RECD Regeneration Not Completing: What Operators Should Record

A regeneration cycle that starts but does not finish is one of the most common RECD service calls. It is also one of the hardest to diagnose after the fact. By the time an engineer arrives, the set has cooled, the controller may have overwritten its short-term data, and the operator remembers only that “the light came on again”.

The fix is not complicated. It is a habit: write down a small set of facts every time a cycle fails to complete. Good records turn a vague complaint into a pattern that can be traced to its cause, usually within one visit.

This article is written for DG operators, facility supervisors and maintenance contractors. It explains what “not completing” actually means, which details matter, and how to record them in a way your RECD supplier can use. It focuses on everyday operation. If you are planning an emission test and want to know how to handle regeneration during sampling, our separate guide on RECD testing during regeneration covers that.

What Counts as “Not Completing”?

Operators often use one phrase for four different outcomes. They have different causes, so record which one you saw.

Outcome	What the operator sees	Typical controller wording
Aborted	Cycle starts, then stops early with an alarm or status change	Regen aborted, Regen stopped, Fault
Timed out	Cycle runs for its full allowed time without reaching its end condition	Regen timeout, Incomplete
Interrupted	Cycle stops because the engine stopped, the load changed or power was lost	Regen interrupted, Engine stop
Completed but ineffective	Controller reports completion, but ΔP stays high or a new request comes soon	Regen complete, followed by Regen request

A fifth situation is often confused with these: a cycle that never starts because a condition blocks it. That is an inhibit, not an incomplete cycle, and is covered in RECD Regeneration Inhibit Signals: When a Cycle Cannot Start.

Why One Failed Cycle Matters Less Than the Pattern

A single incomplete cycle is rarely an emergency. Soot that did not burn this time can burn next time. The risk is cumulative:

• Each failed cycle leaves the filter with more soot than the controller planned for
• The next cycle starts from a heavier load, which raises peak temperature if it does succeed
• Repeated failures can push ΔP towards the level at which the controller blocks regeneration entirely, to protect the filter
• Back-pressure on the engine rises in the meantime; see how exhaust back-pressure affects a DG set

So the aim of recording is not to react to every event. It is to build enough history that a trend becomes obvious.

The Record: Five Blocks of Information

Group what you write down into five blocks. Each answers a different diagnostic question.

Block 1: Identification

Which device, and which cycle?

• Site, DG set number and RECD serial number
• Date and clock time the cycle started and stopped
• DG hour meter reading
• Regeneration counter value from the controller, if shown
• Operator name or shift

Times must be comparable with the DG controller and BMS. If clocks disagree, note the offset. Our guide on aligning RECD sensor timestamps with generator runtime logs explains why a few minutes of drift can hide the real cause.

Block 2: Conditions at the start

Was the device in a fit state to regenerate?

• Engine load in kW (not just percentage) and how long it had been at that load
• Exhaust temperature before the RECD
• ΔP across the RECD
• Time since the previous completed regeneration
• Whether the cycle was automatic, operator-requested or scheduled
• For heater-type devices: heater supply voltage, and whether it came from mains or the set itself

Block 3: What happened during the cycle

Did the filter actually heat and burn soot?

• Inlet and outlet temperature at start, at peak and at stop
• Whether outlet temperature ever exceeded inlet (a sign of ignition)
• ΔP at start and at stop
• Any load changes, transfers or test runs during the cycle
• Any alarms on the DG controller during the cycle

Block 4: How it ended

Why did it stop?

• Exact controller message, written word for word or photographed
• Which of the four outcomes from the table above applies
• Whether the engine was still running when the cycle stopped
• Whether any other equipment tripped or switched at the same moment

Block 5: Afterwards

What state was the filter left in?

• ΔP at the same reference load 15–30 minutes after the event
• Whether a new regeneration request appeared, and how soon
• Any visible smoke from the stack
• Action taken: retry, wait, call service, note only

That is around twenty items. Most come straight off the panel and take less than ten minutes to note.

A Copy-Ready Log Format

The table below can be printed and kept beside the DG panel, or set up as a shared spreadsheet.

Field	Entry
Site / DG no. / RECD serial
Date, start time, stop time
DG hour meter
Outcome (aborted / timed out / interrupted / completed but ineffective)
Trigger (auto / manual / scheduled)
Load at start (kW) and minutes at that load
Inlet temperature at start / peak / stop (°C)
Outlet temperature at start / peak / stop (°C)
ΔP at start / stop / 30 min later (mbar or kPa)
Heater supply voltage and source (if applicable)
Events during cycle (load change, ATS transfer, test run, trip)
Exact controller message
Previous completed regeneration (date, hours ago)
Action taken
Recorded by

If your site has an RECD dashboard or BMS link, many of these fields can be captured automatically. See integrating RECD monitoring with a building management system and what to show on an RECD operating dashboard. Even then, an operator note on what was happening around the set adds context the sensors cannot.

Five Patterns Good Records Reveal

After five to ten entries, look across them. These patterns account for most incomplete cycles.

Pattern A: Failures cluster at light load

If most failed cycles began with the set at low kW, the exhaust was probably too cool for the device to reach its target, or the device expected a minimum load that the building did not provide. This is common on standby sets that only run for short weekly tests. Read choosing an RECD for a generator running at low load and RECD Regeneration Temperature: Why There Is No Universal Setpoint.

