For New York State Public Service Commission, investigated and wrote a comprehensive report on the Causes of a Steam Incident that occurred in New York State. The Incident was determined to be most likely due to a Condensation Induced Waterhammer Event in a steam distribution system. No serious injuries or fatalities occurred. My Report, which was submitted in October 2019, as well as identifying causes, listed recommendations to preclude or lesson the effect of a future accident of the same type.

20" pipe Rupture due to Condensation Induced Waterhammer

For Attorney for Engineering Firm being sued for their design of a steam distribution line of which 500 feet fell off its supports after a plant upset, defended engineering firm in a forensic report by demonstrating that an alternate computer stress analysis program gave the same results as the program they used to predict pipe movement during expansion, and the installing contractor did not cite the pipe supports as directed on the engineering drawings. Therefore, the engineer's pipe expansion program which showed the pipe design "would work" was not actually a model of what the contractor actually built. The Contractor's variance from design was aggravated by the division of contract administration between the Engineer and the Owner. the Engineer did not have field inspection responsibility. This case, which began in February 2015, settled after my report submission in May 2017.

For Newcrest Lihir Mining located on Papua New Guinea, April 2017: Investigated waterhammer incident in the steam piping coming from one of three Autoclaves during warm-up of the Autoclave when cold water was introduced to an isolation loop for instrumentation. The incident startled the Operator but did not rupture any piping component. The incident motivated Operations to invite me to site to investigate the accident and do two 8-hour waterhammer seminars for Supervisors, Engineers, and operators. I left a power point slide presentation for management explaining their incident along with another potential incident found during a plant walk-thru.

April 3, 2017, for attorney representing insurance company, submitted Expert Report on cause of 13 March 2013 Steam Hose Rupture on a tire press at a major tire manufactures plant in which a maintenance contractor's technician received 2nd and 3rd degree burns. The essence of the technician's complaint was that a live steam hose carrying 250 psi steam spontaneously ruptured when he came into its general vicinity after climbing to the top of the press. The Plaintiff claimed that the steam hoses --some 24 per machine-- were worn out and failing. Maintenance records of unscheduled maintenance at the tire press did not support the Plaintiff's contention. Evidence uncovered during depositions plus the burn pattern on the man's body cast doubt on the technician's story that he was nowhere near touching, kicking or entangling distance of the steam hose that he had to step over to get to the item to which he was heading. My suspicions were initially aroused by the description he gave of the steam cloud he says he witnessed at the time of the break. It did not comport with research I'd done on steam jets released into atmospheric conditions.

For attorneys for York University, Toronto, Canada, investigated failures of anchors securing slip-type expansion joints to vault ceilings in 1800 feet steam pipe installed in 2003. Of six slip-type expansion joints in three steam vaults, anchors on all six joints failed within two years of service. My Report identified the two systemic reasons for the failures based on a review of the engineering drawings and the products installed by the Contractor; namely: 1) Pipe was not designed to expand strictly axially into the slip-type expansion joints; and 2) Substitution of spider-type pipe alignment guides for the more robust "combination guides and supports" designed by the Engineer. The substitution deprived the piping system of proper support in accordance with the joint manufacturer's specifications based on the B31.1 Code unless the spider guides were considered pipe supports. But spider type guides are not designed to serve as pipe supports. Using them as such put the weight of 1200 pounds of steel pipe (and sometimes up to 1200 pounds more of water during system warm-up) on the spider of each guide causing it to bottom-out and scrape along the guide housing in order to move during pipe thermal expansion. This weight vastly increased friction to pipe movement compared to the graphite-lubricated "guides and supports" which were originally intended by the Engineer. This case is ongoing.

