The seven stages of chemical engineering technology from concept to industrialization (Issue 45/100) -- PID programming

Thread Content

This post was last edited by xiouxingzhe on 2026-6-17 22:39. Seven stages of chemical technology from concept to industrialization (Issue 45/100) —— Technology refinement: From control objectives to engineering implementation. Dear friends: Hello everyone! In the previous issue, the equipment selection and all the details related to the data sheets were completed; the key parameters for all the main devices have been determined, and the list of equipment has also been finalized. Starting from this issue, we move on to the eighth key task in the preparation of the process package: PID preparation. The PID is the file with the highest information density in the process package, without a doubt. If the PFD is the \"strategic map\" of a plant – determining the path that materials take, where heat is exchanged, and where momentum is replenished – then the PID is the \"general plan for construction and operation\" of the plant; it transforms every line and every control objective shown in the PFD into specific, actionable pipes, valves, instruments, and interlocks. PID is a document that all personnel involved in chemical projects can access and should understand. Process engineers use it to convey the complete design intent; operators rely on it to understand how the equipment operates and how to handle abnormalities; construction workers use it to guide on-site installation and pressure testing; while maintenance personnel depend on it to trace the function of each pipeline and each instrument. From the construction of a system to its operation and then to its renovation decades later, PID remains the most essential technical document. I. The fundamental difference between PID and PFD: The information on a PFD is relatively concise—mainly device data, three-balancing data, and the main control strategies. A PFD may show a dozen or so devices and dozens of main process pipelines, with key parameters and core control loops indicated. PID is different. A stream of material represented by one line on a PFD may require ten lines on a PID for detailed representation – the main line, bypass lines, drain lines, vent lines, sampling lines, purge interfaces, instrument pressure tap lines, and heat tracing lines; each of these has a specific diameter, pipe identifier, material grade, and insulation requirements. The statement on the PFD that reads “TIC-1001 controls the reactor temperature” needs to be translated into a complete set of information in the PID: where the temperature sensor is located, what is its measurement range, which system the signal is fed into, whether the control element is a control valve or an variable-frequency pump, whether the control valve is installed at the refrigerant inlet or outlet, what the approximate opening degree of the valve during normal operation is, what the high-high and low-low alarm values are, whether a shutdown interlock should be triggered upon a high-high alarm, and whether the valve remains open or closed in the event of a loss of pressure. This is the transformation from “control intent” to “engineering implementation” – the PFD specifies “what needs to be controlled”, while the PID specifies “how to achieve this control”. II. Page planning for PID: Before starting to draw the PID, it’s necessary to do some page planning first. To represent the three-balance data, the equipment is drawn relatively compactly in the PFD. However, PID requires a large amount of space around each device – pipes, valves, instruments, control circuits, interlock signals – and if the layout of the PFD is used as a template, the diagram will become overcrowded, leaving no room to add additional pipelines and instruments later on. The basic principle for page division is that complex reactors should be given their own page, or even two pages, as they involve the most control loops and interlocks, and thus have the highest safety requirements. A tower system usually requires two pages; simpler towers can be condensed into one page. Heat exchangers, containers, pumps, etc. are relatively simple and can be arranged flexibly according to space; usually, several pieces of equipment share one page. Devices that are highly correlated should preferably be placed on the same diagram, to avoid pipelines having to move back and forth between different drawings, which makes them harder to locate and understand. After the page layout is completed, first draw the existing pipelines from the PFD onto the PID to distinguish between primary and secondary streams. The distinction principle is simple: the path from the area where the main raw materials enter to the area where the final product exits constitutes the main flow pipeline, which is highlighted with thick solid lines or a distinctive color ； The rest are all secondary pipelines, represented by thin lines. This distinction allows those looking at the diagram to immediately identify the main flow of the process among hundreds of pipelines, without being overwhelmed by the numerous details. The representation of equipment ports is also important. The positions of the pipe outlets should reflect their relative relationships as much as possible—the gas outlet at the top of the tower is at the very highest point, the liquid outlet at the bottom of the tower is at the lowest point, and the side-line outlets are located in the middle section. If these relative positions are accurately reflected in the diagram, those who view it can intuitively understand the flow of materials and the internal structure of the equipment; they can also roughly determine the logic of pipeline routing without having to refer to the equipment’s data sheets. III. Refinement of the control loop: from a single sentence to a set of information. The refinement of the control loop is one of the core tasks in PID programming. For each control loop, the following questions need to be answered: Where are the sensors located – are the temperature measurement points on the reactor vessel or on the inlet and outlet pipes, and are the pressure measurement points at the pump outlet or in front of the control valve? Does the measurement range of the level gauge cover the normal operating levels as well as the alarm levels? Which system does the signal go into – DCS or SIS? Is it a control signal or just an alarm indication? Are there any local indicator instruments available for on-site personnel to check? What is the actuator used for control – a control valve or a variable-frequency pump? Where is the control valve installed, and what is the approximate range of valve opening during normal operation? What are the set values for alarms and interlocks? What are the values for high-high alarm, high alarm, low alarm, and low-low alarm respectively? Should interlock actions be triggered in case of a high-high alarm or a low-low alarm? What is the fault position of the control valve – in what position should the valve be when the power supply is interrupted to ensure the safety of the device? The control valve for the refrigerant flowing into the reactor is usually of the FO type – in the event of a