Carbon Residue Estimation and Control in Vacuum Towers

carbon residue yokogawa

Estimation of the Carbon Residue content at a high frequency in the valuable product is one of the challenges being faced in the Refineries today. Carbon Residue is an indication of Coke forming tendencies in the valuable products.

Reliable and reproducible measurementation of residue in valuable products like Gas oil is challenging, but also very important as inaccurate measurements result in losses - losses in terms of poor operation of the downstream units in Refinery or shutdown of the units in Refinery.

Carbon Residue

Different Carbon Residue values can be defined for a fuel depending on the measurement method used. Ramsbottom Carbon Residue (RCR), Conradson Carbon Residue (CCR), Micro Carbon Residue (MCR) are the most commonly used indexes for estimation of carbon residue.

Estimation of Conradson Carbon Residue

Conradson Carbon Residue is commonly called as Concarbon or CCR is analyzed in the laboratory which provides a coke forming tendency of the oil. The test measures the amount of carbonaceous residue remaining after the process of evaporation and pyrolysis of the sample. A quantity of the sample is weighed, placed in a crucible and subjected to destructive distillation through a fixed period of severe heating. At the completion of severe heating the residue collected is cooled and weighed to be estimated as a % of weight of the original sample.

CCR Estimation Issues

Since the estimation of CCR requires physical sampling, analysis in the lab by severe heating, cooling and analysis of test results to calculate the CCR which is a long process and requires laborious manual work. Because of this time consuming nature of the estimation of CCR the sampling is usually done weekly once, resulting in the improper control of residue in the valuable products in Refinery.

CCR is Vacuum Towers

Vacuum Towers are physical distillation towers operated at a pressure lower than the normal atmospheric pressure as Vacuum distillation works on the principle that boiling occurs when the vapor pressure of the liquid exceeds the atmospheric pressure.


The feed to the Vacuum tower comes from the bottom of the Fractionator or the Crude Distillation column which is heated to the desired temperature for separation of valuable products like Gas oil, Fuel gas from the Fuel oil. Steam jet ejectors will be used to achieve the desired level of negative pressure in the column to maintain a pressure less than the atmospheric pressure. Desirable products like Heavy Vacuum Gas Oil (HVGO), Light Vacuum Gas Oil (LVGO) are side drawn from the tower and is heat exchanged with several upstream and downstream unit heat exchangers to recover the heat. Effective mass transfer is achieved in the column by pump around flows (LVGO and HVGO) recycled back to the tower counter acting with the vapors that are distilled from the bottom of the tower.

Typically model based multi-variable controls used in Vacuum towers try to control and operate the Vacuum tower fired heater, Vacuum tower safely and optimize the complete Refinery plant operation by pushing towards the constraints, like emission constraints on Vacuum tower fired heater, Vacuum Gas oil cut points, CCR in HVGO etc…. CCR estimation being done based on laboratory analysis that are not made in realtime, the result is unreliable Concarbon values in HVGO thereby creating a wrong indication either the vacuum tower has already reached constraint, thus losing all opportunities to explore the plant and optimize the plant operation towards a profitable region or indicating lesser CCR value than actual that was estimated a week before resulting in upset in the operation of downstream units like Fluidized Catalytic Cracker Unit. The emission constraint on Vacuum heaters or Vacuum Gas oil cut points etc… can be directly measured instantaneously whereas the CCR estimation is not instantaneous and the time delay in estimation poses a big challenge for close control and improving the profitability.

Soft Sensor for CCR

Carbon Residue Yokogawa

Reliable online analyzer installation on the HVGO product for measurement of CCR is practically not possible because of the method of analysis. Installation of soft sensor for the measurement of Concarbon will be an effective way of analysis. The CCR Soft sensor can be calculated based on the historical data from operation of the Vacuum tower and the results obtained for the CCR and doing a PCA/PLS correlation based on the independent variables that contribute for the residue in HVGO. Yokogawa’s Platform for Advanced Control and Estimation uses the same engine for the Multi Variable Controller and the Estimation controller (Soft Sensor) and these soft sensors can be estimated either as static model or dynamic model which can be used integrated with the Multi variable controllers. These CCR soft sensors are calculated on real time can be updated by Laboratory sample analysis done on a weekly basis helps to correct the soft sensor model and the availability of soft sensors will help the multi variable control system to push the plant to the constraints realistically, helping the plant throughput to be maximized to the actual constraints rather than to a pseudo constraint condition.


Availability of real time measurements of the CCR in HVGO with proper corrections or model updates using infrequent laboratory analysis helps in maximizing the throughput and minimizing the cost until the actual constraints are hit, thereby indirectly increasing the reliability of the equipment and profit of the plant.

Early detection of CCR results in upstream units like Vacuum Distillation column HVGO samples helps Process Engineers to take corrective action upfront on the Operation of the Fluidized Catalytic Cracking unit thereby improving the reliability of the Refinery operation and improving the quality of the blended products from the Refinery.

Read other articles by Saravanan here.