What Should an RECD Regeneration History Report Include?

Most RECD installations hold a lot of regeneration data and almost no regeneration information. The controller stores events, a technician jots notes on a service sheet, an operator writes “regen done” in a handover log and the supplier keeps its own copy. When a question arises, such as why back-pressure rose last winter, whether a warranty claim is justified or whether the device needs a heavier service, nobody can lay one clear page on the table.

An RECD regeneration history report fixes that. It is a periodic summary, monthly or quarterly, that turns raw cycle records into something a manager, a supplier and an auditor can read in five minutes. This article sets out what to include, section by section, shows a filled sample, gives rules for spotting trouble in the report itself and explains how long to keep it and who should sign it.

A distinction first. Our article on what to record in an RECD event log covers the raw events as they happen. The report described here sits above the log. It summarises, interprets and recommends. You need both.

What a good report must do

A report earns its place only if it can do five jobs.

  1. Show what happened. How many cycles, how long, how effective, which failed.
  2. Show the trend. Is the device getting better, worse or staying steady over months?
  3. Link cause and effect. Connect cycles to load, ambient, maintenance and events.
  4. Support decisions. Say whether to act, wait or ask the supplier.
  5. Stand as evidence. Be complete and consistent enough for warranty, audit and handover.

If a field does not serve one of these jobs, leave it out. If a job has no field to serve it, add one.

The report in nine sections

Section 1: Identification and scope

Purpose. Make the report unambiguous, now and in five years.

Include:

  • Site name, DG set ID, engine make, model and rating
  • RECD make, model, serial number and installation date
  • Controller firmware version and definition of “duration” (heater-on time, cycle time or other)
  • Reporting period and date prepared
  • Preparer and reviewer names
  • Regeneration method: passive only, electric heater, fuel-assisted or operator-initiated

Keep the same layout every period so readers can compare reports side by side.

Section 2: Operating context

Purpose. Cycles make sense only against the duty that produced them.

Include:

  • Run hours, number of starts and average run length
  • Share of run hours in load bands, such as under 20%, 20 to 40%, 40 to 60% and over 60%
  • Longest run and shortest run
  • Number of light-load or no-load test runs
  • Significant events: outages, storms, load changes, shutdowns

Our guide to using DG load logs to assess RECD suitability shows how to build the load-band table.

Section 3: Cycle summary

Purpose. The headline numbers.

Include:

  • Cycles requested, started, completed, aborted, timed out and inhibited
  • Completion rate (completed ÷ started)
  • Number of cycles verified as effective (see Section 4)
  • Interruptions: count and main causes

Section 4: Cycle-by-cycle table

Purpose. Evidence for every cycle, so any summary can be checked.

For each cycle, record:

FieldWhy it matters
Cycle ID, date, start and end timeTraceability
Trigger (automatic, manual, request level)Shows how cycles start
Load at start and average during cycleContext for everything else
ΔP at start and end, and the load at each readingEffectiveness at comparable load
Peak and average inlet and bed temperatureHeat delivered
Heater-on time and heater energy (kWh), or fuel usedEnergy per cycle; see RECD regeneration energy per cycle
Duration (with the definition stated)Trend; see rising regeneration duration
Controller message and reason codeSee interpreting a completion message
Verified outcome (effective, partial, failed, unknown)Your judgement, not the controller’s
Notes (interruptions, smoke, alarms, operator actions)Explanation

The “verified outcome” column is the most valuable. It records whether a person checked the five facts behind the message: ΔP at comparable load, temperature curve, heater behaviour, no interruption and persistence.

Section 5: Trends

Purpose. Show direction.

Include simple charts or small tables for:

  • ΔP at a reference load, per month (see the seasonal tracking guide)
  • Cycle duration and energy per cycle
  • Cycles per 100 run hours
  • Post-cycle ΔP recovery (start minus end)
  • Passive versus assisted share, if relevant

A six-month rolling view is usually enough, with commissioning values marked as a baseline. Our article on what the RECD operating dashboard should show suggests visual layouts.

