Scribd
Upload
126 views5 pages

The Auxiliary Boiler Failures: Tom Herman, Mickey Roberson, Ron Parr

The document summarizes auxiliary boiler failures that occurred during restart of a plant after a 9-month outage. Tubes failed in coils D and E, causing a 2-month delay. Investigation found iron oxide deposits in the boiler downcomers were flaking off during startup and blocking water flow, causing tubes to overheat and fail. All risers and downcomers were hydroblasted to remove built-up iron oxide deposits, and the plant was restarted without further issues.

Uploaded by

varatharajan g r
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
126 views5 pages

The Auxiliary Boiler Failures: Tom Herman, Mickey Roberson, Ron Parr

The document summarizes auxiliary boiler failures that occurred during restart of a plant after a 9-month outage. Tubes failed in coils D and E, causing a 2-month delay. Investigation found iron oxide deposits in the boiler downcomers were flaking off during startup and blocking water flow, causing tubes to overheat and fail. All risers and downcomers were hydroblasted to remove built-up iron oxide deposits, and the plant was restarted without further issues.

Uploaded by

varatharajan g r
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

The Auxiliary Boiler Failures

Tom Herman, Mickey Roberson, Ron Parr


Koch Nitrogen Company

Introduction The A and B coils had been replaced in 1998. The


other three coils were from the original plant construc-

A
t the Koch Nitrogen Complex in Sterling- tion in 1966. All coils are manufactured of carbon steel
ton, Louisiana, on September 6th two tubes and only E coil has fins.
failed in the Auxiliary Boiler. This event
occurred during the re-start of the plant after a 9-month
market driven outage. A decision was made to change The Failure Events and Repair
out three of the Auxiliary Boiler coils that were from the
original construction and re-start the plant. On restart, The first failure occurred on September 6th during
single tubes from two of the new Auxiliary Boiler coils the start-up of the plant. This start-up begain at 1300 on
failed on October 1st. Overall, these events caused a 60- the 6th and the failure happened at 1830. The plant was
day delay in the restart of the plant. at a point during the start-up that the firing on this Aux-
The events that lead up to these two failures, the final iliary boiler was maximized and all of the blowdowns in
failure analysis and the repairs that were made to bring this system were being opened every 2 hours. This
the plant back on stream will be discussed in this paper. “purging” is needed during start-up to transport iron out
of the boiler. A tube on the hot side of “E” coil failed.
This failure then shattered a tube at the upper header
Background connection of “D” coil.
Koch Nitrogen has two Kellogg designed Ammonia A preliminary investigation was held and the deci-
plants at the Sterlington Complex. The first was built in sion was made at this time to replace “C”, “D” & “E”
1966 and the second was built in 1977. This site has the coils in the Auxiliary boiler. The root cause for failure
capability to manufacture over 1,000,000 tons of am- was believed to be “End of Life” for the coils. An as-
monia each year. This site is also located on the Am- sumption was made that the “E” coil overheated and
monia pipeline that allows most all of the ammonia pro- failed and the force of this failure caused the “D” coil
duced in Sterlington to be shipped to the Midwest for tube to shatter. Pictures at the end of this paper show
sale. The failure that will be discussed occurred in the these failures. These pictures also show that the “D”
plant built in 1966. Both of these plants have a standard coil tube had a crack that started in the ID of the tube
Kellogg Reformer with a radiant section where the cata- and went most of the way around the tube. This was the
lyst tubes “Harps” are located, a “Hot” and “Cold” side only tube that exhibited this type of failure, but analysis
convection section and an Auxiliary Boiler for supple- of other tubes in the same area at the header showed
mental steam generation. more evidence of internal cracking. This cracking was
The failures occurred in the Auxiliary Boiler sys- caused by corrosion fatigue. As this type of crack is
tem. The Auxiliary Boiler is a separate firebox that is caused by cyclic tensile stresses, a repair strategy of re-
part of the Primary Reformer system and consists of five placement was appropiate. No good explanation for the
coils (A-E). The coils are arranged in the firebox as overheating of “E” coil was available at this time.
shown in the drawing on the next page.

2002 3 AMMONIA TECHNICAL MANUAL


101-AB (Auxiliary Boiler Coil Lay-out)
Coils
“E” “D” “C” “B” “A”

Burners(5)

“E” Coil four (4) Rows of Finned Tubes, “D” Coil four (4) rows of Tubes, “C” Coil three (3) Rows of Tubes
Plan View

