Sulfur Recovery Unit Reaction Furnaces Waste Heat Boilers Ferrule - - PowerPoint PPT Presentation

Sulfur Recovery Unit Reaction Furnaces Waste Heat Boilers Ferrule Failure Root Cause Analysis

Presented By: Muhammad Bashir Khan – Manager, Technical

SRU and Ferrule Configuration

• The Sulphur Recovery Units (SRU's) have 2 reaction

furnaces in the thermal section of the Claus reaction

SRU and Ferrule Configuration

• The tube-sheet of the waste heat boiler (WHB) is protected

from the high temperature (approximately 1100°C) in the furnace by a lining made up of 2-piece ceramic ferrules (a hexagonal head and a cylindrical stem insertion).

SRU and Ferrule Configuration

• The 4,200 ferrules make up

the tube-sheet lining as can be seen here.

• The pattern is a result of

the beige mortar and the white mastic grouting.

– Both types of grout are white after dry-out

Inspection Findings

• As part of the Warranty Shutdown (WSD) scope, one year

after start-up, the two reaction furnaces of Unit 751 were scheduled to be opened and inspected in January 2016.

– 99% of ferrules in both H-101 and H-102 found damaged

• Circumferential crack of stem at tube to tubesheet interface
• Corrosion deposit on stem at 6 o’ clock at crack point impregnating

into ceramic stem body

• Hot Sulphidation damage to tube end and tube to tubesheet strength

weld

Inspection Findings

Inspection Findings

• The two reaction furnaces of three other units were

inspected over the course of the next 12 months and the results were very similar only varying slightly by magnitude.

Investigation Details

• After the initial inspection of the first 2 units (4 furnaces) in

January and February 2016, representatives from all stakeholders worked together as a joint investigation team. This included:

– Al Hosn Gas / Occidental Petroleum (Owners) – Saipem (as EPC)

• Fluor (as Licensor)
• Siirtec Nigi (Furnace OEM)
• Industrial Ceramics (Ferrule OEM)

– Other 3rd Party Consultants

• Nasato Consulting
• The Welding Institute
• Alberta Sulphur Research Limited
• Porter McGuffie

Investigation Details

• After analysis of the findings, the failure mechanism was

narrowed down to be one of the following four:

– Hot Gas Bypass

• Hot H2S passing between the ferrules, as a result of poor installation, resulting in hot

sulphidation on the tube end.

– Sagging Ferrules

• Due to the weight of the ferrule hexagonal heads, the entire lining of ferrules would

sag and result in compression of the ceramic fiber paper on the stem and creating a hot contact point which then induced hot sulphidation corrosion on the tube end.

– Insufficient Thermal Protection by Ferrules

• The ferrules were not providing sufficient thermal protection to the tube end leading

to hot sulphidation on the tube end.

– Two Phase Process

• This is explained in more detail later in the presentation

Investigation Actions

• The first 2 units (4 waste heat boilers) inspected in January

(751) and February (754) 2016 had all ferrules removed and the tube-sheet cleaned of residual Sulphur (by hand-tools)

• The residual Sulphur was far greater than expected
• The full set of ferrules were replaced with supervised

installation by the OEM

• New ferrules were ordered for the remaining two units (4

waste heat boilers)

• These were scheduled for inspection and repair in October

2016 (752) and January 2017 (753)

Investigation Actions

• Further remedial actions were identified and implemented

to address the various possible mechanisms.

– Re-run of CFD analysis of tube-sheet lining design

• Carried out by various parties

– Review and revision of shutdown procedures

• Increased fuel gas sweep flow rates (from 10% to 30%)
• N2 sweep of acid gas lines

– A follow-up inspection of 751 was planned for February 2017 as there was inconclusive evidence to definitively support one mechanism over all others

Further Investigation Findings

• As mentioned earlier, Units 752 and 753 were inspected in

October 2016 and January 2017 respectively.

• These were shutdown using the revised procedure

developed as part of the investigation.

• When 752 and 753 were inspected, there was negligible

residual Sulphur in the furnace compared to 751 and 754 (original procedure).

• This confirmed that residual Sulphur found in 751 and 754

could have been prevented and that the original procedure was not fit for purpose.

Further Investigation Findings

• A year after reinstatement, the furnaces of Unit 751 were

shutdown and inspected for a second time as per the follow-up actions.

• The inspection process involved taking samples of ferrules
• ut of the tube-sheet lining at pre-determined locations

relevant to the previous inspection and documented:

– Location of ferrules being removed (in clusters of 7 or 19 to maintain the ferrule wall) – Condition of the mastic and mortar (for poor quality in the installation or dislodgement through transient conditions) – Corrosion growth, cracking (circumferential or longitudinal)

Further Investigation Findings

• Representatives from key stakeholders were present for the

duration of the investigation

• The sequence of events was as follows:

Further Investigation Findings

• The main findings were:

– No residual Sulphur – Only 14 and 118 ferrules (out of 4,200 in each furnace) were found affected with corrosion scale at 6 o’ clock position – No correlation between the affected ferrules and their installation condition (i.e. poor grouting) – No correlation between the affected ferrules and the thickness of ceramic fiber paper wrap on the stem – All affected ferrules located at the top and bottom extremities and on the left hand side of the tube-sheet.