Section 6: Failures and anomalies

Purpose. Show what went wrong and what was learned.

Include:

State “none” where there were none. A blank space invites doubt.

Section 7: Maintenance and changes

Purpose. Explain steps and jumps in the trend.

Include:

  • Services, cleanings, sensor calibrations or replacements, heater work
  • Engine work: injectors, turbo, oil changes, repairs
  • Changes to settings, firmware, alarm limits, clock or communications
  • Insulation or exhaust modifications
  • Fuel or oil supplier changes

Each entry should give date, who, what and values before and after. See updating RECD drawings and records after exhaust modifications and RECD firmware update questions before a service visit.

Section 8: Assessment and actions

Purpose. Turn information into decisions.

Include:

  • A short plain-language assessment: healthy, watch, or act
  • The reasons, citing numbers from earlier sections
  • Actions, owners and due dates
  • Questions for the supplier
  • Items carried forward from the last report and their status

Section 9: Data quality and sign-off

Purpose. Tell readers how far to trust the report.

Include:

  • Data sources (controller export, BMS, manual log)
  • Known gaps or uncertainties, including clock problems or missing periods
  • Instrument calibration dates for any portable instruments used
  • Reviewer’s signature and date

A filled sample: one quarter

The following sample is fictional. It shows how the sections work together, using a 400 kVA DG set with a heater-assisted RECD at a pharmaceutical plant.

Identification and scope (extract)

  • DG set: DG-2, 400 kVA. RECD installed 14 months ago.
  • Controller firmware 3.2. “Duration” means start-to-end flag.
  • Period: July to September. Prepared by the maintenance planner, reviewed by the utilities manager.

Operating context (extract)

MeasureJulyAugustSeptemberQuarter
Starts9141134
Run hours31242984
% hours under 20% load18%39%26%28%
% hours over 50% load49%21%36%35%
Longest run (hours)5.23.14.45.2

Cycle summary

ItemCount
Requested12
Started11
Completed (controller)9
Aborted or timed out2
Verified effective6
Verified partial3
Interrupted by DG stop or load drop3

Completion rate by controller: 9 of 11, or 82%. Verified effective rate: 6 of 11, or 55%. The gap between those two figures is the reason to verify, because three “completed” cycles were partial.

Cycle-by-cycle table (extract of four cycles)

IDDateLoad avgΔP start / end (mbar)Peak bed (°C)Heater energy (kWh)Duration (min)ControllerVerifiedNote
R-179 Jul52%58 / 365902.738CompleteEffectiveStable load
R-1922 Aug21%61 / 504522.145CompletePartialLight load, heater current low
R-2027 Aug18%64 / 614101.850TimeoutFailedHeater breaker found tripped
R-2314 Sep48%55 / 375852.840CompleteEffectiveAfter heater lug repaired

Trends (summary)

  • ΔP at the 50% reference load: July 36 mbar, August 44, September 38.
  • Cycles per 100 run hours: July 11, August 17, September 10.
  • Energy per effective cycle: about 2.7 to 2.8 kWh, steady.

Failures and anomalies

  • R-20: timeout. The heater breaker was tripped, found two days later. Cause: a loose terminal lug. Repaired on 3 September.
  • Three interruptions caused by DG stops after grid returned during a cycle.
  • No sensor faults. One data gap of 14 hours in August caused by a gateway power loss.

Maintenance and changes

  • 3 Sep: heater feeder terminals re-torqued, lug replaced; supplier witnessed.
  • 12 Aug: monsoon insulation repair on RECD housing.
  • No engine work, no firmware changes.

Assessment and actions

Assessment: Watch. August saw weak ΔP recovery and one failure, linked to a tripped heater breaker and a month of low-load running. September recovered after the repair and a higher-load pattern.

