Oct-2024

Vacuum heater operational cycle improvement study

Coke formation creates short vacuum heater runs. Controlling film temperature and oil residence time can help reduce coke formation rate inside radiant tubes.

Haytham Al-Barrak, Abdulaziz Mubarak and Mahendran Sella
Saudi Aramco

Article Summary

Plant vacuum heaters have been experiencing short run lengths due to coke formation, which requires shutdown. A comprehensive approach must be followed to evaluate the heaters' performance, condition, and deficiency. This article focuses on vacuum heaters experiencing a short run length of 24 months.

The surveyed two vacuum heaters have four passes each, and the flow rate is distributed equally for each pass. There are two radiant cells for each heater, with both cells sharing one common convection section. Each cell is side-fired at the bottom with 18 burners, divided equally by a wall with nine burners on each side. The process tube inlet is located at top of the radiant section, while the tube outlet is located at the bottom of the radiant section close to the burners.

Coke formation in vacuum heaters is caused by different mechanical and process factors, where radiant tubes absorb the released heat by burners through the tube’s outside surface. Once the film temperature and residence time exceed the oil’s thermal stability, coke formation will occur. Crude oil stability varies and cannot be controlled; however, the oil film temperature and residence time can be controlled through heater design and operation. The main factors of coke formation in vacuum heaters include:
· Crude residence time
· Crude film temperature and thermal stability
· Heater design.

Crude oil residence time depends on feed rate, radiant section tube size and length, and velocity of steam injection. Radiant sections contain between two and five tube sizes from the inlet to the outlet due to oil vapourisation. Oil film temperature increases when the tube size expands because the oil mass velocity decreases; thus, oil residence time increases.

Radiant tubes with high residence time and high film temperature are more prone to coke formation. Injecting velocity steam into radiant section tubes lowers oil residence time, which reduces the coke formation rate. However, injecting steam may increase the pressure drop and have other side effects on the downstream tower.

Despite the changes in the heater's feed composition and mass flow rate, the velocity steam was not adjusted to accommodate those changes.

Crude film temperature determines the susceptibility of a process fluid towards coking, where film temperature varies with composition change. If the film temperature exceeds the limit, the fluid film on the inside tube's surfaces is subject to thermal decomposition, which results in coke deposition at the location. Oil film temperature mainly depends on the critical balance between incoming tube external heat flux and internal oil mass velocity. The heat flux is described as the quantity of heat absorbed by the radiant tube per unit external tube area. Mass velocity is the mass of oil flowing through the heater tube per unit of internal cross-section area and unit time.

Oil thermal stability varies depending on crude type. Some crude oils are less stable than others. The original crude’s maximum allowable film temperature is 510ºC based on crude composition specified at the design stage. However, it is worth mentioning that the current crude feed may have a lower film temperature than the design feed, which means it is susceptible to coking at a lower temperature than the design limit. The current crude blend should be thoroughly analysed in order to identify the new film temperature limit at which coking will start. This new film temperature will determine the new tube metal temperatures (TMT) and heat flux.

Heater design has a major role in influencing coke formation key factors. One of the important criteria in vacuum heater design is radiant heat flux. It is defined as heat transferred per unit of tube (external surface) area. Designing the heater above the standard heat flux limitation may accelerate the coking rate formation. A second major factor of coke formation is the location of the outlet tubes where the feed is at the highest temperature. Vacuum heater outlet tubes have a larger diameter, resulting in lower mass flux combined with higher oil film temperature and, hence, increasing oil residence time. Therefore, this leads to a high rate of coke formation inside the tubes located near the highest heat flux zone.

Methodology
Several simulations of vacuum heaters have been conducted utilising simulation software. This software is well proven to be reliable and trusted industry-wide. The program calculates operating parameters such as firing rate, heater efficiency, and tube skin temperatures. In addition to its calculations, the software can simulate most types of heaters, including complex processes, multiple fire boxes, most coil configurations, and transfer lines.

To maximise vacuum heater run length, reducing coke formation rate and lowering start-of-run TMT are required to be achieved. In other words, reducing oil film temperature and oil residence time is essential to meeting the study’s objective.

The simulation model for the vacuum heaters has been developed using the original process datasheets. Since the vacuum heaters are operating under different parameters compared to the original design, the simulation model needs to be recalibrated to achieve reliable results. An operating case representing the right proportion of crude blend and process input is used as a baseline. The operating case of specific date is used as a baseline for future comparisons. This case is specifically chosen since it represents the right proportion of crude blend, and can be considered as a clean case since it was just after testing and inspection (T&I) after decoking.

Simulation model validation
The calibrated simulation model is compared to the baseline operating case to assess if the program can predict and match the operating conditions. The comparison will only focus on the process feed stream, as it is relevant to the coking issue. However, the steam super heater (SSH) and boiler feedwater (BFW) will mainly affect the overall efficiency. Some parameters have larger variances from the operating data but can be explained as follows:
· Bridge wall temperature: Although the simulation assumes an uniform radiant temperature based on well-stirred box approach, in reality the roof zone of the firebox measures a relatively colder bridge wall temperature compared to the actual temperature of the flue gas near the firebox's middle zone.
· Flue gas stack entry temperature: The SSH and BFW measured data were not reliable; therefore, the original inlet and outlet conditions were used. The variance can also be explained by a dirty convection section that might be limiting the heat transfer to the tubes, resulting in hotter flue gas leaving the convection section.
· TMT – radiant coil bottom pass: The measured data are not reliable since the tubes and temperature elements are wrapped with a ceramic blanket, leading to false temperature readings.
Based on the above explanation, we can safely conclude that the calibrated simulation model is reliable and will be able to simulate future operating cases confidently.

Increasing velocity steam, optimising O₂ level, and replacing convection tubes are selected scenarios to be simulated by the software to maximise vacuum heaters run length by improving heater operations. These scenarios were selected due to their low cost, minor modifications, and best practice recommendations. Other modifications that require major revamps with high cost or extended shutdown beyond the normal T&I period were not discussed in this report.