water tube boiler tube leak that cannot be found 1

[db0]

Our Nebraska D type package water tube boiler has developed a leak that we cannot locate.

This boiler was put into service in 1990. Its MAWP is 200 psi and its max rating is 55,000 #/hr.. We operate it at 90 psi and average 30,000 #/hr. during the winter months.
When the boiler is in standby mode (Kept hot and pressurized by regular, low-fire warming - maintaining pressure between 25 psi and 85 psi until it is needed.) we experience a steady trickle of warm water running out of the front and rear weep holes in the bottom setting framework. The flow rate of the trickle is about 2 ga./hr. and we can tell that it is boiler water and not condensed steam because we read feedwater treatment chemicals when we perform tests on samples of this water - specifically polymers and oxygen scavengers.

We have cut two 1' X 1' windows in the exterior economizer sealing wall along the mud drum and see no signs of water on the tubes or the drum. We see no signs of leakage in the furnace or in the stack breaching or along the soot blowers. Additionally, we cannot hear a leak. All of this leads me to believe that it is not a generating tube that is leaking.

Surprisingly, When we put the boiler on line, the leak stops and the floor under the mud drum dries up.

I don't think that we have to hydro this boiler because we already know that we have a leak. My question is as follows: How can I locate this leak without tearing the boiler apart? Is there an N.D.T. (eddy current test) or dye test that I can apply that is not outrageously expensive? Is there any way to air test each tube individually and can I rig something together with spare parts one would find in an operation such as ours? Please keep in mind that this is a city-owned heating and chiller plant and they maintain very tight purse strings and always expect something for nothing!

I hope someone can give me some ideas because I am fresh out.

 

[metengr]

I would guess that you have one or more tube roll leaks in the lower (mud) drum. These are common in low pressure package boilers after so many start-up and shut down cycles. I would squeeze (pressure test) the boiler, and confirm/check for tube roll leaks.

 

[db000]

Thanks metengr! I'm not quite sure what you mean by "squeeze (pressure test) the boiler". Do you mean a hydro test? I am looking for the specific leaker. We already know that we have a leak. Please forgive my ignorance - I'm just a blue collar working stiff. By-the-way, I'm at a different computer.

 

[metengr]

db000;
Yes, except a pressure test using water is conducted at or below operating pressure.

 

[db000]

Thanks metengr! How can I determine which tube(s) is/are leaking. I suspect the leaker(s) is /are on the firebox side of the drums, either behind the economizer/furnace partition or below the firebox floor - all out of sight.

The reason I need to know this is that the city money people will not pay for a re-tube. They have established this track record with our two other older, Keeler water tubes. The best that I can do is to find the tube(s) and plug it/them.

 

[metengr]

You can try to pressure test the boiler and while the pressure is applied use fiber optics (small borescope) to access the OD surface of the mud drum, and check for leaks. You will need to provide a series of small access openings to observe the drum OD surface with the borescope.

 

[db0]

Thanks very much, metenger! This sounds like a good plan. I believe that we will be able to borrow one from our chemical service rep..

 

[Keith53]

The reason you don't see evidence of the leak when online is because the water flashes to steam and escapes up the stack.

Go back and check your readings to a period before the boiler began wetting the floor......You will see that your exhaust temps were higher at a given firing rate back then.

A leaking tube or rolled joint (if bad enuff) can be detected by lower stack temps and increased feed meter readings.

Keeping good logs and maintaining proper indications are very necessary.....With proper logs you can graph out trends that will pinpoint problems that would otherwise go unnoticed.

In this case, if the boiler was operating, you would never know it had developed a leak until you shut it down for off season and it finally wet the floor.....If instead you had kept very accurate readings and transferred them to graph, you would see a drop in stack temp with a corresponding increase in feed water consumption and that would indicate a possible tube/joint leak.....With good readings and graphs you can see trends that indicate lots more than the daily log sheets alone.

 

[db0]

Thanks Keith 53. We do keep pretty good records. I went back and checked and there was a slight enduring drop in stack temp.- when compared to the beginning of the heating season.
Thanks for the help!

