- October 17, 2026
- Posted by: editor
- Category: RECD & Emission Control Devices
A Retrofit Emission Control Device is tested and accepted against one particular DG set, as it existed on one particular day. Engines and installations don’t stay frozen. Injectors get replaced, turbochargers fail, exhaust pipework is rerouted during a building upgrade, and sometimes the whole set moves to a new site.
Each change raises the same question: is the original test result still valid, or does the RECD need to be retested?
Retesting after every small job is expensive and unnecessary. Never retesting can leave you relying on a result that no longer describes your installation. This guide sets out a practical, change-by-change approach: which changes call for a full retest, which only need a quick functional check, and which need nothing more than an entry in the log.
Three Levels of Retest
Not every change needs a full laboratory test. Three levels of response cover most situations.
Level 1: Log and monitor
Record the change in the DG and RECD log. Continue routine monitoring of differential pressure (ΔP), opacity and back-pressure at your standard check condition. Compare the next readings with the reference values from the last test. See why RECD test results may differ from day-to-day site readings.
Level 2: Functional check
A short, structured check after the change, usually done by the site team or the RECD supplier:
- Leak check of all exhaust joints between the engine and the stack
- ΔP, opacity, back-pressure and exhaust temperatures at the standard check condition, compared with reference values
- Visual inspection of the device, mounts, bellows and insulation
- Controller and sensor checks, for devices with active regeneration
- Short report signed and filed
Level 3: Full performance retest
A gravimetric particulate test by an accredited laboratory, with a matched comparison (upstream/downstream or before/after), following the same protocol as the original acceptance test where possible. See planning an RECD emissions test before and after installation.
The question for each change is which level it needs.
The Change-by-Change Decision Guide
Engine-side changes
| Change | Why it matters | Recommended level |
|---|---|---|
| Routine service (oil, filters) to OEM specification | Restores normal condition; doesn’t change engine design | Level 1 |
| Injector replacement or recalibration | Changes combustion and soot output | Level 2, and Level 3 if a performance figure will be quoted afterwards |
| Fuel pump, governor or timing adjustment | Can change soot and NOx substantially | Level 2, often Level 3 |
| Turbocharger replacement | Changes air supply and back-pressure tolerance | Level 2; Level 3 if a different model or rating |
| Top or major overhaul (rings, liners, valves) | Changes oil consumption and engine-out emissions | Level 3 |
| Engine replaced with the same model | Different engine, even if identical type | Level 2 at minimum; Level 3 recommended |
| Engine replaced with a different model or rating | Device suitability must be re-confirmed | Re-engineering review, then Level 3 |
| Conversion to dual-fuel operation | Changes exhaust composition and temperature | Re-engineering review, then Level 3 |
| Electronic control software update | Can change injection strategy | Level 2; Level 3 if emissions behaviour changes |
The key question for engine changes is whether the exhaust the RECD receives has changed: its soot content, organic fraction, temperature, flow or oil-derived ash. If it has, the original test no longer describes the system. See should you complete an engine overhaul before fitting an RECD and RECD after diesel-to-dual-fuel conversion.
Rating and duty changes
| Change | Why it matters | Recommended level |
|---|---|---|
| Set re-rated (higher kVA) or alternator upgraded | Higher exhaust flow and temperature | Re-engineering review, then Level 3 |
| Duty changes from standby to prime power | Much more running, different load profile | Level 2, plus a review of regeneration and ash cleaning intervals |
| Typical load drops substantially (e.g. load moved to grid or solar) | Cooler exhaust; regeneration may suffer | Level 2; Level 3 if low-load performance is critical |
| New, highly variable or cyclic loads connected | Transient behaviour changes | Level 2 |
See specifying an RECD for future generator load increases, prime power vs standby: same RECD specification? and choosing an RECD for a generator running at low load.
Exhaust system changes
| Change | Why it matters | Recommended level |
|---|---|---|
| Gasket or bellows replaced like-for-like | Possible leak path | Level 2 (leak check essential) |
| Pipework rerouted, extended or resized | Back-pressure, temperature at the device, sampling port validity | Level 2; Level 3 if back-pressure changes noticeably or ports have moved |
| Silencer added, removed or replaced | Back-pressure and possible deposit release | Level 2 |
| Chimney extended or shared with another set | Back-pressure, dilution, mixing with other exhaust | Level 2; Level 3 if comparisons will be made |
| Sampling ports moved | Future results no longer comparable with the original | Level 3 to set a new baseline |
See reusing existing exhaust pipework for an RECD retrofit, RECD exhaust bends and reducers: a layout review and how exhaust sampling location affects RECD test results.
