With the strengthening of exhaust gas regulations such as EU Stage 5 and China’s 4th regulation, the engine such as external EGR and aftertreatment device has become complicated. In addition, Kubota’s industrial engines are used not only in agricultural and construction machines but also in various machines with different applications around the world, there are many variations of intake and exhaust systems, and the engines are diversified.
For an engine adopting an external EGR and a aftertreatment device, a hot wire type mass air flow rate sensor is widely adopted in an intake piping in order to control the EGR valve appropriately and the regeneration control of the DPF. However, it is known that the sensitivity of this sensing system varies depending on the shape of the intake piping. When the sensitivity varies, the engine is controlled based on the incorrect air mass flow rate, so that the exhaust performance may be deteriorated. It is confirmed that the variation of sensitivity does not exceed the limit value by measuring the sensitivity of this sensing system using the prototype and considering the variation of the assembling error and the like. If the sensitivity variation exceeds the limit value, we must change the shape and mounting position of the intake piping. Therefore development lead time will be prolonged in order to design intake piping, prototype, shipment from overseas and experiment again.
In this paper, we investigate the factors that vary the sensitivity of this sensing system by flow rig test of the intake piping and a CFD focusing on the convective heat transfer amount of the hot wire inside the sensor. It was confirmed that the influence by velocity and turbulent kinetic energy were large. CFD is carried out with various intake piping, limit value are set based on the characteristic of physical quantity with large sensitivity variation, and we developed method to predict the possibility that the sensitivity variation of the air mass flow rate sensor exceed the limit value before prototyping.