ActionOwnerDue
Add heater current alarm to the BMSElectrical lead31 Oct
Plan weekly loaded test runs at 50% or moreUtilities managerImmediate
Request supplier review of August cycles R-19 and R-20Maintenance planner15 Oct
Check heater terminals again at next serviceTechnicianNext service

Data quality and sign-off

Controller export plus manual operator log. One data gap noted. Clock checked against the BMS at quarter end, offset under one minute. Signed by the utilities manager.

That is roughly two pages. A reader can see the pattern, understand the cause and see what is being done about it. The same facts, scattered across a controller export and a handover book, would take an hour to assemble.

Five red-flag rules to build into the report

A good report does more than list. It flags. Agree thresholds with your supplier and write them into the template. The rules below are starting suggestions.

RuleExample triggerWhy it matters
Verification gapVerified-effective rate more than 20 percentage points below controller completion rateController says complete, but cycles are partial
Duration driftDuration 20% above baseline for three consecutive cyclesEarly warning; see rising regeneration duration
Recovery shortfallEnd ΔP not returning to within an agreed margin of the baseline after complete cyclesAsh, residual loading or sensor issue
Failure streakTwo or more consecutive failed cyclesTriggers the retry ladder
Frequency jumpCycles per 100 run hours up 50% on the previous periodDuty, engine or catalyst change
Data gapMissing data for more than a stated period during run hoursEvidence weakness

Where a rule triggers, the assessment section must say so and propose an action.

Report formats: choose by audience

One format rarely serves everyone, so consider three layers.

Layer 1: The one-page dashboard

For managers and operators. Four numbers (cycles, verified-effective rate, reference ΔP, open actions), one trend chart and the traffic-light assessment.

Layer 2: The full periodic report

For maintenance, the supplier and auditors. The nine sections above.

Layer 3: The data appendix

The raw controller export, kept unaltered, with a file name that includes the date, DG ID and firmware version. Store it in a read-only location. See preserving RECD data during controller replacement.

Who prepares it, who reads it

RoleTask
OperatorRecords cycles and observations in the shift handover log; see RECD shift handover log
Maintenance planner or technicianExports data, checks the five facts, prepares the report
Utilities or facility managerReviews, signs, assigns actions
RECD supplierReceives report, comments on anomalies, advises
Auditor or compliance leadUses the report as evidence; see audit RECD service report missing information
Insurer or lender (where relevant)May ask for summary figures

Name a single owner for the report. Without an owner, it will not be written.

How long to keep it

There is no universal rule, so decide with your compliance and legal teams. As a working approach:

Using the report with your supplier

A concise report improves supplier conversations in three ways.

  1. Faster diagnosis. The supplier receives cleaned, contextual data rather than a raw dump.
  2. Better warranty position. A consistent record shows that the device was operated and maintained as agreed. See what to record for an RECD warranty claim and RECD warranty terms.
  3. Planned service. Trends help schedule cleaning, heater work or catalyst checks. Our article on catalyst aging and regeneration behaviour shows what to look for over the years.

Ask the supplier to review one report each quarter and to comment in writing on any rule triggers.

Mistakes that weaken a report

  • Copying the controller’s verdict without checking. Include your own verified outcome.
  • Mixing definitions. Do not compare heater-on time in one period with cycle time in another.
  • No load context. ΔP without load is hard to interpret.
  • No baseline. Always include commissioning values.
  • Hiding gaps. Missing data should be stated, not smoothed over.
  • Too long. A twenty-page report will not be read. Use layers.
  • No actions. A report that ends without owners and dates is a diary.
  • Changing the template every quarter. Consistency is what makes trends visible.

The report in use: a supplier meeting

Imagine the utilities manager at the pharmaceutical plant meets the RECD supplier’s service engineer to discuss the quarter in the sample above. Without the report, the conversation might run: “We had some problems in August.” “What kind?” “Some failed cycles, I think.” Twenty minutes later nobody is sure what happened.

With the report on the table, the conversation runs differently.

