Interested in linking to "Level Measurement: The Very Last Resort "?
You may use the Headline, Deck, Byline and URL of this article on your Web site. To link to this article, select and copy the HTML code below and paste it on your own Web site.
By Walt Boyes, editor in Chief
In our February 2007 issue, we looked at how to apply level measurement by considering the application. (“First the Application, Then the Product,”). We talked about a level measurement continuum, and handed out a nifty PDF that you can post on your wall to help you sort out the selection process (See Figure 1).
Now I want to talk about the level product that “dares not speak its name.” It is the level measurement product of last resort—the one you take to your boss after every other one has failed. It’s…nuclear.
Nuclear level gauges will work in nearly any level measurement application. You can see that from the chart below. But nobody uses them for the simple and easy applications. Why? Because they’re relatively expensive, and they’re truly a pain in the posterior because of regulation. On the other hand, they actually work in applications where nothing else does, and they require little or no maintenance.
There are some common (for difficult level applications, anyway) applications that only nuclear level gauges will handle. These include agitated tanks with internals, some interface measurements, moving or vibrating granulars, very hot or cryogenic liquids and thick-walled vessels.
When few devices will work with your application or there is no other way to make the measurement, nuclear may be the choice.
How can you tell if your application is a “nuclear level gauge or nothing” problem? First, if you’ve tried differential pressure, bubblers, capacitance or rf admittance, ultrasonic or radar, and they didn’t work, or they were too high a maintenance headache, you might be a candidate for using a nuclear gauge.
If you can’t stick a sensor inside the vessel at all, you’re probably looking at using a nuclear gauge. If you can’t pick the vessel up and weigh it, you are probably looking at using a nuclear gauge.
If you have a reactor under very high pressure and temperature with a coolant jacket, a couple of inches of insulation, lagging, and an agitator and other internals, you almost certainly are looking at a nuclear gauge application. (See Figure 2).
But how do you know that a nuclear gauge will work? It’s actually much easier to determine success with a nuclear guage than with ultrasonics or radar. The math is simple and clear. You add up all the half-values of all the material between the source and the inside vessel wall, and all of the half-values of the material between the opposite wall and the detector.
For a reactor under very high pressure and temperature, with a coolant jacket, a couple of inches of insulation, lagging and an agitator, a nuclear gauge is the answer.
Basically, an inch of steel is one half-value, and you can calculate the half values of different materials using their thickness and density. Most suppliers will do this for you and have computerized source-sizing programs with which to do it. Knowing the half values and the distance from the source to the detector enables the sizing program to calculate the size of source and source holder required to both make the measurement, and even more important, do it legally. The physics of the measurement are well-known, and once a source and detector assembly have been sized, it is a near certainty that the application will work.
Some of the things that will interfere with the application’s working include buildup of material on the inside or outside wall of the vessel. Sometimes you may need to locate the source and detector on a chord of the diameter of the vessel in order to miss an agitator or some other internal structures in the tank. This can radically increase the apparent thickness of the steel wall in a curved tank and may make the measurement difficult or impossible. Rapid fluctuation of level in the vessel can also cause significant error.
Make sure you know what all the layers are. If you miss just a half-inch of steel on both sides of the vessel, you might not have enough radiation at the detector to make an accurate measurement.
Yes, you have to have the device licensed by some regulatory agency. Somebody from your plant should take a radiation safety course and be licensed to test, move and store sources. Most suppliers provide this as a fee-based service, so it isn’t strictly required. And finally, you can’t throw the device away when you’re done with it. You must dispose of the source legally.
Nuclear gauges work on one of two principles: gamma transmission or neutron backscatter. Most level applications use gamma transmission.