- November 14, 2026
- Posted by: editor
- Category: RECD & Emission Control Devices
How much energy does one regeneration cycle use? Suppliers often answer with a heater rating, for example “6 kW”, but a rating is not a measurement. The heater may not run at full power for the whole cycle, a controller may switch it on and off, voltage may sag, and the cycle may end early. The energy that matters, in kilowatt-hours (kWh) or litres of diesel, is the area under the power curve, and the only way to know it is to measure.
This article is a worked example. We set up a measurement on a heater-assisted RECD, log the data, integrate it by hand, check it three ways, estimate the uncertainty, convert it to cost and fuel, and show what the number can and cannot tell you. The device and figures are illustrative, not from any specific product, and the method is the point. If your RECD uses fuel assistance rather than an electric heater, a section near the end shows how to measure that instead.
Why bother? Three reasons come up repeatedly.
- Comparing offers. Two suppliers may both say “low energy”. A measured kWh per cycle on a comparable duty turns the claim into a number. See electric-heater versus fuel-assisted RECD regeneration.
- Spotting heater ageing. A heater that takes longer or draws less current to do the same job shows up as a change in energy per cycle. Compare with the lesson in what a rising regeneration duration tells you.
- Budgeting and reporting. Operating cost and any energy or carbon disclosure need real figures, not nameplates.
The quantity we are measuring
Energy per cycle is the electrical energy consumed by the regeneration system between the start and end of one cycle, measured at its supply terminals. We split it into two parts.
- Heater energy: the energy drawn by the heater circuit.
- Auxiliary energy: controller, sensors, blower or fan, contactors and any other load that exists only because of regeneration.
Energy is power multiplied by time. If the power is constant, energy is just power times duration. If power changes, you add up many small slices.
For a three-phase heater, the real power in kilowatts is approximately:
P (kW) = √3 × V(line) × I(line) × power factor ÷ 1000
For a resistive heater, the power factor is close to 1, but a power analyser that reads kW directly avoids the need to assume it.
Step 1: Plan the measurement
Before connecting anything, agree the following with the supplier and your electrician. Electrical work on heater feeders must be done by qualified, authorised people following site safety rules.
| Decision | Choice for this example |
|---|---|
| What is measured | Heater feeder, three phases, plus controller supply separately |
| Instrument | Logging three-phase power analyser with clamp-on current probes, reading kW and kWh |
| Sampling interval | Every 1 minute (every 5 minutes shown below for clarity) |
| Window | From heater contactor closing to controller declaring cycle end |
| DG load during test | Steady, about 125 kW, logged from the DG controller |
| Other data recorded | ΔP, bed and inlet temperature, supply voltage, heater contactor status |
| Repeat | Three cycles, if possible, on different days |
Record the instrument’s model, calibration date and accuracy class. For an inexpensive clamp meter the typical uncertainty is larger than for a calibrated analyser. See RECD test instrument calibration records for the kind of documentation worth keeping.
Safety note. Do not open energised terminal boxes or place probes yourself unless qualified. Use clamp-on current probes around insulated cables where possible, and have the supplier’s technician present.
Step 2: Set up and start the log
The analyser clamps go around the three heater phase cables, and voltage leads connect at the heater breaker outgoing terminals. A second channel, or a second small meter, records the controller supply. Time on the analyser is synchronised with the controller clock, so you can line up events. Our article on aligning RECD sensor timestamps with generator runtime logs explains why this step saves headaches.
Start logging before the cycle starts, so you record the pre-cycle baseline, and stop after the cycle ends, so you capture any heater run-on.
Step 3: The raw data
The table shows five-minute readings from one illustrative cycle. In real life you would log every minute, but five-minute points are enough to demonstrate the method.
| Time since heater on (min) | Heater power (kW) | Inlet temp (°C) | Bed temp (°C) | ΔP (mbar) | Notes |
|---|---|---|---|---|---|
| 0 | 5.9 | 360 | 310 | 61 | Heater contactor closes |
| 5 | 6.0 | 372 | 415 | 61 | Full power |
| 10 | 6.0 | 378 | 505 | 60 | Bed heating |
| 15 | 5.8 | 381 | 570 | 58 | Soot ignition likely |
| 20 | 4.4 | 383 | 598 | 52 | Heater begins to modulate |
| 25 | 3.4 | 384 | 612 | 47 | Soot burning, less heater input |
| 30 | 2.7 | 383 | 604 | 43 | |
| 35 | 1.5 | 381 | 575 | 40 | |
| 40 | 0.0 | 380 | 540 | 38 | Heater off |
| 55 | 0.0 | 377 | 430 | 38 | Controller declares complete |
Quick observations from the raw numbers:
- ΔP fell from 61 to 38 mbar, a drop of 23 mbar, at a steady DG load. That supports a successful clean-up. See interpreting an RECD regeneration completion message for the checks behind that statement.
