RECD Emissions Data in mg/Nm³ vs g/kWh: Why Units Matter

Two RECD proposals arrive for the same generator. Vendor A writes that outlet particulate is “25 mg/Nm³”. Vendor B writes “0.05 g/kWh”. Which device is cleaner?

On that information alone, nobody can tell. The two figures measure different things, under conditions neither vendor has fully stated. Yet decisions worth lakhs of rupees, and compliance with State Board directions, often rest on comparisons exactly like this one.

This guide is a working reference for anyone who reads emission data on diesel generators and retrofit emission control devices. It defines the terms, shows where each unit appears in Indian rules, walks through the corrections and conversions step by step, and ends with a checklist for requesting data you can actually compare.

The Short Answer

mg/Nm³ is a concentration: how many milligrams of particulate are in one cubic metre of exhaust at normal conditions. g/kWh is a specific emission: how many grams of particulate the engine emits for every kilowatt-hour of work it delivers. The first describes the gas; the second describes the engine’s performance. You can move from one to the other, but only with extra data: exhaust flow, engine power, and the reference conditions the concentration was reported at.

A Glossary of the Terms on Your Report

Term	Meaning	Why it matters
mg/Nm³	Milligrams per normal cubic metre	The usual unit for stack concentration
Nm³ (normal)	Gas volume at 0 °C and 101.325 kPa	Removes the effect of temperature and pressure on volume
Sm³ (standard)	Gas volume at a stated standard condition, often 15 °C or 25 °C	Not the same as Nm³; always check which is used
Am³ (actual)	Gas volume at the real stack temperature and pressure	Hot gas expands, so actual concentrations look lower
Dry basis	Concentration with water vapour removed from the volume	Common for stack reporting
Wet basis	Concentration including water vapour	Gives a lower figure than dry for the same gas
Reference O₂	The oxygen level a concentration is corrected to	Stops results being lowered by adding air
g/h	Grams of particulate emitted per hour	Mass rate; needed for any efficiency calculation
g/kWh	Grams per kilowatt-hour of engine output	The unit of engine emission norms and RECD certification
Weighted	Combined across test modes using cycle weights	A D2-weighted g/kWh is not the same as a single-load g/kWh

What Each Unit Really Tells You

An everyday comparison helps. Imagine judging how salty a kitchen’s cooking is.

• Concentration (mg/Nm³) is like grams of salt per litre of soup. Add water and the soup looks less salty, even though the same salt went in.
• Specific emission (g/kWh) is like grams of salt per meal served. It describes how much salt the kitchen uses to produce its output, whatever the dilution.

For an RECD, the question is how much particulate the engine and device release for the work done, and what share the device removes. Both require mass, so g/h and g/kWh sit at the heart of RECD evaluation. Concentration is still valuable for stack monitoring, but only when its reference conditions are fixed.

Where Each Unit Appears in Indian DG Set Rules

Engine emission norms use g/kWh. India’s genset norms cover CO, NOx, hydrocarbons and particulate matter, specified as grams of each pollutant per kWh of electricity generated (Powerline). Under the CPCB IV+ norms, NOx and PM limits were cut by 90% from CPCB-II levels in the 56 to 560 kW power band (SAE technical paper). Our guide to CPCB II vs CPCB III vs CPCB IV+ norms explains the stages.

RECD certification uses a percentage derived from mass. State Board directions call for a CPCB-certified RECD with at least 70% particulate capture efficiency for in-use sets up to 800 kW (TeamLease RegTech on the MSPCB directions). That percentage is calculated from weighted particulate mass on the ISO 8178 D2 cycle. We show the method in how to calculate RECD particulate reduction.

Stack limits for large sets use mg/Nm³. In Delhi NCR, a CAQM direction sets a stack PM standard of 50 mg/Nm³ for DG sets of 800 kW and above, corrected to a reference oxygen level (CAQM direction). For the wider picture, see GRAP rules for DG sets in Delhi NCR and our DG set emission regulations page.

