Chinese
Spanish
Russian
Portuguese
French
German
Indonesian
Thai
Vietnamese
Arabic
Italian
Turkish
Polish
Japanese
Radar commissioning errors should be compared across several known conditions before an offset is changed. A nearly constant difference, a difference that grows with level and a sudden jump suggest different checks. Record where the disagreement first appears: physical distance, configured local level or the PLC value.
Establish comparable observations before reading the pattern
Use the same level zero, quantity, units and observation time for each pair. Define the signed difference as indicated level minus reference level, and retain that sign. If the reference is unreliable or observes a different local surface, the resulting pattern may describe the test arrangement rather than an instrument fault. Check reference-measurement capability before drawing a numerical conclusion.
For the METRAVON MV-RD12, retain the installed reference dimensions, selected variable and output configuration with the readings. An analogue/HART unit and an RS485 version require their corresponding receiving-system checks. A model name does not identify which conversion the PLC is applying.
Use radar commissioning errors to prioritise the next check
The following values are hypothetical level comparisons in metres. They are not MV-RD12 test results, calibration data or acceptance tolerances. Assume a trustworthy reference, a fixed level zero and stable paired observations solely to illustrate two error patterns.
| Reference level, m | Illustrative set A, m | Illustrative set B, m |
|---|---|---|
| 0.500 | 0.450 | 0.450 |
| 1.500 | 1.450 | 1.350 |
| 2.500 | 2.450 | 2.250 |
Set A has a signed difference of -0.050 m at all three points. First check the common datum, installed reference-plane height and configured zero. Also inspect offsets in the receiver. A consistent difference does not prove that a particular offset is wrong; a biased reference can produce the same pattern.
Set B reads 90% of the reference level: its differences are -0.050, -0.150 and -0.250 m. Check the selected quantity, units and span or conversion applied at each stage. Do not repair this pattern by adding 0.050 m because that happens to correct the first point; the two higher points would still disagree.
Follow the disagreement through the measurement chain
Compare the raw distance with the physical reference before using a tank table. Then verify the local level calculation. Finally compare the transmitted value with the receiving scale or digital interpretation. Finding agreement at one stage and disagreement at the next narrows the investigation without requiring an immediate hardware replacement.
| Observed pattern | Checks to prioritise |
|---|---|
| Constant difference | Datum, antenna reference position, configured zero and receiver offsets, with reference bias still considered. |
| Difference grows with level | Quantity, units, range endpoints and the calculation applied in the transmitter and receiver. |
| Step near one level | Whether raw distance also jumps, and whether geometry, target selection or a receiver conversion breakpoint coincides with it. |
| Difference changes with time | Paired timestamps, surface motion, filtering, update rate and whether a displayed value is held or stale. |
For a jump at a repeatable height, save observations on both sides of the event. A constant offset cannot remove a target change. If only the PLC value steps while local distance remains credible, examine the receiver before altering echo settings.
More than one cause can coexist. An installation may have a datum error and excessive filtering, or a correct local level with a wrong PLC span. The patterns organise the checks; they do not establish a unique diagnosis from three numbers.
Record the correction and repeat the affected comparisons
Save the as-found settings and identify the evidence supporting each change. Adjust one justified item at a time, then repeat the affected reference and receiving-system checks. Keep the before-and-after signed differences so an apparent improvement at one point cannot hide a larger disagreement elsewhere.
State the actual levels, process states and direction of travel tested. A low-level correction does not verify an unobserved upper interval. Leave any required later observation visible in the commissioning record rather than extrapolating a complete pass from the illustrative pattern.
Questions before applying a correction
Q1. Can a PLC offset be used when the local radar level is already wrong?
It may make one screen agree while leaving the field indication, other outputs or alarm consumers inconsistent. Locate the incorrect reference or conversion first and apply the authorised correction at the appropriate stage.
Q2. Does a constant difference prove that the radar needs calibration?
No. Confirm the reference and installed datum, then compare the underlying distance. A field datum error and an instrument measurement error require different work and cannot be distinguished by a single offset value.
Q3. What if the readings cannot be repeated under the same conditions?
Record the changed conditions and limit the comparison accordingly. Use available trends to plan a suitable later test; do not present observations from different surface or timing conditions as proof that an adjustment worked.
Summary
Use signed error patterns to decide what to investigate, then verify the cause through the physical measurement and receiving system. Submit the MV-RD12 configuration, reference method, paired readings and as-found settings for a commissioning review.