Primary Guide was not robust enough to correct for pipe misalignment

Analysis of Steam Rupture of 24" Externally Pressurized Expansion Joint at Veolia-St. Louis Ashley Street Plant on October 9, 2014. The internal bellows of the expansion joint is protruding out of its casing at right after the joints weld was ruptured by water hammer. On the morning of October 9th, 2014, while the Ashley Street Steam Plant was providing about 65,000 #/hr of 150 psig steam to the downtown area of St. Louis, a 24" expansion joint ruptured allowing steam to escape from M.H. 300-the first Steam Vault outside the Plant thru which the East 24" steam Main was trapped and anchored. The 24" East Steam Main-one of two Mains that leave the Plant-- was valved-off in the Plant at the time but pressurized nevertheless to 150 psi due to backflow from an open 16" interconnecting pipe to the other Steam Main leaving the Plant-the 20" West Header.

I investigated the possible causes of the rupture, concluded it was Condensation Induced Waterhammer after working out the waterhammer scenario and eliminating the other possible causes for the Accident. Reported the cause to Veolia in a slide presentation and Report.

Design Review of Prototype Well Pad Module for Cenovus Energy, Calgary, April 2015.

The purpose of the Review was to point out locations where condensate could accumulate and lead to slugging or Condensation Induced Water Hammer and then suggest design modifications or operational procedures to avoid these possibilities. The design modifications focused on configuration which avoided the need for provision of steam traps.

For Cenovus Foster Creek SAGD Site in Alberta, CA, investigated root causes of support damage in 14" steam lines in the South OTSG Steam Plant due to waterhammer initiated by overflow of boiler feedwater from a steam separator into the steam lines after a plant trip. This was an example of Condensation Induced Waterhammer due to cold water dumping into a hot steam line instead of steam inserting itself into a condensate filled line.

For Veolia Energy Corporation, troubleshot expansion joint failures in condensate return system which was mixing 100oF condensate return from a turbine hot well with the discharge of high-pressure steam traps in the condensate return line. Five externally pressurized bellows joints failed in five separate instances. Diagrammed out reasons for condensation induced waterhammer occurring in the lines, pointed out design flaws, and recommended changes to eliminate or mitigate waterhammer.

For Attorneys representing the University of New Hampshire, investigated and submitted Expert Report on cause of steam release accident due to a water hammer in a new 10" steam line being started up by one of the construction contractor's subcontractor employees. The waterhammer in the 60-psi steam line blew out a gasket at an 8" closed valve that permitted water and steam to escape and fill the 11' 8" deep Vault containing the valve. One worker was badly burned. Neither he nor his helper could remember what they did in the Vault on the day of the Accident, or the day before. In my 10,000-word report, I laid out the scenarios that could have led to, and triggered, the Accident. The drawing at right shows the Vault flooded with condensate 2 days before the accident due to a ruptured check valve in the trap discharge piping. The August 2012 magazine article "Waterhammer in Condensate Return Systems" is about the rupture of the check valve.

For the Comanche Unit III Power Station-- an 850 MW, supercritical, coal fired power plant in Pueblo, CO., troubleshot two hydraulic transient problems --one in the 16" return pipe emptying into the Plant Deareator, and another in the 16" Boiler Drains Tank pipe to the 3 psia Condenser Hot Well. While the transients were not severe enough to rupture the XS pipe or put operators in danger, they were forceful enough to knock pipe slips off supports (see example below at right) and cause damage to valves and structure that had to be repaired.

After doing two seminars for Plant operators, I explained what was happening to the Principal Plant Engineer, why it was happening and, in the case of the Deaerator piping, recommended a simple fix which is being implemented. For the Boiler Drains Tank emptying to the vacuum in the hot well, the damage pattern indicated the Tank was draining below the level of its outlet pipe thereby permitting flashing steam to enter the discharge line. If the water in the horizontal pipe had stood idle and cooled during a plant shut down, the entering steam could cause condensation-induced waterhammer that would match the damage pattern. A leaking level control valve in the Line to the Condenser Hot Well was the most likely cause of the tank draining far below its controlled level.

