To be able to conduct an emission-analysis, the engine must be brought to it's operating temperature. Furthermore, during the whole test-run, the engine RPMs and the four exhaust gases, carbon monoxide (CO ₂), hydrocarbon (HC), carbon dioxide (CO₂) and oxygen (CO₂) should be measured.

Previously the engine RPMs were determined through the terminal 1 on the ignition coil. In the event that ignition cables exist, the measuring with one or more inductive calipers is possible. However, the measuring instrument must be adjusted according to the type of ignition, otherwise it shows, e.g., only half the amount of RPMs. One can also obtain the RPMs from the engine control module directly. However, in most cases, an additional adaptor is necessary and possibly, one may also have to do a calculation conversion. Much simpler is using the block-like sensor which is magnetic and is placed on the engine.

In the case of the diesel engine, the terminal 'W' on the alternator is well known. Indeed, here a calculation conversion must also be done, because the transmission ratio to the crankshaft is not 1. A little more difficult still, would be an optical measurement, although in this case, at least one would not need to do any conversions. One can blacken a point - or place a reflecting strip - on the flywheel and then check it with a light (under no circunstances Neon!). A photodiode then, in a simple circuit, picks up the impulses which are then given out on a display. The simplest method however, is to use an acoustic transducer which is placed, and held in place, e.g., on the valve-cover by a permanent magnet. It evaluates the sound waves from the engine so capably that the engine RPMs can be read off.

The four-gases-measuring instrument is so important for the emission-analysis that it has to be serviced at certain regular intervals, otherwise it refuses to function. It draws off a part of the exhaust gas through a small pump and then radiates this with infrared light. For this reason, one must wait a few minutes after switching on, until the device
- has warmed up,
- has tested it's software,
- has examined the hardware, e.g., for leakages,
- has calibrated itself according to the surrounding air.

Why must the device calibrate itself? The surrounding air does not always contain the same oxygen concentration. One can observe this, when after a warm day, the oxygen concentration rises towards evening. The air in the garage can also contain a certain amount of CO- NOX- and/or HC concentration. This must be considered by the detection device, so that it does not give out an inaccurate reading. Therefore, let it calibrate itself first, only then should it be placed in the exhaust pipe.

During the following engine-run, exhaust gas is pumped out and radiated with infrared light. On it's way through the exhaust gas, one for each exhaust gas portion (CO, HC und CO₂) typical light-wavelength is faded out (absorbed). Sensors, whose filters are installed for the various typical wavelengths, measure on the opposite side, how much of the respective wavelength gets through, thus determining the respective gas portions. At the end, the oxygen concentration is also determined. Through this, the internal resistance of an electro-chemical element is altered. The Lambda value is then calculated according to the four exhaust gas components.

The following part of the emission analysis is repeated again in the same pattern. First of all, a stable condition is reached in the Lambda-control. Through the injection regulation, the Lambda sways back and forth around a certain average value. Now, a disturbance variable is introduced. This could be in the form of an acceleration increase or decrease, the influencing of the air supply or the increasing of the load (e.g., switching on the headlights). What is indeed important, is whether - and when it does - at what point the Lambda signal gets back to a steady regulation.