So a single site may deal with all three: a g/kWh norm for the engine it buys, a percentage for the RECD it retrofits, and an mg/Nm³ limit if it runs a very large set in NCR. Mixing them up is easy, and expensive.

Conversion Drills: From Raw Reading to Comparable Number

Stack readings usually need several corrections before they can be compared with a limit or with another report. Work through them in this order. Each drill uses simple illustrative numbers.

Drill 1: Actual cubic metres to normal cubic metres

Hot gas expands, so the same particulate is spread over more volume. To express an actual concentration at normal conditions:

C_{N} = C_{actual} \times \frac{273.15 + T}{273.15} \times \frac{101.325}{p}

T is the gas temperature in °C at the measuring point, and p is the absolute pressure in kPa.

Example: 18 mg/m³ measured at 180 °C and near-atmospheric pressure. 18 × (453.15 ÷ 273.15) = 29.9 mg/Nm³. Correcting for temperature alone raised the figure by two-thirds.

Drill 2: Wet basis to dry basis

Diesel exhaust carries water vapour from combustion. Removing it from the volume raises the concentration:

C_{dry} = \frac{C_{wet}}{1 - x_{H_2O}}

x is the water vapour volume fraction.

Example: 25 mg/Nm³ on a wet basis with 8% moisture. 25 ÷ (1 − 0.08) = 27.2 mg/Nm³ dry.

Drill 3: Correcting to a reference oxygen level

Diesel engines run with excess air, and lightly loaded engines run with a lot of it. Air leaks and dilution add more. Oxygen correction puts every result on the same footing:

C_{ref} = C_{measured} \times \frac{21 - O_{2,ref}}{21 - O_{2,measured}}

Example: 40 mg/Nm³ measured at 16% O₂, corrected to a 15% reference. 40 × (21 − 15) ÷ (21 − 16) = 40 × 6 ÷ 5 = 48 mg/Nm³.

Notice the corrected value went up. When measured oxygen is higher than the reference, the gas was more diluted than the standard assumes, and the correction removes that advantage. Always confirm which reference oxygen the applicable rule specifies.

Drill 4: Concentration to specific emission (mg/Nm³ to g/kWh)

PM\ (g/kWh) = \frac{C\ (mg/Nm^3) \times Q\ (Nm^3/h)}{1000 \times P\ (kW)}

Use an uncorrected concentration (not oxygen-corrected) with the flow measured on the same basis (both dry, or both wet, both at normal conditions).

Example: 35 mg/Nm³ dry, exhaust flow 700 Nm³/h dry, engine output 120 kW. 35 × 700 ÷ 1,000 = 24.5 g/h. 24.5 ÷ 120 = 0.204 g/kWh.

This gives specific emission at that one load. It is not a D2-weighted result and should not be compared directly with a cycle-based norm or certificate. If you do not have measured flow, start from manufacturer data, as explained in using OEM exhaust flow data to specify an RECD.

Drill 5: The full chain in one example

In real reports, all of these corrections can apply to a single reading. Take a raw measurement of 20 mg/m³, at actual stack conditions of 200 °C, on a wet basis with 7% moisture, at 14% O₂. The applicable rule asks for mg/Nm³, dry, at 15% O₂.

Step	Calculation	Result
1. Actual to normal	20 × (473.15 ÷ 273.15)	34.6 mg/Nm³ wet
2. Wet to dry	34.6 ÷ (1 − 0.07)	37.3 mg/Nm³ dry
3. Correct to 15% O₂	37.3 × (21 − 15) ÷ (21 − 14)	31.9 mg/Nm³ dry at 15% O₂

The raw 20 became nearly 32 once it was put on the basis the rule requires. That gap is large enough to change a pass into a fail, or the reverse, if a report is read without checking its basis. When reading a stack report, find each of these three facts before you compare anything with a limit.