 

[unclesyd]

We have "D" tube boilers (7), actually vaporizers for a thermal heating fluid that developed leaks that had very similar behavior as your problem. After trying to catch the offending tube/tubes I convinced management to just reroll the all the tubes. This turned out to be far cheaper that a production outage to locate and repair the leaks.

 

[db0]

Thanks unclesyd,

This has been discussed. We have also talked about simply running the boiler 'til the offending tube(s) blow(s). Then there will no guessing at which one(s) is/are the leaker(s).

Did you re-roll these tubes with in-house staff or did you contract the job out?

If you contracted this job out, was the cost enough to take you breath away? I'm not asking for a dollar amount - just your reaction to the cost.

 

[Keith53]

Odds are you can locate the leak with a hydro.

Fill the boiler completely to the top with room temp water, then crack in the feed water till it hits working pressure, which I believe you said was 90 PSI, then isolate feed water.

This will allow you to avoid removing or gagging the safetys, and also the little bit of feed water used to generate pressure will not make the boiler hot and unsafe....If you have a chemical injection pump piped to the boiler already, simply fill the chem tank with water and use the injection pump to press it up to 90PSI.

Once pressed up to 90 PSI crawl inside the firebox and look/listen for the leak.

I bet you can isolate it to a very small area down at the mud drum....You may not be able to see it, but I bet you can hear it and maybe see a wetted area developing on the floor refractory.

If so, then you can decide to try rerolling only the tubes in that area or all of them.

Friend, trust me, you are better off hiring a qualified repair concern to reroll your tubes. It takes years of experience and a special feel to get it right and do no harm.

 

[deanc]

Sounds like a tube roll joint to me.

Agree with unclesyd....reroll steam and mud drum. The longer you let it leak the worse the damage. May rot out the underside of your floor tubes.

If you do not have the rolling equipment or experience it would be well to seek a boiler repair firm with a R stamp.

Cost of the equipment alone will cover that,let alone the damage caused by someone who is inexperienced.

 

[unclesyd]

We hired a contractor to reroll the tubes under the watchful eye of a retiree brought back in for the first two jobs. The rerolling was accomplished with no problems.
Even though we had all the equipment and people, none had much experience in rerolling the larger boiler tubes, we elected to go with an outside contractor. The contractor was recommend by our high temperature insulation contractor who later absorbed the repair company.

The amount of reroll was actually gaged by watching the light scale on the drum surrounding the tube. When scale cracked the rolling stopped. Our people gauged several of the tubes before and after with the reroll numbers well within the variability of the original installation.
Because of a fire we had to replace the "D" tubes, that had been rerolled, in one boiler and during this operating the tube holes in both drums were inspected and gauged with all being within OEM specification.

 

[db0]

Thanks to all for taking the time out of your busy lives to help me out with all of this valuable information. I will use it when I talk to the bigshots and the money people.

P.S. This is the first time I have had to pose a question on this web site. I am most grateful to have access to it. Perhaps a donation should be in order!

 

[chief]

An alternative to pressurising your boiler with the feed water pump is to drain down and hook up a plant air hose to your drum. Pressure up to 90 psi and check for your leak . An air leak is always discernible by the sound, or if very small use a soapy solution to brush around the tube joints. You will blow some pretty bubbles in the area of the leak.

 

[db0]

Thanks chief (marine/ocean) - I like this idea. Maybe if we try both tests we should surely be able to discover the leak.

 

[RRhoads]

My sympathies....I have been involved in finding a hard to find leak in a Nebraska "D" Style Packaged Boiler as well. We performed several hydrostatic pressure tests, cut away areas (patches) in the boiler outside casings, etc. looking for this leak. In one particular and frustrating case, we finally found the leak in a tap tube that is a long threaded coupling welded in the top steam drum head (used for chemical feed distribution). This tap tube is located below the drum manway ring. There was a hole in this tap tube between the drum head/tap tube weld joint. The hard part with finding this leak is that you can not see or hear it. Also the outside casings in this area are 100% seal welded to insure no flue gas leakage. The leak then takes the path of least resistance between the refracory wall, insulations and casings to the base of the boiler (nowhere near the problem area). I am not sure if this is true in your case, however it is worth exploring and or keeping in mind.