1.1.1. Gear-type test benches

Currently the industry uses for research in terms of gears and their applications refers to test benches type gears, since the results obtained in the simulations are more related to the behavior of these elements in their actual operation [53].

Generally the gears that are tested are standardized gears, although you can study different types, dimensions, with different surface finishes, materials, heat treatments, lubricants, load regime, the closest to the actual behavior they will be subjected to, to evaluate specifically certain desired service requirements.

There are two fundamental types of power loop in the gear type test banks: open power loop and closed power loop; the latter can be electrical or mechanical, as shown in Figure 2.3.  

                (a)                                                   (b)                                                       (c)

Figure 2.3. Gear test machine with power loop, a) Open, b) closed power loop, electric, c) closed power loop, mechanical [8, 53, 54]

The first case, (a), is very simple, consists of an actuator coupled to the gears under study and a brake, which simulates the load. The second case, (b), an electric motor is coupled to the input shaft of the gearbox, the output shaft of the box is connected to a generator that feeds back energy to the system. In the third case, (c), an electric motor is coupled to the input shaft of the gearbox and this to the output of another gearbox that have the same transmission ratio and both boxes are connected to each other by axes intermediate, forming a closed loop.

To apply the load connects a system that can be included in said loop, this mechanism can be a lever system with weight or hydraulic. This disposition is known as: circulating power, in English it is known as Four-Square.

Table 2.1. Advantages and disadvantages of the three types of test benches.

When designing a test bench, the above must be taken into account, since, depending on the needs or requirements of the client, the designer can use previous knowledge to improve or innovate the product.

1.1.1.1.   Four square system

The Four-Square machine configuration or circulating power described above, has been widely used in industry and in the field of tribological research because the results obtained in the simulations are quite like reality.

There are many patents over the years. All these machines, although of different configurations and design present the common characteristics of the system of circulating power that are: generator of torque, elements of union and transmitter of torque.

Torque generators are generally electric motors, but they can also be hydraulic systems. The elements of union are the systems that transmit the pair from one system to another and can be axes, couplings and so on. The torque transmitters are the gear boxes, both test and auxiliary, that is, they are the elements that transmit torque from one torsion chain to another. Now, there are several systems or mechanisms to apply the load to the gears tested.

Many of the designs employ a simple coaxial planetary reducer where torque is applied to the planetary system by an auxiliary auger transmission mechanism during the study, such as Lanahan, Klinger, Langenbeck and Basedow. This mechanism can be operated manually or controlled numerically by a motor. The test torque is calculated by the difference in the direction of rotation and the relationship between the sun and the internal gear. This system has the disadvantage that the applied torque cannot be controlled. It is necessary to use an additional system to vary the torque during its execution during testing without having to interrupt it [55-58].

In other systems the test torque is applied using two simple coaxial planetary gearboxes connected by satellites, where the input and output shafts rotate in the same direction and the gear teeth numbers are calculated so that the output shaft has an angular velocity deferent to that of the entrance. The problem with this machine is that you cannot determine the power losses of the planetary gearboxes.

Another design consisted in connecting another mechanism of two simple coaxial planetary gearboxes connected by satellites that could be identical to the previous one as compensation mechanism, to try to overcome the disadvantage of the machines previously described, but this system causes additional losses and was more difficult to obtain the power losses of a reducer [59].

Other designers such as Musser and Schröder replaced planetary gear trains with harmonic gearboxes, where it belongs to the group of waveform reducers. The diameter of the flexible sheet is slightly smaller than that of the circular flange since it has two teeth less in its outer circumference. The flexible sheet adopts the elliptical shape of the wave generator and its teeth conform to those of the circular flange through the major axis of the ellipse. As soon as the wave generator begins to rotate in a clockwise direction, the toothed area of ​​adjustment moves solidly to the major axis of the ellipse. When the wave generator rotates 180 degrees clockwise the Flexible Sheet returns with a less tooth position relative to the Circular Flange. Each complete 360 ​​degree turn of the Wave Generator the Flexible Sheet moves, counterclockwise, 2 teeth less in relation to the Circular Flange, thus obtaining the test torque as shown in Figure 2.4.

Figure 2.4. Detail of the interior of a harmonic reducer [60].

