skip to Main Content

Introduction

One Eighty was appointed by an air conditioning contracting supplier and installer to investigate the failure of a heat exchanger unit. The heat exchanger was operated in a butcher’s curing/drying room and was in operation for approximately 18 months before failure.

The heat exchanger allegedly failed due to leakage of the liquid coolant (glycol) during operation. An initial attempt was made to repair the leaks using epoxy putty, but the leaks persisted. The initial leak site was unidentified; however, it was suspected to have originated in the copper tube coils adjacent to the larger copper tube manifold. The unit was clearly corroded; therefore, it was imperative to determine and identify the corrosion mechanism and the root cause for the failure of the heat exchanger. One Eighty was provided with only a section of the heat exchanger for testing, as shown in the images below.

Failure investigation of a corroded heat exchanger
Failure investigation of a corroded heat exchanger

Approach and Methodology

Due to the lack of identification of the failure initiation site, the heat exchanger product datasheet, the paint coating material specification and application procedure, or datasheets of any cleaning agents, we selected a set of metallurgical tests that would aid in the identification of the corrosion mechanism for this root cause failure investigation.

These tests include but not limited to:

  1. Visual inspection
  2. Sampling selection and designation for metallurgical testing
  3. Macro-examination by way of stereomicroscopy
  4. Microstructural analysis by way of light microscopy
  5. Scanning electron microscopy (SEM) and Energy Dispersive Spectroscopy (EDS)
  6. Spectrographic Analysis by way of Optical Emission Spectrometry (OES)
  7. Vickers hardness testing

Results and discussion

From visual inspection, it was evident that the entire heat exchanger was subject to corrosive attack. The most affected areas were where the paint coating was damaged. The corrosion was predominantly on the outer surface.

Corrosion of the copper tubing at the interface with the enclosure plate
Corrosion of the copper pipes at the welded components
Corrosion of the enclosure plates - RHS
Corrosion of the enclosure plates - LHS

Macro-examination through the tube cross-section showed localised pitting corrosion on the outer surface of the copper tubing, suggesting that the copper was exposed to the elements within the environment in which it was operating. It was also noted that the quality of the joints in the copper tubing was inconsistent. The copper tubes were confirmed to have corroded prior to the epoxy putty repair, with sections of the paint layer not providing sufficient corrosion protection.

Macro image - Sample A Location a at 2x magnification
Macro image - Sample A Location2 at 3x magnification

The microstructure of the copper tubing was consistent with copper; however, inconsistent grain sizes indicate that improper heat input may have been present during the fabrication/joining process.

Microstructural image - Sample A Location 1 at 50x magnification
Microstructural image - Sample A Location 1 at 100x magnification

From the microstructural images, it was clearly evident that the copper material had suffered corrosive attack from the exterior side of the tube, which presented itself as pitting. Several of the welded joints also showed to have completely dislodged itself from the copper material, which further indicated poor welding/brazing techniques were implemented in the fabrication.

Microstructural image - Sample A Location 3 at 50x magnification
Microstructural image - Sample A Location 3 at 100x magnification

SEM and EDS analysis confirmed the presence of chlorides near the external surface of the copper tubing, which are typical constituents of paint coatings and/or may have originated from external elements within the environment. The iron that was suspended between the paint coating and material substrate suggested rust contamination was present during the paint application. Elevated levels of sulphur was detected in the paint coating. Sulphur is more harmful to copper than chlorine and would have reacted with the copper to form copper sulphides, which would have likely initiated or accelerated the corrosion of the copper tubes. It is more likely that the chlorides were introduced from the salts used to cure the meat in the processing plant where the heat exchanger is located. These chlorides would react with the copper, resulting in corrosion by forming copper chlorides.

Spectrographic Analysis confirmed that the steel frame plate was AISI 1005, a typical low-carbon steel with low corrosion resistance. It was clear that the current surface coating system was not appropriate due to the lack of corrosion protection noted on the frame. The copper was confirmed to be UNS C18150, a high-copper alloy with good corrosion resistance properties. However, high-copper alloys are susceptible to pitting corrosion when exposed to the wrong environmental conditions. The most corrosive elements to copper are typically chlorides, sulphur compounds and other compounds with heavy-metal salts.

Vickers Hardness Testing showed that the hardness values of the copper tubing in the samples were similar, suggesting that the tubes underwent the same production process. The low hardness values also suggest that the copper tubes did not undergo any hardening treatment/process.

Conclusion

Based on the findings and discussion, the following conclusions were made:

The failure of the heat exchanger was as a result of environmental corrosive attack, most likely from an external source.

The overall construction of the heat exchanger suggests inappropriate material selection and/or fabrication techniques were utilised. A combination of improper material selection, welding, and inappropriate paint selection/application increased the exposure of the copper tubes to the corrosive environment in which it operates. This allowed for corrosion to initiate and propagate throughout the whole heat exchanger, resulting in multiple leaks to occur over time.

Recommendations

Based on the outcome of our investigation, One Eighty provide recommendations with regards to the failed heat exchanger. These include:

  1. Review for appropriate material (material selection).
  2. Review weld procedures.
  3. Review paint coating chemical composition for compatibility.
  4. Review surface preparation and coating application.

Successful implantation of the above-listed would minimize the overall risk for this type of failure from occurring in the future. The utilisation of the appropriate codes and/or standards should always be adhered to in any design in order to prevent unwanted failures due to lack of following the principle guidelines as set forth by the appropriate standardisation organization.

Back To Top