- The heater drew close to its 6 kW rating until the bed was hot, then eased off as soot burned and the bed temperature rose.
- The heater was on for 40 minutes, but the cycle lasted 55.
Step 4: Integrate the power curve
Energy is the area under the power-time curve. With readings at five-minute intervals we use the trapezoid rule: for each interval, take the average of the two readings and multiply by the interval length in hours (5 minutes = 5 ÷ 60 = 0.0833 hours).
| Interval (min) | Start kW | End kW | Average kW | Energy (kWh) = average × 0.0833 |
|---|---|---|---|---|
| 0 to 5 | 5.9 | 6.0 | 5.95 | 0.496 |
| 5 to 10 | 6.0 | 6.0 | 6.00 | 0.500 |
| 10 to 15 | 6.0 | 5.8 | 5.90 | 0.492 |
| 15 to 20 | 5.8 | 4.4 | 5.10 | 0.425 |
| 20 to 25 | 4.4 | 3.4 | 3.90 | 0.325 |
| 25 to 30 | 3.4 | 2.7 | 3.05 | 0.254 |
| 30 to 35 | 2.7 | 1.5 | 2.10 | 0.175 |
| 35 to 40 | 1.5 | 0.0 | 0.75 | 0.063 |
| Total heater energy | about 2.73 kWh |
Heater energy for this cycle: about 2.7 kWh.
Add the auxiliary loads
Suppose the controller, sensors and fan drew a steady 0.15 kW for the whole 55-minute cycle. That is 0.15 × 55 ÷ 60 = about 0.14 kWh. In a real test you would measure this on its own channel. Some designs have no meaningful auxiliary load beyond the controller, which also runs outside cycles, so you might exclude it. Decide the rule before measuring and state it in the report.
Total regeneration energy for the cycle: about 2.9 kWh (2.73 heater plus 0.14 auxiliary).
Step 5: Check the answer three ways
A number is more believable when independent routes agree.
Check 1: The analyser’s own kWh register
Most logging analysers accumulate energy. Read the register at the start and end of the cycle. If it reads 2.7 to 2.8 kWh for the heater channel, it agrees with the hand calculation.
Check 2: Sampling-interval sensitivity
Redo the integration using coarser data to see how sensitive the answer is. Using only the ten-minute points (readings of 5.9, 6.0, 4.4, 2.7 and 0.0 kW at 0, 10, 20, 30 and 40 minutes) gives about 2.68 kWh, which is within 2% of the five-minute result, so the curve is smooth enough that coarse logging does little harm here.
Now try a careless method: take each reading and multiply by the full five-minute slice that follows it (a left-hand rectangle approach). That gives about 2.98 kWh, which overstates the energy by roughly 9% because it ignores the fall in power across each slice. The lesson is to use the trapezoid rule or let the analyser do it.
Check 3: Rating times on-time
A crude test: 6 kW rating × 0.67 hours (40 minutes) heater on-time = 4.0 kWh. The measured figure is only about 68% of that. The difference is the heater modulating down once soot ignited. Without the measurement you would have overstated energy by about 47%, which would distort any cost or carbon figure.
Step 6: Estimate the uncertainty
Every measurement has a margin. A reasonable, simple approach is to list the main contributors.
| Source | Illustrative size | Effect on 2.7 kWh |
|---|---|---|
| Analyser accuracy (about 1% to 2% of reading) | ±2% | ±0.05 kWh |
| Current clamp positioning and cable arrangement | ±1% | ±0.03 kWh |
| Sampling interval (from check 2) | ±2% | ±0.05 kWh |
| Cycle start and end definition (a minute either way) | ±1% | ±0.03 kWh |
Combining independent contributions by adding them in quadrature (square each, add, take the square root) gives about 0.09 kWh, so we could say 2.7 ± 0.1 kWh. The exact method matters less than being honest that the number is not exact. See interpreting uncertainty in an RECD emission reduction claim for the same thinking applied to emissions.