The main disadvantage of this mechanism is that the power losses of the auxiliary motor are unknown, and it is difficult to determine the efficiency of the gears under test. Another similar design by Brüggemann, replaced the harmonic reducer with a cycloidal reducer, thus increasing the torque values. But the radial component only transmits stress to the wheel, internal bolts and eccentric bearings. The cycloidal reducers that have the groove in waveform instead of a defined half-moon tend to handle higher pressure angles. In the case of grooves in wave form, a greater total force is needed to transmit the same tangential force that is finally useful in the reducer. This creates a greater radial force that unnecessarily fatigued the eccentric bearings causing a rapid wear on these elements, being a disadvantage to take into consideration [61-63].

Figure 2.5. Detail of the interior of a cycloidal reducer [64].

Harald and Yano designed machines that allow quick change of torque applied to the test gears by adding additional gears that generate the torque when they are pulled in the transverse direction, being a mechanically simpler system. And Bader built a machine based on the same principle, but the load is applied when one of the boxes of the power loop is rotated on an axis parallel to the transmission shaft [65-67].

Mihailidis presented a test bench in which the load is applied by a planetary gear train previously designed by Wolfrom. Where this mechanism, with the auxiliary motor included, rotates as a block, and this motor is only in operation when a load is being applied or it is wanted to be varied. This system can also use small motors since it has a high transmission ratio, which allows high test torques to be applied. One of the advantages that the designer achieves with his machine is that the efficiency in the gears can be obtained by knowing the torque applied by the main engine. And one of its disadvantages is that the load of essay cannot be changed quickly, which would be an inconvenience to consider at the time of doing studies where it requires this type of changes [53, 68, 69].

A machine that has been very successful in the world market was developed at the Technical University of Munich, Germany, FZG. The applications of this test bench are several, for example: to determine the capacity of load to grasping, the behavior of the coefficient of friction with respect to wear and the formation of micropitting and pitting. Also, to test various materials using the same type of lubricant, which offers results that depend only on the characteristics of the gear material. Lubricants are also tested and how they affect the micropitting and pitting formation on the flank of the gear tooth. The mechanism is relatively simple with a circulating power structure, in which there is a pair of fixed gears closing the cycle, and the gears under study, both pairs of gears with the same transmission ratio. One of the most trees has a device for measuring the torque. In the other tree, the test torque is applied by means of a lever and a counterweight. This principle has also been used in other machines for the testing of hypoid and helical gears of crossed axes. The device reaches a rotation speed of up to 2250 rpm, with a torque of 530 Nm and a contact pressure between the teeth of 2 GPa. In addition, it presents vibration sensors that allow the detection of severe damage during the test.

Figure 2.6. Gear Test Machine FZG [70].

The disadvantages of this machine are: the upper limit of contact stresses on the geared tooth is low, the load is static and is changed at relatively high time intervals [71-88].

Based on the same principle is the IAE straight tooth cylindrical gear testing machine. The torque applied to the teeth of the test gears is applied by a lever mechanism like that of the FZG machine. This test bench allows the evaluation of universal and hypoid transmissions, lubricants and their anti-ripper properties in the operation since it reaches speeds between 4000 and 6000 rpm, and the torque between 20 and 407 Nm. The disadvantages are like that of the FZG test bench [74, 88-91].

However, depending on the method with which the load torque is applied, the test machines for circulating power gears can be classified into mechanical or hydraulic systems [53].

Many of today's gear test bench designs apply the test load hydraulically, replacing mechanically applied load systems. Collins created one of the first machines applying the test load by means of a hydraulic piston coupled to one of the axes of the closed power loop, in which a hydraulic pressure is applied that generates an axial load on helical flutes that transmit the torque of testing. This system was designed to operate as a bidirectional hydraulic piston since the helical flute mechanism is located on both sides of the hydraulic piston, but an auxiliary system is needed to measure the torque that is applied since it becomes very difficult to control the load with hydraulic pressure. Hennings created a similar machine that by means of a hydraulic piston connected to an intoxicating system with helical gears in a drum generated the load of the test, maintaining the same disadvantages of the machine previously explained, the variable, friction, affects the study. However, Schneider based on the same principle designed a mechanism where friction does not affect the applied load, which consisted in connecting the cylinders directly from the shaft to the gear that provides the test torque [91-93].