For Attorneys defending Hydronics Systems Manufacturer, Engineer, and Architect, investigated steam release at Las Vegas hotel where a technician was severely burned while attempting to start up a new steam-to-hot water-heat exchange assembly. During start-up, 110 psig steam was admitted to a steam line which was already filled with subcooled condensate because the Assembly's steam trap and condensate mover had not been activated during a start-up attempt a month earlier by the same contractor. The condensation-induced waterhammer resulting from the mixing of the 344oF steam with 70oF water blew out a flange gasket spraying the worker with hot water and steam. He died a week later in the Hospital. My November 2010 Report on the cause of the accident led to the immediate settlement of the lawsuit brought by the family of the worker against the manufacturer, vendor, hotel, engineer and architect.

The position of the nuts on the PRV pilot proved that the start-up technician had indeed applied steam to the Heat Exchange Assembly, contrary to testimony by his boss.

Forensic engineer assisting plaintiffs' attorney to understand the 2004 steam explosion outside the New Executive Office Building in Washington D.C. which killed two contractors attempting to re-energize steam service to the NEOB after a water main break. Water from the break submerged the steam main serving the building causing increased condensation in the Main and subsequent water hammer in the building. The Contractors received an emergency request to shut the steam feed off in the street. In the morning, when the contractors were attempting to restore steam service to the building, the water hammer repeated in the manhole outside the building where the contractors were opening the valve. The force of the waterhammer blew off the bottom plate of a jury-rigged pipe cap for the drip leg welded in place years ago by an amateur welder. The steam and hot water released from the steam main killed two contractors at the manhole.

Manhole Under Renovation Before Accident

For Suncor Energy, Firebag, Canada--Analyzed water hammer in the common Blowdown Header from OTSG's where flashing steam from superheated water and is mixed with subcooled water. (The diagram at right shows six scenarios for how subcooled water flowing into a steam line can either be stable or cause condensation induced waterhammer). Provided report analyzing all foreseeable water hammer susceptibilities in steam plant producing 11,000 kPa steam for SAGD bitumen recovery at the Firebag site in Northern Alberta. Presented 7 hour seminar at Firebag to Operations Managers and Engineers explaining condensation-induced waterhammer and the origin of waterhammer suffered at their Plant. Advised Suncor's engineers on developing equipment modifications and operating procedures to avoid waterhammer scenarios. Diagram at left is from Aya, Yayame, Nariai's "Effect .. Waterhammer Induced by Injection of Subcooled Water...into Steam flow. " It served as my model for diagrams drawn for Suncor to explain the different waterhammer scenarios that could take place in their OTSG blowdown line.

Forensic engineer in defense of Spence Valve (who was not responsible for the accident) in lawsuit resulting from the 2004 BLEVE of a stainless-steel tank in a Con Agra Plant in Tennessee. A BLEVE is a Boiling Liquid Expanding Vapor Explosion. Although normally associated with accidents involving flammable liquids like propane, a BLEVE can occur in a pressurized hot water tank in which only hot water and steam are released. A BLEVE occurs when superheated water (at elevated temperature and pressure compared to atmospheric conditions) is released through what may start as a relatively small fracture in the vessel. The explosive power comes from the 1600 times expansion of superheated water as a portion of the superheated water flashes to steam at atmospheric pressure. At right is a photo from another accident. It's believed the atmospheric vent on this laundry condensate receiver became plugged allowing hot condensate from high pressure traps draining to it to pressurize the tank beyond its 5 psig intermittent rating. Upon cracking, probably at a weld, a BLEVE blew off both heads of the vessel. The resulting loss of containment created a shock wave in the surrounding space which knocked down and injured two workers roughly 20 and 50 feet away as well as denting duct work, cracking sheet rock, and breaking a glass window. (Click on BLEVE to view the tank that was utterly destroyed in the Con Agra Incident and for more about what causes a BLEVE)

Ends Blasted Off 350 Gallon Condensate Receiver Tank

For Duke University, investigated steam release accident which resulted in the death of a maintenance mechanic trying to reactivate a Building PRV Station. Briefed management to explain the series of events that led to the accident. Wrote revised start-up procedures which management perfected and issued. Conducted two-day seminar for maintenance, safety staff, and management on water hammer in steam systems including what happened in their accident.

