【Daily Question 20090312~15】Instrumentation and Production Safety (four days in a row)

I’ll set the questions. First, I’ll say: 1. The high-pressure compressor has many interlock points; we use electric contact pressure gauges for pressure interlocking, and short circuits caused by aging contacts can lead to the compressor shutting down accidentally. 2. As the shaft vibration of the unit worsened, the DCS system issued a shutdown command; however, due to damage to the 24VDC supply in the DCS unit, the signal sent to the electrical control system was not transmitted, resulting in the turbine blades being scraped. 3. During the expansion and renovation of the urea production facility, and when the instrument control system was upgraded to a DCS system, the design did not take into account emergency shutdowns in urea production, which led to several instances of severe overpressure in the urea synthesis towers. 4. Due to interference, the anti-surge valve of the air separation unit opened on its own, causing the unit to stop operating.

The instrument air from the air compressor flows into the instrument storage tank; there is a bypass control valve used to regulate the pressure of this instrument air. Due to cold weather, the air supply line connected to the control valve freezes, which prevents it from functioning properly. As a result, the pressure of the instrument air becomes excessively high, causing the air compressor to shut down via interlock. It takes about 10 minutes for the backup small air compressor to come online

1. The main air flow rate was measured using a flow meter installed on-site; since no anti-freezing and anti-condensation measures were taken for the pressure guiding pipes in winter, the flow measurement became inaccurate, which led to the shutdown of the device as part of the interlock system； 2. The DI signals of the interlock control system are powered externally; therefore, one end of these signals is connected to the signal ground. Errors in the signals occur due to an improper connection to this ground, which leads to interlock shutdown during testing ；

I remember when I first started working at the factory in 1989, it happened to be during a major maintenance period. At that time, the maintenance technician stepped on the pressure switch while climbing up, causing the mercury contacts to close and stopping the air compressor.

1. Our factory has a device for which online analysis instruments were originally required, but they were later replaced with offline ones; as a result, the data was not available in a timely manner, and eventually the device exploded. 2. Some control systems use PLCs, while others rely on hardwired connections, leading to chaotic management – once the system failed to start because of this. 3. During dust removal, dust falling on the wiring connections caused the system to stop operating.

In the DCS system, a solenoid valve coil at the site burned out, and an operator made a mistake while checking the control cabinet, which caused the system to stop operating. However, due to insufficient expertise, it was not possible to determine the cause of the problem. Once a technical partner arrived at the site, it was found that the fuse had burned out, and the issue was resolved; nevertheless, the system remained shut down for several hours.

After the air compressor stopped, the PLC lost power. Later, when it was necessary to inspect a temperature signal for the instruments, the PLC was started by going to the control cabinet. Shortly after, a staff member rushed over and said that all the oil had leaked out. It turns out that after the air compressor was shut down, the maintenance crew disassembled the oil circuit at the oil station; when the PLC was powered on, it detected low oil pressure and started the oil pump, causing all the oil to flow onto the ground. 5 barrels of engine oil. Later, I helped others use a sponge to soak it up little by little; the only word that comes to mind is “ashamed”. Fortunately, there was no fine.

Such topics are good. It can serve as a reference for everyone. Okay...........!

Therefore, the control system is required to have the FallBank algorithm, so that it can still provide safe control in the event of a transmitter failure

The oil and gas butterfly valve in the catalytic unit suddenly closed completely, causing pressure buildup in the unit and a halt in feed supply. It was found that the positioner was faulty; after replacing it, production was resumed.

The liquefied gas separation tank at the top of the stabilizer, the gasoline separation tank at the top of the distillation tower, and the condensed oil tank at the outlet of the compressor typically use differential pressure type level gauges and interface detectors; float-type gauges are also employed for measurement. Design institutes usually adopt both types of gauges. However, in actual operation, both measurement methods are prone to freezing due to inadequate insulation in cold weather, which poses significant difficulties for normal production and instrument maintenance. We resolved this issue by switching to dual-flange sensors for measuring the level and height of the liquid; the instruments are operating quite normally, which provides favorable conditions for smooth production.

