Failure of an Economizer Bank in operation…A case study

Background information

An economizer part was reported having failed during operation which was ordered as a re- placement item for an existing boiler constructed in accordance with Indian Boiler Regulations 1950(IBR). The process followed for the replacement of this boiler part was as below:

  1. The economizer bank was designed and manufactured at a shop according to the regulations of IBR with inspection by a Competent Person (CP). The basis for de- signing this replacement part was the original design condition of the boiler and economizer.
  2. The certified part was assembled at the site by a “Repairer” approved by the Director of Boiler of the state where the boiler was registered.
  3. Inspection and pressure test witnessed at repair site by a Competent Person
  4. Records of the boiler/ “Memorandum of Inspection“ book was updated following satisfactory completion of repair and tests.

Some details of the item under study:

1. Item: Economizer bank for a cogeneration boiler

  • Maximum allowable working pressure: 30kg/cm2
  • Maximum design temperature: 235deg C
  • Material: Carbon Steel
  • Header dimensions (inlet and outlet): OD 168.3 X 21.95mm
  • Intermediate and drain headers: OD 73 X 7.01 mm
  • Tube dimensions: OD 38.1 X 3.66mm

2. Purchaser/user: A petroleum refinery in India (hereinafter referenced to as user)

3. Manufacturer: A small scale fabricator which manufactured the economizer part hav- ing experience in similar IBR part

4. Quality Plan: Approved by the purchaser/user describing scope for customer inspection as well as scope for IBR inspecting authority

5. Drawing & calculations: Approved by inspecting authority

Chain of events

The manufacturer of the part received the purchase order from the user for the supply of an economizer panel as a replacement part for an existing boiler. The design and drawings were made by the manufacturer based on the design data of the existing boiler. The quality plan was prepared by the manufacturer and approved by the user.

The manufacturing was carried out under the inspection by a competent person according to the scope described in IBR and forms VII and VIII were issued by the inspecting authority and the manufacturer respectively.

The economizer panel was received at the user’s site, assembled by an approved ‘repair’ contractor and hydro tested before being put to use. The repair activities and inspection at the site were carried out by the “Repair” contractor under the supervision of the Inspecting Authority of that state.

The economizer panel failed as reported by the user after fifteen days of operation with multiple leakages found. The leakages were mostly found on intermediate headers and drain headers at locations where stub tubes were welded onto them. A sketch of joints with the location of leakages is given in Figure 1.

Figure 1

It was noted during the investigation that the economizer failed a few times during operation after assembly and testing. In situ repairs were carried out but without much benefit and finally, the economizer bank had to be isolated from the feed water supply.

A study was conducted to identify possible causes of failure. Information and data were collected from the user and the manufacturer of the economizer about its design basis, materials used, fabrication methods, quality control, inspection and test performance. Also received photographs (Figure 2 to 7) of the shop welded joints in the failed economizer which were shared by the user. The photographs were taken several days after the economizer was put into operation and therefore the parts were covered by soot, scales on surfaces. However, these photographs did reveal indications of poor quality of shop welds at certain sections:

  1. Incomplete penetration and inadequate root fusion for the stub to header joints as it appeared
  2. Unusual profile and excessive reinforcement at certain locations of the tube to tube joints
  3. Uneven fillet size
  4. Apparent local repair on joints by adding weld bead
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7

From the shop Quality Control record shared by the Manufacturer, following observations were made :

  1. The QAP prepared by the manufacturer and approved by the user didn’t include fabrication stages like preparation of joints and weld set up an inspection. There was no evidence that the manufacturer QC performed stage inspections to verify weld preparations and set up of stub to header joints..
  2. There was no stage identified for the competent person to inspect weld set up for tubular joints as required by Appendix J of IBR. It was not evident how the stub to header joints (groove cum fillet) was examined since there were no such stages identified for the QC inspector nor any inspection reports available.
  3. The welding log sheet indicated that only one certified welder was engaged to weld all tubular joints close to 650 in number.
  4. All welding is done using Shielded Metal Arc Welding (SMAW) process.

