Tutorials > Combustion analysis

Select the Combustion button and click a blue plus button to add a combustion analysis module. We can set the combustion analyzer in any way we want, but usually the easiest way would be to follow the five steps that are shown in the following image:

Combustion setup

1.	Engine setup

Here we set all engine parameters (geometry, engine type,..). We define the type of engine (gasoline or diesel [2], four stroke, two stroke [1]), cylinder count [3] and engine geometry [6÷10]), which has to be identified for volume calculations. For each cylinder we can also define the piston and crankshaft offset [14, 15], but for most engines these values are zero. We need to define the polytropic coefficient [5], which is a factor for compression (without ignition). This is important for the calculation of thermodynamic zero and for heat release. If you don’t know the polytropic exponent for your engine, take the suggested value for each engine type. We will see more details about how to find this value below.

Next, we define the channels that correspond to the cylinder pressure [12]. Note that we can leave some cylinders without assigned channels (like Cyl. 4 in picture below) and those channels will not be calculated. We also need to define the ignition misalignment. This defines the firing order of the engine. For example, if we take a 6 cylinder engine, the firing order is 1-4-3-6-2-5, so if the whole cycle is 720 deg, the ignition misalignment will be: 0-360-240-600-120-480.

We can also define the start/end of injection channel, trigger levels and number of injections here. We will get the angles of injection as additional channels during the measurement.

Engine setup

2.	Angle sensor setup

The angle sensor setup is a very important part of setting up the combustion analyzer. It defines the performance and abilities of the combustion analyzer module. We have two basic ways of acquiring this data: the internal and the external clock. These two modes totally redefine the acquisition process. Let's take a more detailed look at these two modes.

External clock

External clock means that the data will be acquired in the Angle domain. In other words, if we choose 360 pulses per revolution, it will always acquire 360 points regardless of the current speed of the engine. This procedure is useful for high speed and/or high number of cylinders since it acquires the data directly, which is suitable for the calculation of combustion parameters. However we can't use any additional time domain calculations, for example combustion noise. We also can't acquire a cold start from 60-2 or similar sensors.

In the case of an external clock we need the CA CPU hardware to create the needed number of pulses from any sensor. In the next step we choose the angle sensor from the list (in our case a 60-2). If the sensor doesn't exist, we can add it by clicking the three ellipsis button on the right side of the sensor. CA-CPU supports encoder, CDM sensor as well as the geartooth with missing and double teeth. We define where the input connection of the sensor is on the CA-CPU and the output resolution. This resolution defines the maximum speed (limited by the AD card maximum rate) and the calculation load. If we choose 0.1 degree resolution and we have 500 ks/s AD card, our maximum RPM will be 500000/3600*60=8300 RPM. The output resolution directly defines the number of points per revolutions for all calculations as well as for CA displays.

As soon as we define the engine setup and the angle sensor correctly (and the engine is running), we should see the CA-CPU tracking message and already see the number of pulses, speed and cycle count in the right side of the display.

Angle sensor setup - external clock

Internal clock

Internal clock uses a totally different method. It acquires the data in the Time domain and recalculates by software the angle domain for CA calculations. Since the data is low-pass filtered according to the speed of the engine, the calculation is more demanding than with external clocking, but on the other side it is easily possible to correlate time domain data (like CAN bus, video and so on) and calculate time domain parameters like combustion noise.

We don't have to use the CA-CPU in this case, but we connect the CDM or encoder sensors directly to counters of the Orion card.

If we have a geartooth sensor with missing or double teeth we connect it directly to analog channels and select it (like in the picture below). Additional benefit of the internal clock is that we can measure a cold start. If we use an external clock, the CA-CPU will miss the first few cycles, which are the most important for cold start.

Since the angle sensor math is used the back, it waits until first gap is detected and also calculates the cycle before. To use this, we need to set the trigger levels and direction by clicking the Setup button.

Angle sensor setup - internal clock

The easiest way is to click - the Find button which should set the trigger edge, trigger and retrigger level. On the graph below, we can already see the live data.

