Abstract In the mechanical sealing industry, circumferential seals operate radially (instead of axially) between the shaft and the housing, minimizing leakage by sealing directly against the shaft and creating a tortuous path for fluid to escape during use. There are two primary types of circumferential seals: controlled-gap seals and fractured-ring seals. Controlled-gap seals maintain a consistent clearance during the thermal expansion and contraction of components as they heat and cool. In contrast, fractured-ring seals operate radially with a carbon ring that is fractured into equal segments. As the shaft temperature fluctuates, these carbon segments allow the fractured-ring seal to expand and contract, ensuring constant contact between the seal and the shaft. Defects can be introduced during the production, assembly, handling, and installation of fractured-ring seals. Defects such as nicks, chips, scrapes, or scratches may impact the leakage performance of the seal. However, defects smaller than 0.13 inches (the largest dimension) that occur on the inner diameter edge or fracture line of fractured-ring seals do not significantly affect leakage rates. The 0.13-inch maximum defect size tested is significantly larger than the maximum defect size that is acceptable by the workmanship quality standards. This paper demonstrates that defects smaller than 0.13 inches in the specified locations do not impair the performance of fracturedring seals. The findings are intended to serve as a precedent for future fractured-ring seal applications. To test seal performance, leakage rates and differential pressure across the seal were measured. First, using several defect-free fractured-ring seals, a baseline leakage curve was established by measuring the correlation between leakage rate and differential pressure. Subsequently, fractured-ring seals with varying known locations and defect sizes smaller than 0.13 inches were tested to assess any relationship between defects and leakage. Additionally, two seals were intentionally damaged, with defect size and location noted, and leakage rates and differential pressure were measured both before and after the damage was applied. The results show that the size and location of known and inflicted defects do not affect the seal leakage rates. Testing of the intentionally damaged seals confirms that defects do not impact leakage performance within the limits of size and location evaluated. Through experimentation, it was concluded that defects such as nicks, chips, scrapes, or scratches up to at least 0.13 inches on the inner diameter and fracture line of fractured-ring seals did not affect the performance of the seal. This work also shows that when these defects occur on the ID edge or fracture line of the seal, the specific location of these defects had no measurable impact on seal performance.
Baldauff et al. (Mon,) studied this question.