October 27, 2026
Posted by: editor
Category: RECD & Emission Control Devices

Most RECD warranty arguments are not really about whether a part failed. The failed part is usually sitting there for everyone to see. The argument is about why it failed, and that question is settled by records: how the device was installed, how the DG set was run, and how both were maintained in the months before the fault.

That is why a successful RECD warranty claim is built long before anyone fills in a claim form. It is built from shift logs, commissioning reports, service records and a handful of photographs taken at the right moment. This guide walks through the evidence you need, in the order you will need it, from the minute an alarm appears to the day the replacement part is recommissioned.

One boundary first. This post is about evidence at claim time. If you want to understand what a typical warranty covers in the first place, such as parts versus labour and the usual exclusions, read our guide to RECD warranty terms, parts, labour and exclusions alongside this one. Your own warranty certificate and purchase contract always take precedence over anything general written here.

The three questions every supplier asks

Whatever the device type, a supplier’s warranty team will try to answer three questions before accepting a claim. If your file answers them clearly, most of the work is done.

1. Was it installed and commissioned as specified?

The supplier wants to know the RECD went onto the engine it was sized for, installed as the approved drawings show. Evidence that answers this:

The approved general arrangement drawing and any signed deviations
The installation completion certificate or punch-list closure note
The site acceptance test (SAT) or commissioning report, with baseline readings
Photographs of the completed installation taken at handover

If acceptance was done properly, our explainer on factory acceptance versus site acceptance tests shows the kind of baseline that should already be in your hands.

2. Was it operated within the stated conditions?

Every RECD has an operating envelope: load range, exhaust temperature, run patterns and fuel specification. Evidence that the set stayed inside it:

Shift or run logs showing load, run hours and start/stop pattern
Differential pressure (ΔP) and temperature readings over time, if your system records them
Controller alarm and event history, where the design includes a controller
Fuel purchase records showing the grade bought

Our post on RECD performance guarantees and operating conditions covers how these conditions are usually written into offers.

3. Was it maintained as instructed?

The manual sets cleaning, inspection or regeneration intervals, plus engine conditions such as oil consumption limits. Evidence that answers this:

Dated maintenance records for the RECD itself
Engine service records: oil changes, oil top-ups, injector and turbocharger work
Records of who did the work, and whether they were authorised

A useful reference for the routine side is our RECD maintenance guide.

Stage 1: The moment the fault appears

The first hour is when the most useful evidence exists and is easiest to lose. Pause and record before anyone resets alarms or opens covers.

What to record immediately

Write these down on paper or in your log before anything is reset or touched:

Date and time in ISO 8601 format, for example 2026-10-14T15:42+05:30. A consistent format avoids confusion when your log, the controller clock and the supplier’s ticket are compared later. The ISO 8601 date and time format is the standard reference.
DG load at the moment of the fault, in kW or percentage, plus how long the set had been running in that run.
Every active alarm and warning on the RECD controller (if fitted) and on the DG controller, copied word for word, including codes.
Readings: ΔP, inlet and outlet exhaust temperature, engine coolant temperature and oil pressure, whatever your instruments show.
Visible smoke at the stack: colour (black, white, blue), whether continuous or in puffs.
Sounds and smells: rattling, whistling, hissing at joints, burning odour.
What the operator did in sequence: reduced load, stopped the set, acknowledged alarms, called someone.

If your site already runs a structured log, the fault entry slots into it naturally. Our sibling guide on building an RECD shift-handover log has a format that captures most of these fields by default.

What not to do

Do not dismantle anything before photographing it and speaking to the supplier. Opening a housing or removing an element can destroy evidence of how a failure developed.
Do not bypass or remove the RECD to keep the DG running. Apart from risking the claim, running without the approved device is a compliance risk that can lead to sealing and penalties.
Do not reset and keep re-running the set to “see if it clears”. Repeated runs can enlarge the damage and stress the engine through high backpressure. Our explainer on how exhaust backpressure affects a DG set explains why.
Do not discard anything that falls out: fragments of substrate, gasket pieces, sensor parts, or ash. Bag and label them.
Do not let a third party “have a quick look” without telling the supplier. Unapproved repair is one of the commonest reasons for disputes.

Before anyone approaches the device, make sure the set is stopped, isolated and allowed to cool. Exhaust components stay hot long after shutdown, and the start circuit must be locked and tagged out. The OSHA guidance on controlling hazardous energy is a useful reference for lockout/tagout principles if your site procedure needs a refresher.

Photograph before touching

Take wide shots first, then close-ups, all before any panel is opened. Check the phone’s date and time, because image metadata becomes part of your evidence.