  1. The manager points to the cycle summary: eleven cycles started, nine completed by the controller’s verdict, six verified effective.
  2. The engineer asks about the three that were partial. The cycle table shows they were light-load cycles with low peak temperatures and low heater current.
  3. The manager points to R-20 and the maintenance entry that explains it: a tripped heater breaker and a loose lug, repaired in September.
  4. The engineer checks the trends. Reference ΔP and energy per effective cycle are steady, so the heater and filter are behaving normally when conditions are right.
  5. They agree that the pattern is duty plus one electrical fault, not a failing device. The agreed actions go into the assessment: a heater current alarm, loaded weekly tests and a review of R-19 and R-20 data.

The meeting takes ten minutes, ends with named owners and leaves a record. The next report will show whether the actions worked. That is the difference between data and information.

Paper, spreadsheet or platform?

The best format is the one that will be used. Some practical comparisons follow.

FormatStrengthsWeaknessesSuits
Printed form and logbookWorks anywhere, no power needed, easy for operatorsHard to trend, easy to lose, handwritingSmall sites, operator-level records feeding a spreadsheet
Spreadsheet templateFlexible, charts and rules are easy to add, cheapNeeds discipline, version control and backupMost single-site operations
Controller or BMS export plus report templateAccurate data, less typingOnly as good as the logging; still needs a human assessmentSites with integrated monitoring
Cloud monitoring platformAutomatic collection, alerts and dashboardsSubscription, dependence on connectivityMulti-site fleets; see multi-site standardisation of RECD specifications

Whichever you choose, keep the layout fixed, name files consistently (for example, site, DG ID, period and version) and keep a clear copy of the original raw export.

A short checklist before you sign

  • [ ] Identification block complete, with the definition of duration
  • [ ] Reporting period and data sources stated
  • [ ] Load-band table present
  • [ ] Every cycle listed, with verified outcome
  • [ ] ΔP always paired with load
  • [ ] Trends shown against the commissioning baseline
  • [ ] Failures explained or marked “cause unknown”
  • [ ] Maintenance and changes listed with dates
  • [ ] Red-flag rules checked and results stated
  • [ ] Actions with owners and due dates
  • [ ] Data gaps and instrument calibration noted
  • [ ] Reviewer’s name, signature and date

FAQ

How often should an RECD regeneration history report be produced?

Monthly suits busy sites or the first year after installation. Quarterly is enough for stable, lightly used sets. Produce an extra report after any failure streak, major repair or change in duty.

Can the controller produce the report automatically?

Some controllers or monitoring platforms can export cycle tables and charts. That covers the data layers, but the assessment, verified outcomes and actions still need a human. Use automation for the tables and spend your time on the interpretation.

What if the controller does not log everything on the list?

Report what you have and note what is missing. Fill gaps with manual records where practical, such as operator observations of load and smoke. Then ask the supplier whether a firmware update or additional sensors can fill the gap. See specifying replacement RECD sensors for compatibility.

Is this report required by regulation?

Requirements vary by authority, state and permit conditions. Check your applicable rules and conditions. Our page on DG set emission regulations gives background, but your permit and regulator define what you must keep.

Should the supplier see the report?

Yes, in most cases. Shared records help diagnosis and keep the supplier engaged. Where reports contain commercial or sensitive plant information, share the relevant sections.

How is this different from a service report?

A service report describes one visit. A regeneration history report covers a period and many cycles, with trends and actions. Use both. See auditing an RECD service report for missing information.

Bringing it together

A useful RECD regeneration history report has nine sections: identification, operating context, cycle summary, cycle-by-cycle table, trends, failures, maintenance and changes, assessment with actions, and data quality with sign-off. It records your verified outcome beside the controller’s verdict, always pairs ΔP with load, includes red-flag rules and keeps one consistent template. Produce it on a regular rhythm, assign an owner and store it with the asset record.

If you would like help designing a report template that matches your RECD’s controller, see our retrofit emission control devices, browse the FAQ or contact the Aceget team.

Related Reading



Leave a Reply