Power and Flow Logging

Power and Flow Logging

New completion based on supplied sources.

Measure compatible quantities over the same time

Power and flow logging establishes how electrical input relates to compressed air delivered at a defined boundary. It can reveal operating-state losses, control behaviour and changes in performance. The result is meaningful only when the measurements use compatible boundaries, reference conditions and time windows. Dividing a convenient power number by an unrelated airflow number does not create a valid efficiency indicator.

Define the purpose before selecting instruments. Comparing machines under controlled duty, evaluating a whole station and verifying a maintenance intervention require different boundaries and context. State which question the campaign will answer and what service conditions must be preserved. The analysis should explain the observed system, not merely decorate a report with curves.

Select the electrical boundary

Identify whether power represents the main motor, the complete compressor package or the whole station. List included and excluded auxiliaries such as cooling fans, pumps, dryers and other treatment loads. Mark the electrical measurement point on the system record. Keep this boundary unchanged for a direct before-and-after comparison, or explicitly reconcile any change.

Motor shaft output, motor electrical input and package input are different quantities. A nameplate motor rating is not the measured operating input. Do not mix these values in a performance comparison. Where manufacturer data are used, retain their stated test boundary and operating conditions so the relationship to the site measurement can be assessed.

Measure real power appropriately

Use an instrument and connection method suitable for the electrical system, waveform and measurement purpose. Current alone is not real power. Voltage, phase relationships, power factor and waveform conditions can affect the result. For drives and other non-linear loads, select a method suited to the actual measurement location and equipment characteristics.

Electrical installation and removal belong to qualified personnel under the site's approved procedures. This page does not provide live connection instructions. Record the instrument, sensors, ranges, connection configuration and calibration evidence. Verify channel association and sign convention during commissioning so an incorrect setup is not carried through the full campaign.

Define the delivered-flow boundary

Locate the flow meter relative to the compressors, treatment, purge uses, receivers and distribution. A meter before a dryer may include air later consumed by treatment; a meter after treatment represents a different delivered boundary. Identify branches and bypasses that can move flow around the sensor. Record whether reverse flow is possible and whether the instrument can resolve it.

For a station comparison, include all supply paths or clearly limit the result to the measured path. A second compressor feeding downstream of the meter can make apparent specific power misleading. Boundary completeness should be checked on the physical system, not assumed from an outdated diagram. Record temporary connections and configuration changes during the campaign.

Preserve the flow reference conditions

State the reported flow unit and reference basis. Actual volume at line conditions, standard volume and free-air delivery are not interchangeable labels. Retain the configured pressure and temperature reference, compensation settings and relevant moisture basis. If conversion is required, document the equation, inputs and assumptions.

Use the same basis throughout the analysis and comparison. A nominal conversion between cfm and cubic metres per minute changes units, but it does not resolve different reference conditions. Where a source uses an undefined standard condition, seek clarification before combining it with site data. The report should not hide this uncertainty behind extra decimal places.

Match flow technology to installation

Confirm the instrument's suitability for pressure, temperature, contamination, expected range and available installation geometry. Follow the applicable requirements for straight lengths, orientation, insertion or sensor position. Disturbances from valves, bends or pulsation can affect some measurements. Record actual installation conditions and any departures from the specified method.

Evaluate low-flow performance as well as the peak range. A meter that resolves production demand may not accurately quantify small idle consumption. Range exceedance and below-range readings need quality flags. If the campaign requires a decision in a region the instrument cannot resolve, change the method or state that the evidence is insufficient.

Synchronize clocks and operating states

Power, flow, pressure and compressor-state records need a common time reference. Record time zone, clock source and any observed offset or drift. Check synchronization before and after the campaign where practical. A constant clock offset can make a compressor appear inefficient during transitions because its power is paired with another operating state's flow.

Retain load/unload state, speed or other relevant control information alongside the main signals. State whether those records come from the controller, independent measurement or observation. Do not force an uncertain state into a precise classification. Where several machines interact, identify each machine's state so the station result can be interpreted without assuming equal contributions.

Choose sampling and aggregation deliberately

Distinguish sensor response, sampling interval and stored averaging interval. An averaged signal can be appropriate for energy totals while hiding short events relevant to control diagnosis. Select the resolution from the decision and the dynamics of the system. More samples do not improve accuracy if the instrument or installation is unsuitable.

Preserve enough raw detail to explain transitions and data-quality problems. When aggregating, state whether values are interval averages, instantaneous samples or accumulated totals. Integrating an instantaneous sample as though it represented a long interval introduces an assumption that should be evaluated. Use a consistent processing method and retain its version with the result.

