Radiometric density gauges earned their place by working where nothing else would. Two things have changed: the alternatives got better, and the cost of holding a source went up. This page is about deciding whether a replacement makes sense for a specific measuring point, and how to carry it out.
What the source costs after purchase
- Licensing and inspection for the life of the installation, with an accountable radiation protection officer.
- Source inventory and periodic leak testing, each with records that have to survive an audit.
- Transport under regulated shipment rules whenever the gauge moves — including to a repair facility.
- Decommissioning and disposal. Cs-137 has a half-life of about 30 years and Co-60 about 5.3 years. The Cs-137 case in particular means the source outlives the plant equipment around it, and the disposal cost lands at decommissioning.
- Operational friction: exclusion zones during maintenance, contractor restrictions, and in some jurisdictions a public-consultation dimension that has nothing to do with engineering.
Against that, the reason a source was chosen is usually one specific constraint — normally "nothing may touch the process" or "the duty destroys wetted instruments". The replacement question is really: is that constraint still binding at this measuring point?
Matching the replacement to the duty
| Original reason for the source | Non-nuclear route |
|---|---|
| Abrasive mineral slurry, large bore | Ultrasonic acoustic impedance — sensor face flush with the bore, DN50–DN1000 |
| Cannot tap the pipe at all | Clamp-on ultrasonic attenuation, within DN500 and low-bubble |
| Very high solids, water content is the variable | Microwave |
| Clean liquid, high accuracy wanted | Coriolis or tuning fork |
| Genuinely must stay outside the pipe | Exempt-source Na-22 — still radiometric, but a far lighter regulatory footprint than Cs-137 |
That last row deserves an honest note: it is a reduction in regulatory burden, not an elimination of it. Where a plant's objective is to remove sources entirely, it is not the answer; where the objective is to keep a non-contact measurement while shedding most of the compliance load, it often is.
Running the changeover
- Overlap the two instruments. Install the replacement and run both for a period covering the normal range of process conditions, not just steady state. The overlap data is what settles later arguments.
- Compare against the laboratory, not against the old gauge. The incumbent may have drifted; treating it as truth bakes its error into the new instrument.
- Sample properly. Take samples as close to the new sensor as the piping allows and record the time of each. A sampling point far upstream measures transport lag and calls it instrument error — the single most common cause of a "failed" trial.
- Cover the upsets. Include startup, shutdown, wash cycles and low-flow periods in the comparison. Steady-state agreement tells you little about how the loop will behave at 3 a.m.
- Plan the source's disposal before removal, since the route and cost need arranging in advance.
When to keep the source
Being straight about this is more useful than a sales pitch. Keep it where the process is above the temperature or pressure rating of every wetted alternative; where the medium would destroy any wetted sensor and the line cannot be modified; where the existing gauge is performing and the remaining plant life is short enough that changeover cost will not be recovered. A source that is working, licensed, and inside its useful life is not automatically a problem to be solved.