A single cycle also varies from cycle to cycle with soot load, ambient and load. Measure at least three cycles under similar conditions and report the range.
Step 7: Turn kWh into something useful
Cost and fuel when the DG supplies the heater
The following uses assumed values; replace them with your own.
- Heater energy per cycle: 2.73 kWh
- DG specific fuel consumption at the operating load: 0.30 litres per kWh (assumed)
- Diesel price: ₹90 per litre (assumed)
Fuel per cycle = 2.73 × 0.30 = about 0.82 litres. Cost per cycle = 0.82 × 90 = about ₹74. For 50 cycles a year = about 41 litres and about ₹3,700.
If the heater could be powered from the grid on some occasions, the cost per kWh would follow your tariff instead, but during outages the DG supplies it.
Energy per mbar of recovery
The cycle recovered 23 mbar of ΔP. Energy per mbar is 2.73 ÷ 23 = about 0.12 kWh per mbar. This is a rough efficiency indicator, useful for trending. If the same heater needs 0.18 kWh per mbar a year later, either the heater is delivering less heat, heat losses have risen or the cycle is clearing less soot.
Share of the DG’s output
The DG averaged 125 kW over the 55-minute cycle, so it produced about 125 × 55 ÷ 60 = 115 kWh. The regeneration energy is about 2.4% of that. Small, but not zero, and it matters when the DG is near its limit. See does an RECD change a generator’s usable power output.
Compare with the energy locked in the soot
Soot is mostly carbon, and carbon releases about 32.8 megajoules per kilogram when burned, which is about 9.1 kWh per kg. If the cycle burned an assumed 150 g of soot (an illustrative figure from a supplier’s ΔP calibration), the chemical energy released is about 0.15 × 9.1 = 1.4 kWh. The heater supplied about twice that. This is normal, since much of the heater’s energy warms the exhaust gas and the housing and is lost, and the soot’s own heat helps once it ignites. The comparison shows why regeneration is fundamentally a heat-delivery problem, not an energy-supply problem. DieselNet’s page on particulate filter regeneration explains the heat balance concepts.
Repeatability: three cycles, three numbers
One cycle is an anecdote. Here is what three consecutive cycles on the same device might look like, again with illustrative figures.
| Cycle | DG load (kW) | Start soot proxy: ΔP (mbar) | Heater energy (kWh) | ΔP recovered (mbar) | kWh per mbar |
|---|---|---|---|---|---|
| A | 125 | 61 | 2.73 | 23 | 0.119 |
| B | 118 | 57 | 2.61 | 20 | 0.131 |
| C | 130 | 66 | 2.88 | 27 | 0.107 |
The mean heater energy is 2.74 kWh and the sample standard deviation is about 0.14 kWh, or 5% of the mean. Two patterns are worth noticing. First, the energy rises with starting ΔP, because a more heavily loaded filter needs the heater on for a little longer before the soot ignites and takes over. Second, energy per millibar falls as the filter starts more loaded, because the soot’s own heat does more of the work. A report that quoted only Cycle B would understate the typical energy by about 5%, and one that quoted only Cycle C would overstate it by about the same.
For reporting, give the mean, the range and the number of cycles, such as “2.7 kWh (range 2.6 to 2.9, three cycles)”. If your measured range is much wider than this, look for a cause: heater supply voltage changes, wet insulation, very different loads or inconsistent start and end definitions.
What the number can and cannot tell you
It can tell you:
- The real operating cost and carbon contribution per cycle, with a margin.
- Whether a cycle is behaving like earlier cycles.
- Whether a supplier’s claim is plausible.
- Whether heater ageing or supply problems are developing.
It cannot tell you:
- Whether the filter is clean. That needs ΔP at a reference load and persistence over time.
- Whether the soot burned evenly.
- Whether the same cycle would use the same energy at another load, ambient temperature or soot load.
Treat energy per cycle as one column in a regeneration record, next to duration, ΔP and temperatures. See what an RECD regeneration history report should include.
Measuring a fuel-assisted system
If the RECD uses a burner or dosing system, the method has the same shape but a different instrument.