Calculate energy and delivered volume

For interval-average power P_i in kW over duration Δt_i in hours, electrical energy E is the sum of P_i × Δt_i in kWh. For interval-average flow Q_i in m³/min, volume V is the sum of Q_i × Δt_i × 60 when the same durations are expressed in hours. The volume retains the meter's stated reference basis.

Use only matching valid intervals for a paired energy-per-volume result. If flow is missing during an interval with recorded power, do not divide the complete energy total by a partial volume total and call it a complete-period performance figure. Report coverage, exclusions and any estimate separately. Missing values are not equivalent to zero flow or zero power.

Distinguish instantaneous and integrated indicators

Instantaneous specific power P/Q has units kW per m³/min when those are the input units. It can help compare stable operating points on compatible boundaries. Near zero flow, the ratio becomes unstable or undefined. Do not use an arbitrary small denominator to generate a dramatic but meaningless number.

Integrated energy per delivered volume is E/V in kWh/m³ over a stated period. It includes the effect of low-delivery and unloaded states within that period. The arithmetic average of instantaneous P/Q values is generally not the same as E/V. Use the indicator suited to the question and label its units, time window and boundary explicitly.

Worked integration example

Consider invented interval averages for a one-hour period. For 30 minutes, power is 100 kW and delivered flow is 15 m³/min. For the next 30 minutes, power is 40 kW and flow is 3 m³/min. Energy is 50 + 20 = 70 kWh. Delivered volume is 450 + 90 = 540 m³. Energy per volume is approximately 0.130 kWh/m³.

The two instantaneous ratios are approximately 6.67 and 13.33 kW per m³/min. Their simple average is 10, while the ratio of average power to average flow is 70/9, approximately 7.78 kW per m³/min. Dividing that latter value by 60 gives the integrated energy-per-volume result. The example illustrates weighting and does not represent a measured project or a performance target.

Interpret unloaded and stopped periods

Retain power consumed while delivery is low or absent. Such intervals can be important to the station's actual energy performance, but their interpretation needs the operating state and measurement boundary. A downstream flow meter may continue recording demand supplied from storage while a compressor is unloaded or stopped. That does not mean the machine is delivering the measured flow at that instant.

Separate machine analysis from station analysis where storage and multiple machines intervene. Do not assign all station airflow to the one machine whose power happens to be logged. If individual machine flow is unavailable, use an appropriately bounded method and disclose the limitation. A system-level energy result may still be valid even when individual machine efficiency cannot be resolved.

Include pressure and environmental context

Power and delivered flow depend on the operating duty. Record pressure and relevant inlet or ambient conditions when comparing periods or equipment. A result obtained at a lower delivery pressure should not be ranked directly against a higher-pressure result without explaining the difference. Likewise, a changing production profile can alter the integrated result even when the machines are unchanged.

Do not apply an unsupported universal correction to remove these effects. Use appropriate manufacturer information, a justified model or measured comparable states. Where reconciliation is inadequate, present separate bounded results. The comparison should make its conditions visible rather than implying a universal leaderboard.

Assess measurement uncertainty

Identify uncertainty contributors from the instruments, installation, range, compensation, clock alignment and processing. Calibration evidence addresses part of the chain; it does not eliminate installation effects or incomplete boundaries. Report uncertainty in a way suited to the decision and avoid presenting a nominal instrument accuracy as the uncertainty of the entire audit.

For an illustrative first-order ratio estimate with independent small relative uncertainties, the relative uncertainty of P/Q can be approximated by the square root of the sum of the squared relative uncertainties of P and Q. Correlation and other contributions can change that result. A competent reviewer should select the appropriate method and confidence convention for the actual measurement.

Worked uncertainty interpretation

If an illustrative stable power measurement has 2% relative uncertainty and flow has 3%, an independent first-order combination gives approximately 3.6% for their ratio before other contributions are included. These invented values do not establish the uncertainty of any supplied instrument. Installation, timing and boundary issues may add material effects.

If two reported performance values differ by less than the relevant comparison uncertainty, the evidence may not support a ranking. The correct response is to improve the method, collect suitable additional data or report that the difference is unresolved. More decimal places do not create a meaningful distinction. Keep that limitation visible in any investment conclusion.

Control data exclusions and corrections

Retain raw files and create a separate quality log for changes. Record the reason, affected intervals and treatment of each exclusion. Typical issues include instrument disconnection, range exceedance, clock faults and periods outside the agreed boundary. Do not exclude an inefficient operating state merely because it makes the result less favourable.

