DANA Technical Services
[DANA Technical Services]

Process Design, Gas Processing and Sulphur Recovery Expertise


-----> Introducing the DAD (Double Air Demand) System Now available from DANA <-----

REDUCE UPSET SO2 EMISSIONS SUBSTANTIALLY by Improving the Dynamic Behaviour of Your Sulphur Plant!

What is DAD ?

DAD is an innovative control system upgrade for Sulphur Recovery Units (SRUs) that uses two air demand analyzers for sulphur plant feedback control, one located at the back end of the SRU (as usual) and an additional one located between the mainburner and first sulphur condenser outlet. This combination, along with the accompanying control logic upgrade, results in a much faster response to changes in acid gas composition. In fact over 75% of major plant upsets are easily handled using this approach due to the much faster response. In SRUs that have a Superclaus stage, the DAD system minimizes the number of bypasses that occur due to high H2S events that are caused by acid gas compositional upsets.

Typical Sulphur Recovery Unit (SRU) Control System

The purpose of a Claus unit is to recover as much elemental sulphur from the feed H2S as possible. The key factor to achieve this is to burn approximately one third of the H2S to SO2 in order to obtain an H2S to SO2 ratio of 2:1 in the Claus tail gas. This requires very fast and accurate control of the combustion air flow rate. The combustion air flow has to follow the flow changes of the acid gas feed as well as the compositional changes. Consequently the control system consists of two parts: (1) The actual "ratio control part" which is a flow-flow ratio control (to maintain the required air to acid gas ratio), which can act fast and accurately by utilizing advanced techniques (such as with the ABC system used by Comprimo Sulfur Solutions (Jacobs)); and, the "compositional change part" which requires a tail gas analyzer (or ADA) where H2S and SO2 are measured, and the results of this measurement are used to correct the ratio setting of the "ratio control part", in order to satisfy 2SO2 - H2S = 0 in the Claus tail gas. In the case of an SRU that consists of a Claus section and a downstream Superclaus stage, the Claus section control system works the same way using a different ratio set point. While for a Claus unit the tail gas must contain twice as much H2S as SO2 on a molar basis for optimum recovery, the feed gas to the Superclaus stage (which is the tail gas from the Claus section) must contain a minimum mol percent of SO2 and about 0.7 to 1.0 mol percent H2S. The above described system works very well, if the controllers are correctly tuned. Because the ratio control part acts fast and accurately, its contribution to the deviations from the optimum conditions is very small. The weak point of the system is the control of the ratio set point which involves the corrective action taken after compositional changes that are caused by upsets in the upstream units, e.g. switching of gas wells upstream of a natural gas processing plant, switching of cokers in refineries, etc. The final control elements (flow control valves) are positioned at the inlet of the Claus plant while the measurement of the tail gas composition takes place downstream of the Claus unit. This latter circumstance introduces a dead time combined with a number of first order delays of typically 2 to 3 minutes to the control loop at full plant load. Consequently the control action from the tail gas analyzer (ADA) must be slow in order to avoid control instability. It takes 20 to 30 minutes to come to equilibrium again after a compositional change with this typical approach. During this time period the tail gas composition is not optimal, which decreases the average sulphur recovery efficiency and thus increases the SO2 emission.

Details of the DAD System

Substantial improvement in the speed of response can be achieved by using the DAD (Double Air Demand) control system over the conventional single Air Demand system described above. DAD is the invention of Paul E. d'Haene (DANA Technical Services, Calgary, Canada) and Peter S. Schermann (Process Control Consultants, Austerlitz, The Netherlands), and, a patent application has been filed for this invention in the USA. The system improves the dynamic behavior substantially by placing a second analyzer downstream of the first Claus condenser (or even downstream of the mainburner and upstream the first condenser) cutting the time delay to 10 to 20 seconds (or less). This added loop is much faster than the typical tail gas loop on the Claus section, so the resulting duration of the deviation from optimum conditions is substantially less, and, hence the resulting amount of SO2 emission is also much smaller during any compositional upset. Also, the absolute accuracy of the additional air demand analyzer can also be less than normal because the final correction is still done by the tail gas analyzer (the only requirement for the additional analyzer is good reproducibility). Another operational advantage of the DAD system is that the presence of two analyzers in the unit means that if one of the analyzers fails, the plant can still be controlled automatically (with somewhat less control quality, which is still much better than using manual control as is required with only a tail gas analyzer in the system). Figure 1 shows the schematic diagrams of the two systems (in this example in combination with the Comprimo ABC system).


Figures 2 and 3 show the results of comprehensive dynamic simulations, displaying the comparisons of the dynamic behavior of the conventional ABC with the DAD system approach, where Figure 2 is for the Claus control mode and Figure 3 for the Superclaus control mode. They show the dynamic system responses and performance parameters for the same disturbance, 18% ratio change in 15 minutes in both directions. The black lines represent the ABC system, the red lines the DAD system. In Figures 2A and 3A the H2S content of the Claus tail gas (that is the feed to the Superclaus stage) has been plotted against time. In Figures 2B and 3B the actual sulfur emission as a percentage of the total sulfur feed has been plotted against time. In Figures 2C and 3C the equivalent time of the additional emission caused by the disturbance, has been plotted (this equivalent time refers to the time required to obtain the extra SO2 emission when the plant is operating at optimal steady state conditions). In Figure 3C this time is 1800 seconds for the ABC system and 80 seconds for the DAD system. This means that five such disturbances during one day decreases the daily average recovery efficiency by 0.1% for the ABC and by 0.005% for the DAD system. They show clearly that the control is much faster when DAD is applied, so the plant runs much more quietly and eliminates most of the upsets, thereby minimizing recovery losses due to compositional changes. The average recovery efficiency of sulphur plants with inherently frequent gas composition changes (e.g. natural gas sweetening units that have little control over the plant feed gas composition) will then closely approach the expected (guaranteed) recovery.

Figure 2a

Figure 2b

Figure 2c

Figure 3a

Figure 3b

Figure 3c


DANA Technical Services Ltd.
Paul E. d'Haêne, President
208, 3016 - 19 Street N.E.
Calgary, Alberta, Canada T2E 6Y9
Telephone: (403) 571-0390 or 571-0392 ; Fax: (403) 571-0399
Cellular: (403) 680-0687
E-mail: paul@danats.com


Process Control Consultants
Peter S. Schermann
De Beaufortweg 12,
3711 BW
Austerlitz, The Netherlands
Telephone: (31) 34-349-1562 ; Fax: (31) 34-349-1143
E-mail: pcc.p.schermann@ziggo.com