Down-sizing is very popular in engine design as it reduces fuel consumption while keeping the same power output as the bigger engine by using a turbocharger. A downsized engine uses a smaller cylinder and a higher boosting pressure to yield a higher cylinder pressure over a smaller displacement volume and more work per cycle by the cylinder volume. The benefit of down-sizing in improving fuel economy however comes with the price of higher knock tendency at higher loads. A retarded spark timing can be utilized to reduce the knock tendency, but this will cause a deterioration of the thermodynamic efficiency and a higher turbine inlet temperature. A further fuel enrichment to reduce the turbine inlet temperature will lessen the fuel economy gain.
Knock is described as uncontrolled auto-ignition within the unburnt gas ahead of the flame, where fuel air mixtures are compressed due to flame propagation and chemical reaction occurs uncontrollably. The auto-ignition leads to very high local pressure with very high frequency which damages the engine wall. In this work, down-sizing from a 1.6 L 4 cylinders engine (Ford Zetec 1.6 Sigma) to a 1 L 3 cylinders turbocharged engine (Ford Fox) shows higher knock tendency apparent in the 1 L engine at full load, where the knock and over-fuelling problems are more common. This is demonstrated by comparing the two engines at the same load and speed points, same AFR (Air-Fuel Ratio) and SA (Spark Advance), with the exception that a turbocharger is used to produce a higher boosting pressure for the smaller engine to reach the required load points. There are various methods for suppressing the knock, such as GDI, cooled EGR or the Atkinson cycle, all attempts to reduce the combustion temperature to mitigate the knock.
In this blog post, to reduce the knocking tendency a cooled Exhaust Gas Recirculation (EGR) is used for the 1 L engine. By keeping sparking timing and AFR the same as the non-EGR 1 L engine and benchmarking the results at the same BMEP point, it shows that the 1 L engine with cooled EGR allows to slow down the combustion rate, both Mass Fraction at 50 percent of burn (MFB50) and 10 to 90 percent of Mass Fraction Burn (MFB 10 to 90). The slower combustion rate leads to reduced knocking tendency which allows spark timing to be advanced. The advanced spark timing increases the thermodynamic efficiency and fuel economy gain. A higher boosting pressure is required to cover the load penalty induced by the EGR and that reduces the pumping loss. The earlier combustion due to spark advance also lowers the turbine inlet temperature, which is a substitute for fuel enrichment at high speed and high load points, where exhaust temperature reaches more than 1000 degree Celsius. It is shown an average of 3.86% BSFC gain can be obtained at 3500 rpm at full load across different AFRs using EGR up to 13%. An average of 35-degree Celsius reduction of turbine inlet temperature can be gained. There is a nearly 8% BSFC gain at 1500 rpm and down to 0.57% BSFC gain at 5500 rpm at full load, because pumping losses are higher with increasing engine speed.
Acronyms
AFR: Air fuel ratio
ATDC: After TDC
BMEP: Brake mean effective pressure
BSFC: Brake specific fuel consumption
EGR: Exhaust gas recirculation
GDI: Gasoline direct injection
KLSA: Knock limited spark advance
MFB50: Mass fraction burned at 50%
MFB10to90: Mass fraction burned between 10 to 90%
PMEP: Pumping mean effective pressure
SA: Spark advanceVVT: Variable valve timing
Section 1: knock associated with downsizing a 1.6 L engine to a 1 L engine
The following shows an example where a 1.6 litre four-cylinder naturally aspirated engine is downsized to a 1 litre three-cylinder turbocharged engine. The two engines are run to produce the same amount of brake power at different engine speed where boosting pressure and Knock Limited Spark Advance (KLSA) are adjusted.
1. 1.6 L naturally aspirated engine
| Model | 4 cylinders naturally aspirated |
| Displacement | 1596 |
| Stroke/Bore | 81.4mm/79mm |
| Connecting rod length | 137mm |
| Compression ratio | 11:1 |
Table 1: 1.6 L engine data (Ford Zetec 1.6 Sigma)
The engine is running at full load, spark advance and air fuel ratio are adjusted to reach the knock limit.
The knock intensity factor is classified as follows
The speed and load points are shown in the following table
| Engine speed (rpm) | MFB50 (degree C) | BMEP(Bar) | BSFC (g/kWh) | AFR | SA(ATDC) | Knock intensity factor | Brake torque (Nm) | Brake power (kw) |
| 1500 | 25 | 11.8 | 235.6 | 14.7, | 12 | 0.5 | 149.6 | 23.5 |
| 2500 | 19.67 | 12.6 | 270 | 11.9, | 6 | 0.495 | 160 | 41.9 |
| 3500 | 17.67 | 12.68 | 295.9 | 10.7, | 3 | 0.5 | 161 | 59 |
| 4500 | 15.65 | 12.6 | 295.1 | 10.6, | 0.3 | 0.51 | 160.7 | 75.5 |
| 5500 | 12.75 | 12.51 | 303.8 | 10.2, | -3.4 | 0.5 | 158.8 | 91.54 |
Table 2: simulation results of 1.6 L engine
The plots of the variables in Table 2 are shown in the following figures. Figure 1 below shows due to the knock limitation the MFB50 must be retarded at lower engine speeds.
In addition, the AFR, Figure 2, must be particularly enriched at high engine speeds mainly to maintain the TIT so that the turbine is not damaged due to exceedingly high TIT.
Figure 3 and 4 show BMEP and BSFC for the knock onset. The highest BMEP occurs at around 2500 RPM. BSFC increases as engine speed increases.
