【正文】 esults obtained from vehicle tests are shown in Figure 5. Reported are the average steeringwheel torque Ts and the average corrective opposite steering input required to maintain a straight course during acceleration on a splitμ track with an open and a viscous differential. The differences between the values with the open differential and those with the viscous coupling are relatively large in parison to each other. However, they are small in absolute terms. Subjectively, the steering influence is nearly unnoticeable. The torque steer is also influenced by several kinematic parameters which will be explained in the next section of this paper. 4 FACTORS AFFECTING STEERING TORQUE As shown in Figure 6 the tractive forces lead to an increase in the toein response per wheel. For differing tractive forces, Which appear when accelerating on splitμ with limitedslip differentials, the toein response changes per wheel are also different. Unfortunately, this effect leads to an undesirable turnin response to the lowμ side, . the same yaw direction as caused by the difference in the tractive forces. Reduced toein elasticity is thus an essential requirement for the successful frontaxle application of a viscous limitedslip differential as well as any other type of limitedslip differential. Generally the following equations apply to the driving forces on a wheel ?VT FF ? With ?TF Tractive Force ?VF Vertical Wheel Load ?? Utilized Adhesion Coefficient These driving forces result in steering torque at each wheel via the wheel disturbance level arm “e” and a steering torque difference between the wheels given by the equation: △ eT = ? ?loHhiH FFe ?? ??? ?c o s Where △ ?eT Steering Torque Difference e=Wheel Disturbance Level Arm ?? King Pin Angle hi=highμ side subscript lo=lowμ side subscript In the case of frontwheel drive vehicles with open differentials, △ Ts is almost unnoticeable, since the torque bias ( loHhiT FF ?? / ) is no more than . For applications with limitedslip differentials, however, the influence is significant. Thus the wheel disturbance lever arm e should be as small as possible. Differing wheel loads also lead to an increase in △Te so the difference should also be as small as possible. When torque is transmitted by an articulated CVJoint, on the drive side (subscript 1) and the driven side (subscript 2),differing secondary moments are produced that must have a reaction in a vertical plane relative to the plane of articulation. The magnitude and direction of the secondary moments (M) are calculated as follows (see Figure 8): Drive side M1 = vv TT ?? ? tan/)2/ta n (2 ?? Driven side M2 = vv TT ?? ? tan/)2/ta n (2 ?? With T2 = dynT rF? ?T = ? ?systemJoTf in t,2 ? Where v? ?? Vertical Articulation Angle ? ?? Resulting Articulation Angle dynr ?? Dynamic Wheel Radius ?T ?? Average Torque Loss 3 The ponent ?cos2?M acts around the kingpin axis (see figure 7) as a steering torque per wheel and as a steering torque difference between