top of page

Transmitter Adjustment: Save As-Found Results Before Trimming the Output

Nov 2
4 min read

Updated: Nov 5

A transmitter returns from maintenance with an excellent final result, but nobody recorded how it performed before adjustment. The equipment may now be ready for a formal return-to-service review, yet an important question remains unanswered: what evidence supports the measurements made before the work?


Preserve the as-found comparison before trimming, changing an offset or altering the measuring range. Keep the subsequent work and as-left results separately linked to it. This guide explains how to specify that record for an analogue transmitter, using a hypothetical 4–20 mA pressure example without prescribing a live test or an adjustment procedure.


Three records answer different questions


Beamex’s calibration terminology separates calibration from adjustment. Its as-found calibration establishes the instrument’s condition before adjustments; its as-left calibration establishes the condition afterwards. A trim changes the instrument. A comparison records its response against a reference. Keep those activities identifiable even if a work order uses ‘calibration’ as an umbrella term.


Ask the service provider to return three clearly labelled parts: the original results, a description of any intervention, and the final verification results. If no adjustment was made, state that explicitly and follow the approved reporting procedure. Do not invent a separate post-adjustment test or label one set of readings as two independently performed tests.


Fluke’s multimeter-calibration guidance describes calibration-only, adjustment-first, and before-and-after sequences. Although that example concerns multimeters, it highlights a useful purchasing question for transmitter service: does the requested scope include measured as-found data, or only a final check? Agree this before authorising work, not after the original state has been changed.



Capture configuration as well as readings


An as-found record needs enough context to explain the numbers. Record the asset tag, model, serial number, configured input and output ranges, units and output law. Identify whether the test covers the transmitter alone, an attached sensing element or additional downstream equipment. Record the procedure revision and the reference instruments used.


Save the observed configuration before changing it. A transmitter configured for a different span can produce a consistent-looking output that is inappropriate for the application. If configuration is corrected before any comparison, describe that intervention and the evidence available beforehand. Do not present later readings as proof of the original configuration’s performance.


Beamex’s calibration overview explains the need for traceability, measurement uncertainty and documented results. A high-resolution display is not sufficient evidence that the reference and method can resolve the error of interest. The approved test must account for the relevant reference capability, method and conditions.


Worked example: one point, two different states


Assume a hypothetical linear transmitter is configured for 0–10 bar represented by 4–20 mA. At an applied reference value of 5.000 bar, the ideal output is 4 + (5 ÷ 10 × 16) = 12.000 mA. An as-found observation of 12.080 mA is therefore 0.080 mA above that ideal output.


Expressed against the 16 mA output span, the difference is 0.080 ÷ 16 × 100 = 0.50% of span. Its equivalent pressure difference is 0.080 ÷ 16 × 10 = 0.050 bar. These are calculated differences for this example, not specifications or measured results for the pictured calibrator.


Suppose an authorised adjustment is then performed and the same point gives 12.010 mA. That later observation belongs in the as-left record. It does not replace 12.080 mA in the as-found record. Nor does an improved midpoint alone establish acceptable performance across the range.


Keep the actual applied reference value and observed output at every required point, including any required ascending and descending runs. Use the approved procedure to define points, stabilisation, repeats and acceptance criteria. Do not turn this illustrative midpoint calculation into a universal calibration method.


Plan acceptance before seeing the results


Beamex’s out-of-tolerance guidance asks both which tolerance applies and whether the calibration uncertainty supports the decision. Define the application requirement and the rule for interpreting results with uncertainty before testing. A manufacturer’s headline accuracy, a process tolerance and a complete measurement-system requirement need not be interchangeable.


The person responsible for acceptance should receive the original and final results, reference details, test conditions, uncertainty information and intervention record. This guide does not assign a pass/fail verdict to the numerical example or authorise an instrument’s return to service.


What to request on the work order


1. State that measured as-found results are required before adjustment. Identify the asset, range, units, function and test boundary, along with the approved procedure and acceptance basis.


2. Require a retained original record. Link each result to its applied reference, timestamp and test run. Keep corrections to documentation traceable instead of overwriting the original evidence.


3. Separate the intervention. Record whether work involved configuration, trim, repair or replacement, and identify relevant settings before and after. Use only the exact equipment’s authorised procedure.


4. Require final verification and a clear disposition. Keep unresolved limitations visible. Where an original result is unacceptable, refer the possible effect on earlier measurements to the responsible quality or engineering team; a successful final test does not decide that question.


Avoid a tidy report with a missing history


Common mistakes include retaining only a final ‘pass’, recording nominal rather than actual reference values, and merging pre-adjustment and post-adjustment points into one table. Another is treating a screenshot of an improved reading as a complete calibration record.


Calibration work can disturb process controls, pressure systems and electrical circuits. Competent people must plan isolation, equipment selection and validation under the applicable site procedures and current requirements. Do not trim a live operating instrument just to collect an example. Preserve evidence first, then make and verify any authorised change.


Sources and pictured equipment







Need help with your application?




For instruments, sensors, data loggers and test equipment, visit ProSense Instruments, email sales@prosenseinstruments.com.au or call 1800 560 854.




For complete monitoring and automation projects, visit ProSense Solutions or email sales@prosense.com.au.




For software enquiries, visit ProSight Software or email info@prosightsoftware.com. Device and protocol integration must be confirmed for your application.

Comments


bottom of page