Figures 5 and 6 show brake torque, brake power at the knock onset. Figures 7 and 8 show spark advance after TDC and the knock intensity factor. The knock intensity factor classifies knock intensity into four levels, namely no knock, trace knock, medium knock and strong knock [1]. For knock intensity factor larger than 0.5, trace knock occurs. It can be seen from the figure 8 knock intensity factor is maintained in a proximity to 0.5.
2. 1 L three cylinders turbocharged engine (Ford Fox)
The engine’s geometrical data is shown in table 3
| Model | 3 cylinders turbocharged |
| Displacement | 999 |
| Stroke/Bore | 82mm/71.9mm |
| Connecting rod length | 137mm |
| Compression ratio | 10:1 |
Table 3: 1 L engine data, Ford Fox
Comparing 1 L engine to 1.6 L engine, 1 L engine has
- less cylinder number
- longer stroke
- smaller bore
- lower compression ratio
Below, a P-V diagram of the two engines at the same power output, speed point, SA, and AFR reveals downsized engine produces higher IMEP.
The boosting pressure is increased by the turbocharger at full load while keeping AFR and SA unchanged as in table 2, to keep the brake torque and power at the same level as the 1.6 L engine in table 2, see figure below.
The testing data is shown in table 4.
| Engine speed (rpm) | MFB50 (degree C) | BMEP(Bar) | BSFC (g/kWh) | AFR | SA(ATDC) | Knock limit factor | Brake torque (Nm) | Brake power (kw) |
| 1500 | 23.9 | 18.6 | 171.2 | 14.7, | 12 | 0.73 | 148.7 | 23.6 |
| 2500 | 18.4 | 20.2 | 200 | 11.9, | 6 | 1 | 162.5 | 42.3 |
| 3500 | 16.2 | 20.6 | 221 | 10.7, | 3 | 0.6 | 162.5 | 59.3 |
| 4500 | 14.17 | 20.5 | 221 | 10.6, | 0.3 | 0.65 | 163 | 77 |
| 5500 | 11.4 | 19.6 | 229.5 | 10.2, | -3.4 | 0.77 | 158 | 91 |
Table 4: simulation results of 1 L engine
The plots for the variables shown in Table 4 are given below.
The BSFC improves by around 25% compared to the 1.6 L engine across all speeds, Figure 17, however this improvement comes with higher knock tendency, see figure 18. The highest knock tendency increment occurs at 2500 rpm where maximum brake torque is.
Section 2: Cooled EGR in mitigating the knock and BSFC improvement
Cooled EGR allows to reduce the knock intensity. The recirculated gas temperature is around 433K (160 degrees C). An experiment is performed to keep all the control variables the same as to the 1 L non-EGR engine in Table 4 except enabling EGR to be 10% this time and increasing the boost pressure (from 1.2 bar to 1.5 bar) to yield the same BMEP.
| Engine speed (rpm) | MFB50 (degree C) | BMEP(Bar) | BSFC (g/kWh) | AFR | SA(ATDC) | Knock limit factor | Brake torque (Nm) | Brake power (kw) |
| 3500 | 16.9 | 20.8 | 232 | 10.7 | 3 | 0.22 | 165 | 60.6 |
Table 5: simulation results of 1 L engine with EGR at 3500 RPM
Comparing Table 4 and Table 5, it is seen that with EGR, MFB50 occurs later than in the non-EGR case (16.9 CA compared to 16.2 CA) and as a result the knock limit factor is decreased from 0.6 to 0.22. However, the longer burning duration causes higher exhaust gas temperatures. Experiments are performed for different AFRs and EGR rates at 3500 rpm to examine the EGR effect on the BSFC and other variables at the same BMEP point. For each EGR rate and AFR, Knock Limited Spark Advance (KLSA) and boosting pressure are adjusted until the knock tendency factor reaches its limit which is 0.5.
An EGR increase always involves a better fuel consumption due to increased boosting pressure and more importantly the more advanced spark timing. BSFC mainly reduces as the A/F ratio increases which is trivial to see Figure 20. The latter, however, affects the combustion phasing, Figure 19, since less over-fuelling causes higher combustion temperature which requires retarded spark timing to reduce knock tendency, Figure 21. The regarded spark timing due to less over-fuelling features longer combustion duration MFB10to90, Figure 22. The longer combustion duration due to less over-fuelling determines higher exhaust temperature, Figure 23.
Figure 21 shows Spark timing advance with increased EGR rate allowing increased combustion thermal efficiency. Figure 22 highlights further effects caused by EGR on combustion duration. As the EGR rate increases, combustion duration becomes longer despite more advanced spark timing. Less over-fuelling also determines longer combustion duration. In Figure 23, turbine inlet temperature decreases as EGR rate increases and the spark angle is further advanced, Figure 21. Turbine inlet temperature increases due to less over-fuelling. Figure 24 underlines that an increase in boost pressure is required to recover the load penalties induced by the EGR. Figure 25 shows PMEP decreases for higher EGR rate and higher boosting pressure.
It has been shown that downsizing improves BSFC but can also induce knock. Cooled EGR is shown to be effective to reduce the combustion temperaturs and mitigate the knock. In this example, there is 3.7% of BSFC improvement achieved using EGR at full load.
Reference
[1]: X. Han, Knock modelling: analysis of its sensitivity to spark advance, https://www.claytex.com/tech-blog/knock-modelling-analysis-of-its-sensitivity-to-spark-advance/, 2018
Written by: Xiaoran Han – Project Engineer
Please get in touch if you have any questions or have got a topic in mind that you would like us to write about. You can submit your questions / topics via: Tech Blog Questions / Topic Suggestion.