the wheels as follows: ])t a n/2/t a n()s i n/2/t a n[(c o s 22 liwhiw TTTTT ?? ?????? ??????? ????? where ????T Steering Torque Difference W?? Wheel side subscript It is therefore apparent that not only differing driving torque but also differing articulations caused by various driveshaft lengths are also a factor. Referring to the momentpolygon in Figure 7, the rotational direction of M2 or ?T respectively change, depending on the position of the wheelcenter to the gearbox output. For the normal position of the halfshaft shown in Figure 7(wheelcenter below the gearbox output joint) the secondary moments work in the same rotational direction as the driving forces. For a modified suspension layout (wheelcenter above gearbox output joint, . v? negative) the secondary moments counteract the moments caused by the driving forces. Thus for good patibility of the front axle with a limitedslip differential, the design requires: 1) vertical bending angles which are centered around 0?v? or negative ( 0?v? ) with same values of v? on both left and right sides。 (figure 19,middle). Since most vehicle and ABS manufacturers consider 90176。在歐洲和日本前輪驅(qū)動轎車產(chǎn)量的施用已經(jīng)證明黏性連接器不僅對于光滑路面的汽車牽引，而且在正常行駛條件下對于操縱性和穩(wěn)定性都有所改善。 轉(zhuǎn)彎試驗展現(xiàn)出黏性連接器在前輪驅(qū)動的汽車上獨立轉(zhuǎn)彎時的影響。 黏性連接器是根據(jù)液體摩擦的原理和依靠速度差來運轉(zhuǎn)的。外面的方式如圖 2所示。其次，差速器架和轉(zhuǎn)送軸套只需要很小的修改。 這種特殊的設(shè)計也為有實際意義的重量和黏性單元費用的降低給出了很好的可能性。 3 牽引力的影響 作為一個扭轉(zhuǎn)力平衡裝置，一個開的差速器提供相等的力到兩個驅(qū)動輪上。 例如，當(dāng)一個車輪傳遞的最大扭轉(zhuǎn)力超出表面滑動系數(shù) ? 允許值或者以一個高的側(cè)面加速度轉(zhuǎn)彎時，兩個車輪的速度是不同的 .在黏性連接器中產(chǎn)生的自鎖扭轉(zhuǎn)力抵抗速度差的增加并且傳遞合適的扭轉(zhuǎn)力到車輪上它具有更好的牽引力潛能。然而，在絕對條件下它們是小的。因為帶有限制滑動差速器的車輪在滑動系數(shù) ? 的路面上加速時會出現(xiàn)不同的牽引力，所以從頭到尾反應(yīng)每個車輪的變化也是不同的。 c o s ( )ioe H h H lT e F F? ??? ? ? ? ? 這里 eT? — 扭轉(zhuǎn)力矩差值 e— 車輪干擾常數(shù) ? — 主銷傾角 ih — 高滑動系數(shù)一側(cè)下標(biāo) ol — 低滑動系數(shù)一側(cè)下標(biāo) 9 在帶有開式差速器前輪驅(qū)動汽車的情況下， ST? 是很不明顯的，因為扭轉(zhuǎn)力基數(shù) ( / )H hi H loFF??是不大于 。 當(dāng)扭轉(zhuǎn)力通過鉸接“ CV連接”傳遞時，在主動一側(cè)（下標(biāo) 1）和從動一側(cè)（下標(biāo) 2），必須反應(yīng)垂直平面相對于連接平面的不同的第二個力矩產(chǎn)生了。由于改進的懸掛裝置設(shè)計（車輪中心高于變速箱輸出點，也就是說， v? 為負值）第二個力矩抵消了由驅(qū)動力引起的力矩。然而，事實上（有扭轉(zhuǎn)力損失），第二個力矩出現(xiàn)不同： 21DW D WM M M? ? ? 22DWT T T??? 第二個力矩不同點是： 22( ) ta n / 2 / sin ta n / 2 / ta nD W W W V D W v w W v wM T T T T T? ? ?? ? ? ?? ? ? ? ? 為了簡化 應(yīng)用給出 fT? VD VW V? ? ???和 DT T w T? ???? ( ta n / 2 1 / sin 1 / ta n )D W v v vMT ? ? ? ?? ? ? ? DWM? 需要在兩個連接處都有抵抗反應(yīng)的力這里 /DW DWF M L?? 。 如圖表 10：前輪驅(qū)動力的汽車穩(wěn)定狀態(tài)下轉(zhuǎn)向時的牽引力。 如圖表 11：安裝有開式差速器的汽車餓安裝有黏性連接器的汽車在穩(wěn)定狀態(tài)下轉(zhuǎn)彎時的對比 如圖表 10所示在轉(zhuǎn)彎時不對稱的牽引力干擾也會改進汽車的直線行駛。 安裝有 開式差速器的高動力前輪驅(qū)動汽車當(dāng)以低檔加速離開緊急轉(zhuǎn)角時通常旋轉(zhuǎn)它們的內(nèi)側(cè)車輪。 圖表 13 和 14 顯示了裝有開式差速器和裝有黏性限制滑動差速器在穩(wěn)定狀態(tài)下轉(zhuǎn)彎過程中加速試驗的結(jié)果。前輪傳遞側(cè)偏力潛能降低的原理是由于重心移到后軸車輪并且在驅(qū)動輪上增加了縱向力。 安裝有限制滑動前差速器的汽車在轉(zhuǎn)彎過程中加速時具有一個更穩(wěn)定的最初反應(yīng)比裝有開式差速器的汽車，降低它的操縱狀態(tài)。在非常低的摩擦力表面，例如 雪或者冰，當(dāng)裝有限制滑動差速器的汽車在曲線路面上加速時更強的操縱性被期望因為通過黏性連接器連接的驅(qū)動輪更容易旋轉(zhuǎn)（動力轉(zhuǎn)向裝置