- Fuel flow. Measure fuel consumption by the regeneration system with a flow meter on the burner supply line, a weighed or graduated day tank that feeds only the burner, or controller data if it logs fuel dose and the supplier can show how it is calibrated.
- Cycle boundaries. Use the same start and end definition.
- Convert. Diesel contains about 10 kWh of heat per litre (lower heating value). Multiply litres by 10 for thermal energy, then by the burner’s efficiency if you want heat delivered to the exhaust.
- Add auxiliaries. Pumps, blowers and igniters draw electricity; measure them as in the heater case.
- Repeat. Fuel dosing can vary with exhaust conditions, so measure several cycles.
For an example of how the two methods compare on cost, see our buyer guide to electric-heater versus fuel-assisted RECD regeneration.
A reporting template
Keep a short record for each measurement campaign.
| Field | Entry |
|---|---|
| Site, DG and RECD identification | |
| Date, ambient, DG load during cycle | |
| Instrument model, serial, calibration date, accuracy | |
| Measurement points (heater feeder, controller, other) | |
| Cycle start and end definition | |
| Sampling interval | |
| Heater energy (kWh) | |
| Auxiliary energy (kWh) | |
| Total energy (kWh) with uncertainty | |
| Duration (heater-on and total) | |
| ΔP before and after, load at each | |
| Peak and average temperatures | |
| Fuel and cost per cycle (assumptions stated) | |
| Number of cycles measured | |
| Observations, deviations, who witnessed |
Store it with the commissioning file and the regeneration history. It also supports warranty and performance discussions. See RECD warranty claim: what to record.
Common mistakes
- Using the nameplate instead of a measurement. The heater rarely runs at full rating throughout.
- Measuring at the wrong point. Measuring upstream of a transformer or at the DG output includes other loads.
- Mixing in the building load. Use a dedicated channel for the heater circuit.
- Ignoring power factor. For non-resistive loads, a clamp meter that reads only current gives the wrong kW.
- Too coarse sampling. Fast heater switching needs one-minute or finer logging.
- Inconsistent cycle boundaries. Define start and end once and use it every time.
- One cycle, one conclusion. Measure several.
- Forgetting the DG fuel step. Heater kWh from a DG costs more than grid kWh per unit, so state which one you used.
FAQ
How much energy does a typical RECD regeneration use?
It depends entirely on the design, soot load and conditions. The 2.7 kWh in this article is an illustrative figure, not a typical one. Ask your supplier for measured values and measure on your own device.
Can I read energy from the controller instead of a meter?
Some controllers report heater on-time or current but not energy. You can estimate energy from on-time and current if you know voltage and power factor, but a direct measurement is better, especially if the heater modulates.
Does a longer cycle always use more energy?
Not always. A longer cycle with the heater off for much of the time may use little more than a shorter one. Measure rather than infer from duration.
Is the energy cost significant?
Per cycle it is usually small compared with the DG’s main fuel use, as the example shows. It becomes important when cycles are very frequent, when the DG has little spare capacity, or when comparing designs at scale across a fleet.
How do I include this in an ESG report?
Report measured kWh or litres for regeneration separately from the DG’s main fuel use, state your method and assumptions, and note that figures are site-specific. Our article on reporting RECD outcomes in ESG disclosures discusses suitable metrics.
What should I do if energy per cycle suddenly rises?
Check the heater supply voltage and current, insulation condition, soot load at start and cycle logs. A sudden rise can point to a degraded heater, heat loss, a changed set-point or a different load pattern. Read our guide on what a rising regeneration duration tells you.
Bringing it together
Measuring RECD regeneration energy per cycle takes seven steps: plan the measurement, set up the logger, record the data, integrate the power curve, check the answer several ways, state the uncertainty and convert the result into cost, fuel and trend indicators. In our worked example, a 6 kW heater used only about 2.7 kWh in a 55-minute cycle, roughly two-thirds of what a rating-times-time estimate would suggest, and the result was reliable to about ±0.1 kWh.
Once you have the number, keep it in the regeneration record, repeat it periodically and use it to compare designs, spot ageing and support budgets and reports.
If you would like help planning a measurement on your own DG set, or comparing the energy needs of different RECD designs, explore our retrofit emission control devices, see our certifications or contact the Aceget team.