Further Investigation Findings

• Locations of affected ferrules in each furnace

Failure Mechanism Analysis

• For each possible failure mechanism, the findings from the

follow-up inspection of 751 concluded the following:

• affected. No further hot

• affected. No further hot

Failure Mechanism Analysis – Two Phase Process

• Step 1

– Starts with wet Sulphur corrosion where moisture condensed and collected (being coalesced by the ceramic fiber paper) at the tube end and in the presence of Sulphur after a shutdown of the unit to ambient conditions. – This results in a corrosion scale to grow which impregnated the ceramic fiber paper and the ceramic ferrule stem.

• Step 2

– After the furnace is then restarted, the corrosion scale deposit forms a thermal bridge allowing a heat transfer to the tube end and starts hot sulphidation corrosion. – The cracking in the ferrule stem was due to the dissimilar properties of thermal expansion between the ceramic stem and the impregnated scale which on cooldown resulted in mechanical stress from the scale “wedge” and caused the brittle ceramic stem to crack circumferentially.

Failure Mechanism Analysis – Two Phase Process

Normal Running Conditions

• No Hot Gas Bypass
• Robust Thermal Protection
• No Mechanical Stresses on Ferrules
• Tubesheet Lining Intact

Failure Mechanism Analysis – Two Phase Process

Shutdown follows procedure

• Acid Gas Cut-off
• Fuel Gas Sweep (10% of Normal Flow)
• Sulphur Rundown Ceases
• Cool-down

Failure Mechanism Analysis – Two Phase Process

Whilst Shutdown

• Residual Sulphur Present (ineffective FG sweep)
• Moisture Present (process & forced air)
• Carbon Steel Present (tube and tube-sheet)
• Time (dependent on duration of outage)

The wet Sulphur corrosion process is aggressive

• Laboratory proven
• Does not require oxygen
• Can occur at ambient conditions

Failure Mechanism Analysis – Two Phase Process

Corrosion Scale Deposit Grows

• Moisture Coalesces in Ceramic Fiber (stem wrap)
• Gravity Pools Water at Base of Ferrule (6 o’ clock position)
• Wet Sulphur Corrosion Scale Grows
• Up from tube end
• Through ceramic fiber wrap
• Impregnates into ceramic stem of ferrule

Failure Mechanism Analysis – Two Phase Process

Not Known That Wet Sulphur Corrosion Has Occurred

Failure Mechanism Analysis – Two Phase Process

Running Conditions Are Now:

• No Gas Bypass
• Compromised Thermal Protection Due To Thermal Bridge
• Through ferrule stem
• Through fiber wrap
• All the way to tube and tube to tube-sheet weld
• No Mechanical Stresses on Ferrules
• Ferrule Lining is Now compromised

Failure Mechanism Analysis – Two Phase Process

Heat Conduction To Tube End

• Temperatures in excess of 250 °C
• H2S Present
• Carbon Steel Present
• Time (dependent on duration of operation

Heat Affected Zone

Failure Mechanism Analysis – Two Phase Process

Shutdown follows procedure

• Acid Gas Cut-off
• Fuel Gas Sweep (10% of Normal Flow)
• Sulphur Rundown Ceases
• Cool-down

Failure Mechanism Analysis – Two Phase Process

Dissimilar Thermal Expansion Properties Between

• Ceramic Stem
• Scale Wedge

On Cooldown, Forces Act On Brittle Ceramic

Failure Mechanism Analysis – Two Phase Process

Results In Circumferential Crack To Ferrule Stem

• Significant hot gas path to tube
• Unfit for re-use
• Ferrule replacement required
• Scale build-up not visible until ferrule is removed
• Removal & inspection of tube-sheet lining required

Conclusions

• The only failure mechanism that came close to meeting the expected

conditions found in the follow-up inspection was the “2-Phase Mechanism”

– The magnitude of the failures in each pair of furnaces correlates to the shutdown procedure and duration as well as time in service. – The reason for the minimal number of failed ferrules in the follow up inspection is assumed to be due to their location and the residual Sulphur had not been fully removed and resulting in further wet Sulphur corrosion prior to restart.

• The design of the waste heat boiler and tube-sheet lining (specifically

the ceramic ferrules) is robust and requires no modifications.

• All focus must be on ensuring that no residual Sulphur is present if a

furnace will be taken to ambient conditions for any period of time.

Q & A