Angle sensor channel setup

The output resolution and the sample rate define the maximum speed of the engine where the CA still works correctly. We can easily see this from the message shown to the right side of the resolution.

Top dead center detection

Regardless if using internal or external clock, we will see some data already in the CA cycle graph when we finish with the settings. To see the p-v diagram correctly, we need to define the trigger offset. Trigger offset is the angle between the trigger and the top dead center of the reference cylinder (noted as ref. cyl. in the engine setup).

This angle can be physically measured or it can be also defined with TDC (top dead center) detection. The engine must not be fired for this procedure. In the test bed this is easily achieved since we have an option to switch off the ignition. We can use the combustion analyzer inside the vehicle by driving to a certain speed, leaving the car in gear and releasing the gas pedal so that the engine brakes the vehicle. Then the engine will also run in the so-called motored cycle, meaning only compression and expansion with no work. The right picture shows a typical motored cycle, where we clearly see that the compression fits the expansion exactly, therefore no work is done inside the cylinder.

We enter the number of cycles for averaging and the thermodynamic loss angle. This is the delay between the top dead center and the pressure peak. The pressure peak happens a bit after top dead center and we can compensate for this by entering the loss angle. Then we select the Start TDC detection procedure. The pre-defined number of cycles will be acquired and the trigger offset will automatically be calculated and set.

3.	Calculation setup

Here we define the output channels and the pressure correction method. The last thing to set is the Zero point correction. As was already mentioned, the charge sensors will drift over a longer time, so we need to calculate the absolute pressure. This can be done in three ways: with Thermodynamic zero, which looks at the compression and assumes that if we were to expand the volume to infinite, the absolute pressure should be zero.

Calculation setup

At this point we need to take a closer look at the p-v diagram. In a logarithmic p-v diagram the polytropic compression and expansion is a straight line and the steepness of that line is defined by the polytropic coefficient. We can easily see this during the measurement, as shown in the picture below. If the polytropic coefficient doesn't fit, the zero correction and heat release will be calculated in the wrong way.

To come back to the zero point correction, we need to define two angles in the compression part of the cycle where the compression fits with the polytropic exponent. If we define the points where the ignition already starts or where the intake valves are not closed, then we will have a wrong calculation.

Other options are to enter the known or measured pressure at a certain angle (usually at intake).

Now we are finished with setting up the system and we only need to choose which parameters we would like to see during measurement. The statistic values like maximum pressure and position of maximum pressure will always be calculated. We can choose the derivation of the pressure signal as an option and we get the value and the position of maximum pressure rise. The MEP option will calculate IMEPn (for the entire cycle), IMEPg (for the working part of the cycle) and PMEP (for the pumping part of the cycle). The MEP value is the mean effective pressure and is actually the area inside the p-v diagram (energy of the engine), normalized to the displaced volume of the cylinder.

Additionally we can get the total average values (for the entire run) for each cylinder by selecting Statistics, MEP, Derivation and Heat release. This will allow us to see averaged data in CA-scope, as well as adding the total statistical values.

4.	Thermodynamics setup

In this section we turn on the calculation of the heat release. Heat release is quite intensive mathematically, so we need to set it up with care. We set the Staring point and the Ending point of the calculation, because heat release only makes sense around the top dead center of the working part of the cycle. The result from this calculation is the start of combustion, end of combustion and the degrees at a certain percentage of the curve.

Thermodynamics setup

First let's take a look what heat release is. Heat release is indicated as the pressure rise above the polytropic compression and expansion part of the cycle. We have already shown a difference from the motored cycle to the working cycle in p-v diagram, but now let' repeat this in a CA graph. In the picture below the motored cycle is drawn below the working cycle and the area between those is the work produced by the engine. The part of the cycle where those two curves don't fit is the part where the heat is released from the fuel.

In this section we can also choose a temperature calculation. By selecting gas mass or defining intake pressure, intake temperature and volumetric efficiency, the temperature in the cylinder can be calculated.

5. Knock detection setup

Knock detection is the procedure that filters the data with a high and low pass filter and compares the expansion to the compression stroke. The result is the knocking factor, which shows the amount of knocking inside the engine.

Knock detection setup