At what temperature is a wasteheat steam generator identical to a fired tube boiler?

[SREisme]

I am trying to understand the differences between unfired waste heat steam generators and fired boilers as it applies to a low water cutout requirement. NFPA requires a low water trip on all fired boilers, presumably due to the number of boilers that have exploded when water was lost. I can not find a similar requirement for waste heat steam generators. I found one manufacturer of WHSGs that publishes a run dry capability up to 840F (450C), but no reason why.

Can anyone point me to an industry standard, research report, or study that identifies the maximum safe operating temperature of a dry boiler tube to avoid the potential for a steam explosion in the boiler on the reintroduction of water to a dry surface? It stands to reason that there is a limit somewhere between 536F (minimum homogenous nucleate boiling temp) and fired boilers (2000-3500F), but I have not found one in my search yet.


Thanks

SREisme

 

[macmet]

The dry run temperature of heat recovery steam generator is dictated by the GT. Introducing feedwater to a hot boiler must be done in a controlled manner to prevent damage to the boiler. Controlling the feedwater allows control of the "steam explosions". If the flow is allowed into the boiler to quickly you can cause major damage.

I am on a site now that we are running the HRSGs dry up to 860F before introducing water without problems.

 

[SREisme]

Macmet- thanks for the reply. My application is in recovering heat from hot process gases or hot oils. We are well aware of the potential for a steam explosion due to injecting water into a hot oil pool due to a major tube leak. What we don't know is the potential for steam explosions caused by an intermittent loss of boiler feedwater to a steam generator caused by a faulty level control valve that frees itself and opens to restore water level. In this scenario you would have no control on feed rate to the dry tube.

There is a theoretical lower limit of 536F for liquid to liquid steam explosions and the NFPA requirement for low water cutoffs for fired boilers, but I am looking for a published upper limit on dry, waste heat steam generators. In my estimation, at some point the waste heat gases can be as hot as a fired heater and the only property that makes sense to me to differentiate on is temperature. Therefore, there must be an upper limit on safe operating temps of waste heat exchangers.

Do you have any data on exchangers that were damaged by putting water in too fast? Any idea on what temp they were at?

 

[macmet]

Well, I know that if we did not properly control our water flow our boilers would be badly damaged at the temperatures we're working at. I don't think that is useful to you though. We have do have some fired boilers that were damaged by running dry, but I think that we are talking about a different type of firing. But, just to give you a number, we do not allow our boilers to be dry-run above 1050F, but our standard limit is lower than that.

 

[chicopee]

Insurance companies will require at least two low water cutoffs on either fire tube or water tube boilers for obvious reasons; annually when these boilers are opened for internal inspections, the low water cut off devices must be dismantled for cleaning and for the inspections of the boiler inspectors.

 

[SREisme]

A damaged boiler, ie twisted tubes, bent stays, minor leaks, etc are a problem but not my primary concern relating to the original post. My primary concern is the perceived risk of a boiler explosion resulting in flying debris and release of the hot process gases to atmosphere. I want to understand if such an event is possible with an unfired steam generator, and why or why not.
I have read failure investigations for fired boilers that list reintroduction of feedwater to a dry tube as the cause for severe explosions resulting in the loss of containment I am concerned about. As previously mentioned, the only significant difference I see between fired and unfired boilers is the gas temperature, unless I am missing something (please tell me what, if so). Therefore, there should be an upper operating limit on inlet gas temperature to reduce the risk of a steam explosion caused by rewetting a dry tube, unless that is impossible. 1000F seems like a good guideline for an upper limit, but I would like to understand the physics behind why.
Thanks for the posts in response.


SREisme

 

[macmet]

Can you maybe give us some details about your boiler? Style, working pressure, etc... Maybe that will help.