1. The air filtration pressure reducing valve burst, causing the reactor’s feed control valve to close and leading to a shutdown. 2. The liquid chlorine submersible pump was sealed using instrument air; during the shutdown, chlorine gas mixed into this instrument air, resulting in corrosion of the positioner and causing losses of several million

I noticed that the question allows for an answer up to the 15th day, so I’d like to share a few of my thoughts as a sign of respect for the original poster. 1. During the upgrade of the Foxboro system, the PS2 mice and keyboards used in the engineers’ workstations were replaced with USB ones; this caused the systems to crash. The issue was that there was no response when those devices were connected. Eventually, I simply connected USB keyboards and mice and restarted the systems, keeping them connected so that they could be used at any time. Everyone should consider doing the same. 2. Our HIMA interlock cards have aged and start to malfunction. 3. In some cases with HIMA’s auxiliary interlocks, loose wiring leads to shutdowns of the system. 4. The adjustment range of the reactor feed valve is very small, around 2% only; the valve stem moves only a tiny amount, which results in the formation of a small pit. Any attempt at a larger adjustment doesn’t work – when we set it to 5%, nothing happens, and this leads to a shutdown of the system.

On the DCS operation screen, when adjusting the control valve, the rate of change in its operation was not controlled; as a result of an operator’s mistake, the system came to a stop. Later, the rate of change in valve operation was controlled within the specified range on the operation screen, resolving the issue.

1. There is interlock for the minimum main wind pressure. It was originally the low-limit DI signal from the on-site electrical contact pressure or pressure controller. There have been instances where the pressure-transmitting copper tube of the pressure controller broke due to prolonged vibration of the pipeline, resulting in the activation of the low-pressure limit interlock mechanism of the pressure controller ； Short circuit of the wiring terminals of the electric contact pressure gauge triggered the interlock mechanism ； Later, pressure sensors were used for the low-limit alarm and as an AND gate for the on-site low pressure DI signal; that is, an interlock would be triggered only when both a low pressure alarm and a low pressure condition were present ； 2. When checking the natural gas flow orifice plate, no communication was held with the operators (there is a low-flow interlock); the transmitter was flushed, which resulted in an interlocked shutdown ； 3. The main air flow control valve, a Fisher ball valve model 8560, would automatically open from time to time during normal operation, resulting in a sudden increase in air intake and affecting product quality; the issue was resolved after replacing the valve actuator. The replacement valve positioner could not have its zero and range adjusted through communication, as the 375 unit had not been upgraded; as a result, there was a difference of about 20 valve positions between the output and the feedback. Communication-based adjustment was only possible after taking the 375 unit to Shanghai for upgrading and bringing it back, and production had to proceed under great concern during those three days. 4. The interlock temperature sensor, which is a K-type thermocouple, became damp and covered with dust; the negative terminal came into contact with ground, causing the temperature reading to soar to over 1000 degrees, which led to an interlocked shutdown of the system ； Subsequently, the rate of change of the temperature signal is monitored; if the change exceeds 30 degrees per second, it is considered a faulty point, and the interlock is automatically disabled. After repair, manual confirmation is required to restore the interlock. 5. The DCS cabinet is powered by two circuits; when Circuit A was shut down, it was confirmed that there was power at the input terminal for Circuit B (on the back side). However, it was forgotten that there was an additional circuit breaker that needed to be turned on, and as a result, Circuit A was shut down. . . . . . . :(6. There are also cases where the control valve gets stuck and does not function, causing the automatic control circuit to keep sending signals to keep it open; as a result, the valve operates suddenly. . . Many, many

A fault in the boiler level transmitter caused the PLC to receive signals indicating an extremely low water level, resulting in automatic shutdown of the boiler.