Why the Same Exhaust Can Look Clean in One Unit and Dirty in Another

Load changes the picture dramatically. Take one illustrative 160 kW engine at two load points:

Load point	PM mass rate	Output	Exhaust flow	Concentration	Specific emission
10% load	4 g/h	16 kW	450 Nm³/h	8.9 mg/Nm³	0.25 g/kWh
100% load	30 g/h	160 kW	1,300 Nm³/h	23.1 mg/Nm³	0.19 g/kWh

At light load the exhaust looks cleaner by concentration, less than half the full-load figure. Yet per unit of useful energy, the engine is dirtier at light load. A standby set that spends its life at 10 to 20% load can pass a casual concentration check while performing poorly on a g/kWh basis.

This is one reason low-load operation deserves attention when selecting an RECD. See choosing an RECD for a generator running at low load and understanding generator load factor and duty cycle.

The reverse distortion comes from dilution. A downstream sample with air leaking into the duct reads a lower concentration even if the device captured nothing extra. That is why efficiency is always calculated from mass rates, as our particulate reduction calculation guide shows with a worked upstream and downstream example.

Back to the Two Vendor Quotes

Vendor A’s “25 mg/Nm³” and Vendor B’s “0.05 g/kWh” can now be read properly. To compare them you would need:

• From Vendor A: the temperature basis, dry or wet, reference oxygen, load, engine power and exhaust flow.
• From Vendor B: whether 0.05 g/kWh is D2-weighted or a single load, and on which engine.

Suppose Vendor A’s figure was at full load on a 100 kW engine with 900 Nm³/h of flow, on the same dry basis. Drill 4 gives 25 × 900 ÷ 1,000 ÷ 100 = 0.225 g/kWh. That is a single-load figure, so it still is not directly comparable with a weighted 0.05 g/kWh. Neither number on its own tells you the capture efficiency either, because that needs the baseline. The useful outcome is a precise list of what to ask both vendors for.

A Data Request Checklist for Labs and Vendors

Copy this into your RFQ or email when asking for emission data. It also works as a section in a technical RFQ for an RECD project.

• Particulate results in g/kWh, weighted on the ISO 8178 D2 cycle, for baseline and with RECD
• Mode-by-mode PM mass rate in g/h, engine power in kW, and weighting factors
• The resulting capture efficiency (%) and the laboratory that certified it
• Tested engine make, model, rated power and swept volume
• For any concentration: mg/Nm³ with normal conditions stated, dry or wet basis, and the reference O₂
• Measured O₂, moisture, temperature and exhaust flow at each sampling point
• Sampling method, including whether sampling was isokinetic
• Fuel specification used during testing

When reports arrive, our guide on how to read an RECD test certificate helps with the rest.

Writing Units into Your RECD Purchase Documents

Most unit problems start long before a test report arrives. They start in the tender or purchase order, when performance is described loosely. “Device shall reduce emissions significantly” or “outlet PM shall be below 30” leaves every important question open.

Write performance requirements so that only one reading is possible:

• For capture efficiency: “Minimum 70% particulate matter capture efficiency, weighted on the ISO 8178 D2 cycle, certified by a CPCB-recognised laboratory, on an engine within the accepted power and swept-volume range of the DG set listed in Schedule A.”
• For any site stack limit: “PM concentration not exceeding X mg/Nm³, dry basis, at Y% reference O₂, measured by isokinetic sampling at rated load by an accredited laboratory.”
• For reporting: “Test reports shall state mode-by-mode PM mass rates in g/h, engine power in kW, gas flow in Nm³/h, measured O₂ and moisture, and the basis of every concentration.”

Clear wording also protects the supplier. A vendor who knows exactly how performance will be judged can design and quote accordingly, and disputes at handover become far less likely. Our guides on writing measurable RECD tender specifications, preparing a bill of quantities for an RECD retrofit and setting trial installation acceptance criteria show how to carry this through the whole project.

Common Reporting Errors and How to Spot Them

These are the unit mistakes that turn up most often in emission reports, proposals and internal summaries.