 

[SREisme]

I sure can, Macmet.
The boiler in question is a sulfur condenser in an oil refinery. It is a horizontal shell and tube steam generator with hot sulfur/ SO2/H2S vapors flowing through the tubes at 600-750F and ~10psig with boiler feedwater on the shell side making 70# saturated steam. This condenses the sulfur vapors and cools molten sulfur to approximately 320F on the outlet end of the tubes. The tube bundle is flooded by water and controlled by a local pneumatic level controller operating the BFW makeup valve. The 70# steam leaves the shell side and passes through a non return valve before entering the 70# header. The bundle length is approximately 20' long, and 44-3/4" diameter, with 363 1.5" OD tubes. The shell diameter is 45-1/4" ID, so there is not much steam disengagement space in the shell.
A question came up about the potential for steam explosions in waste heat boilers caused by losing level, drying out the tubes, and then readmitting water to the dry tubes. We hired a consultant to review the published research information on steam explosions and got an inconclusive answer.

As I have mentioned previously, the liquid to liquid steam explosion potential is well documented for water entering hot liquids above 536F. There is an interesting report of a foundry blowing out the walls several hundred feet from the smelt spout when a small amount of water was dropped into moltem metal. The instantaneous boiling of water results in a pressure impulse that rips through structures, similar to a TNT detonation.

Our consultant was unable to come to a conclusion on the potential or probability of a dry tube (metal to liquid) steam explosion due to lack of published research, industry experience, and the presence of surface roughness providing numerous nucleate boiling sites to propagate boiling before the feedwater could superheat to above 536F. They closed the report saying that if dry surface explosions are credible, they should be considered to be catastrophic in nature. Hence my unending quest to determine at what temperature a steam explosion could be credible. I believe the problam is bracketed between 536F and the firebox temp of a fired boiler at ~2500F, but I don't know where to draw my line in the sand in between.

Thanks

 

[macmet]

Well I was hoping I would be able to help more, but I'm not familiar enough with shell and tube steam generators to comment any further.

Good luck

 

[rmw]

I worked around hundreds, and that might be understatement, of boilers over a 40 year period and I can number the boiler explosions I ever heard of on one hand. Most of those explosions were on the fuel side, not the water side. The days of those types of explosions pretty much went by the way side with the advent of ASME.

What is your end game, to determine if you need a low water cut off, or something else?

Mistreat your equipment badly enough, and you might rip a weld seam in your shell wall or blow an end off, but I can't picture a failure of you steam generator making the local nightly news.

rmw

 

[dougarthur42]

From your original question, difference between a fired or unfired vessel? there is no difference. If you are operating a fired boiler or un fired heat recovery unit the low water cut off should work the same way. You should not introduce heat to the unit if it has no water in it. Just because it is not fired doesnt mean its not a boiler producing steam. A low water alarm/trip should shutoff the heat going into the unit to prevent overheating and tube damage or rupture. i hope i understood the question. thanks Doug

 

[SREisme]

Thanks for the comments. My end goal is to determine the point at which a low water cutout must be installed on a waste heat boiler to prevent a steam explosion caused by reintroducing water to a dry boiler. We currently operate a number of waste heat steam generators with no low level safety systems, as it appears to be an industry practice to not be concerned with steam explosions with hot gas temps in the 700F range or less. Although I agree there is little likelihood of a steam explosion with a hot gas temp less than 700F, I don't know where to insist on installing low water trips, cutouts, alarms, etc if gas temps are above 700F. It seems there is a "magic number" somewhere between 800 and 2500F, but where is it? I've seen a manufacturer of steam generators claim they provide run dry capability to 840F. Macmet provides experience to 860F and an upper limit of 1050F.

As for ASME eliminating steam explosions of boilers, refer to the paper at

http://www.doli.state.mn.us/ccld/PDF/tenn_boiler_report.pdf

that documents the Tennessee Dept of Labor's investigation of a boiler explosion at an extrusion plant in June 2007. Root cause was determined to be reintroducing water back into a fired boiler that had run dry, due to poor maintenance of the low water cutouts.

This report, as well as other more historical reports on steam ships and steam locomotives, lead me to believe that it is
very possible to explode a fired boiler if water is reintroduced to dry tubes, not just from fuel explosions or overfiring/overpressuring. Therefore, it should be possible to explode a "unfired" waste heat boiler if the process gas approaches the temperature of a fired boiler, what I've seen to be in the 2000+F range.