failure, the valve remains fully open to ensure a continuous supply of refrigerant and prevent overheating and loss of control in the reaction ； The reactor feed valve is usually selected as FC – to cut off feeding in case of a failure and prevent loss of control. These markings may seem simple from a PID perspective, but they are based on the results of process safety analysis. The selection of the measurement point location also directly affects the accuracy of the instrument readings. Where should the temperature measurement point of a thermometer be located to best represent the true temperature of the fluid? Should it be inserted deep enough to reach the center of the pipe, and at a time when the material has been thoroughly mixed? Whether the pressure tapping of a pressure transmitter should face upward or to the side depends on whether the medium is gas or liquid, and whether there is condensation. How long a straight pipe section is required upstream of the flow meter? Different types of flow meters have different requirements for the straight pipe sections before and after them, and these needs must be taken into account in the PID design. IV. Representation of interlock circuits: SIS interlock circuits and DCS control circuits differ fundamentally in terms of safety levels and maintenance requirements; it is necessary to clearly distinguish between the two in terms of PID. The sensor tag numbers, interlock logic relationships, actuator tag numbers, and operating status of all SIS interlock circuits should be clearly shown in the diagram. Once interlock is triggered, it overrides all DCS control commands and carries out safety actions directly – this difference in priority is indicated by markings on the PID. The reset method for interlocking should also be indicated on the PID. Automatic reset or manual reset? Is manual reset done on the DCS operator station, or must a person go to the site to press the reset button? The details of these reset designs directly affect the recovery efficiency and safety of the device after an interlock is triggered. For high-risk interlocks—such as reactor over-temperature interlocks and combustible gas high-level alarm interlocks—manual reset is usually required, and the operator must confirm that safety conditions are met before performing the reset. For low-risk interlocks, they can be designed to reset automatically, reducing the operator’s workload. The sensors and actuators of SIS and DCS must remain independent, which is a basic requirement for layer analysis. If the SIS and DCS share a sensor or a actuator, then once that shared component fails, both control and interlock functions are lost, leading to out-of-control risks. During the PID programming phase, it is necessary to verify the independence of each interlock circuit one by one to ensure that there are no shared components. V. A detail that is easily overlooked: the fault position. Marking the fault position of control valves is a task that appears simple but is actually very crucial in PID programming. FC stands for fault off, FO stands for fault on, and FL stands for fault hold. The selection of the faulty bit is not arbitrary, but is based on the results of safety analysis. What position should the valve be in in order to bring the device into a safe state when the power source—usually instrument air or electrical power—is interrupted? The reactor feed valve is usually selected as FC – it cuts off the feed in case of a failure, preventing the reaction from getting out of control. The reactor coolant control valve is usually of the FO type – in the event of a failure, the valve remains fully open to ensure a continuous supply of coolant, thereby removing the heat generated by the reaction. The bottom discharge valve of a distillation tower is usually selected as an FC type – it cuts off discharge in the event of a failure, thereby preventing the liquid level at the bottom of the tower from dropping to such an extent that vaporization in the pump occurs. For emergency shut-off valves in safety interlocks, FC type must be selected to ensure that the material flow can be stopped when the interlock is activated. On a PID, the indication of a fault position is simply FC or FO, but behind these letters lies a complete set of analysis logic. If the issue of fault positions is not carefully considered during PID development, and labels are assigned based on habit or by simply copying those from similar devices, it will surely lead to questions during the HAZOP analysis. VI. Developing a PID is not a one-time task. It is a process that involves successive stages, and it is not completed or finalized in just one go. In the base layer, first draw the device nozzles and the main process pipelines, and label them with pipe numbers, diameters, material grades, and flow directions. The control layer involves installing instruments on equipment and pipelines, as well as designing control circuits and interlock circuits. Auxiliary layer, supplementing start-up pipelines, shutdown pipelines, emergency discharge pipelines, purging and displacement pipelines, and defective product pipelines. Sampling layer, equipped with a sampling system and on-line analysis instruments. Labeling layer: performs pipe diameter calculation, pipeline numbering, insulation design, and instrument tag compilation. After each layer is added, it is necessary to go back and check the integrity and consistency of the previous layers. After adding the auxiliary pipelines, will there be enough pipe connections on the equipment? Is it necessary to add additional pipe connections to the equipment data sheet? After adding the sampling points, will there be enough space for the operation channel and the sampling area? After adding the interlock circuits, are the I/O counts of the DCS and SIS within the cabinet capacity limit? The entire process involves repeated iterations, with the PID gradually evolving from a \"skeleton\" into a fully developed final product. This iterative process requires patience as well as close communication with downstream specialties – the piping specialty needs the pipe diameter and material grade, the instrumentation specialty needs the tag numbers and measurement ranges, the equipment specialty needs the pipe connection numbers and specifications, and the safety specialty needs the interlock logic and discharge pathways. The PID engineer is the hub and distributor for all this information. Preview for the next issue: Issue 46 – PID compilation: Complete representation of eight types of information. In the previous issue, we discussed the page layout for PID, the refinement of control loops, and the marking of fault locations. In the next issue, we will delve deeper into PID programming, detailing the eight types of information that need to be included in a PID: pipeline information, process valve information, special fitting information, instrument information, control scheme information, interlock safety information, equipment port information, and sampling and analysis information. Each type of information corresponds to the key inputs for the subsequent specialty, as well as the details that are prone to being overlooked. To be continued in the next issue.