Power Plant Fire In Missouri

[GaryM]

It was reported on the news tonight that the coal-fired Ameren power plant in Labadie Missouri had caught fire and a portion would be off line for some time. The cause was reported as a hydrogen leak.

Anthracite, is the Hydrogen a by-product of combustion or is this perhaps a gasification plant? I think that’s the correct term. Or perhaps the news reports are just incorrect?

[Una_Persson]

Nope. The hydrogen is used to cool the generators. Hydrogen works great as a gaseous coolant (it’s not liquid hydrogen, it’s just normal hydrogen gas) and is not explosive…until it leaks and combines with air.

I know quite a bit about that plant - I’ll be calling there tomorrow in fact to see what the “Straight Dope” is on the incident…

[Una_Persson]

First off, you are wondering…why use hydrogen to cool the generator? Simple:

1. A gaseous coolant is needed, and
2. Hydrogen gas has a very high specific heat. Meaning, it can soak up more heat per pound than any other common gas that can be used. Thus, it makes for efficient, high-performance coolant.

Now…what does a hydrogen cooling system consist of? Well…I’m glad you asked.
(public domain document, US Nuclear Regulatory Commission)

An example from a nuclear plant’s generator cooling, but coal systems are identical.

… The generator assembly is cooled by hydrogen gas, with the hydrogen gas being cooled by cooling water flow through the tubes of a heat exchanger. A manual throttle valve in the discharge line from each cooler is adjusted to balance flow through the coolers. Total cooling water flow can be controlled from the control room by throttling a control valve in the common cooling water header to the heat exchanger.

Inside the generator, a shaft mounted blower circulates the hydrogen through the coolers and the various regions of the generator. The blower is a high head, low volume unit. The blower discharges into the hydrogen cooler and from there the hydrogen is distributed back to the generator at the exciter end. Cold gas leaving the coolers is typically carried by duct work to the opposite end of the generator. The hydrogen is pulled by the blower through the generator. The hydrogen then passes through the stator core and is discharged at the turbine end of the generator back into the blower, where it is directed back through the coolers.

In addition, part of the hydrogen leaving the coolers is directed through the rotor. Some of this hydrogen enters the rotor at each end, flows through the ventilation passage provided in the rotor winding, where it is discharged into the air gap at the center of the rotor. The hydrogen in the air gap is drawn back into the blower and returned to the coolers. A portion of the cold hydrogen gas entering the exciter end of the generator is diverted into the main lead connectors and bushings. The intimate contact of the hydrogen with the conductors in both the rotor and the stator facilitates maximizes heat removal from the generator.

Purity of the hydrogen gas must be maintained at a high level to prevent an explosive mixture and to maintain its effectiveness as a cooling medium. Impurities such as moisture, hydrocarbons, and particulate can have catastrophic effects on the generator if they are not adequately removed. Moisture, hydrocarbons, and other contaminants generally enter the hydrogen cooling system through leakage from the generator shaft seals. Moisture in the hydrogen can lead to internal structural concerns, coil shorting and arching, and insulation breakdown. A small portion of the cooling hydrogen is routed through desiccant dryers and filters to continuously remove moisture and particulate contaminants from the system.

The hydrogen condition is monitored on a hydrogen control panel. Hydrogen purity with relation to air is monitored. In the event the hydrogen is required to be purged from the generator, the relationship between air and carbon dioxide, the purge gas, is monitored by this control panel. The internal generator hydrogen gas and circulating fan pressure are monitored by the hydrogen control panel. The alarms associated with this panel are alarmed and annunciated locally and in the main control room.

Hydrogen makeup to the generator, to either improve purity or increase pressure to the generator is a manual operation. Once a day the hydrogen purity in the generator is monitored by plant operations personnel. The amount of hydrogen makeup to the generator is monitored daily. If a makeup flow rate of 500 to 1500 cubic feet of hydrogen per day is exceeded, plant procedures generally require the cause of leakage to be investigated.