Pattern B: Failures coincide with the set stopping

If the outcome is usually “interrupted” and the engine stopped mid-cycle, the cause is often the site’s run pattern: grid returning, a test-run timer ending, or an ATS retransfer. That has its own risks, explained in What Happens When a Generator Stops During RECD Regeneration?. Coordinating ATS and test timers with the RECD is covered in RECD fitment and ATS testing coordination.

Pattern C: Heater-type cycles fail with low supply voltage

On electrically heated devices, a weak or interrupted supply can stop the heater reaching temperature. Look at Block 2’s supply voltage entries. Undersized cables, a supply taken from a loaded DB, or a breaker that trips under heater current are common findings. See RECD controller and heater power supply requirements.

Pattern D: Cycles “complete” but ΔP barely moves

If the controller reports success yet ΔP after the event is close to ΔP before it, the restriction may not be soot. Ash, a blocked sensor line or a drifted pressure sensor are the usual suspects. Read Soot vs Ash in an RECD and separating a sensor fault from an RECD performance problem.

Pattern E: Failures increase over months

A slow rise in failure rate, along with more frequent requests, points to a changing engine or a filter approaching its ash limit. Check oil consumption, injector condition and the last cleaning date. Our article on How Frequently Should a Filter-Based RECD Regenerate? explains how to read the trend.

What Not to Do While Recording

Recording is not the same as intervening. Avoid these common reactions until the cause is understood:

• Do not repeatedly force manual cycles on a filter that keeps failing. Each attempt adds heat stress and may start with an even heavier soot load.
• Do not reset alarms without noting them. A cleared alarm with no record is lost evidence.
• Do not change controller settings to stop requests. Settings changes should follow your site’s access control rules; see RECD controller settings access control.
• Do not bypass the device to relieve back-pressure without written approval. A bypass can affect compliance and warranty.
• Do not open hot housings or sensor lines. Let the device cool and follow the supplier’s isolation procedure.

When to Escalate

Set simple escalation rules so operators know when a note becomes a call. A practical starting point:

Situation	Action
One incomplete cycle, ΔP within normal band afterwards	Record only
Two incomplete cycles within a week	Record, inform supervisor, check load pattern
Three or more incomplete cycles, or ΔP rising after each	Call RECD service provider with the log
ΔP approaching the controller’s high alarm	Call service; follow supplier instructions on running the set
Any over-temperature alarm, smoke from housing or burning smell	Stop regeneration if safe to do so, follow site emergency procedure, call service

Adjust the thresholds with your supplier. Many sites fold these rules into a wider RECD alarm response matrix.

Handing the Record to Your Service Provider

When you call for service, send the log in advance. A useful handover includes:

1. The completed log entries for each failed cycle
2. Photos of the controller screen showing the message and counters
3. A one-week extract of DG run hours and load, if available
4. Dates of the last RECD service, last ash cleaning and any engine work
5. Any changes on site, such as a new load, a new ATS setting or a fuel supplier change

With that information, an engineer can often narrow the cause before arriving and bring the right parts. It also creates the paper trail needed if a warranty claim becomes necessary. Engine-side causes should be passed to the DG service team too; see coordinating DG engine repairs with the RECD service team.

Building the Habit Into Shift Routines

Records only work if they are kept consistently. Three small steps help:

• Add an RECD line to the shift handover sheet, so the incoming operator knows whether a cycle failed. Our RECD shift handover log template includes one.
• Train every operator who can start the set, not just the senior technician. Our article on RECD operator training before handover lists the skills worth checking.
• Review the log monthly, even when nothing seems wrong. A quiet month is useful baseline data.

Where sites are unattended, remote alerts can capture failed cycles automatically; see remote RECD alerts for unattended sites. Make sure missing data is spotted too, using the checks in detecting missing data in RECD monitoring records.

Frequently Asked Questions

Why does my RECD regeneration keep stopping halfway?

Common causes are low exhaust temperature at light load, the engine stopping mid-cycle, an inadequate heater supply, or a controller protection limit. Your records of load, temperature, ΔP and the exact message will show which.

Is one incomplete regeneration cycle dangerous?

Usually not. The concern is repeated failures, which let soot build up and make the next successful cycle hotter.

Should I start a manual regeneration after a failed cycle?

Only if your supplier’s procedure allows it and the conditions that caused the failure have been corrected. Repeated manual attempts on a failing filter can cause damage.

What is the single most useful reading to record?

ΔP across the RECD at a known load before and after the cycle. It shows whether soot was actually removed.

How long should I keep regeneration records?

At least as long as the warranty and your compliance record-keeping period. Many sites keep them for the life of the device.

Records Turn Repeat Calls Into One Fix

An incomplete regeneration is a symptom, not a diagnosis. Writing down the same facts every time lets the pattern point to the cause, whether that is load, supply, sensors, ash or engine condition.

Aceget supplies CPCB type-approved Retrofit Emission Control Devices and supports clients with operator training and log reviews. Visit our FAQ or talk to our service team if your device is not completing its cycles.