Possible causes

The investigation pointed at a number of lapses at the manufacturing stage and inadequate quality control performance leading to poor workmanship. But the study also brought out the importance of weld joint design and its impact on the ease of fabrication and improved workmanship.

The detail of the header to stub weld groove showed in the drawing is in Figure 8:

Figure 8

The weld joint chosen was for a partial penetration groove cum fillet weld similar to Fig. 365/17 in IBR. According to this design, the clearance between the OD of the tube stub and the ID of the groove on which the stub rests is 0.5mm. From the fabrication point of view, it is rather difficult to maintain unless precision machining and close quality control checks are applied. The tube OD may vary over its nominal dimension and larger the OD means tighter joint unless the OD is machined to suit the groove dimensions. The accuracy of the groove dimension and its finish again would largely depend on the type of machining, tooling and degree of quality control interventions. Accessing the root of such a joint and fusing it with a stick electrode is difficult unless there is enough room at the location where the stub end is resting on the header. The unfused root of the stub would lead to crevices and cracking eventually under thermal stresses during operation.

The focus of this study has been to identify improvement in the weld design process which is critical to workmanship. For the weld joint design selected as above, the designers should provide more specific details about the minimum clearance required, tolerance and dimension control of the tubes OD. Consideration should be given to the OD of the header to en- sure sufficient land at the face of the groove for good fusion. A suitable manufacturing plan should be made based on the design, fabrication process, materials, resources with identified QC intervention stages. It is essential for the shop QC to inspect the groove after ma- chining and inspect again when the weld set up of stub to the header is ready. It is also important to ensure that the welders are qualified for welding such joints in a position to suit the assembly set-up. Selection of the correct size of SMAW electrode with an appropriate cur- rent for the complete fusion of the root is vital. The Quality Plan must identify the stages of inspection in detail for groove machining, stub tube end preparation, weld set up and inspection of welds. Preparing a mockup assembly to replicate such joints and visual inspection of welds after sectioning is a good way to assure workmanship.

As part of this study, we also looked at some alternative weld designs and practices for improved performance. The selection of joint design will however depend on the criticality of product design, pressure, temperature, medium, machining capacities etc.

Alternative#1

Figure 9

Complete penetration and fusion are ensured. However, the preparation would require additional machining capabilities and therefore cost of manufacturing would go up. Suitable for very high-pressure applications.

Alternative#2

Figure 10

This type of preparation would ensure complete penetration and fusion at the root. Relatively easier to prepare however , the tubes would require machining .

Alternative#3

Figure 11

This type of preparation would facilitate fusion at the tube ends for partial penetration joints. Easy to machine with some adjustment in tooling to provide a taper on the face where the stub tube rests.

Conclusion:

Good workmanship is not an outcome of the operator’s skill alone. It is a cumulative output of effective control right through processes like designing, materials, fabrication, testing and inspection. Weld joint design is critical to the quality of fabrication and the performance of a product. A drawing with details of joint dimensions, tolerances, size and shape of the weld is the key to compliance and better quality. A designer should take into consideration the shop capabilities of machining, ease of fabrication, welding process and skill demands of a welder while designing weld joints. Good engineering is the foundation for a sound manufacturing and quality planning process which are vital for the good performance of a product.

References

  1. Recent experience in condition assessment of boiler header components and sup- ports by James P. King DB Riley, Inc. Worcester, Massachusetts
  2. ASME Sec I Rules for construction of power boilers, 2019 Edition
  3. Indian Boiler Regulations, 1950

Parthapratim Brahma, Area Technical & Quality Manager, South Asia Middle East and Africa, Lloyd’s Register Marine & Inspection Services India LLP

    Leave Your Comment

    Your email address will not be published.*

    8 + 11 =