This column is moderated by Béla Lipták, automation and safety consultant and editor of the Instrument and Automation Engineers' Handbook (IAEH). If you have an automation-related question for this column, write to [email protected].
Q: Our project involves a reactor 5 m in diameter and about 25 m in height. Presently, its level is being measured by a differential pressure (DP) gauge, capillary type, and I was told to install, in addition, a radiation level detector. Is the addition of a radiation-type unit advisable, and what are the advantages and disadvantages of these two options? Is the radiation-type a better choice? Can radiation sensors be dangerous if the reactor walls are thick and hot?
Rahim Salamat
[email protected]
A: For a couple of decades, I was the chief instrument engineer at C&R, and during that period, we must have designed nearly a hundred polymer reactor control systems. So your question is familiar, but it is lacking the key information: Is this a batch or a continuous reactor?
If it is a batch reactor and if you have good flowmeters on the charging side, you might not measure the level at all, but just depend on the batch flowmeters for recipe formulation and add a high-level interlock for safety. If the accuracy or reliability of the flowmeters is insufficient, and if the full weight of the reactor is more than four times its empty weight, you might put the reactor on load cells.
Figure 1: In differential pressure (DP) applications, extended diaphragm seals can prevent plugging, and with equal-length capillaries and temperature compensation for ambient temperature and sun exposure variations, DP can give reasonable performance.
To consider nuclear sensors, you need an NCR license, an on-site certified radiation officer, and if you have heavy coating, it will still affect accuracy. In addition, you must also arrange for source disposal. For DP, extended diaphragms (Figure 1) with equal-length capillaries and temperature compensation for ambient temperature and sun exposure variations can also give reasonable performance.
While in batch reactors the residence time is measured by a timer, in continuous reactors, residence time is a ratio of reacting volume divided by the outflow (V/F), where V is a function of level. Therefore, in controlling continuous polymer reactors, level measurement is not optional, but a must.
Figure 2 illustrates such a control system. The selection of the level sensor should consider the comments I made in connection with the batch reactors, and some people might also consider the use of self-diagnosing laser (up to 300 °F, if the transmittance in the vapor space and the reflectance of the polymer surface is acceptable) or noncontacting and self-diagnosing radar (up to 500 °F, if there is no coating, condensation or crystallization on the antenna).
Béla Lipták
[email protected]
A: The biggest reasons not to use radioactive measurement are:
1. The instrument rays have to shine through the walls of the reactor so the receiver can absorb them, but the radioactive beam is not concentrated in one point like a laser, so sometimes you will have scattering of the radioactive beam, which could harm personnel.
2. Generally, reactors have extremely thick walls requiring a very high-energy source, which, over time, may make the reactor walls radioactive around the the beam area.
3. Most radioactive systems need to be close if not in contact with the surface of the walls of a hot reactor. This could damage the source, which could cause radiation to leak.
Alex (Alejandro) Varga
[email protected]
A: I have not come across DP level used in polymer reactors. Even with designs from 40 years ago, we were using nucleonic/radiation level, however, that is a small sample of the total number of polymer reactors in the world.
Most of the reactors I worked with did not use level. They just batched in a certain quantity and called it good. In the cases I am familiar with, there is no real benefit to knowing level in a batch reactor. Perhaps most of the modern polymer reactors are now semi-continuous.
There are regulatory difficulties involved with nucleonic installations. In the jurisdictions I know about, you have to have a trained and certified radiation officer on-site. There is also a perception that such devices are very dangerous and difficult to monitor/maintain/control.
Simon Lucchini, CFSE, MIEAust CPEng (Australia)
Chief controls specialist, Fluor Fellow in Safety Systems
[email protected]
Figure 2: In continuous reactors, residence time is a ratio of reacting volume divided by the outflow (V/F), where V is a function of level, and maximum production rate equals maximum volume divided by minimum residence time. Therefore, in controlling continuous polymer reactors, level measurement is not optional, but a must.
A: I lived with this question for 15 years. The only successful approach was a blow-back dip tube. This did plug now and then, and we used a long rod with a drill bit welded to the end to clear it. It plugged because the blow-back air was supplied at too low a pressure. The reactor normally ran at a high vacuum but we discovered (by watching the operators late at night) that when the valves became plugged, the operators would turn off the vacuum and pressurize the vessel until the valve plugging cleared. And the blow-back tube was filled with polymer.
The proper installation sequence would be: air supply header with filter (not regulator) > check valve > needle valve > rotameter > dip tube.
The polymer would not be pushed into the dip tube because the pressure downstream of the needle valve would rise to block backflow. The standard air supply regulator has downstream pressure protection, and when the downstream rises, the regulator vents and allows backflow.
It took a long time to discover this. Our installation did have a radioactive source in the agitator shaft, and the detector outside. A "micro-micro-ammeter" amplified the signal. Most of the time it worked well. The problem was the way the various parts were electrically grounded, so it appeared to be unreliable. Any welding in the building caused a panic.
Cullen Langford
[email protected]