On the other hand, Ryder, designed a compact equipment of a single box maintaining the system Four-Square, where the two pairs of gears have the same ratio of transmission, two pairs of gears: cylindrical of straight teeth and the other helical. Its system replaces helical fluted shafts with helical gears from test benches designed by Collins, Hennings and Schneider. The test load is applied by supplying a determined oil pressure to a helical gear and this by an axial movement to the other helical gear. By controlling the hydraulic pressure, it is possible to vary the load accurately during the test. The main disadvantage is that you cannot determine the efficiency of the gear pair under study and you cannot test gear trains. The testing capabilities of the machine are very good as it has a high rotation speed of up to 10,000 rpm and a range of torque applied between 0 to 270 Nm; which makes the Ryder machine has had good acceptance in the market and the research of lubricants and gears tests especially [53, 74, 88, 94, 95].

Shipley on the other hand creates a mechanism for the application of the test pair consisting of a drum and inside a rotor equipped with radial fins, where between the walls of the drum and the fins pressure chambers are formed, thus generating the torque of trial, by means of oil pressure in the chambers. This was one of the first test benches to use pressure chambers for the application of the load and has as advantages that it minimizes friction and the torque is controlled by conveniently varying the oil pressure in the chambers. In addition, it has test torques of up to 8000 Nm. On the other hand, Kugler created a more compact variant of this system; inside one of the fixed gears of the Four-Square system designed a hydraulic rotary cylinder where the pressure chambers are located [96, 97].

The NASA test bench developed by this same company in the United States of America. Its principle of operation is like the previous described. It can reach a speed of 10,000 rpm, since it has an engine connected to a belt drive. In addition, it presents a powerful hydraulic system that can reach a maximum pressure of 690x104 N / m² and obtain contact voltages close to the two GPa on the tooth surface of the gear under test. The hydraulic system transmits the test load on one of the standard gears, and these are connected to the gears under study. In the machine, four tests are performed for each pair of gears since the gears are tested with an axial displacement with which the desired contact voltages are reached, with a lower torque [54, 98].

Figure 2.7. NASA Gear Test Machine [54]

Benches of tests also used to evaluate bending fatigue in the gears are the machines of circulating power, which have a configuration mechanically like the machine of Ryder gear tests. The tests are designed for a certain number of cycles at speeds below 1000 rpm, with specific loads for the type of study that you want to perform, which requires controlling all possible variables that may affect or falsify the results, that is to say that product to external or internal factors, occur during the test the appearance of another type of failure such as wear, contact fatigue, seizure or other, which were not foreseen in the previous design of the trial. The main disadvantage of this type of mechanism is precisely the appearance of another failure before the number of cycles required for the running test is met. Although if the programmed testing is completed without being affected by another factor, the results obtained are accurate [99, 100].

The systems of circulating power or Four-Square are a mechanical configuration, used basically in the fatigue test of components [101].

Another interesting test bench is the push-button type, it is a universal electrohydraulic machine, controlled by servomotors. This machine is designed for cylindrical gears with straight teeth, in which several teeth are removed so that the tool that applies the test load has better access to them. The test load in these equipment is between 45 and 90 kN and is applied through a rod that comes into contact on the flank of a tooth and the reaction is supported through another rod that is in the same direction, but in the opposite direction in contact with another tooth of the gear on the opposite flank since the gear is placed in a device rigidly supported on an axis, so that a tooth can be subjected to the study as shown in figure 2.8.

Figure 2.8. Push-button gear test bench [102].

The teeth of the gear are tested independently, in which the test load is applied on a single point on the tooth. This apparatus is generally used to test the gearing to fatigue failure by bending at a relatively low economic cost. The advantages of this equipment are that you can measure the bending stress at the root of the tooth while the test is running, the machine automatically stops when the tooth fails, and there are no variables such as tree wear, bearings, deviations in the line of passage of the gears, the gap between teeth, the errors in the construction of the profile of the tooth, among other variables that tend to distort the study that is carried out. The main disadvantages are only trials to cylindrical gears of straight teeth, the contact is a single point of the tooth by which few types of faults are tested that affect the gears and are not recreated in the test the actual operating conditions of the gear [99, 102-104].

Other designs such as that of Guille Abreu [8] or Pedro Nel Martínez [54] use designs similar to those previously explained, adding methodological and / or economic concepts, continuously improving the equipment with new components, materials and innovations.

Figure 2.9. Gear test bench proposed by Martínez [54].

The figure shows a simple, economical and functional design. The more complex a system, the more chances of failure.