Work commenced on replacing these coils and in- Failure Inspection and Final Repair
formation on the failure was gathered and reviewed dur-
ing several meetings. The coils were replaced with the The “D” coil was removed from the firebox for re-
same design as was originally installed in 1966. Tube pair. While the coil was on the ground, the bottom header
sections from all three coils were sent to an independent was inspected and a pile of flaked iron rust was noticed
metallurgical lab for analysis. This analysis showed that just opposite of the tube that failed. This lead to an un-
“E” coil failed due to short-term overheating and the derstanding of how the tubes were overheating. It was
fracture of “D” coil was due to the presence of internal thought that the iron oxide that collected on the down-
corrosion fatigue. At this time, no definitive reason was comers was broken lose by the heating up of the system.
found that explained “E” coil failing. This data seemed This oxide then flaked off and fell into the bottom headers
to confirm the “End of Life” theory for the failures. of the Auxiliary boiler and the waste heat boilers in the
The plant was in the process of starting up after the system. The start-up procedure for the boiler requires that
repair when the Auxiliary boiler ruptured again on Oc- the individual blow down valves on each coil be opened
tober 1st. This time, a tube in “D” coil overheated. The two to three times during the start-up. The downcomers
force exhibited by the tube rupturing caused the tube enter the bottom header at the opposite end of the header
isolators for “C” coil to separate causing cracking in from the blowdown connection. This meant that the iron
several tubes. This failure happened at about the same oxide entered the bottom header and collected near where
place in the start-up when the boiler was being fired at it entered. As the blowdowns were opened, the oxide mi-
its maximum level. An investigation was continued into grated toward the blowdown valve. This oxide continued
the root cause of these failures. At this time, a system to build up in a pile near the blowdown valve until the
failure was suspected. pile was large enough to block the 5/8” hole that allows

AMMONIA TECHNICAL MANUAL 4 2002


water into the tubes for the “D” coil (see drawing below). In addition to the cleaning of the risers, downcomers
This drawing also shows how the oxide flakes built up and bottom headers, we had to assure ourselves that
and blocked the hole for the tube that failed. none of these chunks of iron oxide made it into the
The blockage was only for a short time, long enough bayonet tubes in the 101-C’s (waste heat boiler down-
for the tube to be starved for water and overheat. This stream of the Secondary Reformer). The two bundles
was verified by the pile of oxide that was still left in “D” were removed from their shells and sent to Gulf Engi-
coil header under the tube that failed. neering for inspection. The inspection entailed remov-
In order to follow up on the system failure scenario, ing all of the tubes and separating the bayonet from the
a robotic video camera was used to inspect all of the ris- outside tube. Some iron oxide chunks were found in the
ers and downcomers. The downcomers to the 101 and bottom of one of these exchangers.
102 and 103-C’s looked clean. When the “Rovver” was After all of the cleaning was completed, the video
sent down the Auxiliary Boiler downcomer, a thick rover was sent back down the risers, downcomers and
layer of iron oxide was noticed on the horizontal runs. bottom headers to assure that the cleaning was com-
This confirmed the failure mechanism. Iron oxide was pleted. No rework was required. All of the boiler pip-
caked on the horizontal runs of the downcomers to the ing was then reconnected, the 101-C’s were reinstalled
Auxiliary boiler. and the plant was re-started without incident.
It was apparent that at this time all of the risers and
downcomers needed cleaning. Chemical cleaning of the
boiler was investigated, but the quantity of rust in the Root Cause
boiler was so large that this method of cleaning would
The root cause for these failures is imbedded in how
not work. The only way to properly clean the system
all of this iron oxide was allowed to deposit on the boiler
was hydroblast all of the piping. A 3500-psi machine
piping and then flake off during start-up. Two factors al-
was used with a “pineapple” sprayhead. The hose that
lowed this oxide to deposit and flake off. They are:
the sprayhead was mounted on could only effectively
maneuver beyond two (2) turns; so many cuts were
1. Not properly laying up the boiler during
made in the risers and downcomers to access all of this
times the plant is shutdown.
pipe. The bottom headers of the coils were cleaned by
2. History of not being able to maintain a tight
welding on a full size butterfly valve on the blowdown
control of the chemistry in the boiler and
end of the header. After the coil was flooded with
other past chemistry issues in the boiler
demin water, the butterfly valve was opened which al-
system.
lowed all of the water to flush through the bottom
header. This was done several times.

2002 5 AMMONIA TECHNICAL MANUAL


Lessons Learned soon as the plant is secured and mechanically intact and
then maintaining this nitrogen purge until start-up.
The most important things these events teach are Inspect risers and downcomers after 15 years of op-
just how important it is to properly protect the boiler eration. That is the most important point that can be
system when the plant is not operating. It is imparitive taken from this paper. Check boiler piping for signs of
that a nitrogen purge be placed on the boiler system as corrosion. This one was a surprise. There will not be a
next one.

Figures 1 and 2. First failure of coil – “E” Coil thin lipped fish mouth failure.

AMMONIA TECHNICAL MANUAL 6 2002


Figures 3 and 4. First Failure - “D” coil that fractured after the failure of the
tube in “E” coil. These pictures show that this tube had a crack in it that
started from the ID and went most of the way around the tube.

2002 7 AMMONIA TECHNICAL MANUAL

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy