Nov-2020

Level measurement in coke drums (TiA)

If heavy components of crude oil are to be converted into lighter ones, cracking processes come into play.

Sabrina Nees
Berthold Technologies

Article Summary

The desired conversion effect is achieved by overheating the distillation residues under pressure. Such thermal cracking can be done in a coker. Therefore, a mixture of pre-heated fresh feed (residual oil) is passed from the fractionator via a heater to one of the large coker drums. Here the cracking and expansion of the fluids is allowed. The lighter cracked hydrocarbons are taken off at the top of the drums to the fractionation tower, while cracking heavy hydrocarbons results in their conversion into a vapour due to the high temperature.

As the vapours escape the viscous liquid, they tend to create a foam layer and solid coke remains in the drum which slowly fills up. The foam layer can vary depending on many parameters such as operating temperature, pressure of the drum, type of crude, or charging rate. In a worst case scenario, this leads to a foam over, which can be a costly event (up to $2 million), not only causing loss of production with downtimes of 2-6 weeks but also requiring a lot of manual labour to clean coke out of the overhead lines and fractionation tower. To prevent such foam overs and reduce carryover, but also to improve the utilisation of drum capacity, a level measurement in a coke drum is necessary. Also, to increase throughput of the unit, one of the most important objectives is to fill the drum higher, safely, and reliably. However, to achieve this, one must have a reliable measurement method for the foam front in the drums.

But level measurements in coker units are quite different when compared to typical level measurements. Common measurements are installed on continuous processes with no changes in process conditions, which means that the lower fluid density as well as vapour density is relatively constant, and one calibration is enough. But due to the continuous switched feed between two drums in a time based cycle, delayed coking is the only semi-continuous batch process in a refinery with different process conditions during a cycle and possible build-ups.

A simple calibration is not sufficient for this measurement task, calibration changes depending on the cycle. Due to the conditions in the coke drum, a level measurement is a very challenging task. Where other measuring technologies tend to fail or end up being extremely unreliable, radiometric level technology is ideal for monitoring the coke level due to its non-intrusive nature. Thereby, use of gamma continuous level measurement has many advantages. Measuring a continuous level instead of single point information with neutron backscatter measurements helps to determine the rate of change and thus to increase the throughput of the unit. By using gamma radiation instead of neutron radiation, a radiation area is not necessary. The typical dose rate of a gamma continuous measurement 1 m from the drum at the detector side is 0.1-0.2 µSv/h (0.01-0.02 mrem/h) or at the source side 0.5-1.5 µSv/h (0.05-0.15 mrem/h) compared to 85 µSv/h (8.5 mrem/h) from a neutron backscatter measurement. A point measurement only in proximity to the probe does not detect any foam, which rises faster outside the reach of the point neutron backscatter probe, but with a gamma level it is possible to reliably detect non-uniform rising of foam front as the measurement covers the entire drum diameter. This not only reduces the foam overs, but also the use of anti-foam chemicals. Minimising the use of anti-foam saves a considerable amount of money and increases hydrotreater catalyst life.

Berthold’s measurement systems for coker units consist typically of two or three fan beam point sources, a continuous level measurement (up to 32 m) by use of cascaded TowerSENS detectors, a top density measurement with gas property compensation, and a detector for auto calibration. With this system, it is possible to measure the level during the normal filling of a drum as well as the outage measurement, which has completely different conditions. This measurement is unaffected by vapour channels inside the drum, product falling from the trays, different product densities, or scaling/coking. Since large temperature variations through the whole operational cycle of a coke drum are a known issue to the operators, Berthold’s automatic stabilisation technology has proven to be the most important feature to guarantee a stable and reliable level measurement without the need for recalibration. Separate level alarms ensure additional process safety. Other benefits of this solution are significantly lower source activities due to the use of solid scintillators with high sensitivity and minimised influences of interference radiation during weld inspections with the aid of internal algorithms such as X-ray interference protection.

With Berthold’s solution for delayed coker units, operators have a high-repeatable and long-term stable measurement, not only increasing the throughput but also the reliability of the drum. In the long term, this can increase profits while maintaining plant safety at the same time.

This short case study originally appeared in PTQ's Technology In Action feature - Q4 2020 issue.

For more information: sabrina.nees@berthold.com