The use of hydrogen as a coolant in the generator requires that the shaft be sealed. This shaft seal is necessary to prevent the loss of coolant and the exposure of hydrogen to both ignition sources and air. The seal oil system is divided into subsystems. An air side subsystem which seals the generator shaft from the outside atmosphere and a hydrogen side subsystem that seals the shaft from the generator side. Seal oil is provided by the separate subsystems to the gap between the shaft and the generator housing. The hydrogen side oil is supplied at pressure slightly higher than the hydrogen pressure and drains towards the generator internals. The air seal is provided oil at approximately the same pressure as the hydrogen side and is designed to flow outward on the generator shaft. The seal itself is a mechanical gland type seal. …

[Bawdysurfer]

The business of consulting has involved me with much of the activity that surrounds a turbine deck during maintence periods. The pickle (rotor) will usually only be removed when recorded vibration has neared certain limits. Bearings, seals and such can be serviced without removing the pickle.

When a full disassembly is required, plus the servicing, and then reassembly, it is absolutely captivating for those who love mechanisms. Huge cranes, tremendous rigging, frightfully heavy components, and large burly men (petite women sometimes too) who manage to work as if repairing a Swiss watch.

Many pairs of eyes watch all aspects of the reassembly. Truly precise measurements are taken. Vast documentation occurs. The “what ifs” and “how abouts” require charts and tables and calculations to resolve. Items are x rayed. New, and verified hardware replaces some of the old. Asses, customers, and the NRC are covered and satisfied seven ways from Sunday. After final button-up, tests are performed.

But “stuff” happens.

Electronic “sniffers” can catch gas leaks in time, almost every time.

Sometimes, fault cannot be placed on human shoulders. The sequence of events that lead to mechanical failure are not always found in their entirety.

Looking forward to Anthracite’s findings.

[GaryM]

Thank you both for the answers. I had no idea that the generators were cooled by hydrogen. I have passed the Labadie plant many times, usually across the river and it’s quite a landmark. Do most power plants use the same type of system?

As soon as I heard the report on the news, I knew where to go for answers.

I have visited both Bagnell Dam and Hoover Dam and cooling the generators was never mentioned, although a little thought would have made me realize it’s required.

Thanks again and I await further developments.

[Una_Persson]

Most plants use hydrogen, but not all. The high-performance ones do.

My “sources” tell me there was a hydrogen fire and small explosion near the boiler feed pump turbine at the plant, and (reputedly) there is a hole in the turbine casing. If so, the unit will be down for some time…if I hear anymore info that I can repeat, I will post.

[Bob_Scene]

I’m glad I read this because I’m supposed to get samples of fly ash from Labadie on a regular basis (so they can keep selling it for TxDOT jobs) and I’ve had a whole lot of trouble in the past getting them to send me the samples they were supposed to (or getting them to return my phone calls even), and now I know why they won’t be sending me any for a while.

[Una_Persson]

Well, it looks like the damage is a lot less severe than thought at first. The “hole” in the turbine magically changed into a hole in the hydrogen system, and supposedly there is a section of the turbine hall roof that is missing. The plant says they may get the Unit back online by the end of the week, however.

[Una_Persson]

OK…my last post here unless someone else has a question. This is “unofficial” (but not confidential) from a friend at Amren, with some editing for clarity and emphasis added:

"…The fire occured Saturday around 11:30 AM when a pressure gauge and its associated root valve broke loose, allowing 150 PSI control oil, for the boiler feed pump, to spray oil all the way to the turbine building roof. This oil ignited on hot pipes sending a plume of fire up to the turbine building roof (Note: The operator at the plant told me that the fire plume was more than 100 feet high!), igniting the roof. The plant operators isolated the oil and extinguished the main feed pump fire within about five minutes. The roof fire was finally extinguished by the fire department around 4:30 PM on Saturday. No injuries occured as a result of the fire.

During the fire, Unit 2 tripped offline due to electrical bus failures on the common bus caused by water from the fire fighting effort. Unit 2 was restarted and put back online around 10 PM Saturday night…two roof trusses sustained significant fire damage and will have to be replaced…8000 square feet of roof decking was destroyed. Approximately, 26,000 square feet of roofing material burned and will have to be replaced. This is expected to take around 2 weeks to complete…

…damage to date includes a 4160 Volt breaker and its cubicle that failed due to water intrusion, failure of a structural hanger for a steam line, control cabling for the boiler feed pump, control cabling for the D A tank and a cable tray carrying all power, phones,and fiber optics to the service building…Unit 1 was scheduled to come off this Thursday for a four day tube leak outage…this work will be completed during this fire recovery outage."