THE DYNAMIC BEHAVIOUR OF AUTOMATIC TOOL FEEDING SYSTEMS FOR DIFFERENT TRANSITORY REGIMES

The total machining time for a part produced on machining centers is influenced, besides the actual machining time, by the auxiliary times which include the automatic changing of the tools. In order to increase the productivity of machining centers, the research presented in this paper has focused on enhancing the movement speed of an automatic tool feeding system. Increasing the movement speed of the mechanism`s components has led to the appearance of unwanted side effects, such as shocks and vibrations. In order to minimize these effects, we have proposed and tested a new solution to adjust the braking regime.


INTRODUCTION
The continuous trend in the manufacturing industry using machining centers is to increase the productivity, leading the researches towards the development of new methods and solutions in order to minimize the nonproductive auxiliary times [1,2].The researches aimed at finding new solutions for reconfiguring tool changing systems in order to reduce the number of movements needed to perform the tool change [3].A method of structural synthesis of tool changing systems based on three graphs is able to minimize the number of movements needed to change the tool from the spindle to the tool magazine of an existing machining center [4].In [5] Turkay developed a new method based on the genetic algorithms, for the redistribution of the entering order of the tools in the machining process, in order to minimize the non-productive machining time.Ecker and Gupta [6] presented a new method of minimizing the auxiliary times by using the balanced graphs.In this paper, using the proposed and researched solution of adjusting the deceleration path, the movement speed of the mobile elements was highly increased by improving the dynamic behavior at the end of the stroke.

Method description
Because the tool feeding systems which equip machining centers can perform translational and rotational movements in order to exchange the tool in the machining process [7,8], the main objective followed in this paper is to increase their functional performances.One of the components which can be adapted to increase the functional performances of these tool feeding systems is to reduce the times consumed with the tool exchange, which imposes a speed increase for each linear and rotational displacement sequence of the feeding system [9].
In this paper we considered that the tool feeding system is hydraulically driven through the use of hydraulic cylinders; we also considered that the deceleration of the hydraulic cylinder is assured through its construction.The original braking device of the cylinder is replaced with the proposed solution which consists in a tracking valve, this replacement having the role to improve the dynamic behavior at the end of the hydraulic cylinder`s piston stroke.
The researches followed the speed variation in time of the moving element regarding the dynamic parameters of the component elements: vibration, load, pressure, etc.Because the tool feeding systems perform movements during and outside the machining process the level of vibration range was recorded in the system`s bed and needed to be held between 0.002 ÷ 0.004 m/s according to IRD 10816 Vibration Severity Standard [10].

Experimental set-up and data acquisition tools
The experimental set-up was built to analyze the possibilities of time reduction of the mobile component linear displacement performed by a hydraulic drive system.In Figure 1 the experimental set-up is presented, the proposed solution, namely the tracking valve 3, is mounted in the hydraulic circuit on the exhaust pipe of the linear hydraulic cylinder 1 (at the piston`s withdrawal).The hydraulic oil which flows through the exhaust pipe of the cylinder is routed to the tracking valve through pipe 2 and then to the tank through pipe 6.The construction of the tracking valve is normally open, the transitory regime being assured when the cam 5, mounted on the mobile element 4, steps on the valve`s rod; thus the exhaust circuit is gradually closed, obtaining the deceleration transitory regime.In order to obtain different deceleration paths of the transitory regime, and also different deceleration times, besides varying the input flow, the cam 5 position can be modified through rotating it at different angles (other than the original angle).
In the case of the classic hydraulic cylinder deceleration (Figure 2) the transitory regime is made by making use of an hydraulic resistance 1 and the directional valve 2 which are mounted in the cylinder`s construction.At the displacement of the piston, the oil exhaust orifice from the cylinder is obstructed by the piston`s 4 extensions 3 which are diametrical opposite.Thus the hydraulic oil will be routed through the resistance 1 which can be adjusted.
For acquiring the measured data, three different measurement tools were used.The way the three data acquisition tools are mounted on the experimental set-up and also their components are presented in Figure 3.We used a laser interferometer, a pressure transducer and a vibration analyzer, all synchronized to acquire the data simultaneously.

RESULTS AND DISCUSSION
The experimental tests when the transitory deceleration regime was assured through the built-in braking device of the cylinder aimed to obtain a minimum total time to cover the whole displacement range by the mobile element, taking into account the range of the vibration speed which occurs on the experimental set-up bed.In order to obtain minimum times the input flow was varied with the following values: 26, 30, and 34 L/min; the load on the mobile element was held constant at 12 kgf.In Table 1 we present the values obtained in the case of braking through the built-in system of the cylinder.In Figure 4 we present by overlapping the three speed variations against time which correspond to the input flows.We can notice that all three diagrams have a quasi-similar characteristic, but with important implications on increasing the speed of the mobile element.In Figure 5 were overlapped the speed variation of vibration range that occur in the experimental set-up bed which correspond to the displacement speed obtained for the mobile element.
Following this sets of tests one can emphasize that the time of the transitory regime at deceleration (T dec ) and acceleration (T acc ) increased once the input flow is increased, because of the inertia forces which occur at start and braking.
The time of the constant speed regime (T frp ) decreases with the increase of the input flow; the total time, given by summing the times corresponding to each transitory regime, decreases once the input flow is increased.

Fig. 4 .
Fig. 4. Speed variation against time for linear displacement at variable input flow and constant flow of 12kgf.

Fig. 5 .
Fig. 5. Speed variation of vibration range against time that occurs in the experimental set-up`s bed.

Table 1 .
Experimental test values at constant load of 12 kgf.