If a time offset is corrected, preserve the original timestamps and document the basis for the correction. If a flow conversion is revised, retain the prior processing version and the approved new one. The final figures should be reproducible from the identified dataset and calculation, with enough context for another reviewer to understand what changed.

Verify an intervention using comparable conditions

Define the baseline and post-change boundaries before implementation. Record production, pressure, operating hours and other material variables in both periods. Select comparable states or a justified normalization method. A lower total electricity use during a quieter week is not, by itself, evidence that the intervention improved efficiency.

Report the measured outcome, uncertainty and service performance together. If demand falls but power does not measurably fall, report the airflow change and investigate controls rather than claiming proportional energy savings. The owner's instruction excludes unsupported savings from project conclusions. A result can remain unverified even when the maintenance work was completed correctly.

Logging acceptance matrix

Check the treatment of accumulated meter totals

Some instruments export accumulated energy or volume rather than interval averages. Calculate interval use from differences between valid successive totals and record how resets, rollover and data gaps are handled. A negative difference may indicate a reset rather than negative consumption. Do not combine cumulative totals and interval rates in the same integration formula without first converting them to a consistent representation.

Compare the integrated rate data with available accumulated totals as a plausibility check where their boundaries and periods match. A discrepancy may reveal missing samples, unit errors or different aggregation conventions. Investigate the cause rather than applying an unexplained scaling factor. The final calculation record should state which representation was authoritative and why.

Examine startup and shutdown separately when useful

Transitions can consume energy while delivery changes rapidly, and storage can separate the timing of generation from use. Retain these intervals in an overall operating-period result when they belong to the defined duty. For a stable-point machine comparison, identify and exclude transitions only through a stated selection rule that fits that different question.

Do not use the best stable interval to represent a whole shift without explaining the choice. Conversely, do not interpret a transient ratio as a steady performance characteristic. Present the operating-period result and the stable-state analysis separately where both are useful. This allows the owner to distinguish machine behaviour from the consequences of the actual demand and control schedule.

Set a decision threshold before comparing alternatives

Before testing, consider what size of difference would change the purchase or operating decision. Select the measurement method to resolve that difference with useful confidence. If two candidates are expected to differ only slightly, a coarse site setup may be insufficient even when it is adequate for detecting a major fault. State the intended decision and the measurement capability together.

After testing, evaluate whether the observed difference is material relative to uncertainty, operating variation and service performance. A small apparent energy advantage may not outweigh a reliability or maintenance constraint. The measurement report should supply the evidence for that tradeoff without pretending that one rounded efficiency number determines the whole lifecycle decision.

Keep financial conversion downstream of verified energy

When an energy change has been established, any financial calculation still needs the relevant tariff, operating period and cost structure. Demand charges, time-varying rates and changes in operating hours may affect the result. Record the actual inputs and distinguish measured-period cost from an annual projection. Do not label a projection as an already realized annual saving.

For this gap completion, all numerical examples remain instructional. No site tariff, operating-hour assumption or measured saving has been supplied for these pages. The resulting documents therefore make no project-specific financial claim. A future project can populate the method with verified inputs and retain the calculation as part of its own evidence record.

Questions about performance figures

Can motor current be used as power

Current can support diagnostics, but it is not automatically real power. A justified electrical method must account for the relevant voltage and power characteristics. Label estimates and do not use them as measured audit-grade power without appropriate validation.

Why does the ratio spike when flow falls

The denominator may approach zero while the system still consumes power. Check the operating state, timing and boundary. Use integrated energy and volume for period performance where appropriate, and avoid interpreting unstable ratios as meaningful stable operating points.

Can package data be compared with station data

Only after the different inclusions and conditions are reconciled. Station power may include treatment and auxiliaries, while package data may describe a specific test duty. State the limits if a direct comparison is not supported.

Does a calibrated meter guarantee a good audit

No. Correct installation, suitable range, reference settings, time alignment, boundary completeness and representative operating coverage also matter. Calibration is necessary evidence for many decisions, but it does not replace the complete measurement method.

Enquiry information and next action

Provide the electrical and air-system diagrams, compressor inventory, proposed meter locations, production schedule and the decision to be supported. Include existing logger files with units, reference settings and timestamps where available. Reddy Compressors can use this information to define the logging boundary and required instrumentation review.

Preserve PROJECT ENGINEERED and PROJECT-SPECIFIC status. The legacy release requires both measurement/audit engineering and instrumentation/QA review. Record actual reviewer approval before treating the completed analysis as an accepted project result.