RECD-side changes
| Change | Why it matters | Recommended level |
|---|---|---|
| Ash cleaning by the supplier’s procedure | Restores capacity; disturbs the substrate | Level 2 |
| Substrate or catalyst replacement | Essentially a new filter element | Level 2 after break-in; Level 3 if a performance figure is needed |
| Controller, heater or sensor replacement | Affects regeneration behaviour | Level 2, including a regeneration check |
| Repair after a crack, overheating or uncontrolled burn | Efficiency may have been lost | Level 3 |
| RECD moved to a different DG set | Different engine and installation | Suitability review, then Level 3. See can an RECD be transferred to another diesel generator |
| Whole DG set and RECD relocated to another site | New exhaust layout, possibly new consent | Level 2; Level 3 if consent or contract requires |
Fuel and operating changes
| Change | Why it matters | Recommended level |
|---|---|---|
| New fuel supplier meeting the same specification | Minor variability | Level 1 |
| Switch to a biodiesel blend | Changes particulate make-up and soot burning | Level 2; Level 3 if a figure will be quoted |
| Long lay-up (months unused) | Corrosion, moisture, stuck components | Level 2 before returning to service |
See using biodiesel blends in an RECD-equipped generator.
Non-physical triggers
Some retests are needed even when nothing physical has changed:
| Trigger | Recommended level |
|---|---|
| Routine indicators (ΔP, opacity) breach agreed trigger values | Level 2, then Level 3 if the cause isn’t found |
| Consent renewal or a Pollution Control Board request | As the consent or the board specifies |
| A warranty claim or performance dispute | Level 3, ideally witnessed |
| A sale, audit or lender’s due diligence asks for current performance | Level 3 if the last test is old or conditions have changed |
| A long time since the last full test, with significant operating hours | Level 3 at an interval agreed with your supplier |
See how to set up a witnessed RECD performance demonstration and how often to renew a DG set pollution certificate.
A Simple Rule of Thumb
When you’re unsure, ask three questions:
- Has the exhaust entering the RECD changed? Different engine, combustion, fuel, flow or temperature means at least Level 2.
- Has the path through or around the RECD changed? New joints, pipework, ports or device components mean at least Level 2, with a leak check.
- Will anyone rely on a performance figure after the change? A contract, regulator, auditor or ESG report means Level 3.
If the answer to all three is “no,” Level 1 is usually enough.
Worked Example: A Year of Changes on One Set
The events below are illustrative, for a 320 kVA set at a manufacturing site, accepted with a 76% PM reduction at 50% load. Reference values from acceptance: ΔP 48 mbar, opacity 0.8% at 50% load, warm.
| Month | Change | Level chosen | Outcome |
|---|---|---|---|
| 2 | Oil and filter service | 1 | Next routine readings normal |
| 4 | Two injectors replaced after misfire | 2 | Opacity 0.6%, ΔP 45 mbar; leak check passed |
| 6 | Exhaust rerouted for a new mezzanine; two extra bends | 2 | Back-pressure up but within limit; ΔP unchanged; ports unaffected |
| 8 | Turbocharger replaced with a different model | 2 → 3 | Functional check normal, but the different model triggered a full retest: 74% |
| 10 | ΔP drops to 30 mbar with no cleaning, opacity 3.1% | 2 → 3 | Inspection found a cracked substrate; replaced; retest after break-in: 75% |
| 12 | Annual consent renewal | Consent monitoring | Downstream concentration consistent with previous year |
Only two full retests were needed in the year. One was triggered by a significant engine change, the other by monitoring indicators. All other changes were handled with logs and functional checks.
Building a Change-Control Routine
Many sites already have change control for production equipment. Extend it to RECD-fitted DG sets:
- Change request. Whoever plans the work, such as maintenance, a contractor or a project team, notes it on a simple form.
- Classification. The EHS or utilities lead decides on Level 1, 2 or 3, using a matrix like the one above, agreed with the RECD supplier.
- Pre-change readings. Where practical, take ΔP, opacity and back-pressure at the standard condition before the work, so you have a fresh comparison.
- Post-change action. Carry out the agreed level of check or retest.
- Record and file. Keep the form, readings and any test report with the RECD’s history.
This is particularly important in regulated sectors. See RECD for pharmaceutical plants: change control and RECDs for data centres: retrofit and redundancy testing.
A Change-Control Form You Can Adapt
| Field | Entry |
|---|---|
| DG set ID and RECD serial number | |
| Description of planned change | |
| Planned date and contractor | |
| Does it change exhaust entering the RECD? (Y/N) | |
| Does it change the exhaust path or device components? (Y/N) | |
| Will a performance figure be relied on afterwards? (Y/N) | |
| Level assigned (1/2/3) and by whom | |
| Pre-change readings (ΔP, opacity, back-pressure, load) | |
| Post-change readings | |
| Leak check result | |
| Retest report reference (if Level 3) | |
| Reference values updated? (Y/N) | |
| Sign-off |
The Cost of Getting It Wrong, in Both Directions
Retesting too little carries real risks:
- A compliance claim based on an old test may no longer be true, which matters during a Pollution Control Board inspection or if a complaint is made
- Damage from engine work or exhaust changes can go unnoticed, so the RECD may be filtering far less than you think
- A warranty claim may be weakened if changes weren’t recorded or checked
- An excessive back-pressure problem introduced by exhaust work can harm the engine
Retesting too much also has costs:
- Accredited laboratory fees, load bank hire and staff time for each full test
- Downtime or disruption while the set is tested
- Comparisons become harder if retests use inconsistent protocols
The three-level approach keeps full retests for the changes that need them, and uses cheap, quick checks for the rest.
