Diesel Emission Testing: Opacity, NOx and Loaded Cycles
Diesel emissions are usually reported on a lane as one number, and that number is almost always opacity. Opacity is a light measurement, it responds to visible smoke, and it is blind to the nitrogen oxides that diesel engines produce most of. Understanding that gap is what separates a useful inspection result from a compliance stamp.
Opacity is a light measurement, not a mass measurement
A partial-flow opacity meter shines a light across a sample chamber and reports how much of it reaches the detector. The result is expressed as an absorption coefficient per metre, or as a percentage of light blocked. Both describe how much visible smoke is in the sample, and neither reports how many grams of particulate the engine emitted.
The two are related but not proportional. A change in fuel, injector condition, turbo boost or ambient conditions can move opacity without moving particulate mass by the same factor. This is why opacity is used as a screening and enforcement figure rather than as an emissions inventory number.
Free acceleration and loaded acceleration answer different questions
Free acceleration testing revs the engine in neutral and takes the peak opacity from each snap. It needs no load hardware, takes seconds, and reproduces the visible puff a worn injector or a bad turbo produces. What it does not reproduce is sustained fuelling under load, which is where a modern engine spends most of its working life.
Loaded testing runs the driven wheels against an absorber and applies load at defined speeds, so the engine reaches the fuelling and exhaust temperature of real operation. It needs a chassis dynamometer, takes far longer, and is a much better predictor of what the vehicle does on a grade with a load behind it. Where a regulation allows either method, the choice is a statement about how strictly the fleet is being checked.
| Method | Hardware | Catches | Misses |
|---|---|---|---|
| Free acceleration | Opacity meter only | Smoke from fuelling faults, worn injectors, blocked air path | Behaviour under sustained load |
| Loaded acceleration | Opacity meter plus chassis dynamometer | Faults that only appear with load and heat | Anything the opacimeter cannot see, including NOx |
| Steady-state sampling | Absorber plus gas analyser | Gas concentrations at defined operating points | Transient behaviour between points |
NOx needs a different analyser, not a better one
Non-dispersive infrared analysers, the standard fit for carbon monoxide and carbon dioxide on a petrol lane, respond poorly to nitric oxide and effectively not at all to nitrogen dioxide. A lane equipped only with NDIR can report a clean diesel result while the vehicle is emitting several times the NOx limit.
Measuring NOx means chemiluminescence or non-dispersive ultraviolet, plus a heated sample line to stop the sample changing before it reaches the cell. That is a different instrument class with different running costs and calibration gases, and it is the reason many lanes test opacity only. The limitation should be stated to the customer rather than hidden behind a generic "emissions passed" label.
What a combined lane looks like in practice
A lane that handles both petrol and diesel usually shares the dynamometer and switches analysers per vehicle type. Petrol vehicles get carbon monoxide, carbon dioxide and hydrocarbons at idle and at raised speed; diesel vehicles get opacity under free or loaded acceleration, and NOx only where the equipment and the regulation require it.
The practical constraint is time. A loaded diesel test occupies the dynamometer for minutes rather than seconds, so the station that sets the lane cycle time is usually the dynamometer, not the brake tester. Adding NOx sampling lengthens it further and needs the analyser to be ready before the vehicle arrives, not warmed up while it waits.
The vehicle condition that invalidates a result
An engine that is cold, or that has been idling for an hour before the test, will not produce the smoke or the gas concentrations of one at operating temperature. A blocked air filter, a restricted exhaust, or a diesel particulate filter partway through a regeneration cycle all move the result more than most faults the test is looking for.
The procedure should therefore fix the warm-up condition and record it, and the operator should note visible smoke, a DPF warning lamp or an obvious exhaust leak as a reason to stop rather than a reason to write down a number.
- Engine at normal operating temperature before sampling, verified rather than assumed.
- No active regeneration or warning lamp on the dashboard at the time of test.
- Sample probe fully inserted to the depth the meter specifies, not resting at the pipe mouth.
- Analyser zeroed and spanned on the day of use, with the record kept.
- Ambient conditions noted: altitude, temperature and humidity all shift opacity readings.
Common questions
Does a smoke test check NOx?
No. An opacity meter measures visible smoke by light absorption. Nitrogen oxides are colourless and require a chemiluminescence or ultraviolet analyser with a heated sample line. A vehicle can pass an opacity test and still exceed a NOx limit.
What is the difference between free acceleration and loaded smoke testing?
Free acceleration revs the engine in neutral and takes the peak smoke reading from each acceleration. Loaded testing applies resistance through a chassis dynamometer so the engine works under sustained load. Loaded testing takes longer but reproduces the conditions the vehicle meets on a road.
Why do two opacity readings on the same vehicle differ?
Engine temperature, ambient conditions, probe depth, analyser zero state and how hard the accelerator is snapped all change the peak value. The test procedure fixes these variables; when they are not fixed, the readings are not comparable.
Need the equipment side of this?
Tell us the vehicle classes and the volume you need to handle, and we will map these options onto a station list with the specifications for each unit.
Request a specification