with a hand and say aye. Opposed, there are none. A new recommendation to Washington Gas has been adopted. And you have another proposal for Washington Gas.
Yes. Another recommendation for Washington Gas, the addition of number 13. Relocate existing interior service regulators outside occupied structures whenever the gas service line comma, meter comma, or regulator is replaced. In addition, multifamily structures should be prioritized over single-family dwellings period. Okay, there’s a motion. Is there a second? Second. We had seconds and thirds on that. So, Member Weener seconded that one. And to be very clear that is an exact mirroring of what we’re proposing for PHMSA. Yes. Is there any further or is there any discussion on that? Staff concurs. Great. It’s been moved and seconded. No further discussion. All in favor of adopting the motion, the motion to add that new recommendation to Washington Gas, please signal with the hand and say aye. Opposed, there are none. The new recommendation to Washington Gas has just been adopted. Now, let’s go back and adopt all of the recommendations that we have just, all of them including those that we have just amended. Do I have a motion to do so? So moved. Vice Chairman has moved to–
Second. adopt the recommendations as amended and Member Homendy has seconded that. Any discussion about this? Seeing none, all in favor of adopting the recommendations as amended please signal with a hand and say aye. Opposed, there are none. The vote to pass the recommendations has passed unanimously. So, we now move to the adoption of the final report. Does anyone have any additional issues to discuss related to this matter? Is there a motion to adopt the report as revised? So moved. Member Weener has moved. Vice Chairman has seconded. All in favor of adopting the report as revised, please signal with a hand and say aye. Opposed, there are none. The report has been unanimously adopted. Do any members wish the right to file a concurring or dissenting statement? Yes. We have Member Homendy. We have Vice Chairman Landsberg. Okay. So noted. In closing, I do want to thank the staff for your hard work. I know that this has been a, as Mr. Hall said probably one of the most complex investigations that you’ve seen in your 40 years. I think I fairly well characterized that. They’re all complex but this, the physical evidence was largely destroyed. So I wanna thank the staff for your hard work on the investigation. I also want to thank Member Weener, Vice Chairman Landsberg, Member Homendy for all of your preparation that goes into it. What the public may not realize is that the staff, the board members do meet individually one on one with staff beforehand to go over any concerns that they may have and I know I can tell from the questions that a lot of hard work has gone into this. So thank you for that, the good debate and the good discussion. Special thanks to Rachael Gunaratnam, I almost said that right, the investigator-in-charge who took over a year ago. So thank you for picking up the ball and carrying it to the goal line. I’ve always said though that nothing gets done around here with just one person. It takes an entire team. We have our support personnel. We have, everybody helps this product come together and so on behalf of my colleagues on the board, a sincere thank you not only to the investigative staff but to the support and program staff as well. As we all know, pipelines provide heating and cooking fuel to millions of homes and businesses throughout the nation. In fact it fuels our nation’s economy. But when pipelines fail, as we’ve seen today, the results can be tragic as they were on the night of the explosion and fire in Silver Spring, Maryland. This agency’s employees know that by the time they arrive at the accident site nothing that they can do can stop the worst from happening. All that they and we on the board can ever do is to prevent the same thing from happening to others. Today’s recommendations, if acted upon, will prevent such tragedies through system-wide improvements to equipment and procedures. We stand adjourned.

10 Comments

Nicholas L

Apr 4, 2019, 8:21 pm Reply

Natural gas is very dangerous, even explosive. We all need to go solar.

william coe

Apr 4, 2019, 9:18 pm Reply

I used to live about 45 min.from where this took place…in my neighborhood there were 3 houses in the neighborhood in a cul-de-sac that blew up the same way.the 3 houses were completely destroyed.the only thing left were the concrete slabs.thankfully it happened in the middle of the day when nobody was home so there were no injuries.some of my neighbors said they felt the earth move when it happened….there is no way I will ever live in a house with gas again.it is way too dangerous.

Bienvenido Espinal

Apr 4, 2019, 10:32 pm Reply

Service regulators are required to be inspected every 20 years. When was the last time Washington Gas inspected this regulator? Was the inspector trained and qualified to identify that this was a Mercury regulator?

Keith Pedersen

Apr 4, 2019, 4:54 am Reply

Kudos to the NTSB for posting these on YouTube for permanent hosting. The old, in-house hosting was terrible. One critique; can we edit out the lunch breaks?

jfsa380

Apr 4, 2019, 2:44 am Reply

Speaking for our fire department and those surrounding ours with which we work, if we are called for a gas leak, and we need access to a service room, we are getting in. We are not requiring knox boxes, many do, but if a keyholder is not found, or else has an extended ETA (usually over ten or twenty minutes) we’re popping the door. I cannot recall thinking of getting a corner open, we’re getting completely in. It’s not a request. I am typing this as i listen, so answering Rachel’s comment, if we have a reasonable belief of an incident of any scale (a gas call, needing access to the utility room is more than reasonable) we have every right to completely break down the door (in practice just forcing through the lock).

WPLU572 Trunked Radio

May 5, 2019, 8:04 pm Reply

This happened 2 years ago for anyone wondering. Thank you to the NTSB for your hard work and dedication.

othername1000

May 5, 2019, 1:54 am Reply

Lunch break ends and the meeting resumes at 1:38:47

othername1000

May 5, 2019, 3:12 am Reply

It's interesting that a private person is expected to notify more people than professional firefighters and dispatchers are expected to notify. You'd kind of hope that the professionals would be held to a higher standard, and be required to notify the gas company. The gas company has the better technology and experience to properly detect , trace, and stop leakages.
Within the past couple months, I smelled gas in a larger building, and called METCAD. The fire department responded, they too could smell the gas but with their equipment were not able to get a reading.
In this case however, METCAD and Fire did notify the gas company. Gas company showed up and found a significant leak where contractors had connected to run a flexible line up a chimney.
This was on a Friday evening, that thing could have leaked all weekend. Even though the FD couldn't get it to show up on their instruments.

jefflewis4

May 5, 2019, 3:16 am Reply

Did they determine why the locks to the meter room were changed ??
Locks are not usually changed on a whim. They must have had a compelling reason to do so. Perhaps people outside the maintenance staff previously had access to that room ?

Tel H

May 5, 2019, 1:45 pm Reply

No mention of service pipes coming into the building. Gas leak especially with Iron pipework, there could have been a rusty pin hole leak external to the building, and the leak could be making its way beside, external to the pipe access into the building, No mention if the gas pipework external, was buried up to the external wall before it entered the building, or whether the pipework raised from the ground before entering the building.
Because the socket and nut was disconnected from 1 Pressure regulator, if either safety regulator failed, top or bottom, gas would leak into the room.
The weather, was it Hot or Cold? was it Windy or Calm? Barometer pressure, High or Low?
The service boiler, was it spark ignition or permanent pilot light? was it balanced Flue, Air supply directly from external to the room via duct, and exhaust away from the room from a sealed flue,
Meter room lock, a DSK8000 Door Security Bolt Spline Key type lock could be standard fitting to meter rooms.

Leave a Reply