Planning Retests Around Operations
Full retests need the DG set running at defined loads for several hours, which isn’t always easy on a working site. Some practical ways to fit them in:
- Combine with consent monitoring. If an accredited lab is already visiting for consent stack monitoring, ask whether it can also do an upstream/downstream test under your protocol on the same day.
- Use planned outage or maintenance windows. Schedule retests when the site can run on the DG and a load bank without disrupting production.
- Avoid restricted periods. In Delhi NCR, plan around GRAP stages that limit DG operation. See GRAP stages 1 to 4: what changes at each level.
- Allow break-in time. After engine work or substrate replacement, plan the retest after a short period of normal operation, not on the same day.
- Book a load bank early. Load bank availability is often the bottleneck. A load bank also allows precise, repeatable load points.
- Group changes where possible. If several changes are planned within a few weeks, a single retest after all of them may be more efficient, as long as the set isn’t relied on for compliance in between.
Practical Tips for Retests
- Use the same protocol as the original test: same load points, stability criteria, reporting basis and, ideally, same ports. Otherwise results won’t be comparable. See how to compare RECD test results from different laboratories.
- Allow a break-in period after substrate replacement or engine work before testing. See can RECD efficiency change between new and aged conditions.
- Report uncertainty, especially if the result is close to 70%. See how to interpret uncertainty in an RECD emission reduction claim.
- Update the fingerprint. After every Level 3 retest, record new reference values for routine monitoring.
- Check warranty terms. Some warranties require the supplier to be informed of, or to approve, changes to the engine or exhaust. See will installing an RECD affect the generator’s OEM warranty and RECD warranty terms: parts, labour and exclusions.
Retest Triggers to Write Into Your Supply Contract
The best time to agree retest rules is when the RECD is ordered, not when the first change happens. Useful clauses include:
- Notification: the owner will inform the supplier of planned changes to the engine, rating, fuel type or exhaust system before they’re made.
- Classification: changes will be classified as Level 1, 2 or 3 using an agreed matrix attached to the contract.
- Functional check support: the supplier will carry out or support Level 2 checks within an agreed response time, at an agreed cost or included in the service agreement.
- Retest protocol: Level 3 retests will follow the same protocol as the acceptance test, including load points, reporting basis, uncertainty reporting and decision rule.
- Warranty effect: which changes, if made without notification or checks, affect the warranty.
- Monitoring triggers: ΔP and opacity values that automatically trigger a Level 2 check.
Our guides on RECD service response commitments, RECD performance guarantees and operating conditions and writing a technical RFQ for an RECD project show where these clauses fit.
What About the Type-Approval Certificate?
The RECD’s type approval covers the device model for a defined range of engines. It isn’t tied to your particular engine’s condition, so routine engine work doesn’t affect it. However, if the engine is replaced with one outside the range covered by the device’s approval, or re-rated beyond it, the device may no longer be a covered application. In that case, confirm suitability with the supplier and check the approval scope. See how to read an RECD test certificate and how to verify a CPCB-tested RECD on your DG set. The CPCB website is the authoritative source for the current framework.
Related Reading
- Can RECD efficiency change between new and aged conditions?
- RECD particle number vs particle mass: what each test shows
- How to interpret uncertainty in an RECD emission reduction claim
- Why RECD test results may differ from day-to-day site readings
- How to manage scope changes after an RECD order
Frequently Asked Questions
Do I need to retest the RECD after every engine service?
No. A routine service to the engine maker’s specification only restores normal condition. Log it and carry on with routine monitoring. Retest when the change affects combustion, air supply, oil consumption or the exhaust path, or when someone will rely on a performance figure afterwards.
My engine was replaced with an identical model. Is the old test still valid?
It’s a different engine, with its own emission characteristics and condition, so a functional check is the minimum and a full retest is recommended if a performance figure is needed. If the new engine is a different model or rating, confirm the device’s suitability before retesting.
Who decides whether a retest is needed?
Usually the site’s EHS or utilities lead, using a matrix agreed with the RECD supplier. Contracts, warranties and consent conditions may also specify retest triggers. Check them before deciding.
Can I move an RECD to another generator without retesting?
Not if you want a valid performance figure. The device must suit the new engine’s size, flow and duty, and the installation is new. A suitability review followed by a full retest is the safe approach.
What should I do if routine readings suddenly change with no known cause?
Treat it as a trigger. Carry out a functional check, including a leak check and inspection. If no cause is found, or the readings point to damage, arrange a full retest. Record each step, so the history is clear if anyone asks about it later.
Keep Your RECD’s Evidence Current
A test result describes the system that was tested. When the system changes, the evidence needs to keep up. Aceget supplies CPCB type-approved Retrofit Emission Control Devices for DG sets from 25 kVA to 2500 kVA, and we help clients set up change-control matrices, functional check procedures and retest protocols. See our certifications, learn how an RECD works, or contact our team before your next DG modification.
[…] When should an RECD be retested after changes to the DG set? […]