Stage 2: The first 48 hours

Notify the supplier in writing

A phone call is fine for speed, but follow it with written notice the same day, by email or the supplier’s service portal, to the contact named in your warranty certificate or service agreement. The written notice should include:

Site name, address and contact person
DG make, model, kVA rating and engine serial number
RECD model and serial number (from its nameplate)
Date of commissioning and current run hours
A short factual description of the fault, with your recorded readings
A statement that the device has not been opened or altered
A request for a complaint or ticket reference number and a proposed inspection date

Ask for the reference number in writing and quote it on every later email. If your purchase order included response-time commitments, note the clock start. Our guide to RECD service response commitments explains what those commitments usually look like.

Keep the evidence secure

Leave the DG locked out with a tag that says “Under warranty investigation, do not operate or open”.
Store any loose fragments in labelled, sealed bags with date and finder’s name.
Export or photograph the controller logs before they roll over. Some controllers keep only a limited number of events.

Interim compliance considerations

A failed RECD often means the DG cannot be used in a compliant way until it is repaired. Do not assume you can simply run without it. Depending on your location, consent conditions, and any CAQM directions in force in Delhi NCR (see caqm.nic.in for current orders), operating a set without a working approved device could expose you to action.

If you need backup power meanwhile, consider a compliant hired set or load shedding. Where there is genuine doubt, speak to your State Pollution Control Board and record what was discussed. The consequences of getting this wrong are covered in our post on what happens when a DG set is sealed for non-compliance.

Stage 3: Building the RECD warranty claim file

With the supplier notified, assemble one claim file, ideally a single folder (physical and digital) with an index. This is the heart of your RECD warranty evidence. The table below lists what typically belongs in it.

Document	What it proves	Where to find it
Invoice and purchase order	Ownership, purchase date, model, warranty start trigger	Accounts or purchase department
Warranty certificate and terms	Coverage period, conditions, claim procedure	Supplied with the device or in the contract
Approved GA drawing and deviation records	Installation matched the agreed design	Project file, installer handover pack
Commissioning / SAT report	Device worked correctly at start; baseline values	Supplier or installer handover pack
ΔP and temperature baseline	Normal reference values at known loads	SAT report, first weeks of logs
Shift or run logs	Load, run hours, start/stop pattern, readings over time	Operator logbook or digital log
Run-hour history	Total and recent duty, to compare with service intervals	Hour meter photos, DG controller, logs
RECD maintenance records	Cleaning, inspection or regeneration done on time	Maintenance register, AMC reports
Engine service records	Oil changes, oil consumption, injector and turbo work	DG service provider reports
Fuel quality records	Fuel grade and source; no contaminated batches	Fuel purchase receipts, delivery notes
Controller event and alarm logs	Alarm history and when the fault developed	Exported from controller, if fitted
Sensor calibration or replacement records	Readings were trustworthy; correct part numbers used	Service reports, calibration certificates
Records of any third-party work	Who touched the system and whether approved	Work orders, gate passes, invoices
Transport and storage records	Device condition before installation	Delivery inspection report, storage log
Photos and video	Condition at the time of fault	Your shot list (below)
Correspondence	Notification date, reference numbers, promises made	Email, portal screenshots

Engine service records. Many RECD faults begin upstream. High oil consumption, a leaking injector or a worn turbo seal can load a filter with ash or wet soot faster than intended, and suppliers commonly check this first. Our post on engine oil consumption and RECD suitability explains the link, and DieselNet’s page on ash in particulate filters gives the technical background on how lubricant ash accumulates.

Third-party work. If a service company other than the supplier has been maintaining the device, include their authorisation (if any) and their reports. Our sibling post asking whether a different service company can maintain an installed RECD covers how to set that up so it does not undermine warranty later.

Sensors. If a ΔP or temperature sensor was replaced during the warranty period, record the part number and who supplied it. A non-equivalent sensor can produce misleading readings or alarms. See our guide to specifying replacement RECD sensors without losing compatibility.

Transport and storage. If the fault appeared at or soon after first start, the supplier may question handling before installation. Your delivery inspection report and storage notes become important. Our sibling guide on how to package and transport an RECD without damaging its internals lists what good handling evidence looks like.

A warranty claim summary sheet you can copy

Put a one-page summary on top of the file. Copy the layout below.