Error	How to spot it	What to do
Concentration given with no temperature basis	“mg/m³” with no N, S or A and no stated conditions	Ask whether it is normal, standard or actual
No dry or wet basis stated	Moisture not reported anywhere	Ask for the moisture content and basis
No reference oxygen	Concentration compared with a limit that specifies O₂	Ask for measured O₂ and recalculate
Oxygen-corrected value used in a mass calculation	g/h worked out from a corrected concentration	Use the uncorrected concentration with matching flow
Single-load g/kWh compared with a weighted norm	No mention of test cycle	Ask whether the figure is D2-weighted
Efficiency from concentrations only	Upstream and downstream mg/Nm³, no flows	Ask for flows, or for O₂ at both points
µg/m³ ambient readings mixed with mg/Nm³ stack data	Values differ by a factor of about 1,000	Separate ambient air data from emission data

The last row is worth stressing. Ambient air quality is reported in micrograms per cubic metre, and stack emissions in milligrams per normal cubic metre. One milligram is a thousand micrograms. Putting the two in the same table, or the same sentence, invites confusion. Our post on whether a handheld AQI monitor can verify an RECD explains why ambient readings cannot stand in for emission data.

Quick Reference: Which Unit for Which Job

Job	Unit to use	Must be stated alongside
Checking an RECD’s certified capture efficiency	% from weighted g/kWh	Test cycle (ISO 8178 D2), laboratory, tested engine
Comparing two engines or two RECDs	g/kWh, weighted	Same cycle and test conditions
Calculating efficiency from site data	g/h at each point	Concentration, flow, basis and load
Checking a stack limit for a large DG set	mg/Nm³ at reference O₂	Temperature basis, dry or wet, measured O₂
Explaining local air quality to residents	µg/m³, AQI	Location, time, background conditions

Pin this table to the front of your compliance file. It answers most of the questions that come up when a new report arrives.

How Precise Should the Numbers Be?

A final point on presentation. Emission measurements carry real uncertainty, especially for particulate at low levels. Reporting a result as 27.1834 mg/Nm³ suggests a precision the method cannot deliver.

As a working rule, two or three significant figures are usually enough for concentrations and specific emissions, and whole numbers or one decimal place for percentages. Keep full precision inside your spreadsheet so rounding errors do not build up through a chain of corrections, and round only the final figures you report.

When comparing a result with a limit, look at the margin as well as the pass or fail. A result just below a limit, from a single test, deserves follow-up monitoring. A result well below it, repeated over time, gives real confidence. Our guide to RECD performance guarantees and operating conditions shows how to build that thinking into a purchase contract.

Frequently Asked Questions

Is a lower mg/Nm³ figure always better?

Only when both figures share the same reference conditions and load. A lower number can result from extra dilution, higher temperature basis, a wet basis, or simply lighter load.

Can I convert g/kWh to mg/Nm³ to check a stack limit?

Yes, by rearranging Drill 4: C = PM (g/kWh) × P × 1,000 ÷ Q. You still need the exhaust flow at that load, and you must then apply the oxygen correction the limit specifies.

What is the difference between Nm³ and Sm³?

Nm³ refers to 0 °C and 101.325 kPa. Sm³ refers to whatever standard temperature the report or rule defines, often 15 °C or 25 °C. The same gas gives a slightly higher concentration in mg/Nm³ than in mg/Sm³, because a normal cubic metre holds more gas.

Why do RECD certificates use g/kWh rather than mg/Nm³?

Because capture efficiency compares the mass of particulate emitted with and without the device across a weighted load cycle. g/kWh ties mass to the work the engine does, and dilution cannot change it.

Do smaller DG sets ever get tested in mg/Nm³?

Site stack monitoring of any set may report concentrations, and some consent conditions use them. For RECD evaluation, though, the certified capture efficiency is the reference. Check your State Board’s conditions; our state-wise pollution control norms for DG sets is a useful starting point.

How do I convert mg/Nm³ to µg/m³?

Multiply by 1,000 to change milligrams to micrograms, then adjust the volume basis if the conditions differ. But be careful about why you are converting. Stack concentrations and ambient air readings describe different things, and comparing them directly is rarely meaningful.