Blown out Strainer upstream of 120 psi PRV Station that was being re-energized after a brief shut down for maintenance.

For MIT in Cambridge, MA Nov. 2008, Investigated 200 psi steam release accident where anchors were jerked from their ceiling mounts allowing a slip-type expansion joint to be yanked apart by a condensation-induced waterhammer. Condensate accumulated in the steam line due to an isolated trap which was hidden from plain view by an unrelated equipment modification. The accumulation apparently sat in a piping depression undetected and undisturbed for more than a year because of near equal steam pressure on either side of the University's looped steam network. The accident was finally initiated by a major pressure upset at the University's Power Plant. There were no personnel injuries, but extensive property damage. In addition to reconstructing and explaining the cause of the Accident, provided an 8-hour seminar to steam fitters and University's engineers.

Slip Yanked from 8" Expansion Joint by Condensation-Induced Waterhammer when anchors broke.

University of Massachusetts Expansion Joint Failure, Amherst, MA. February 2008, Investigated and reported on root cause of externally pressurized expansion joint failure in one of two 20" steam mains from new Central Heating Plant. Found that condensation-induced water hammer had occurred due to misunderstanding of inverted bucket trap operating characteristics by the start-up engineering company. As a consequence, a "parallel two-phase flow instability" between the two nearly identical 20" lines allowed 7500 gallons of condensate to accumulate in one 20" line unbeknownst to operators. The condensate was able to accumulate without notice because the system was "looped" and under low-flow conditions so that all steam was readily delivered through one line while the other was blocked with condensate. Advised University on identifying other expansion joints which may have sustained damage. Dr. Peter Griffith assisted in this investigation. Clogged trap strainer pictured at right s explained below.



Strainer removed from trap assembly draining the low point in the 2000' of 20" Mains. Clogging material is pipe scale loosed from pipe walls as a result of pipe cleaning by the start-up contractor who, afterward, neglected to clean the strainers. While not the cause of this Accident, the pictures speak to the residual pipe scale present in new pipe and the need for strainers to protect trap orifices. This strainer on the lowest elevation trap drained the most liquid and so caught the most crud.

Expert Witness for Attorney defending Industrial Campus against OSHA Citation and Fine in case where steam worker was killed during start-up/warmup of a branch steam main. OSHA cited management for violating OSHA's Lock-Out/ Tag-Out Standard CFR 1910.147. My report and affidavit argued that employees' procedures exceeded requirements of OSHA's Lock-Out/ Tag-Out Standard and that the accident was caused by circumstances not included in the Citation or anticipated by the Standard. The Federal Judge upheld this argument among others made by the attorney and dismissed this citation. At right is a photo of the 8" Class 250 cast iron valve which ruptured while being opened to 220 psig steam by the worker. (The valve bypass line has been removed.) My investigation found that the actual root cause of the accident was Shut-down/Start-up Procedures which did not require a drain be opened and LEFT OPEN at the low point in the isolated pipeline. A closed gate valve under full steam pressure leaked steam into the isolated section of line which, over five days, filled the line with condensate. Even though there was a trap in the isolated line section, there was insufficient pressure to discharge condensate from the trap. At start-up, full steam pressure was applied to the water-logged line without noticing the condensate accumulation. When the downstream isolation valve was opened, draining condensate allowed steam to intrude into the subcooled condensate and hammer.