Field	Entry
Claim / ticket reference
Site name and address
Contact person, phone, email
DG make, model, kVA, engine serial no.
RECD model and serial no.
Date of commissioning
Run hours at commissioning / at fault
Date and time of fault (ISO 8601)
Load at fault (kW or %)
Alarms and codes displayed
ΔP and temperatures at fault
Baseline ΔP and temperatures at similar load
Last RECD maintenance date and type
Last engine service date and work done
Any third-party work since commissioning
Device opened or altered after fault? (Yes/No)
Interim power arrangement
Documents attached (list)
Requested remedy
Prepared by, date

Photo and video shot list

Use this numbered list on the day. Take each item before any panel, clamp or cover is moved, and again after the supplier opens the device if you are allowed to.

Wide shot of the DG room or enclosure showing the RECD in position.
RECD nameplate, readable, with serial number.
DG nameplate and engine serial plate.
Hour meter or controller screen showing current run hours.
Controller screen showing each active alarm, one photo per screen page.
ΔP gauge or display and temperature displays, if fitted.
Inlet and outlet flanges, clamps and bellows from two angles.
Any visible soot streaks, oil wetness or discolouration on the housing or joints.
Supports, hangers and brackets, including any cracks or loose bolts.
Sensor connectors, cables and their routing near hot surfaces.
Stack outlet, plus a short video of the exhaust if the set is safely run under supplier instruction.
Any loose fragments or debris, beside a ruler or coin for scale.
A 30 to 60 second walk-around video with spoken commentary of date, time and what is visible.
After opening by the supplier: the inlet and outlet faces of the element, and any cracks, melting or plugging patterns.

Keep the original, unedited files with metadata intact; send copies.

Stage 4: The inspection visit

Treat the supplier’s inspection as part of your evidence, not just a service call.

Preparing for the visit

Confirm the date, the inspectors’ names and what they plan to do.
Arrange safe access: lockout/tagout in place, cool equipment, lifting gear or scaffold if the device sits high, and a permit to work if your site requires one.
Make sure the person who saw the fault, usually the shift operator, is available to explain what happened.
Agree in advance whether the set will be run during the visit and under whose control.

Attending and recording

Have your own engineer present throughout. Note in writing:

Who attended and their roles
What was checked: visual inspection, ΔP and temperature readings, sensor checks, controller download, element removal
Whether parts or samples were taken away, with a signed receipt that lists them
Any readings taken and the load at which they were taken
Any verbal opinions about cause

Ask for written findings

Ask for a signed visit report and a date for written findings. A verbal “looks like an engine problem” is not a decision. If the supplier attributes the failure to a cause outside warranty, ask for the supporting evidence, such as oil analysis, ash composition or controller data.

Stage 5: The decision and what comes after

There are broadly three outcomes.

Accepted

The supplier agrees the fault is covered. Confirm in writing what will be replaced, who pays for labour, freight and recommissioning, and the expected date. Our sibling comparison of replacing an RECD filter core versus the complete assembly can help you understand which option is being offered and why.

Partially accepted

Typically the part is covered but labour or freight is not, or a goodwill share is offered. Compare the offer with your terms and, if they differ, point to the clause in writing.

Rejected

If the claim is refused, ask for:

The written reason, referring to the specific warranty condition.
The evidence used to reach that conclusion.
Return of any parts taken, or confirmation they will be kept for independent examination.

Then review your file against their reasoning. Sometimes the supplier is right; sometimes your logs answer the point. Keep the exchange factual and dated.

If discussion stalls, escalate in steps: the supplier’s service head, then the commercial or management contact named in your purchase order, then the dispute mechanism in your contract. For contractual remedies, take proper legal advice; this guide does not replace it.

After any outcome

Record the replacement part’s serial number and date fitted.
Recommission and record new ΔP and temperature baselines at known loads.
Update the maintenance register and set the next service date.
Note whether the replacement part carries its own warranty period.
Keep the claim file for the life of the device.

Why claims commonly get delayed or refused

Every warranty is different, but the same warranty claim rejected reasons come up repeatedly.

Missing baselines. Without commissioning ΔP and temperature values, there is nothing to compare the fault readings against, so “abnormal” becomes a matter of opinion.
Gaps in maintenance records. A missing cleaning entry or an overdue service, even if the work was actually done, is hard to defend later.
Engine faults upstream. Excess oil consumption, injector faults or turbo seal leaks are commonly treated as outside RECD warranty because they change what enters the device. Signs such as persistent blue smoke from the generator should be investigated and recorded promptly.
Unauthorised modification. Changing piping, removing insulation, relocating sensors or altering controller settings without written approval.
Third-party repairs without approval. Work by someone not authorised under the warranty, especially if the device was opened.
Physical or transport damage. Dents, cracked housings or impact marks commonly point the discussion towards handling or site damage.
Operation outside stated conditions. Long periods of very low load, repeated short runs, out-of-specification fuel, or loads above the rating.
Slow notification. Waiting weeks, or running on with an active alarm, makes cause harder to prove.