Coronado North Island Naval Air Station, San Diego, Sep 2004. Comprehensive Report on 8 in.steam valve rupture in a manhole which blew apart releasing 125 psi steam during an attempt to open the valve. The civilian steam worker opening the valve, received second and third degree burns over 20 to 25 % of his body. Had he not been in the process of evacuating the manhole in response to a loud bang which preceded the rupture, it's doubtful he would have survived the incident. Found that the direct cause of the rupture was condensation induced waterhammer resulting from nucleate boiling of ground water in a manhole 300 feet from the accident site. A heat transfer calculation showed that completely submerged steam piping within the downstream manhole (as had been, on occasion, observed to be the case) would generate enough condensate within the pipe to overcome the capacity of the steam trap in the Manhole and fill the 300 feet of 8" pipe between the two manholes. This is a somewhat rare but recurring problem at the base because of the Environmental Protection's office objection to discharging ground and rainwater to the street level by sump pumps and the resultant abandonment of sump pump maintenance and replacement.

Concurrent with accident analysis, taught three half-day sessions for civilian steam fitters on condensation-induced water hammer and briefed supervisors and managers on the systemic causes of the accident. In another phase, inspected 8 Piers at Naval Station San Diego for steam waterhammer potential and reported on hazards found.

Boiling ground water in Steam Pit

For Motorola, now Freescale Semi-Conductor, in Austin, TX, June 2004, investigated cause of steam rupture which shattered a 10" cast iron strainer releasing steam and condensate into the mechanical room of the Plant. Found that one of two 16" steam mains that made up the "loop system" serving the Plant was completely blocked by condensate resulting in an unstable equilibrium in the system. Closing any of several isolation valves between the East and West 16" mains would again accelerate the slug and repeat the accident. Advised Freescale engineers to immediately locked-out valves. Helped them devise a first-of-its-kind method to defuse the situation without down shutting steam production . Taught two-day training class to workers, supervisors, and engineers on condensation- induced water hammer and what caused their accident.

Ruptured 10" Cast Iron Strainer

For Houston, TX law firm representing industrial concern, investigated cause of steam accident resulting in operator death when a hydraulic shock stripped the nuts from the studs of a valve flange allowing it to separate releasing a spray of scalding water. The release burned the operator to death.



I concluded that the accident was due to condensate which accumulated atop the valve and subsequent water hammer which was initiated by the operator when he attempted to open the valve. The root cause of the water hammer was not operator error, but the improper original design configuration by the engineer of the trap assembly with respect to the valve.

For NASA Langley Research Center, 2002, investigated anchor failures and seized joints in 4900 ft. long Steam Tunnel No.4 carrying 350 psig steam. Found joint misalignment due to deficient 1964 engineering drawing detail resulting in widespread improper anchor installation by the original contractor. (It should be obvious that the arrangement shown at right cannot work). Problems were not due to water hammer as had been postulated. Analyzed system vulnerability to waterhammer as a precaution. Documented root causes of tunnel problems in Report to NASA Engineers August 2002. Click to view Report.

Can you spot the obvious engineering error above made by the original Design Engineer?

Investigated cause of steam accident at Metcalf Lumber in Metcalf, Georgia in which a Class 250 cast iron valve exploded killing a worker as he passed by the valve. My accident report found that the water hammer was due to the lack of rudimentary engineering design for the "in-house installed" addition of an additional drying kiln. The installers didn't understand the importance of steam traps.

Author and chief investigator retained by the manufacturer for the report on The Expansion Joint Failures at Le Moyne College in Syracuse, NY in which three flexible-hose type joints catastrophically failed during start-up of the steam system. Report found water hammer was not responsible for failure.

Forensic Analyst retained by the Georgia Department of Human Resources to investigate the cause of a steam accident which resulted in an operator's death at Northwest Georgia Regional Hospital. Authored the report: "Analysis of Conditions and Events which led to the Valve Failure in Steam Pit 3A ...". Retained as the expert witness defending the engineering design firm from the lawsuit which arose from the accident. Report found that the accident was due to a combination of operator error and the inadvertent disabling of a trap during a building demolition which had taken place years earlier. The magazine article "What Caused the Steam Accident that killed Jack Smith" which appeared in the July 1995 issue of Heating/Piping/Air Conditioning magazine and the Feb. '96 issue of National Engineer was based on this accident.