Almost every item here is a documentation problem as much as a technical one.

Illustrative example: a 250 kVA set with fast-rising ΔP

The following is an illustrative example. Numbers are chosen to show the method and arithmetic, not to represent any particular product, supplier or warranty.

The situation

A commercial building runs a 250 kVA DG set fitted with an RECD of the filter type, with ΔP and exhaust temperature displayed on a small controller. (For context on sizing at this rating, see our page on RECDs for 250 kVA DG sets.)

At commissioning, the SAT report recorded a ΔP of 6 kPa at 75% load. Over the next months, the shift log shows ΔP drifting slowly between 6 and 8 kPa at similar loads, with cleaning done on schedule.

At about 400 run hours, the operator notices ΔP climbing much faster than usual: 8 kPa to 14 kPa over roughly 20 run hours, with a high-ΔP warning on the controller. The stack remains clear. The operator records readings, stops the set, and locks it out.

What went into the claim file

Purchase order, invoice and warranty certificate
SAT report with the 6 kPa baseline at 75% load
Shift logs covering all 400 hours, showing average load between 55% and 80%, with few runs under 30 minutes
RECD maintenance register with all scheduled cleanings dated and signed
Engine service records showing oil changes on time, oil top-ups within the engine maker’s normal range, and an injector check at 250 hours
Fuel delivery notes from a single regular supplier
Exported controller event log showing the ΔP trend and the warning time
14 photographs and a walk-around video per the shot list
The summary sheet on top

The timeline in days

Day	Event
0	Fault observed, readings recorded, set locked out
1	Written notice sent; ticket reference received
2	Controller log exported; claim file assembled
3	Claim file and summary sheet emailed to supplier
6	Supplier inspection; element removed and a localised crack found
9	Written findings received: defect accepted under warranty
13	Replacement element fitted, recommissioned, new baseline recorded

From fault to recommissioning took 13 days. The logs were what made the decision quick: they showed load and run pattern inside the stated envelope, maintenance on time, and no engine-side cause for abnormal loading.

The cost picture

Assume an illustrative replacement element cost of ₹1,80,000 plus ₹25,000 for labour and recommissioning, both covered under this warranty. Covered value: ₹1,80,000 + ₹25,000 = ₹2,05,000.

The building hired a compliant rental set for backup during the 13 days at an illustrative ₹8,500 per day, which the warranty did not cover. Owner’s interim cost: 13 × ₹8,500 = ₹1,10,500.

If the claim had dragged on for an extra 20 days because baselines or maintenance records were missing, rental alone would have added 20 × ₹8,500 = ₹1,70,000. That is close to the price of the element itself. Good records did not only win the claim; they cut the downtime that the warranty never pays for.

FAQ

How soon should I raise an RECD warranty claim after a fault?

The same day, once the set is safe and readings are recorded. Send written notice and ask for a reference number. Delays make the cause harder to establish.

Can I clean or open the RECD before the supplier inspects it?

Generally no. Photograph everything and ask the supplier first. Unapproved opening can destroy evidence and is a common ground for refusal.

What photos are most important for an RECD warranty claim?

The nameplates, hour meter, controller alarm screens, ΔP and temperature displays, joints and supports, and any debris. Take them before touching anything and keep the original, unedited files.

Will engine problems affect my RECD warranty?

They can. Many suppliers commonly exclude damage caused by excess oil consumption, injector faults or fuel contamination. Keeping engine service records alongside RECD records is the best way to show the engine was in good order.

Can I keep running the DG set without the RECD while the claim is processed?

Do not bypass or remove the RECD to keep running. That is a compliance risk. Arrange compliant interim power or speak to your SPCB if you are unsure what is permitted at your site.

What if the supplier rejects my claim?

Ask for the written reason, the evidence behind it, and preservation of any parts taken. Respond with specific evidence and escalate through your purchase order contacts.

Need Help Assembling an RECD Claim File?

Aceget supplies CPCB type-approved RECDs for DG sets from 25 kVA to 2500 kVA, and we encourage every customer to keep the records described in this guide from the day of commissioning. If you want to understand the documentation to keep for your device, or to discuss a fault, start with our retrofit emission control device range, review our certifications and type approvals, and contact our team with your RECD serial number and run hours. For the general regulatory picture, the Central Pollution Control Board website remains the primary source.

What Should You Record When Raising an RECD Warranty Claim?